You are on page 1of 195

Childrens functioning

following parental cancer

Annemieke Visser

Ik was heel erg gesroken toen ik

hoorde dat mijn mama kanker hat.

Ik kan niet gelooven dat een mens het ik kan iedereen wel immekaar slaan. Ik heb een klub en als iemand van de

zomaar kan krijgen. Ik baal er zo van,

klub lagt dan ik nooit meelagen. Want mijn mama heeft kanker en dat voelt heel hart aan. Ik hou van je mama. Egt waar mama.
(zoon, 8 jaar)

Er komen veel vremde mensen, dat vind ik nooit zo leuk. Wand dan word er veel over ziek zijn gepraat. Dan ga ik gewoon naar mijn ponys en dan praat ik tegen ze. Ik kan er niet goed tegen en dan word ik snel boos!
(dochter, 8 jaar)

Ik ben er geestelijk sterker van geworden, ik weet met welke mensen ik goed kan praten en ik heb ontdekt dat ik van alle dagen moet genieten.
(dochter 17 jaar)

Lieve mama gaa niet dood, ik wet dat je heel veel pijn hept. Lieve heer wilt u mijn mama en de rest weer beter makem
(zoon, 9 jaar)

Ik had moeite om mijn concentratie lang vast te houden, maar school/hobbys heb ik toch altijd volgehouden, ook als afleiding. Echt lol hebben gaf me vaak een schuldgevoel.
(dochter 19 jaar)

Omdat ik van alle karweitjes en de geestelijke druk heel moe werd, deed ik ook veel minder leuke dingen zoals sporten, uitgaan, winkelen.
(dochter 17 jaar)

Ik dacht er steeds aan, dus op school kon ik niet zo goed werken.


(dochter 11 jaar)

Mijn ouders willen praten over de gevolgen van de ziekte, maar daar heb ik niet echt behoefte aan.
(dochter, 19 jaar)

Ik moet me wat stiller houden, ik moet niet ergens over doorzeuren.


(dochter, 14 jaar)

Teveel informatie is lang niet altijd goed. Wat ik wou weten wist ik.
(dochter,17 jaar)


(zoon, 8 jaar) (zoon, 9 jaar)

Toen ik hoorden dat ze kangker hat ik snags gehuld om mama. Ik kan er wel weer om huilen. Ik hoop dat ze niet dood gaat Ik hou van mijn moeder. Ik vint mijn moeder ook zo knap dat ze dit allemaal aan kan ... En nu worden de kuren veel zwader en als ze dood gaat kan ik me zelf ook wel dood maken.

Maar ik en mama hebben geluk lieveheersbeesjes. Geluk lieveheers-beesje zijn lieve beesjes die je zorgen opnemen. Zeg maar ik wil dit kwijt... doe je dat in 2 lieveheersbeesjes
(dochter, 9 jaar)

Ik hoob dat papa ooit beter word Ik ben niet bang voor de dood want We heben het er over gehat.
(zoon, 9 jaar)

Lieve mama Ik ben heel erg verdrietig geweest, en bang dat je dood zou gaan, maar ik ben blij dat het nu goed gaat. Ik vind het moeilijk om er over te praten.

Mama was toen geoperert aan haar boresten die waren weg omdat die ziek waren mama heeft mij verteld dat het kanker heete en mama hat 2 zakken aan haar bed hangen en ze deden BLUP BLUP mama kwam zaterdag thuis met twee tassen aan haar zeiden en dat heet een dreen.
.....

De vlinder zat op een bloempje. Dat was een zonnebloempje. De vlinder zat vol kleurepragt en vlarderde naar een ander bloempje.
(dochter, 9 jaar)

Maar nu denk ik soms nog wat er gebeurt is, maar meestal niet meer. Ik brobeer er zo weinig mogelijk aan te denken.
(8-jarig meisje)

Elke ervaring is anders. Ik zou wel willen praten,

maar voor mijn verwerking praat ik liever met mensen die niet steeds dat had mijn moeder ook zeggen.
(dochter, 19 jaar)

Childrens functioning
following parental cancer

Annemieke Visser

Visser, Annemieke Childrens functioning following parental cancer Thesis University of Groningen, the Netherlands With ref. With summary in Dutch. ISBN 9077113509

Financial support for the study in this thesis was obtained from the Dutch Cancer Society (grant number 2000-2333)

The printing of this thesis was supported by: Dutch Cancer Society University of Groningen Northern Centre for Healthcare Research Groningen Stichting Werkgroep Interne Oncologie Comprehensive Cancer Centre North Netherlands Amgen B.V. Breda

Cover design and lay out: Eelkje Eppenga Printed by: Gildeprint Copyright A. Visser, 2007 All right reserved

RIJKSUNIVERSITEIT GRONINGEN

Childrens functioning following parental cancer


Proefschrift
ter verkrijging van het doctoraat in de aan de Rijksuniversiteit Groningen Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op woensdag 4 april 2007 om 16.15 uur door Annemieke Visser te Leeuwarden op gezag van de Medische Wetenschappen

geboren op 21 oktober 1974

Promotores

Prof. Dr. W.T.A. van der Graaf Dr. J.E.H.M. Hoekstra-Weebers Prof. Dr. J.W. Groothoff

Prof. Dr. H.J. Hoekstra

Copromotor Beoordelingscommissie

Prof. Dr. J.L.N. Roodenburg Prof. Dr. P.H.B. Willemse

Paranimfen
Gea Huizinga Andre Visser

Voor mijn ouders

10

Contents
1 2
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 The impact of parental cancer on children and the family: a review of the literature Cancer Treatment Reviews 2004; 30: 683-694 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Emotional and behavioural functioning of children of a parent diagnosed with cancer: a cross informant perspective Psycho-Oncology 2005; 14: 746-758 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Parental cancer: characteristics of parents as predictors for child functioning Cancer, 2006; 106: 1178-1187 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Temperament as a predictor of internalizing and externalizing problems in adolescent children of parents diagnosed with cancer Supportive Care in Cancer, 2006 August, 30 (Epub ahead of print) . . . . . . . . . . . . . . . . . . . 105 Emotional and behavioural problems in children of parents recently diagnosed with cancer: a longitudinal study Acta Oncologica, 2007; 46: 67-76 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 Samenvatting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 Dankwoord . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Articles related to the project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 Northern Centre for Healthcare Research and previous dissertations . . . . . . . . . . 191

4 5

7 8 9

11

12

General introduction

13

Introduction

Various studies have shown that stressful events in a family can cause functioning disorders in children.1 The consequences of divorce for children have been given

much attention2,3 as have parents with complaints of depression.4 Increasingly, there

is a focus on the consequences for a child of having a parent with a physical illness. Lately there have been publications about the consequences for children of a par-

bowel disease9 and human immunodeficiency virus.10,11 The publications show that

ent with multiple sclerosis,5,6 cerebrovascular accident,7 epilepsy,8 inflammatory

these children run a higher risk of developing emotional and behavioural problems than children of healthy parents. The problem risk, however, depends on various illness-related, individual and family variables.12-15

Chronic diseases can vary in the effects they have on children by variation in onset (fatal/shortened life expectancy/no long-term consequences) and individual restric-

(acute/gradual), course (progressive, constant or episodic/recurrent), outcome tions.15 Irrespective of the type of illness individual characteristics of parent and child (age/gender) and family characteristics (parenting/family relationships) might correlate with childrens functioning.12,13

So far no research has been performed in the Netherlands into the functioning of children of a parent diagnosed with cancer. This thesis focuses on the consequences of the diagnosis cancer and its treatment for the functioning of children. In this of cancer for the parent will be discussed. Using the system theory as model, the possible influence of the event on the patient and also on other family members, will be described. Finally an outline of this thesis is presented.

chapter the medical and psychosocial consequences of the diagnosis and treatment

14

Medical aspects of cancer


More than 72,000 people were diagnosed with cancer in the Netherlands in 2003.16 cent of the patients in this age group is female. Breast cancer has the highest preva-

About 9000 of them have children living at home.17 Approximately sixty-five perlence in women between 30 and 60 years of age. In men in the age range 30 to 44

years testicular cancer has the highest prevalence and between 45 and 60 years of gery, chemotherapy, radiotherapy, hormonal therapy, immunotherapy or a combination of these treatments. The treatment chosen depends on the type of cancer, the phase of the illness and the general condition of the patient. The patient can experience physical problems such as fatigue, malaise, loss of appetite and pain combined treatment, can have a large impact on the patients functioning. An operation plaints as a consequence of the illness. The treatment, in particular in case of commay lead to permanent mutilation (amputation of the breast) or handicap (stoma). Radiotherapy may cause fatigue, nausea, skin problems, diarrhoea and loss of hair, depending on the radiation field. Chemotherapy often involves fatigue, loss of hair, mucositis, nausea and vomiting. It can cause bone marrow damage in such a way to concentration problems and loss of memory. The first year after diagnosis is age the diagnosis lung cancer scores highest.16 The treatment of cancer can be sur-

that the defence against infections decreases. In some cases the treatment may lead often characterized by diagnostic examinations followed by one or more treatment modalities. After this first year most of the acute side-effects associated with the treatment received have disappeared. Results of previous research among patients patients still reported physical problems and limitations in social activities.18 A

younger than 65 years of age show however that eight years after diagnosis these report of the Netherlands Institute for Research on Health Care (NIVEL) showed that 20-40% of the people were suffering from fatigue that seriously impedes daily functioning years after they were diagnosed with cancer.19

Early diagnosis and improvements in treatment have increased the survival rate of

many patients in the past 30-35 years. In 1998 the mean five-year survival of all
15

types of cancer and for all ages totals about 54% and this survival rate still increases.20 The survival rate of cancer patients however strongly depends on the type of cancer and the phase in which the cancer is diagnosed. The 5-years survival rate of breast cancer patient in the Netherlands is approximately 80%.21 Young women

often have a more aggressive type of cancer than elderly women which makes the survival rate in the first group lower.22 Men with testicular cancer have a five-year survival rate.21 survival rate of approximately 95%, whereas lung cancer patients have only a 18%

Psychosocial consequences
Cancer has a high impact on the patients psychosocial functioning. A considerable stress.23,24 The percentage of patients having clinically elevated levels of psycholog-

part of the patients confronted with cancer experience depression, anxiety and/or ical distress appeared to be dependent on the type of cancer and varied from 13%

cancer patients.24 Patients may lose trust in their bodies and experience uncertainty

in testicular patients25 to 30% in gynaecological cancer patients and 43% in lung

about the future. Although most studies show that the seriousness of the problems

problems years after diagnosis.27,28 The capacity of persons to adapt to the illness and its treatment, under similar circumstances, can vary individually. Factors such as, personality, social support received, coping, but also illness related variables such as time since diagnosis, treatment and phase of illness26,29-31 may affect the persons adaptation capacity. Younger patients prove to have more psychosocial problems when having cancer than older patients.18,32,33 The confrontation with a serious

decreases over time,26 a considerable part of the patients (15-25%) still experiences

illnesses seems to cause a greater disruption of everyday lives of younger persons, both occupationally and socially, than that of older persons. Parents with children experience feelings of imperfection as to their parental tasks.34 By far the most stud16

living at home worry about the consequences of the event for the children and may

ies have been performed on the psychological consequences of breast cancer in women. Next to that, research often focuses on other homogeneous groups of cancer patients such as prostate and testicular cancer patients. Only few studies have compared differences between male and female parents in the impact of the cancer found these were attributed to confounding effects such as age or variation in depends on the phase of the illness and the treatment the patient had undergone. Highest distress was found in the acute phase of the illness and in people who had diagnosis. In general, no gender differences were found.23,24 If differences were

types of cancer and prognosis.35 There are indications that psychological distress

received higher doses of chemotherapy and experienced more side-effects.36 Other patients in an advanced phase of cancer than in patients in the onset phase.26,37 On and psychosocial functioning were unrelated. 25,38

studies have found that a higher prevalence rate of depression was found among the other hand there are also studies that found that illness-related characteristics

Family system theory


The diagnosis cancer not only has consequences for the patients life but also for

the other family members. In the system theory families are seen as hierarchically organised systems, comprising subsystems (parent-child relationship, marital relationship). Every individual in a family is both a system in itself and a subsystem of the family total. The hypothesis of this theory is that an event in the life of a memmembers. ber of a family can also touch the lives and plans for the future of other family The spouse was found to experience as much stress as the patient23,39 and in the with breast cancer proved to be worried about the physical problems of their wives and about sexual intimacy, about the unpredictability of the illness and the disruption of the personal and family life.41 In case of the patient being parent of young
17

spouse as well problems may continue to exist for a long time.40 Spouses of women

children living at home household tasks and child care often fall to the responsibility of the spouse. For children the confrontation with cancer means that parents and family prove not to be as inviolable as always thought. The illness often leads to many changes in family life.42 In particular shortly after diagnosis and during treatment cancer will

be very prominently present in family life. In this phase the patient, but the healthy parent as well, can be less accessible for children. Tasks and responsibilities within the family may have to be rearranged. Children are confronted with the possibility of death of the parent. Various emotions such as fear, sorrow and anger may arise.42 Children may experience the fear of losing the parent or of becoming ill

cancer, which disrupts family life, or from feeling abandoned now that all attention facing the fact that they can do nothing for the parent in pain or the sick parent.

themselves.43 Anger may result from the feeling of injustice of the parent having

focuses on the parent.44 On the other hand they may experience impotence when Children may have feelings of guilt, thinking that they are the cause of the illness

or they may feel responsible personally for the parent being quickly irritated or payor to talk about their concerns related to the illness of their parent. Children do not themselves by complaints of a physical nature, sleeplessness, withdrawal or aggressive behaviour and irritability.43 Phase of illness

ing less attention to them.43 Children often find it very hard to explain their feelings

easily express their sorrow, fear, anger and worry verbally, they more often express

Rolland developed the family system health model to help explain family adapta-

tion to serious illness as a developmental process over time. This model hypothesizes a variation in demands upon the family members in the various phases of the illness (crisis, adaptation and terminal phase).15 The crisis phase, the period shortly

is characterized by an abrupt disruption of day-to-day life. Its life threatening character disrupts plans for the future. Acute anticipating decisions must be made about
18

after diagnosis and during treatment is often seen as a stressful period.46 This phase

treatments and hospital admissions and the subsequent high impact on emotions and practical affairs. The period of time and the intensity of the treatment and demands upon the family member.47 prevalence of complications influence the restrictions a parent experiences and the The adaptation phase comes after treatment. Sometimes the patients condition

improves in this phase, sometimes the cancer can lead to permanent restrictions and family members will be confronted with changes in lifestyle and role patterns. In case of a recurrence, a family is again confronted with often physically and emoeffect of the treatment and often families have to face a worse prognosis.46

tionally distressing treatments. Again fear and uncertainty will come up about the In the terminal phase, family members must come to terms with the death of the patient, a loss that may have been more or less anticipated. The physical condition care tasks.46,48 of the patient worsens and in general family members will have to take on more

Developmental phase of the family

The phases of the illness, according to Rolland, interact with the developmental phase of the individual family member and with the family as a unity.15 These phases are often characterized by means of the childrens age and the subsequent concept of life and death. To these children the absence of the parent and the disruption of the family routine in particular have the largest impact. Children in this age group moreover are totally dependent on the parental care. Families with children in this age group face the task of continuation of care for the children. When tute this care will have to be found. a parent is (temporarily) not capable of giving this care, ways and means to substiPrimary school children (6-11 yrs) are more self-reliant already and function more own outside the family, in school and with friends. One of the most important tasks in this phase is the developing of social and cognitive skills. A parent with cancer
19

changes in family life. Preschool children (2-5 yrs) are not yet familiar with the

independently from the parents. Children of this age have a social network of their

to them mostly brings about worries about the day-to-day functioning of the family and the consequences for their own life.49 Children in this age group may feel guilty about causing the illness of their parent themselves or feel personally responsible when the parent is quickly irritated or angry. They may think that cancer, just as a cold, is contagious, and avoid contact with the parent.

Adolescents (12-18 yrs) are physically and psychologically breaking away from the

parent and developing their own identity. Adolescents are cognitively and emotionconcept of life and death and they consequently have more empathy with the parinvolved in taking over the tasks of the parent and the care of the siblings. These to become more autonomous.42,49,50

ally further developed than younger children and have a very good notion of the ent. Another characteristic of this phase is that in general adolescents are sooner tasks may conflict with the developmental task to break away from the parent and To summarize, the consequence of the diagnosis cancer in a parent for children depends on their age-related developmental needs and the possibilities to realise them.51

The current study


Background

The first studies that focus on the consequences for children of a parent with canbeen any serious attention for this group of children.
23,47

cer were published some thirty years ago.52,53 Not until the past ten years there has

States and in Canada various studies have been performed investigating the conseresearch has been performed in this field. This is the subject of this thesis. The con-

Although in the United

quences of the diagnosis cancer in a parent for children, in the Netherlands no ceptual framework presented in Figure 1 hypothesizes first of all that the diagnosis cancer in a parent has consequences for the functioning of children. And it hypothesizes that the reason why some children experience more problems than others
20

may correlate with the following factors: a) Illness-related variables. As described earlier in this chapter the impact of the illness on children may depend on a variety of illness-related characteristics. The period shortly after diagnosis is often hectic pick up their normal life. Parents may experience more restrictions when treatthe demands on the other family members will increase as well. In case of recur-

and brings more uncertainty than the period after treatment when families try to ment takes longer and is more intensive with more complications. In this situation rence a family is confronted again with the illness, the impact of treatment and the often subsequent worse prognosis. b) Characteristics of the parent/spouse. Whether it is the father or the mother having cancer can have different consequences for the

child. In Dutch families there often is a traditional role pattern between fathers and

mothers, the mother mainly taking care of the household tasks and the care for the children.54 In case of the mother being diagnosed with cancer there will have to be a restructuring of household and child care tasks. Next there is the hypothesis that the gender of the parent interacts with the gender of the children. One study has shown that children of the same gender as the diseased parent are more problemother demographic variables such as age and educational level of the parent. The prone because they identify with this parent.55 Little research has been done into

quality of life a parent experiences as a result of illness and treatment, but also the adaptation capacities of children.12,56,57 c) Characteristics of the child. As seen in the

quality of life of the spouse are considered factors playing an important part in the previous paragraph the consequences of the diagnosis cancer in a parent can vary Although most studies do not include the gender factor where emotional and behav-

per age group. Therefore the age factor is important and should be included. ioural functioning after a stressful event is concerned, those who did so have shown

that girls are more vulnerable.23,58 Also, the differences between children in personality might affect the way in which they handle stress.59,60 The relationship between studied in children of a parent with cancer. d) Family structure. Being able to cope the functioning of children and the personality of children has not previously been

with a high impact event may depend on the family structure, such as one or two
21

parent families, number of children in a family and the childs position in the fam-

ily. The more family members the more the opportunities to distribute the practicalities of family life and the better the chance of family member support. e) Stressful life events. Having experienced more negative life events may have a cumulative

is that in situations where the parent is diagnosed with cancer and family members face several recent stressful events the chance of problems in children is larger. related, such as illness of another family or loss of a friend. Aim of the thesis

effect and enlarge the chance of psychosocial problems.1 The hypothesis therefore

These stressful events can be related to the cancer, e.g. financial problems, or non-

The main purposes of this thesis are:

1. To examine the prevalence of emotional and behavioural problems in primary school (aged 4-11 years) and adolescent (aged 12-18 years) sons and daughters of a parent diagnosed with cancer,

2. To investigate risk factors for the prevalence of problems in children, namely the childs socio-demographic variables and temperament; ill parents and spouses events. demographics and quality of life; illness-related variables, and experienced life

Two studies were performed: a prospective, longitudinal study and a cross-section-

al study. The first study was executed among families with a parent recently diagtimes during the first year after diagnosis. The second study focused on families one to five years after the cancer diagnosis in the parent. These families were asked the parent had an expected survival time of at least one year.

nosed with cancer. Family members were asked to fill out a questionnaire three

to fill out a questionnaire once. In both studies families were only approached when

22

Outline
Chapter 2 contains a review of all English language studies published between 1980 and 2004 on children of a parent diagnosed with cancer and on variables affecting childrens functioning. Chapter 3 investigates the emotional and behav-

ioural functioning of Dutch children whose parent was diagnosed with cancer one on child functioning was obtained from multiple sources. The effects of the childs age and gender, ill parents gender and illness related variables on childs functioning were examined. Attention is paid also to the extent to which different informants agree or differ in their perception of the functioning of children. The impact of

to five years previously by comparing them with norm group children. Information

ill parents demographics, family characteristics, illness-related variables, as well as both parents physical and mental functioning on the functioning of children, was the focus of chapter 4. Chapter 5 describes the effect of temperament on adoles-

cents psychosocial functioning, in addition to the effects of socio-demographics,

illness-related variables and number of negative life events children experienced behavioural problems in children during the first year after the parents cancer diagnosis is explored. Chapter 7 discusses the most important findings of this thesis. during the year prior to assessment. In chapter 6 the prevalence of emotional and

Methodological shortcomings and limitations are discussed and implications for clinical practice and future research are offered. Chapters 8 and 9 summarize the findings presented in the various chapters in English and Dutch, respectively.

23

Figure 1: Conceptual model

24

References
1. 2. 3. 4. 5. 6. 7. 8. 9. Forehand R, Biggar H, Kotchick BA. Cumulative risk across family stressors: short- and longterm effects for adolescents. J Abnorm Psychol. 1998; 26: 119-128. Hetherington EM, Stanley-Hagan M. The adjustment of children with divorced parents: a risk and resiliency perspective. J Child Psychol Psychiatry. 1999; 40: 129-140. Wallerstein JS. The long-term effects of divorce on children: a review. J Am Acad Child Adolesc Psychiatry. 1991; 30: 349-360. Wals M, van Os J, Reichart CG, Hillegers MHJ, Ormel J, Verhulst FC et al. Multiple dimensions of familial psychopathology affect risk of mood disorder in children of bipolar parents. Am J Med Genet B Neuropsychiatr Genet. 1997; 127: 35-41. Judicibus MA de, McCabe MP. The impact of parental multiple sclerosis on the adjustment of children and adolescents. Adolesc. 2004; 39: 551-569. Yahav R, Vosburgh J, Miller A. Emotional responses of children and adolescents to parents with multiple sclerosis. Mult Scler. 2005; 11: 464-468. Visser-Meily M. Childrens adjustment to a parents stroke: determinants of health status and psychological problems, and the role of support from the rehabilitation team. J Rehabil Med. 2005; 37: 236-241. Aldenkamp AP, Suurmeijer TP, Bijvoet ME, Heisen TW. Emotional and social reactions of chil-

10. Armistead L, Klein K, Forehand R, Wierson M. Disclosure of parental HIV infection to children in the families of men with hemophilia: description, outcomes, and the role of family processes. J Fam Psychol. 1997; 11: 49-61. 11. Rotheram-Borus MJ, Weiss R, Alber S, Lester P. Adolescent adjustment before and after HIVrelated parental death. J Consult Clin Psychol. 2005; 73: 221-228. 12. Armistead L, Klein K, Forehand R. Parental physical illness and child functioning. Clin Psychol Rev. 1995; 15: 409-422. 13. Korneluk YG, Lee CM. Childrens adjustment to parental physical illness. Clin Child Fam

dren to epilepsy in a parent. Fam Pract. 1990; 7: 110-115. Mukherjee S, Sloper P. Understanding the impact of inflammatory bowel disease on parents and their children. York: Social Policy Research Unit, University of York, 2001.

Psychol Rev. 1998; 1: 179-193. 14. Lewis FM, Hammond MA, Woods NF. The familys functioning with newly diagnosed breast cancer in the mother: the development of an explanatory model. J Behav Med. 1993; 16: 351-370. 15. Rolland JS. Chronic Illness and the life cycle: a conceptual framework. Fam Process. 1987; 26: 203-221. 16. Visser O, Siesling S, Dijck JAAM (eds). Incidence of cancer in the Netherlands 1999/2000, eleventh report of the Netherlands Cancer Registry. Utrecht: Association of Comprehensive Cancer Centres, 2003. 17. KWF Kankerbestrijding [Dutch Cancer Society, internet]. Cited February 2006. Available from www.kwfkankerbestrijding.nl. 18. Schroevers MJ. Short- and long-term adaptation to cancer: a comparison of patients with the general population. Dissertation, University of Groningen, 2003.

25

19. Jong A de, Rijken M. Zorg- en maatschappelijke situatie van mensen met kanker in Nederland [Care and social situation of people with cancer in the Netherlands], NIVEL, 2005. Available from url: http://www.nivel.nl. 20. Kennisnet Integrale Kankercentra [Dutch Comprehensive Cancer Centre, internet]. Cited August 10, 2006. Available on http://www.ikcnet.nl. 2006. 21. Coebergh JWW, Janssen-Heijnen MLG, Louwman WJ, Voogd AC, (eds.). Cancer incidence, care and survival in the South of the Netherlands, 1955-1999: a report of the Eindhoven Cancer Registry with cross-border implications. Eindhoven: Comprehensive Cancer Centre South (IKZ), 2001. 22. Dubsky PC, Gnant MFX, Taucher S, Roka S, Kandioler D, Pichler-Gebhard B et al. Young age as an independent adverse prognostic factor in premenopausal patients with breast cancer. J Clin Breast Cancer. 2002; 3: 65-72. 23. Compas BE, Worsham NL, Epping-Jordan JE, Grant KE, Mireault G, Howell DC et al. When mom or dad has cancer: markers of psychological distress in cancer patients, spouses, and children. Health Psychol. 1994; 13: 507-515. 24. Zabora J, BrintzenhofeSzoc K, Curbow B, Hooker C, Piantadosi S. The prevalence of psychological distress by cancer site. Psychooncology. 2001; 10: 19-28. 25. Fleer J. Quality of life of testicular cancer survivors. Dissertation. University Medical Centre Groningen, 2006. 26. Nordin K, Berglund G, Glimelius B, Sjoden P-O. Predicting anxiety and depression among cancer patients: a clinical model. Eur J Cancer. 2001; 37: 376-384. 27. Bleiker EMA, Pouwer F, van der Ploeg HM, Leer JWH, Ader HJ. Psychological distress two years after diagnosis of breast cancer: frequency and prediction. Patient Educ Couns. 2000; 40: 209-217. 28. Maguire P. Psychosocial interventions to reduce affective disorders in cancer patients: research

priorities. Psychooncology. 1995; 4: 113-119. 29. Carver CS, Smith RG, Petronis PVM, Weiss S, Derhagopian RP. Optimistic personality and psychosocial well-being during treatment predict psychosocial well-being among long-term survivors of breast cancer. Health Psychol. 2005; 24: 508-516. 30. Arora NK, Gustafson DH, Hawkins RP, McTavish F, Cella DF, Pingree S et al. Impact of surgery and chemotherapy on the quality of life of younger women with breast carcinoma: a prospective study. Cancer. 2001; 92: 1288-1298. 31. Eton DT, Lepore SJ. Prostate Cancer and health-related quality of life: a review of the literature. Psychooncology. 2001; 10: 1-20. 32. Northouse LL. Breast cancer in younger women: effects on interpersonal and family relations. J Natl Cancer Inst Monogr. 1994; 16: 183-190. 33. Wenzel LB, Fairclough DL, Brady MJ, Cella D, Garrett KM, Kluhsman BC et al. Age-related differences in the quality of life of breast carcinoma patients after treatment. Cancer. 1999; 86: 1768-1774. 34. Kelley SDM, Sikka A, Venkatesan S. A review of research on parental disability: implications for research and counseling practice. Rehabil Couns Bull. 1997; 41: 105-121. 35. Van t Spijker A, Trijsburg RW, Duivenvoorden HJ. Psychological sequelae of cancer diagnosis: a meta-analytical review of 58 studies after 1980. Psychosom Med. 1997; 59: 280-293.

26

36. Hurny C, Bernard J, Coates AS, Castiglione-Gertsch N, Peterson HF, Gelber RD et al. Impact of adjuvant therapy on quality of life in women with node-positive operable breast cancer. International Breast Cancer Study Group. Lancet. 1996; 347: 1279-1284. 37. Edwards B, Clarke V. The psychological impact of a cancer diagnosis on families: the influence of family functioning and patients; illness characteristics on depression and anxiety. Psychooncology. 2004; 13: 562-576. 38. Blank TO, Bellizzi KM. After prostate cancer: predictors of well-being among long-term prostate cancer survivors. Cancer 2006; 106: 2128-2135. 39. Davis-Ali SH, Chesler MA, Chesney BK. Recognizing cancer as a family disease: worries and support reported by patients and spouses. Soc Work Health Care. 1993; 19: 45-65. 40. Northouse LL, Templin T, Mood D, Oberst M. Couples adjustment to breast cancer and benign breast disease: a longitudinal analysis. Psychooncology. 1998; 7: 37-48. 41. Cochrane BB, Lewis FM. Partners adjustment to breast cancer: a critical analysis of intervention studies. Health Psychol. 2005; 24: 327-332. 42. Armsden GC, Lewis FM. The childs adaptation to parental medical illness: theory and clinical implications. Patient Educ Couns. 1993; 22: 153-165. 43. Christ GH, Siegel K, Freund B, Langosch D, Hendersen S, Sperber D et al. Impact of parental terminal cancer on latency-age children. Am J Orthopsychiatry. 1993; 63: 417-425. 44. Hilton BA, Elfert H. Childrens experiences with mothers early breast cancer. Cancer Pract. 1996; 4: 96-104. 45. Helseth S, Ulfsaet N. Having a parent with cancer: coping and quality of life of children during serious illness in the family. Cancer Nurs. 2003; 26: 355-362. 46. Sales E, Schulz R, Biegel D. Predictors of strain in families of cancer patients: a review of the literature. J Psychosoc Oncol. 1992; 10: 1-26. 47. Lewis FM, Hammond MA. The fathers, mothers, and adolescents functioning with breast

cancer. Fam Relat. 1996; 45: 456-465. 48. Siegel K, Karus D, Raveis VH. Adjustment of children facing the death of a parent due to cancer. J Am Acad Child Adolesc Psychiatry. 1996; 35: 442-450. 49. Lewis FM, Ellison ES, Woods NF. The impact of breast cancer on the family. Semin Oncol Nurs. 1985; 1: 206-213. 50. Stein JA, Riedel M, Rotheram-Borus MJ. Parentification and its impact on adolescent children of parents with AIDS. Fam Process. 1999; 38: 193-208. 51. Marshak L, Seligman M, Prezant F. Disability and the family life cycle: recognizing and treating developmental challenges. xiii ed. New York, NY, US: Basic Books, Inc., 1999. 52. Grandstaff NW. The impact of breast cancer on the family. Front Radiat Ther Oncol. 1976; 11: 146-156. 53. Wellisch DK. Adolescent acting out when a parent has cancer. Int J fam Ther. 1979; 1: 230-241. 54. Sociaal Cultureel Planbureau [Social and Cultural Planning Office of the Netherlands]. Cited February, 2006. Available on URL: http://www.scp.nl. 55. Connell AM, Goodman SH. The association between psychopathology in fathers versus mothers and childrens internalizing and externalizing behaviour problems: a meta-analysis. Psychol Bull. 2002; 128: 746-773. 56. Armsden GC, Lewis FM. Behavioural adjustment and self-esteem of school-age children of women with breast cancer. Oncol Nurs Forum. 1994; 21: 39-45.

27

57. Lewis FM, Woods NF, Hough EE, Bensley LS. The familys functioning with chronic illness in the mother: the spouses perspective. Soc Sci Med. 1989; 29: 1261-1269. 58. Compas BE, Worsham NL, Ey S, Howell DC. When mom or dad has cancer: II. Coping, cognitive appraisals, and psychological distress in children of cancer patients. Health Psychol. 1996; 15: 167-175. 59. Compas BE, Connor-Smith JK, Saltzman H, Thomsen AH, Wadsworth ME. Coping with stress during childhood and adolescence: problems, progress, and potential in theory and research. Psychol Bull. 2001; 127: 87-127. 60. Muris P, Ollendick T. The role of temperament in the etiology of child psychopathology. Clin Child Fam Psychol Rev. 2005; 8: 271-289.

28

29

30

The impact of parental cancer on children and the family: a review of the literature

Winette TA van der Graaf, Harald J Hoekstra, Josette EHM Hoekstra-Weebers

Annemieke Visser, Gea A Huizinga,

Both first authors contributed equally to this article

Cancer Treatment Reviews 2004; 30: 683-694

31

Abstract

this review was to survey present knowledge on the impact of parental cancer on children and the family. Design. Studies published between January 1980 March 2004 addressing emo-

Objective. Children of cancer patient may go through a distressing time. The aim of

tional, social, behavioural, cognitive and physical functioning of children of a parent diagnosed with cancer, as well as the association with child, parental and familial variables were reviewed.

dren (11 years) were reported in several qualitative studies, but in only one quan-

Results. Fifty-two studies were found. Emotional problems in primary school chil-

titative study. Quantitative and qualitative studies reported anxiety and depression Quantitative studies generally showed no behavioural and social problems in primary school children and adolescents. One quantitative study found physical complaints in primary school children. However, qualitative studies revealed behavtive and physical functioning in children of all ages. The most consistent variables related to child functioning appeared to be parental psychological functioning, marneeds of children and their families reported positive effects.

in adolescents (12 years), in particular in adolescent daughters of ill mothers.

ioural problems in primary school children and also described restrictions in cogni-

ital satisfaction and family communication. Intervention studies directed to the Conclusion. While quantitative studies reported especially disturbed emotional

functioning, qualitative studies reported problems in all domains of child functionfunctioning of children of cancer patients to develop tailored care.

ing. Well-designed studies are needed to gain more insight into the psychosocial

32

Introduction
The impact of cancer on patients psychosocial functioning has received considerable attention in the literature during the past two decades. A growing number of studies have addressed the psychosocial consequences for the spouse. However, limited attention has been paid to the effects on children when a parent is diagnosed and may result in the development of psychosocial problems, such as anxiety, conwith cancer. Confrontation with parental cancer can be very threatening for children fusion, sadness, anger, and feelings of uncertainty with respect to the outcome of the illness. They may face many changes in daily family routines due to repeated hospital admissions, hospital visits and care of the parent when at home.

This study reviews the current state of knowledge on psychosocial consequences for these childrens functioning. The findings will be organized around the following

children who have a parent diagnosed with cancer, and on variables that influence questions. Firstly, what is the impact of parental cancer on children, in terms of their emotional, social, behavioural, cognitive and physical functioning? Secondly, is there evidence that child, parental or familial variables are associated with the functioning of children who have a parent diagnosed with cancer? Thirdly, are there with this major life event? evidence-based interventions described which may help parents and children cope

Methods
A comprehensive search of the literature published after 1980 was conducted, using used in this search were: neoplasm, parental cancer, mothers and cancer, fathers

MEDLINE, EMBASE, PsycINFO, CINAHL, and CancerLit databases. The keywords and cancer, parent-child-relations, child functioning, quality of life, children and anxiety or depression, family functioning, and cancer and offspring. This search was supplemented with manual searches of the reference lists of extracted arti33

cles. The initial search yielded a total of 90 studies. Studies were excluded because it were dissertation abstracts, they were not in English, reported on the consequences for adult children of a parent with cancer, focussed on related topics (e.g., parenting), or that dealt with pre-death as well as post-death adaptation of children, only pre-death information was used.
1,2

described the bereavement of children of parents who had died of cancer. Of studies The remaining 52 studies addressed the psychosocial func-

tioning of children aged 0-20 years of parents diagnosed with cancer, and comprised

quantitative, qualitative and intervention studies. Those studies were reviewed indeincluded may vary, the quality of the quantitative studies was assessed using the guidelines of the Cochrane Library. Studies were considered as methodologically stronger or poorer on the basis of: design, representativeness of the sample, reliability of measurements, and use of control or norm groups. 3 The methodological qualGuba.
4,5

pendently by the first two authors. Because the methodological quality of studies

ity of the qualitative studies was evaluated using procedures described by Lincoln and

tive data: credibility, dependability, confirmability and transferability.

They suggested four criteria for establishing the trustworthiness of qualita-

To assess the methodological quality of the studies a standardized form was used for the authors.

data extraction. In case of disagreement consensus was achieved by discussion among Because quantitative and qualitative research approaches are methodologically differ-

ent, the results of studies will be reported separately. Results reported in the quantitawas reached.

tive studies reviewed are considered to be significant only when a level of p 0.05

Results
Study characteristics

A total of 52 studies met the inclusion criteria. Sample size, informants and illnessrelated information in the quantitative studies (n = 14) and in the qualitative studies
34

(n = 18) are summarized in Tables 1 and 2, respectively. Mixed-method studies (n = 13)

are summarized in Table 3. Intervention-studies (n = 7) were described in the text only. of children, 1,2,6-34 eight studies focused on family functioning or parent-child relationships, the role of school support,
35-42

The aim of the studies differed: 31 studies reported on the psychosocial functioning four on family communication,
47 43-46

described intervention programs for families that were designed to help children cope In almost half of the studies (46%) only mother with breast cancer were included. In with their parents cancer. 49-55

and one on care-provision by children.

one on the adolescents perceptions of


48

Seven papers

the remaining studies an overrepresentation of mothers with breast cancer was found. The majority of studies used a cross-sectional design with the exception of five studies 6-8,33,37 that used a longitudinal design. Twenty studies used normative data for comparison purposes,
2,7-9,14,-23,35-38,41,42

community sample of comparable subjects served as the control group in four studies.
12,13,32,39

while a

respondents on a standardized questionnaire. The manual of a questionnaire provided those norm scores. Data on child functioning and related variables were obtained from different inform-

Normative data comprised the scores of a large group of randomly selected

nine from one of the parents,


1,7,8,14,15,18,22-24,33-35

ants: eighteen studies gathered information from the child only, 2,10,12,13,16,17,19,20,25-29,31,32,46-48 four studies from both parents
11,21,37-40,43-45

twelve studies from the child and parent(s),


6,30,36,42 9,41

and/or parent(s) and another closely related person.

and two studies from the child

Based on the above mentioned quality criteria nine of the 27 quantitative studies and mixed-method studies 8,12,14,17-19,23,34,37 were qualified as methodologically stronger studies. Eighteen of the 30 qualitative and mixed-method studies were qualified as trustworthy. have a strong qualitative and a poor quantitative part or vice versa. The references the other references in italic.
6,10,11,17-19,23,24,29-31,33,40,43,44,46-48

It appeared that some mixed-method studies

of methodologically stronger (parts of) studies are presented in the text in bold and

35

The reviewed studies reported on children of various ages. Although a few studies

presented the results without making a distinction between age groups, most stud(about 6-11 years) and adolescent children (about 12-18 years) separately. examined the functioning of pre-school children,
6,30,40,42

ies focused on specific age groups or presented the results for primary school Therefore, this classification will be used in this review. Four qualitative studies describe the results for pre-school children separately. but generally did not

Furthermore, it may be argued that the experience of a stressful life event may have

also positive consequences (e.g., deepening of relationships). Most studies have were reported by studies, these will be described.

focused on the negative impact of parental cancer, but if positive consequences

The impact of parental cancer on the psychosocial functioning of children Emotional functioning. Most quantitative studies reported that primary school some other studies found increased scores. children scored within the normal range on emotional problems,
12,14

studies reported more emotional problems in adolescents when compared to control or norm groups. 2,8,12,13,14,15,18 Yet, there were also a few studies that found simi16,20,21

With regard to adolescents, most

8,15,18,20,21

while

response symptoms (avoidance and intrusive thoughts) were also observed in primary school and adolescent children.
18

lar emotional problems in adolescents than found in norm groups.

Stress-

school children reported fear of cancer symptoms, side effects of treatment, the parent dying and of the vulnerability of the well parent. They reported feelings of guilt, because they considered themselves responsible for the occurrence of their

Qualitative studies showed that primary

parents cancer, for their parents anger, withdrawal, or lack of affection. Besides,

they were distressed about loss of their usual activities and loss of contact with symptoms and mood disturbances. They also reported fear of developing breast cancer themselves, fear of relapse, fear of losing their mother, anger, and guilt,
36

their peers. 25 Adolescent daughters were found to have increased psychosomatic

Table 1: Quantitative studies 1. Diagnosis / 2. Stage / 3. Time since diagnosis

Study

Respondents

1. Armsden & Lewis (1994) 2. Barnes et al. (2002)45 3. Birenbaum et al. (1999)14

48 children (194, 295) aged 6-12 years; 13 ill mothers; 11 nurse observers 32 mothers of 56 children aged 5-18 years (204, 355) 115 youngsters (31 primary school children, 84 adolescents) 66 ill parents (524, 145) and 54 partners

4. Harris & Zakowski (2003)32

27 adolescents (184, 95) aged 12-19 years from 22 families; 23 controls

5. Heiney et al. (1997)15

1. Breast / 2. Not terminal / 3. Range 6-96 mo 1. Breast / 2. I or II / 3. Range 4-6 mo 1. Breast (65%), genital/urinary/reproductive (18%), haematological (17%), others (9%) / 2. I: 47%, II: 24%, III: 8%, IV: 12%, missing: 9% / 3. Range 3-64 mo 1. Breast (55%), gynaecological (18%), other (27%) / 2. N.i. / 3. 1-5 years 1. Cancer (n.i.) / 2. N.i. / 3. N.i. 1. Breast cancer or benign biopsies / 2. I: 48%, II: 44%, III: 7% / 3. Mean 6.7 mo 1. Breast / 2. Non-metastastic / 3. Median 35.3 mo 1. Breast / 2. 0-II: 93.7%; III, IV: 6.7% / 3. Median 27.4 mo 1. Breast / 2. 0-II: 88.6%; III, IV:11.4% / 3. Median 13 mo 1. Breast / 2. 0-II: 97,6 %; IIIA: 2.4 % / 3. Mean 23.6 mo 1. Cancer (n.i.) / 2. Terminal disease / 3. N.i. 1. Cancer (n.i.) / 2. Terminal disease / 3. N.i. 1. BRCA 1-2 mutation carriers: 80% symptomatic / 2. N.i. / 3. N.i. 1. Breast (37%); gynaecological (20%); others (43%) / 2. I: 29%; II: 36%; III: 22%; IV: 13% / 3. T1 Mean 9.7 weeks: T2: 4 mo post-diagnosis

6. Hoke (2001)20

7. Lewis et al. (1989)38 8. Lewis & Hammond (1992)37

9. Lewis et al. (1993)36 10. Lewis & Hammond (1996)35 11. Siegel et al. (1990)39

12. Siegel et al. (1992)12

13. Tercyak et al. (2001)16 14. Welch et al. (1996)8

33 children (164, 175) aged 5-12 years (N=21) and 13 to 17 (N=12) 22 ill parents (194, 35) 35 children (214, 14 5) of 28 mothers with breast cancer 34 children (174, 17 5) of 24 mothers with benign breast biopsies Children both groups aged 8-16 years, mean age 11.6 years 19 well fathers of 19 primary school children 111 ill mothers with one or more primary school or adolescent children (number n.i.) 40 ill mothers and 40 partners with primary school children 70 adolescents (56%4, 44%5); 70 ill mothers; 70 well partners 27 well parents (154, 125); Community sample: 44 parents (254, 195) Reports on children aged 7-16 years 62 children (aged 7-16) from 42 families Community sample: 556 students (aged 7-16 years) from 434 families 20 children (144, 65, mean age 14.9) of 15 mothers 34 primary school children (50 % 4); 55 adolescents (60 % 4) 54 ill parents (80 % 4) and 36 spouses (33 % 4)

37

N.i.= no information; 0, I, II, III, IIIA, IV: stage of disease as described by authors (stage 0 = in situ)

Table 2: Qualitative studies 1. Diagnosis / 2. Stage / 3. Time since diagnosis 1. Breast / 2. I or II / 3. Range 4-6 mo 1. Cancer (n.i.) / 2. Terminal disease / 3. N.i. 1. Breast / 2. I: 32%, II: 6%, remission: 39%, unknown: 23% / 3. N.i. 1. Cancer (n. i.) / 2. Terminal disease / 3. N.i. 1. Cancer (n.i.), / 2. Terminal disease / 3. N.i.

38
1. Breast 36%, haematological 19%, gynaecologic 12%, others: 33% / 2. N.i. / 3. N.i. 1. Multiple myeloma, lung, breast, pancreas / 2. N.i. / 3. N.i. 1. Cancer / 2. Not terminal / 3. N.i. 1. Breast / 2. Early stage / 3. T1 at diagnosis 1. Breast / 2. N.i. / 3. 2 years 1. Solid tumour 70%; leukaemia 30% / 2. N.i / 3. Range 4 weeks - 7 yrs 1. Breast / 2. Not terminal / 3. Mean one year 1. Breast / 2. I-II: 29%; III-IV: 7%; Recurrence: 13%; Remission: 39; Advanced or terminal: 13% / 3. N.i. 1. Breast / 2. N.i. / 3. Range 2-3 years 1. Breast / 2. 0: 6%; I: 33%; II: 39%; III: 17%; IV: 6% / 3. Mean 12.6 mo, range 3-32 mo 1. Breast / 2. N.i., 2 recurrences / 3. N.i. 1. Breast / 2. 0-II / 3. Mean 39 mo, range 23-56 mo 1. Breast / 2. 0-II / 3. Mean 56.6 mo, range 43-64 mo

Study

Respondents

1. 2. 3. 4.

Barnes et al. (2000) Berman et al. (1988)1 Chalmers et al. (2000)47 Christ et al. (1993)25

43

5. Christ et al. (1994)26

6. Fitch et al. (1999)40

32 mothers of 56 children aged 5-18 years (204, 355) 10 adolescents aged 11-17 years; 7 surviving parents (44, 35) 31 adolescents (224, 95) aged 12-20 years from 27 families 87 primary school children (aged 7-11 years) from 76 two-parent families with a terminally ill parent 120 adolescents (aged 11-17 years) from 86 two-parent families with a terminally ill parent 47 ill mothers of 1-4 children aged 4-18 years

7. Gates & Lackey (1998)48

8. Helseth & Ulfsaet (2003)33

9. Hilton & Elfert (1996)6

10. Hilton & Gustavson (2002)31 11. Hymovich (1993)30

12. Issel et al. (1990)10

13. Kristjanson et al. (2004)46

11 adolescents (84, 35) aged 10-19 years (mean: 14) caring for adults from 7 families 10 families: patients (74, 35), spouses (34, 55) and children (44, 75) aged 7-12 years 3 families with pre-schoolers and primary school children 5 families of primary school children and younger adolescents 4 families with older adolescents 11 children (74, 45), aged 7-21, from six families 10 families: patients (34, 75), and spouses (14, 25) of 26 children aged 9 weeks - 20 years 35 children aged 6-12 and 46 adolescents aged 13-20 years from 50 families (1-3 children per family) 31 adolescents aged 12-18 years

14. Rosenfeld et al. (1983)27 15. Shands et al. (2000)44

8 adolescent daughters aged 12-20 years from 7 families 19 mothers of 30 children aged 7-12 years

16. Spira & Kenemore (2000)28

17. Zahlis & Lewis (1998)11 18. Zahlis (2001)29

adolescent daughters aged 12-19 years (number n.i.); referred to social worker 26 ill mothers of 36 children aged 8-12 years (194, 175) 16 adolescents (84, 85, aged 11-18 years) from 11 families

N.i.= no information; 0, I, II, III, IV: stage of disease as described by authors (stage 0 = in situ)

Table 3: Mixed-method studies 1. Diagnosis / 2. Stage / 3. Time since diagnosis

Study

Respondents

1.

Compas et al. (1994)

18

2.

Compas et al. (1996)17

50 adolescents (58% 4), 26 primary school children (42% 4) 117 ill parents (72%4); 76 spouses (36% 4) 32 primary school children (47% 4); 59 adolescents (54% 4)

3.

Grant & Compas (1995)19

55 adolescents (334, 225), aged 11-18 years

4. 5.

Howes et al. (1994)21 Huizinga et al. (2003)24

6.

Lewis et al. (1996)7

7.

Lichtman et al. (1984)41

19 ill mothers of 32 children (184, 145), aged 3-18 years 15 children (104, 55), aged 7-18 years from 14 families 14 ill parents (13 mothers, 1 father); 12 spouses (11 fathers, 1 mother) 22 single ill mothers; 25 primary school children and adolescents 101 partnered ill mothers; 106 primary school children and adolescents 78 ill mothers; 63 significant others (spouses 73%); 3 physicians

1. Breast 32%, gynaecologic 21%, brain 12%, haematological 10%, others 25% / 2. I: 33%, II: 28%, III: 22%, IV: 17% / 3. Mean 8.6 weeks 1. Breast 28%; gynaecologic 20%; haematological 10%; lung 7%; others 35% / 2. I: 36%, II: 24%; III: 21%; IV: 19% / 3. Mean 9.8 weeks 1. Breast, ovarian, leukaemia, Hodgkin and others / 2. I: 36%; II: 24%; III: 21%, IV: 19% / 3. Mean 8.6 weeks 1. Breast / 2. I-IV / 3. Mean 32.8 mo, range 7-117 mo 1. Breast 71%; germ cell tumor 7%; soft tissue sarcoma 7%; ovarian 7%; testicular 7% / 2. N.i / 3. 2-52 months post-treatment 1. Breast / 2. 0-II / 3. Median 18 mo

8. 9.

Mireault & Compas (1996)2 11 adolescents (6 4, 5 5) 24 adolescents (8 4, 16 5), aged 11- 21 years from 16 families Nelson et al. (1994)13

10. Nelson & While (2002)22

1. Breast / 2. I: 31%; II: 55%; distant metast.: 14% / 3. Mean 25.5 mo; range 1-60 mo 1. Cancer (n.i.) / 2. N.i / 3. T1 at diagnosis 1. Hodgkin 43.6%; Non-Hodgkin 12.6%; breast 43.8% / 2. II: 3; III: 1; IV: 3; Incurable: 2; High risk relapse: 2; ongoing disease: 1; unknown: 4 / 3. Range 2-6 years 1. Breast 63%; lymphoma 11%, leukaemia 6%, gynaecological 4%; testicular 4% / 2. N.i. / 3. One year 1. Cancer (n.i.) / 2. Terminal disease / 3. Mean 2 years 1. Breast cancer / 2. Metastatic vs. non-metastatic / 3. N.i. 1. Cancer (n.i.) / 2. N.i. / 3. N.i.

11. Siegel et al. (2000)23

12. Sigal et al. (2003)34

13. Vess et al. (1985)42

34 primary school children and 46 adolescents (464, 345) 80 ill parents (684, 125) 64 spouses 119 children (50 % 4, aged 7-16 years) from 77 two-parent families 77 well parents (57 % 4) 42 women with metastatic and 45 women with non-metastatic breast cancer with children aged 6-18 years 54 patients (304, 245) and 54 spouses of children under age 20 years living at home

39

N.i.= no information; 0, I, II, III, IV= stage of disease as described by authors (stage 0 = in situ)

because they wished to continue their own lives. 26-28,29 Adolescents were afraid of mistakes in the care of this parent. 48

being left alone with their ill parent, because they were worried about making

Social functioning. Quantitative studies did not show any differences in social parents with cancer and a norm group. 14,20,21 Qualitative studies focused mainly on dren reported to have no one to help them cope with the situation.
10

competence (skills in social contacts and leisure activities) between children of relationships of children with family members and friends. Primary school chilAdolescents

reported to have more people (parent, school nurse or counsellor, teacher) to rely

on than younger children. 26 One study found that adolescents perceived the home need for more support from inside their family. 27 School was an important source of support for adolescents 47 and served as a haven away from care-provision. 48 environment as supportive, 48 while another found that adolescent daughters had a

Behavioural functioning. Quantitative studies using self-report data from youngsters 2,10,29 and/or data from parents 8,9,14,20,21 did not show any differences in the prevalence of behavioural problems (e.g., externalizing: delinquency or aggression)

between primary school or adolescent children of cancer patients and norm group children. Qualitative studies reported various results. Increased crying, clinging and drens behavioural reactions included a change in the intensity of talking, trying to distance themselves from cancer, increased checking on how the ill mother was doing, taking over the mothering role, seeking physical closeness or withdrawal, 11 having increased conflicts with parents, siblings and peers, 25 and paying more than usual attention to the mothers needs and wanting to support her. 10 difficulty in sleeping were found in pre-school children.
30

Primary school chil-

primary school children were unable to concentrate and complete assignments at school. 25 Some adolescents showed a decline in school performance and attendance
40

Cognitive functioning and school performance. Qualitative studies reported that

lescents functioned better at school. 48

(truancy, coming late or leaving earlier to pick up siblings), 13,27,28 while other ado-

ents reported somatic symptoms, such as sleeping difficulties and headaches, in selves also reported sleeping problems. 25 Adolescent daughters indicated that they suffered from a variety of symptoms, including headaches, abdominal pain, dizziness, sleeping problems and loss of appetite. their ill parent reported fatigue.
48 27,28

Physical symptoms. According to one quantitative and two qualitative studies par30

their pre-school

and primary school children.

6,15

Primary school children them-

Youngsters who were caring for

Relationships between study variables and child functioning Child variables

Age. Quantitative studies found more emotional problems in adolescents than in primary school children.
2,8,18,26.

stress-response symptoms than adolescents. 18 Qualitative studies documented that pre-school children were reacting on nonverbal and stressful behaviour of the parent and separation from the mother. 6 Primary school children were more affected ing and loss of hair. 25,31 Complications and emergency hospitalisations were espewith the well being of their parent
15

Primary school children, however, showed more

by the visible symptoms of the illness and side effects of treatment, such as vomitcially disturbing for primary school children. 25 Adolescents were more preoccupied their thoughts and feelings about cancer than primary school children. 10

and were more inclined to talk openly about

Gender. The methodologically stronger quantitative studies found more emotional problems in adolescent daughters of mothers diagnosed with cancer than in daughters of fathers with cancer, or in sons of mothers or fathers diagnosed with cancer.
8,18,19

independent of the ill parents gender.

Adolescent daughters reported also the highest scores on aggressive behaviour,


8

Otherwise, the methodologically poorer


41

studies found no gender differences for emotional problems, behavioural problems adolescent sons than in adolescent daughters. 13,22 Parental variables and social competence,
21

or found higher anxiety-scores and lower self-esteem in

and treatment modalities.

functioning and type and stage of cancer, time since diagnosis,


20

Illness-related variables. Quantitative studies found no relationship between child Children whose parents suffered from advanced stage
8,18

illness severity

disease and a poor prognosis seemed to perceive their parents illness as more seridren were reported to have fewer externalizing symptoms than children of parents with non-advanced stage illness.
34

ous and stressful, and avoided thinking about their parents cancer. 17,18 These chilQualitative studies revealed a negative impact
35,41

on the mother-child relationship when the mother had a poor prognosis, extensive surgery, and suffered more side-effects from radiotherapy and chemotherapy.

The period of diagnosis and treatment, and when the illness situation decreased seemed to be most difficult for primary school and adolescent children, because of the uncertainty and the diminished availability of their mother. 6,11,27,33,46

Five studies paid attention to the impact of gender-specific cancers or hereditary risks on children. One quantitative study found no differences in anxiety/depression and stress-response symptoms between daughters of mothers who had gender-specific cancer (breast or gynaecological cancer) and daughters of mothers who had developing cancer, however, showed more withdrawal and somatic problems. nerability when their mother and grandmother had breast cancer.
25 16

non-gender-specific cancer. 18 Adolescents who worried about their own chances of Qualitative studies reported that primary school daughters were aware of their vul-

daughters showed increased high-risk behaviours, such as delinquent behaviour and the use of drugs, as a consequence of fear of getting the disease themselves. 28 were at risk for breast cancer, but did not perceive this as a continuous threat. 6 However, another study reported that most adolescent daughters knew that they Parent psychological functioning. Quantitative studies found that better psycholog42

Adolescent

ical functioning of the ill parent was associated with better psychological functioning of the child, ship. 41 However, another study found no relationship between the parents psycho15,20,21,22

a higher self-esteem,

23

and a better mother-child relation-

ill parent was also related to positive effects: adolescent children of more anxious and distressed mothers were found to be socially more competent. 20 Family variables

logical functioning and that of the child. 18 Worse psychological functioning of the

Parent-child relationship. Results of the qualitative studies concerning the consequences of the parents illness on the parent-child relationship varied within studconflicts.
25,26 6,13,25,41

ies from an improvement in the parent-child relationship, to no change, to increased with both parents before the diagnosis found it more difficult to adapt to the illness. Contradictory results were found for the effect of the gender of the child. Two
40,41

Adolescents who had a poor relationship with the well parent or

studies showed that mothers experienced deterioration in the relationship with their daughters, but an improvement in the relationship with their sons, ters than with their sons. 6 another study parents indicated that they talked more sensitively with their daughwhile in

Marital functioning. Greater marital satisfaction had a positive impact on the childs psychological functioning, on family functioning, family coping, quality of the parent-child relationship 36,37,38 and on the adolescents self-esteem. 35
35,36,37,38

Family structure. Quantitative studies found that primary school children of single seemed to have higher scores on behavioural problems 7 and stress levels 6 than chil-

mothers had lower scores on global self-esteem and social acceptance, and they

dren of two-parent families. Results obtained on adolescents of single mothers, however, showed that the quality of the parent-child relationship and self-esteem were equal to those in adolescents living in two-parent families. 7 Adolescents from

two-parent families and those with siblings had less involvement in the illness process than adolescents from single-parent families or only children, and their normal daily lives were less disrupted. 42,48

43

Changes in role. Qualitative studies revealed that although parents attempted to roles in families. continue the daily life of their children as far as possible, 24,33 the illness blurred the
6,33,44

ceived increased care responsibilities for siblings and the ill parent. home health care was an additional burden for them.
24

Adolescents had to do more household chores

1,13,48

depended on the severity of the illness and the number of available care-providers, and were experienced as hard work, but also as gratifying. 48 The way the new roles were divided was important: tasks performed voluntarily, instead of being comwhich resulted in better family functioning.
42

Care-provision duties

13,48

Absence of

and per-

pelled, had significant positive effects in terms of less role strain and role conflicts, daughters had a more intimate nature than that provided by sons. Care for the ill parent provided by
6

caused anxiety in adolescent daughters, because they were afraid that this changed role would definitively alter their relationship with their mothers, 28 and also in adolescent sons. 13

Taking care

Family functioning. Quantitative studies reported that a high number of illnessrelated demands had a negative effect on family functioning. 35,36,38 In families that
24,32,35,38 35,36,38

the parent-child relationship was better.

were functioning fairly well, parents and/or children functioned better,

more organised than families with primary school children or children of both agegroups; they experienced more family cohesion, less family and role conflicts, and less role strain. 42

Families with adolescents only were

and

ness was of particular concern to parents. It was difficult for parents to decide what to tell their children about the illness, when and by whom. 30,44 It was a stressful task for parents to talk with the children, because they lacked knowledge about the illness themselves and were afraid they could not maintain emotional control in front of the dren were given simple information about the illness. Parents avoid to use words like ation, 6 but often appeared to be misinformed or had misconceptions about their parchildren. 30 The type of information children received varied. 1,22,43,44,45 Pre-school chil-

Family communication. Qualitative studies showed that communication about the ill-

cancer and dying. 6 Primary school children were generally informed about the situ-

44

ents illness and treatment, probably due to their limited cognitive development.

increased cognitive capacities of adolescents allowed them to understand the impli-

Adolescent children were informed more extensively than younger children. 6,45 The

25

cations of cancer, and many of them searched for information about cancer and treat-

reported that adolescents had a need for more information and support from family protected from negative test results, such as new lumps, 6 although adolescents tendmembers and persons outside of the family.
46

ment in addition to that received from their parents. 26 In spite of this, another study

Generally, children of all ages were

ed to be informed about the possibility of death when the parent became terminally ill. 1 Reasons to withhold information from children were that parents wished to avoid childrens questions about cancer and death, and belief that children were too young to understand.
6,43

although they perceived at the same time that they communicated openly with their children about the disease. 40,43 Exchanging information with their children and talking with each other served as a means of decreasing distress. 43,44
22

Parents tried to protect their children from fear and worries,

Whereas one study did not find any relationship between family communication tion or non-communication increased the risk of problems in children, 25,28 and in the parent-child relationship. 13 and child functioning, other studies demonstrated that poor family communica-

Informant agreement

Parental agreement. Fathers observed similar levels of anxiety/depression or aggression in their children as compared to mothers. drens social competence. 14 agreed the most about adolescents externalizing symptoms and the least about chil8

Ill parents and partners

Intergenerational agreement. Parents and children agreed moderately on childrens emotional and behavioural functioning, particularly regarding externalizing behavthan parents-reports.
8,9

iour. 14 Self-reports of children revealed more emotional and behavioural problems because children hide their emotions. 15,21 Problems of the child may escape the parents attention

45

Intervention studies Intervention studies were aimed to help family members to communicate more ed positive effects of the interventions, including less anxiety and more open communication. openly with each other and to increase their coping strategies. 49-55 All papers report-

Discussion
The first aim of this study was to examine the impact of parental cancer on children, in terms of emotional, social and behavioural, cognitive and physical funcal functioning of children who have a parent with cancer. Results for primary functioning in comparison with their peers. Nearly all studies reported that adolestioning. The majority of quantitative studies were aimed at evaluating the emotionschool children were inconsistent, varying from more emotional problems to equal cents had more emotional problems than found in the norm group. Adolescents may

have an increased vulnerability because of the conflicting demands of on the one hand the developmental task to separate from the family and the need to direct to practical, psychological and social tasks demanded by the illness. 2,6,26 relationships outside the family, but on the other hand the confrontation with the In the domain of social and behavioural functioning, primary school children and that children were doing well on these domains. Otherwise, children may try to pro-

adolescents were not found to differ from control or norm group peers. It might be tect the parent by showing less behavioural problems. None of the quantitative describing sleeplessness and headaches.

studies focused on cognitive functioning, and only one on physical functioning, The qualitative studies gave a different view on the four domains of primary school childrens and adolescents functioning. In the emotional domain, fear, mood disturbances, feelings of distress and guilt were described. Besides, several qualitative stud46

ies found a variety of behavioural, cognitive and/or physical problems in children.

The second aim of this study was to examine relationships between child, parental respond to parental cancer in various ways. Firstly, the reactions of children may be affected by their developmental level. Although studies reporting on children between the 0 and 20 years of age were included, results on the functioning of pre-

and familial variables and child functioning. It may be assumed that children

school-children were limited or were not described separately from other age of parental cancer on pre-school children. Comparison with other age groups showed that adolescents were reported to have more emotional problems than pri-

groups. This means that no general pronouncement can be made about the impact

mary school children. This may be due to their cognitive capabilities, as a result of adolescent daughters of mothers with cancer seemed vulnerable: they had more emotional problems than adolescent sons in general and adolescent daughters of fathers with cancer. Probably, adolescent daughters are more vulnerable due to the identification with their mothers and increased role responsibilities. 18,56 Emotional problems may be affected by the childs perceptions of the seriousness and stressfulness of the illness and a poor prognosis rather than other objective disease characteristics (such as type, stage, and time since diagnosis). The majority of the studies reviewed found a positive relationship between the psyof a meta-analysis on maternal depression and childs functioning.
57

which adolescents are more aware of the consequences of the illness. 18 Particularly,

chological functioning of the parent and the child, which is in line with the results level, open communication between the family members and greater marital satisOn family

faction between the parents had a positive effect on child functioning. Varying role patterns within the family, family structure and family functioning on childrens functioning.

results were found regarding the effects of parent-child relationships, changes in

The non-uniformity in results may be due to the heterogeneity in research questions, methodology, illness-related characteristics and different informant perspectives. The majority of studies evaluated the psychosocial functioning of the child, but in
47

some studies family functioning, family communication, school support or careprovision played a central role. Quantitative studies used a variety of questionnaires to measure psychosocial functioning in children (e.g., internalizing problems versus anxiety alone). However, it ure the specific problems children encounter when a parent has cancer. may be questioned whether the questionnaires used were sensitive enough to measA number of studies had fewer than 50 respondents, which may have lead to type

II errors. Moreover, in the majority of studies, cross-sectional data were described, which means that no conclusions could be drawn about causal relationships. Furthermore, over half of the studies did not give any information about the response rate, or the response rate was low. This raises the question as to whether has cancer.

the populations can be considered representative of all families in which a parent In a number of qualitative studies the methods of analysis were described only briefly. All qualitative studies used (semi)structured interviews, but it remained ity and confirmability of those studies. unclear what had exactly been performed, which limits the credibility, dependabilMany of the studies (quantitative and qualitative) focused specifically on patients with one certain type (e.g., breast cancer) or stage of cancer (stage I/II or terminal disease), whereas other studies included various diagnoses and stages of disease. In addition, the time since diagnosis varied widely, from a few days to nine years. A number of studies included children of considerably different ages, but did not

make any distinctions regarding age or developmental level when presenting the results. This may have limited the generalizability and transferability of the results. Different informants (parent/child) did not always have the same perceptions of on behavioural problems than on emotional problems. This may not be surprising because behavioural problems are easier to detect. child functioning. Parents as a whole tended to show a higher level of agreement

Finally, the third aim of this study was to examine whether evidence-based interventions are described for families in this situation. Though the reviewed interven48

tion studies reported all positive outcomes, these results were based on impressions of the facilitators, verbal feedback from participants and on self-constructed, nonvalidated questionnaires. The effectiveness of these interventions has not been evidence-based.

examined in randomised controlled trials and may therefore not be considered as

Future directions

In view of the diversity of results, as shown in this extensive review, it is extremely important to perform higher quality research into the psychosocial functioning of children who have a parent with cancer. Quantitative studies with large numbers of

respondents have greater power. In addition, larger samples offer the opportunity to

compare subgroups, for example differences between children whose parent has a good prognosis and children whose parent has a poor prognosis. Longitudinal studies are needed to gain more insight into the causal relationships between child functioning and the above-mentioned variables and into the long-term consequences.

The majority of the studies were performed among families of breast cancer patients. Although breast cancer is the most common disease in parents with children, more information is needed to gain insight into the functioning of children of fathers diagnosed with cancer. Further research is also needed about the functioning of pre-school children in this situation.

It is important to develop and validate an instrument that specifically measures the psychosocial functioning of children whose parents were diagnosed with cancer. With respect to the differences in outcomes between quantitative and qualitative studies, it seems advisable to combine these study methods (method triangulation). For instance, the results of a large quantitative study on child functioning can gain

in strength when combined with the results of a qualitative study with in-depth interviews with those children (and parents) reporting extremely high or low levels of functioning. There is no golden standard regarding who is the best informant of child functioning. It is therefore worthwhile to triangulate perspectives, not only from the parents
49

and children, but also from a significant other (such as schoolteachers).

Some children may be more vulnerable than others. It is therefore important to needed to establish the role of child characteristics (such as gender, developmental

identify factors that may act as facilitators or as barriers. Consequently, studies are phase, personality), parental characteristics (such as psychological functioning,

marital satisfaction, up-bringing style), family characteristics (such as parent-child communication, role changes within the family) and illness and treatment related variables. A theoretical model can serve as a guide to gain insight into the complexity of child functioning within families confronted with cancer. With more strucfor children and families at risk.

tured and well-grounded knowledge appropriate interventions may be developed

50

References
1. 2. 3. 4. 5. 6. 7. 8. Berman H, Cragg CE, Kuenzig L. Having a parent die of cancer: adolescents' grief reactions. Oncol Nurs Forum. 1988; 15: 159-163. Mireault GC, Compas BE. A prospective study of coping and adjustment before and after a parent's death from cancer. J Psychosoc Oncol. 1996; 14: 1-18. The Cochrane Collaboration open learning material for reviewers. Available from URL: http://www cochrane-net org/openlearning 2002. Lincoln YS, Guba EG. Naturalistic inquiry. Newbury Park, CA: Sage, 1985. Polit DF, Hungler BP. Nursing Research, 6th ed. Philadelphia: Lippincott Williams & Wilkins, 1999. Hilton BA, Elfert H. Children's experiences with mothers' early breast cancer. Cancer Pract. 1996; 4: 96-104. Lewis FM, Zahlis EH, Shands ME, Sinsheimer JA, Hammond MA. The functioning of single women with breast cancer and their school- aged children. Cancer Pract. 1996; 4: 15-24. Welch AS, Wadsworth ME, Compas BE. Adjustment of children and adolescents to parental

cancer. Parents' and children's perspectives. Cancer. 1996; 77: 1409-1418. Armsden GC, Lewis FM. Behavioural adjustment and self-esteem of school-age children of women with breast cancer. Oncol Nurs Forum. 1994; 21: 39-45. 10. Issel LM, Ersek M, Lewis FM. How children cope with mother's breast cancer. Oncol Nurs Forum. 1990; 17: 5-12. 11. Zahlis EH, Lewis FM. Mothers' stories of the school-age child's experience with the mother's 9.

breast cancer. J Psychosoc Oncol. 1998; 16: 25-43. 12. Siegel K, Mesagno FP, Karus D, Christ G, Banks K, Moynihan R. Psychosocial adjustment of children with a terminally ill parent. J Am Acad Child Adolesc Psychiatry. 1992; 31: 327-333.

13. Nelson E, Sloper P, Charlton A, While D. Children who have a parent with cancer: a pilot study. J Cancer Educ. 1994; 9: 30-36. 14. Birenbaum LK, Yancey DZ, Phillips DS, Chand N, Huster G. School-age children's and adolescents' adjustment when a parent has cancer. Oncol Nurs Forum. 1999; 26: 1639-1645. 15. Heiney SP, Bryant LH, Walker S, Parrish RS, Provenzano FJ, Kelly KE. Impact of parental anxiety on child emotional adjustment when a parent has cancer. Oncol Nurs Forum. 1997; 24: 655-661. Tercyak KP, Peshkin BN, Streisand R, Lerman C. Psychological issues among children of hereditary breast cancer gene (BRCA1/2) testing participants. Psychooncology. 2001; 10: 336-346. Compas BE, Worsham NL, Ey S, Howell DC. When mom or dad has cancer: II. Coping, cognitive appraisals, and psychological distress in children of cancer patients. Health Psychol. 1996; 15: 167-175. Compas BE, Worsham NL, Epping-Jordan JE, Grant KE, Mireault G, Howell DC, et al. When mom or dad has cancer: markers of psychological distress in cancer patients, spouses, and children. Health Psychol. 1994; 13: 507-515. Grant KE, Compas BE. Stress and anxious-depressed symptoms among adolescents: searching for mechanisms of risk. J Cons Clin Psychol. 1995; 63: 1015-1021. Hoke LA. Psychosocial adjustment in children of mothers with breast cancer. Psychooncology. 2001; 10: 361-369.

16. 17. 18. 19. 20.

51

21. Howes MJ, Hoke L, Winterbottom M, Delafield D. Psychosocial effects of breast cancer on the patient's children. J Psychosoc Oncol. 1994; 12: 1-21. 22. Nelson E, While D. Children's adjustment during the first year of a parent's cancer diagnosis. J Psychosoc Oncol. 2002; 20: 15-36. 23. Siegel K, Raveis VH, Karus D. Correlates of Self-esteem among children facing the death of a parent to cancer. In: Baider L, Cooper CL, Kaplan De-Nour A eds, Cancer and the family. Chichester: 2000, 223-237. 24. Huizinga GA, Van der Graaf WTA, Visser A, Dijkstra J, Hoekstra-Weebers JEHM. Psychosocial consequences for children of a parent with cancer: a pilot study. Cancer Nurs. 2003; 26: 195-202. 25. Christ GH, Siegel K, Freund B, Langosch D, Hendersen S, Sperber D, et al. Impact of parental terminal cancer on latency-age children. Am J Orthopsychiatry. 1993; 63: 417-425. 26. Christ GH, Siegel K, Sperber D. Impact of parental terminal cancer on adolescents. Am J Orthopsychiatry. 1994; 64: 604-613. 27. Rosenfeld A, Caplan G, Yaroslavsky A, Jacobowitz J, Yuval Y, LeBow H. Adaptation of children of parents suffering from cancer: a preliminary study of a new field for primary prevention research. J Prim Prev. 1983; 3: 244-250. 28. Spira M, Kenemore E. Adolescent daughters of mothers with breast cancer: impact and implications. Clin Soc Work J. 2000; 28: 183-195. 29. Zahlis EH. The child's worries about the mother's breast cancer: sources of distress in school-age children. Oncol Nurs Forum. 2001; 28: 1019-1025. 30. Hymovich DP. Child-rearing concerns of parents with cancer. Oncol Nurs Forum. 1993; 20: 1355-1360. 31. Hilton BA, Gustavson K. Shielding and being shielded: children's perspectives on coping with their mother's cancer and chemotherapy. Can Oncol Nurs J. 2002; 12: 198-206. 32. Harris CA, Zakowski SG. Comparisons of distress in adolescents of cancer patients and controls. Psychooncology. 2003; 12: 173-182. 33. Helseth S, Ulfsaet N. Having a parent with cancer. Cancer Nurs. 2003; 26: 355-362. 34. Sigal JJ, Perry JC, Robbins JM, Gagne MA, Nassif E. Maternal preoccupation and parenting as predictors of emotional and behavioural problems in children of women with breast cancer. J Clin Oncol. 2003; 1155-1160. 35. Lewis FM, Hammond MA. The father's, mother's, and adolescent's functioning with breast cancer. Fam Rel. 1996; 45: 456-465. 36. Lewis FM, Hammond MA, Woods NF. The family's functioning with newly diagnosed breast cancer in the mother: the development of an explanatory model. J Behav Med. 1993; 16: 351-370. 37. Lewis FM, Hammond MA. Psychosocial adjustment of the family to breast cancer: a longitudinal analysis. J Am Med Womens Assoc. 1992; 47: 194-200. 38. Lewis FM, Woods NF, Hough EE, Bensley LS. The family's functioning with chronic illness in the mother: the spouse's perspective. Soc Sci Med. 1989; 29: 1261-1269. 39. Siegel K, Raveis VH, Bettes B, Mesagno FP, Christ G, Weinstein L. Perceptions of parental competence while facing the death of a spouse. Am J Orthopsychiatry. 1990; 60: 567-576. 40. Fitch MI, Bunston T, Elliot M. When mom's sick: changes in a mother's role and in the family after her diagnosis of cancer. Cancer Nurs. 1999; 22: 58-63.

52

41. Lichtman RR, Taylor SE, Wood JV, Bluming AZ, Dosik GM, Leibowitz RL. Relations with children after breast cancer: the mother-daughter relationship at risk. J Psychosoc Oncol. 1984; 2: 1-19. 42. Vess J, Moreland JR, Schwebel A. An empirical assessment of the effects of cancer on family role functioning. J Psychosoc Oncol. 1985; 3: 1-16. 43. Barnes J, Kroll L, Burke O, Lee J, Jones A, Stein A. Qualitative interview study of communication between parents and children about maternal breast cancer. BMJ. 2000; 173: 385-389. 44. Shands ME, Lewis FM, Zahlis EH. Mother and child interactions about the mother's breast cancer: an interview study. Oncol Nurs Forum. 2000; 27: 77-85. 45. Barnes J, Kroll L, Lee J, Burke O, Jones A, Stein A. Factors predicting communication about the diagnosis of maternal breast cancer to children. J Psychosom Res. 2002; 52: 209-215. 46. Kristjanson LJ, Chalmers KI, Woodgate R. Information and support needs of adolescent children of women with breast cancer. Oncol Nurs Forum. 2004; 31: 111-119. 47. Chalmers KI, Kristjanson LJ, Woodgate R, Taylor-Brown J, Nelson F, Ramserran S, et al. Perceptions of the role of the school in providing information and support to adolescent children of women with breast cancer. J Adv Nurs. 2000; 31: 1430-1438. 48. Gates MF, Lackey NR. Youngsters caring for adults with cancer. Image J Nurs Sch. 1998; 30: 11-15. 49. Call DA. School-based groups: A valuable support for children of cancer patients. J Psychosoc Oncol. 1990; 8: 97-118. 50. Greening K. The 'Bear Essentials" Program: helping young children and their families cope when a parent has cancer. J Psychosoc Oncol. 1992; 10: 47-61. 51. Taylor-Brown J, Acheson A, Farber JM. Kids can cope: a group intervention for children whose parents have cancer. J Psychosoc Oncol. 1993; 11: 41-44.

52. Bedway AJ, Smith HJ. For Kids Only: development of a program for children from families with a cancer patient. J Psychosoc Oncol. 1996; 14: 19-28. 53. Heiney SP, Lesesne CA. Quest. An intervention program for children whose parent or grandparent has cancer. Cancer Pract. 1996; 4: 324-329. 54. Hoke LA. A short-term psycho educational intervention for families with parental cancer. Harv Rev Psychiatry. 1997; 5: 99-103. 55. Davis Kirsch SE, Brandt PA, Lewis FM. Making the most of the moment: when a child's mother has breast cancer. Cancer Nurs. 2003; 26: 47-54. 56. Korneluk Y, Lee C. Children's adjustment of parental physical illness. Clin Child Fam Psychol Rev. 1998; 1: 179-193. 57. Beck TC. Maternal depression and child behaviour problems: a meta-analysis. J Adv Nurs. 1999; 29: 623-629.

53

54

Emotional and behavioural functioning of children of a parent diagnosed with cancer: a cross-informant perspective

Ed C Klip, Elisabeth Pras, Josette EHM Hoekstra-Weebers

Harald J Hoekstra, Winette TA van der Graaf,

Annemieke Visser, Gea A Huizinga,

Psycho-Oncology 2005; 14: 746-758

55

Abstract

This study investigates emotional and behavioural problems in children of parents diagnosed with cancer and examines the relationship with demographic and illnessrelated variables. Furthermore, agreement and differences between informants

regarding childs functioning were examined. Members of 186 families in which a parent had been diagnosed with cancer participated. More emotional problems were reported for primary school sons (ill parents) and adolescent daughters (ill parents; self-reports), whereas also better functioning was reported in adolescent

children (spouses), compared to the norm group. Age and gender-effects were found: primary school sons were perceived as having more emotional problems than adolescent sons (ill parents); adolescent daughters as having more emotional

and behavioural problems than adolescent sons (ill parents; self-reports). Results mother was (spouses and self-reports). The treatment intensity affected adolescent

indicated a higher prevalence of problems when the father was ill than when the daughters functioning (spouses), whereas adolescent sons functioning was affectble levels of problems, but fathers perceived problems in children to be less preva-

ed by relapsed disease (self-reports). Adolescents and mothers perceived comparalent. Findings suggest that adolescent daughters and primary school sons are at risk for emotional problems following the diagnosis of a parent with cancer. The perto informant. ception of childs functioning and potential influencing variables varied according

56

Introduction
Cancer has a profound impact on patients, but may also be a significant emotional

stressor for children.1 Children may experience stress when confronted with the symptoms of the illness, the consequences and side effects of the treatments, and visits and admissions can also be stressful for children, who may manifest such the threat of a parents death.2 Alterations in daily family routines due to hospital stress in increased levels of emotional and behavioural problems. Primary school children have been shown to function emotionally and behaviourally similar to other children.3,4 A number of studies, however, have shown that the adolescent adolescents.5-8 Adolescent daughters whose mothers were ill appeared to be particchildren of cancer patients appear to have more emotional problems than do other ularly vulnerable.6,8 Other studies have suggested that the functioning of the adoles-

adolescents.10

cent children of cancer patients is similar to4,9 or even better than that of other

Inconsistencies in the results reported in prior studies may be due to variability in study design. Studies with small samples,e.g.,3,4 may suffer from bias. Studies vary breast cancer patients10 and others focusing on patients with various types of canfurther according to illness-related characteristics, with some studies focusing on cer.e.g.,6. In addition, time since diagnosis varied substantially. Additionally, various

tant informants, observing childrens behaviour over time and in many situations.

informants on childrens functioning introduce incongruency.5,8 Parents are impor-

Parent reports, however, are based on observable behaviour and the verbal reports

of children.11 The demands and uncertainties of cancer may make it difficult for a about their functioning. The self-reports of adolescents reflect their emotions and behaviours across different situations as well as their internal states.11 Yet, children

parent to recognize the needs of the children and to provide accurate information

may tend to deny their symptoms.12 Information obtained from a single source may give a one-sided view, while multiple perspectives may provide a more complete picture of the functioning of the child.

57

Large-scale research is needed to gain a better understanding of the consequences of parental cancer for the behavioural and emotional functioning of children, and to The goals of this study are 1) to examine the emotional and behavioural functionidentify individual differences among children in the prevalence of these problems. ing of sons and daughters of parents diagnosed with cancer by comparing them to children in a norm group; 2) to investigate whether emotional and behavioural functioning differs according to the age and gender of the child and the gender of

the ill parent; 3) to assess the impact of illness-related variables; and 4) to examine the extent to which different informants agree or differ in their perceptions of the functioning of children.

Methods
Procedure

Cancer patients and their family members were approached at the University

Medical Centre Groningen between January 2001 and February 2003. Physicians and oncology nurses offered written information to all eligible patients and an adapted version for their children. Patients were eligible for study participation they had children between 4 and 18 years of age who had resided with or had fre-

when diagnosed with cancer between one to five years prior to study entry and if quent contact with the parent diagnosed with cancer. To be eligible to participate,

both parents and children had to be fluent in Dutch. Informed consent was obtained from family members separately, according to the regulations of the Medical Ethical Committee of the University Medical Centre Groningen. After obtaining written informed consent, separate questionnaires and prepaid return envelopes were sent to each participating family member. Family members were family members.

instructed to complete the questionnaires independently and not to consult other

58

Instruments

Parents were asked to complete the Child Behaviour Checklist (CBCL)13,14 to assess

the emotional and behavioural functioning of children over the preceding six months. Adolescents completed the self-report version of the CBCL, the Youth Self-Report (YSR), designed for children between 11 and 18 years of age.15,16 The

CBCL consists of 120 and the YSR of 102 problem items, each of which has three

response options (0 = not true, 1 = somewhat or sometimes true, 2 = very true or of parents diagnosed with cancer, the present study used the internalizing, external-

often true). To provide a generalizable picture of the problems occurred in children izing, and total problem scale of the CBCL/YSR. The internalizing scale reflects

the internal mental states of children and consists of the narrowband syndromes of withdrawal, somatic complaints, and anxiety/depression. The externalizing scale represents socially unacceptable behaviour and consists of the syndromes of delinderived from the sum of all items and consists of the internalizing and externaliz-

quent and aggressive behaviour. The total problem scale represents the total score ing scales, thought, social, attention problems as well as the scale other problems. In addition, self-reports for boys contained the syndrome of self-destructive probproblems. Higher scores denote more problems. Normative data of the CBCL and lems, and parents reports for primary school children included the syndrome of sex YSR are available, with separate norms for primary school children (aged 4-11

years) and adolescents (aged 12-18 years), and for boys and girls.14,16 The norm data

of the CBCL were based on a randomly selected Dutch sample of 1241 parents school children (623 boys, 618 daughters) and 986 parents (95% mothers) who

(95% mothers) who provided information about the functioning of their primary rated the functioning of their adolescent children (493 boys, 493 daughters). The YSR norm group consisted of a random selection of 1124 adolescents (560 boys, 564 daughters). Raw scores were used to compare the mean scores of children and the percentage children who were clinically disturbed in the present study with the norm group. Raw scores were transformed into T-scores, based on normative data, to assess possible differences between age and gender groups, and between inform59

ants, beyond expected differences in the general population. The manual of the CBCL/YSR defines cut-off points for T-scores to differentiate youngsters considered to function normally from those considered to have clinically elevated problems (T-score > 63). The reliability and validity of the CBCL/YSR has been supported in a wide number of international and national studies. Cronbachs alphas in CBCL and YSR ranged from 0.84 to 0.94. Analysis

the present study for the internalizing, externalizing, and total problem scales of the

One Sample t-tests were conducted to compare raw scores on the CBCL and the of primary school children and adolescents scoring above and below the cut-off points with the norm group.14,16

YSR with normative data. Chi-square tests were used to compare the frequencies

Independent t-tests were performed to test for differences in T-scores between primary school children and adolescents. Analyses of variance (ANOVA) were conducted to test for differences in T-scores on the CBCL and YSR as a function of the these two variables.

gender of the child and of the ill parent, or as a function of the interaction between Independent t-tests, ANOVA and post hoc tests (Scheff) were computed to examillness-related variables. Length of time since diagnosis was categorized (1 to 2;

ine whether differences in childrens functioning (T-scores) could be explained by 2 to 3; 3 to 4 or > 4 years after diagnosis). A dichotomous variable was created for the treatment parents received (surgery only and non-surgical or multimodal was categorized (0-3 months; 4-6 months, 7 months) and a dummy variable was treatments, combining two or more treatment regimens). Duration of treatment created for recurrence of illness.

Pearsons product-moment correlations were performed to assess relative agreement (T-scores) between informants.17 Intraclass correlation coefficients (ICC)

were also calculated to assess pair-wise agreement between informants (absolute agreement). A Pearson correlation coefficient lower than 0.30 indicates poor
60

agreement, a coefficient between 0.30 and 0.50 indicates moderate agreement, and suggests low agreement, a coefficient between 0.40 and 0.75 suggests moderate to a coefficient higher than 0.50 indicates good agreement.18 An ICC lower than 0.40

good agreement, and ICC above 0.75 suggest excellent agreement.41 Finally, Paired t-tests were conducted to assess differences in the mean scores of various informants.

Results
Participants

Of the 476 cancer patients and family members informed about the study, 205 consented to participate (response rate 43%). In 22 percent of the families that declined to participate, the parents indicated that they were too emotionally distressed them-

selves, did not want to stir up emotions again, or that they wanted to move on and children lacked interest in the study, parents were afraid for emotionally distress of

leave the illness behind. In 20 percent the reason not to participate was because the the children, or expected that effects of cancer were small because minimal treat-

ment was needed (for example melanoma), or children were not informed about the of the children or the parents, or other illnesses in the family. The remaining 33 percent of the families specified no reasons for non-participation. Parents in the famients concerning gender of the ill parent, type of cancer, and time since diagnosis. lies that declined to participate did not differ significantly from participating parSome children had been 18 years of age at the time of diagnosis, but were 19 years of age or older during the study period. Because the instruments used were devel-

diagnosis. Twenty-five percent mentioned a variety of reasons, including business

oped for children between the ages of 4 and 18 years, children above the age of 18 years did not complete the questionnaires. The sample for the study therefore consisted of 180 ill parents, 145 spouses, 114 primary school children (4-11 years), and

222 adolescents (12-18 years). Child-rearing activities in participating two-parent


61

families were performed by the mothers in 61 percent; by both parents in 35 pertion are summarized in Table 1.

cent, and by fathers in 4 percent. Demographic characteristics of the study populaPatients were diagnosed with various types of cancer: breast (53%); gynaecologi-

cal (11%); skin (10%); haematological (9%); soft tissue and bone tumours (5%); urological (5%); gastrointestinal (2%); and other cancers (including central nervous system or head and neck cancer 5%). The mean time since diagnosis was 2.7 years the ill parents had initially received treatments involving only surgery, and 85 perradiotherapy, or multi-modal treatments combining two or more treatment regimens. The mean duration of treatment was 3.8 months (SD 3.0).

(SD 1.2). Thirty-four patients (19%) had suffered from relapses. Fifteen percent of cent had received more intensive treatment regimens consisting of chemotherapy or

Ill fathers did not differ significantly from ill mothers in age, educational level, number of children and one versus two-parent families, or in time since diagnosis, treatment received or recurrence of illness. Emotional and behavioural functioning

ly more internalizing problems in their sons than did parents from the norm population. No other significant differences were found (Table 2). According to reports from ill parents and their spouses, 23 percent of sons had internalizing problems in the clinical range, as compared to 10 percent in the norm population (2 = 10.9, p 0.001; 2 = 7.4, p = 0.006, respectively). Ill parents rated percentage significantly higher than the 8 percent found in the norm population (2 = 10.3, p = 0.003). According to reports from ill parents, 20 percent of daugh-

Parent reports regarding primary school children. Ill parents reported significant-

20 percent of the sons above the clinical cut-off on the total problem scale, a

ters had externalizing problems above the cut-off, as compared to 10 percent of

were scored above the cut-off on the remaining CBCL scales were comparable with those found in the norm group.
62

girls in the norm population (2 = 6.1, p = 0.013). The percentages of children who

Table 1: Demographic characteristics Parent characteristics N Ill parents Mothers Fathers Mean age = 44.3, SD = 5.1 Age range = 32 57 yrs Highest level of education completed by the ill parent Low1 Middle2 High3 56 74 56 30 40 30 180 145 34 81 19 % Spouses Mothers Fathers Mean age = 44.8, SD = 7.3 Age range = 31 65 yrs Highest level of education completed by the spouse Low1 Middle2 High3 43 57 45 30 39 31 N 145 32 113 22 78 %

Child characteristics N Primary school children Daughters Sons Mean age = 7.8 yrs, SD = 1.6 114 55 59 48 52 % Adolescents Daughters Sons N 222 117 105 Mean age = 15.0 yrs, SD = 2.3 52 48 %

Family characteristics N One-parent families Number of children in a family 1 2 3 >4


1 2 3

% 7

13 28 90 45 17

Number of children participating in the study per family 16 50 25 9 1 2 3 >4 69 79 23 9 38 44 13 5

Primary school or lower vocational degree Lower general secondary education or intermediate vocational education or high school degree Higher vocational education or university degree

63

more internalizing problems in adolescent daughters than were reported for girls in in sons, and externalizing and total problems in both sons and daughters as those levels of internalizing and total problems in both adolescent sons and daughters and lower levels of externalizing problems in sons (Table 2).

Parent reports regarding adolescent functioning. Ill parents reported significantly

the norm population. Ill parents reported similar scores for internalizing problems reported for the norm population. Spouses, however, reported significantly lower

According to the ill parents, 17 percent of adolescent daughters had scores above the cut-off on the internalizing scale, which was significantly higher than the 8 percent found in the norm population (2 = 12.5, p 0.001). No further significant difplaced them within the clinical range and those found within the norm population.

ferences were found between the percentages of adolescents whose parents scores Adolescent self-reports. Adolescent daughters reported significantly more internal-

izing and total problems than did their peers in the norm group (Table 3). Sons reported no more problems than did boys in the norm group.

Compared to adolescent girls in the norm group, a significantly higher percentage

Table 2: Descriptive statistics for raw scores on the Child Behaviour Checklist (CBCL) and t-tests for differences between parental scores for primary school children and adolescents and those from the norm group Ill parent Sons M (SD) Primary school children Internalizing Externalizing Total Problems Adolescents Internalizing Externalizing Total Problems
a

Spouse Sons M (SD) Daughters M (SD) M (SD) (4.9) (6.6)

Norm group Boys (SD) Girls M (SD)

Daughters M 6.2 6.8

7.1 (6.9)a 8.9 (6.6) 24.7 (18.6)

5.7 (6.5) 7.6 (6.0) 20.8 (16.0)

5.3 (5.7) 6.2 (7.2) 18.6 (18.0)

4.8 (4.7) 8.6 (6.7) 22.5 (15.2)

5.4 (5.5) 6.2 (5.9) 20.0 (16.1)

20.4 (14.3)

5.4 (5.5) 6.4 (5.7) 18.0 (14.2)

8.5 6.1

(7.2)b (5.9)

4.1 (3.9)b 4.7 (4.2)b

5.3 (5.0)a 5.2 (4.9)

5.7 (5.7) 7.1 (7.1) 20.1 (16.5)

6.5 (6.0) 5.5 (5.8) 18.7 (14.6)

21.5 (17.0)

14.2 (9.7)b 15.2 (12.6)a

p 0.05; b p 0.001

64

of adolescent daughters of cancer patients had scores above the clinical cut-off on both scales; 2 = 38.6, p 0.001; 2 = 27.3, p 0.001, respectively). The percent-

the internalizing (23%) and total problem scales (21%) (norm group girls: 8% on ages of sons in the clinical range were all at norm group levels. Gender and age differences (using T-scores)

Age of child. Ill parents reported that primary school sons had significantly more nificant differences were found between primary school children and adolescents.

internalizing problems than did adolescent sons (t = 2.3, p = 0.023). No other sigGender of primary school child and parent. Results of the ANOVA performed on

the reports of ill parents concerning their primary school children revealed no sig-

nificant gender effects for either the child or the ill parent. For the reports of spouses, however, ANOVA showed significant effects for the gender of the ill parent: more internalizing (F = 8.2, p = 0.005) and total problems (F = 9.6, p = 0.003) were mother was ill.

reported among primary school children when the father was ill than when the Gender of adolescent and parent. Results of the ANOVA performed on the reports

of ill parents showed significant effects for the gender of the adolescent on inter(F = 8.6, p = 0.004). In addition, interaction effects were found for the gender of

nalizing (F = 7.3, p = 0.007), externalizing (F = 6.5, p = 0.012), and total problems both the adolescent and the ill parent on externalizing (F = 8.4, p = 0.004) and total

Table 3: Descriptive statistics for raw scores on the Youth Self-Report (YSR) and t-tests for differences with the norm group Adolescents Sons M Internalizing Externalizing Total Problems
a

Norm group M 8.6 11.5 33.7 (SD) (5.8) (6.7) (16.8)

Daughters M 13.7 10.6 39.6 (SD) (9.6)b (6.1) (21.6)a

Norm group M 10.8 10.0 34.5 (SD) (7.1) (6.1) (18.0)

(SD) (6.8) (5.7) (17.3)

8.7 10.6 32.4

p 0.01; b p 0.001

65

problems (F = 8.2, p = 0.005). This means that, according to ill parents, daughters particularly those whose fathers were ill were perceived to have more problems than sons. Results of the ANOVA performed on the reports of spouses showed significant effects for the gender of the ill parent: adolescents had more internaliz-

ing (F = 24.3, p 0.001), externalizing (F = 4.8, p = 0.030), and total problems

tion, the gender of the adolescent was found to have a significant effect on externalizing problems (F = 5.6, p = 0.019), suggesting that, according to the spouses, adolescent daughters had more externalizing problems than did sons.

(F = 14.5, p 0.001) when the father was ill than when the mother was ill. In addi-

Table 4: Informant agreement between fathers and mothers regarding primary school child and adolescent functioning using Pearson product-moment correlation coefficients, intraclass correlation coefficients, and Paired t-tests Fathers Mothers Mean differences - 3.9 - 1.5 - 2.8 95%Confidence intervals of the differences - 8.05 to 0.16 - 4.05 to 1.05 - 6.51 to 0.78

T-scores Primary school sons Internalizing Externalizing Total Problems

Mean (SD) 49.1 (12.8) 48.3 (10.6) 47.1 (13.4)

Mean (SD) 53.1 (10.6) 49.8 (10.7) 49.9 (11.3)

r 0.48b 0.75
c

1
0.45b 0.75c 0.61c

t - 1.59 - 1.97 - 1.59

0.63c

Primary school daughters Internalizing Externalizing Total Problems Adolescent sons Internalizing Externalizing Total Problems 45.1 46.1 44.6 (9.3) (8.9) (9.5) 51.4 (9.3) 50.8 (9.0) 50.6 (9.6) 0.52c 0.61
c

48.8 (10.7) 49.6 (10.4) 47.3 (12.1)

52.3 (10.7) 50.3 (12.5) 50.7 (11.6)

0.66b 0.59
c

0.63c 0.58
c

- 2.46a - 0.39 - 2.11a

- 3.5 - 0.6 - 3.4

- 6.36 to - 0.62 - 4.11 to 2.77 - 6.73 to - 0.14

0.63c

0.61c

0.43c 0.54
c

- 5.84c - 5.03c - 6.41c

- 6.2 - 4.7 - 6.0

- 8.35 to - 4.11 - 6.54 to - 2.83 - 7.88 to - 4.14

0.65c

0.54c

Adolescent daughters Internalizing Externalizing Total Problems


a

47.5 (10.1) 49.9 (9.8) 46.6 (11.0)

52.7 (10.2) 51.6 (10.0) 51.2 (10.6)

0.55c 0.55
c

0.49c 0.54
c

- 4.87c - 1.64 - 4.06c

- 5.1 - 1.7 - 4.6

- 7.25 to - 3.04 - 3.76 to 0.37 - 6.79 to - 2.33

0.56c

0.52c

p 0.05; b p 0.01; c p 0.001

66

Results of the ANOVA performed on adolescents self reports showed significant izing (F = 9.1, p = 0.003; F = 5.0, p = 0.027, respectively) and total problems

differences as a function of the gender of the ill parent and of the child for internal(F = 5.4, p = 0.021; F = 4.3, p = 0.04, respectively). This means that adolescent chiladolescent children whose mothers were ill, and that adolescent daughters reported significantly more problems than did adolescent sons. Illness-related variables

dren whose fathers were ill reported having significantly more problems than did

Independent t-tests of the reports of ill parents yielded no significant differences in

the level of internalizing, externalizing, and total problems experienced by primary school and adolescent children of parents whose treatment had consisted of surgery alone and those experienced by children of parents who had received chemotherapy, radiotherapy, or multi-modality treatment, nor did the self-reports of adolescents. The reports of spouses indicated that adolescent daughters of parents whose total problems (t = 2.7, p = 0.005) than did adolescent daughters whose parents had adolescent children whose parents had suffered from recurrent illness were not

treatments had involved only surgery had less externalizing (t = 2.9, p = 0.008) and received other treatments or combinations of treatments. Primary school and perceived by their parents to have more problems than were the children of parents who had experienced no recurrent illness. Self-reports from adolescent sons, howp = 0.019) when the parent had experienced recurrent illness. Time since diagnosis and duration of treatment did not have significant effects on problems in primary school or adolescent sons and daughters. Informant agreement ever, revealed more internalizing (t = 2.8, p = 0.006) and total problems (t = 2.4,

fathers and mothers1 regarding the internalizing, externalizing, and total problems of
1

Agreement between parents. High correlations were found between the reports of

Pearson correlations computed for ill parent spouses yielded similar results.

67

their primary school children (r = 0.48 to r = 0.75) and adolescents (r = 0.52 to

r = 0.65). Comparable levels of agreement were found for parents in the norm group Intraclass correlation coefficients (ICC) showed similar results among primary school children and adolescents (1 = 0.45 to 1 = 0.75; 1 = 0.43 to 1 = 0.54, respectively). Paired t-tests showed no significant differences between reports of fathers and mothers with regard to the internalizing, externalizing, and total prob-

(primary school children: r = 0.54 to r = 0.81; adolescents: r = 0.52 to r = 0.74).

lems of primary school sons. Mothers reported significantly more internalizing and

total problems in their primary school daughters and in both adolescent sons and

Table 5: Informant agreement between adolescent sons and daughters and the fathers and mothers using Pearson product-moment correlation coefficients, intraclass correlation coefficients, and Paired t-tests Adolescents Parents Mean differences 4.97 2.66 4.94 95%Confidence intervals of the differences 2.46 to 7.49 0.47 to 4.86 2.43 to 7.44

T-scores Internalizing Externalizing Total Problems

Mean (SD) 49.9 (9.4) 48.7 (8.5) 49.3 (8.7)

Mean (SD) 44.9 (9.4) 46.0 (8.9) 44.4 (9.7)

r 0.31b 0.38
c

1
0.27b 0.37
c

t 3.94c 2.42a 3.92c

Adolescent sons - fathers

0.28a

0.25b

Adolescent sons - mothers Internalizing Externalizing Total Problems 50.8 (10.5) 49.5 (9.1) 50.2 (9.8) 51.6 (9.4) 51.1 (9.2) 50.7 (9.7) 0.36c 0.37
c

0.36c 0.37c 0.36c

-0.68 -1.52 -0.43

-0.80 -1.63 -0.49

-3.15 to 1.54 -3.76 to 0.50 -2.79 to 1.80

0.35c

Adolescent daughters - fathers Internalizing Externalizing Total Problems 53.2 (11.2) 51.3 (9.7) 52.4 (11.1) 47.4 (10.0) 49.0 (9.7) 45.9 (11.3) 0.34c 0.36
c

0.29c 0.35
c

4.65c 2.02a 5.05c

5.81 2.29 6.52

3.33 to 8.29 3.50 to 4.55 3.95 to 9.09

0.36c

0.31c

Adolescent daughters - mothers Internalizing Externalizing Total Problems


a

53.6 (11.9) 51.6 (9.9) 52.7 (11.3)

53.0 (10.5) 50.9 (9.6) 51.0 (10.9)

0.39c 0.40
c

0.38c 0.40c 0.34c

0.52 1.64 1.46

0.60 0.68 1.71

-1.67 to 2.87 -1.27 to 2.62 -0.61 to 4.04

0.34c

p 0 05; bp 0.01; cp 0.001

68

daughters than did fathers. According to the mothers, adolescent sons also had more externalizing problems (Table 4). T-tests for norm group parents showed one sigters than did fathers (p 0.05).14 nificant difference only: mothers reported more total problems in adolescent daughParent-adolescent agreement. Pearson correlations and ICC showed low to moder-

ate agreement (varying from r = 0.28 to r = 0.40 and 1 = 0.25 to 1 = 0.40) between the reports of parents and those of adolescents. Levels of agreement between parents and adolescents in the norm group ranged from moderate to high (parents-sons: r = 0.45 to r = 0.55; parents-daughters: r = 0.50 to r = 0.63). Paired t-tests showed no significant differences between the mean scores of the mothers and those of adolescent sons and daughters. In contrast, adolescent sons and daugh-

ters reported significantly more internalizing, externalizing, and total problems than did fathers (Table 5). In the norm group, adolescents reported significantly more problems than did their parents (p 0.05). (Agreement/differences between parents and adolescents in the norm group were not examined for fathers and mothers separately.11)

Discussion
A parents life-threatening illness can have far-reaching consequences for the functioning of children. The current study is the first large-scale research project in the children whose parents were diagnosed with cancer between one and five years prior to the study. The study involves separate analyses conducted for primary Netherlands to address the incidence of emotional and behavioural problems in

school children and adolescents, and for sons and daughters. In addition, various developed of the functioning of the child.

sources of information are considered, through which a comprehensive image is The primary goal of the study was to examine the emotional and behavioural functioning of sons and daughters of parents diagnosed with cancer by comparing them
69

to a norm population. The results of the present study showed that ill parents report-

ed more internalizing problems for their primary school sons than were reported for

norm group peers. In addition, the percentage of primary school sons reported by izing problems and total problems was higher (approximately one in five) than that

both ill parents and their spouses as having scores in the clinical range for internalof primary school boys in the norm population. Furthermore, reports from ill parfive) had externalizing problems in the clinical range. The results of the present

ents revealed that a greater percentage of primary school daughters (also one in study are not in line with previous studies among the primary school children of

parents diagnosed with cancer, which have reported the functioning of these chilonly on the CBCL subscales for anxiety/depression and aggression, while the interdren to be similar to that of norm group children.4,6,8 The studies cited here focused

nalizing and externalizing scale used in the present study includes a wider range of problems. Another explanation might be that the studies cited here were based on childrens self-reports, while the present study used parents reports. Children in the earlier studies also had high scores on a lie scale, which may suggest that the chil-

dren had attempted to present themselves in favourable or socially desirable ways. Although, the ambiguous results between the current study and prior studies among gested that boys in the primary school period are at risk for developing problems language and social-emotional functioning are not yet matured. As a consequence, boys might be incapable to express personal feelings and preferences efficaciously, which may lead to more distress. More research is needed in order to gain a more ability of primary school sons. when confronted with stressors.19-21. These studies hypothesized that boys skills in children of cancer patients, studies aimed at the development of children have sug-

thorough picture of the factors influencing and mechanisms underlying the vulnerThe present study also showed adolescent daughters to have higher mean scores on internalizing problems and that a greater percentage of adolescent daughters had clinically elevated scores on internalizing and total problems than was the case among girls in the norm population. This finding is consistent with other studies.6,8,22
70

The heavier responsibility for household or care-taking tasks in the family experi-

enced by girls during a parents illness may account for the higher incidence of problems among daughters.2,6,22 An alternative explanation for the greater vulnerability of adolescent daughters is the tendency of mothers to share their emotions with their daughters and to lean on their daughters for support.23 This emotional involvement may be too great a burden for the adolescent daughters.2,6,22

One remarkable finding was that the level of functioning reported by spouses concerning their adolescent sons and daughters was better than that reported by the parents of adolescents in the norm population. More than three-quarters of the spouses in the current study were fathers, while information on the functioning of chil-

dren from the norm group was obtained from a majority of mothers. In general, fathers tend to report fewer problems in their children than do mothers.24-26 It is posan artefact of the overrepresentation of male spouses in the present study. sible that the apparent underreporting of problems in children by the spouses was The results described above suggest that parental cancer may have more to do with anxiety/depression) than with externalizing problems (e.g., delinquent or aggres-

the prevalence of internalizing problems (e.g., withdrawal, somatic complaints, and sive behaviour). A study of children of divorced parents showed that children expetypes of stressors trigger problems in different areas, and confrontation with cancer children to turn inward emotionally. rienced more externalizing than internalizing problems.27. Apparently, different

does not necessarily lead to outward-directed behavioural problems, but may lead The second purpose of the present study was to investigate whether the prevalence

of problems was related to the age and gender of the child or to the gender of the ill parent. The results of the present study indicated that primary school sons suffered more internalizing problems than did adolescent sons. This is in contrast to

previous studies, which have found adolescents to experience more emotional be due to differences in the informants that were consulted. The present study was

problems than primary school children.2,6,8 The contradictions in these results may

based on the reports of parents, while the earlier studies relied on the self-reports of
71

children.

Regarding the gender of the adolescents and that of the parents, effects depended

on the informant who provided the information about childs functioning.

According to the ill parents and adolescents themselves, adolescent daughters experienced more problems than did adolescent sons. In addition to the explanations offered above, differences between sons and daughters may be due to the higher sensitivity of adolescent daughters (as compared to sons) to interpersonal concerns and stressful life-events involving others.28-30 It has also been suggested that adolescent daughters are more likely than sons to respond to stressful events with ineffec-

role socialization, daughters are encouraged to express emotions, but sons are stimulated to control their emotions.32,33

tive coping strategies.31 Furthermore, it is possible that, as a consequence of gender

Reports from spouses demonstrated that primary school children and adolescents mother was ill. Adolescents self-reports also showed that adolescents had more vious research that showed adolescent daughters of ill mothers to be the most vul-

had more problems when the father was diagnosed with cancer than when the problems when the father was ill. These findings were not in concordance with prenerable.6,8,22,23 Because of the small number of ill fathers in the present study, these

results may be due to coincidence and must be interpreted with caution. More attention will be paid to these gender effects in parental reports later on in this paper. The third goal of the study was to study the relationship between illness-related variables (time since diagnosis, initial treatment regimen, duration of treatments, and recurrence of illness) and the functioning of children. According to parents, the functioning of primary school children appeared not to have been affected by these

illness-related variables. In general, this was also found for adolescents. The selfreports of adolescent sons, however, revealed more problems when the parent had experienced a recurrence of the illness, and spouses reported that adolescent daughters functioned less well when the parent had received a more intensive treatment than surgery alone. The literature on the impact of illness-related variables is limited. The results of the few studies conducted on this topic indicated that not the
72

objective characteristics of the illness (e.g., stage of illness, 5-year survival rates, time since diagnosis) but the childs perception of the severity and stressfulness of ness and an intensive treatment regimen may be indicative of the adolescents perthe illness were related to emotional problems.6,34 For this reason, recurrence of ill-

ceptions of the severity or stressfulness of a life-threatening illness such as cancer. The fourth purpose of the present study was to examine agreement among the reports of mothers and fathers, and between parents and adolescents. Interparental

correspondence was found to be moderate to high, which corresponds to findings from other studies.14,25 Further analyses demonstrated significant discrepancies in mean scores between fathers and mothers. In general, mothers reported more inter-

nalizing and total problems in primary school daughters and in adolescent sons than did fathers was consistent with the results of a meta-analysis.25 but was in

and daughters than did fathers. The finding that mothers reported more problems conflict with other studies that found fathers and mothers to report similar levels of problems.14,35,36 The low to moderate agreement between the reports of parents and adolescents found in the present study was also documented in studies among parents and adolescents in other situations5,37,38 Furthermore, adolescents reported experiencing more problems than their fathers had reported, while the level of problems reportbecause studies among a community sample found that adolescents reported expeed by adolescents and mothers was comparable. This last finding is remarkable, riencing more problems than their parents perceived them to have.11,35,36,39 These studies, however, did not examine differences between reports of adolescents and those of fathers and mothers separately.

Our study suggests that fathers underreport the problems of their children. In genmothers.25 which may obscure the fathers perceptions of the emotional and behav-

eral, children tend to behave more obediently toward their fathers than toward their ioural problems their children truly have. The fathers in the present study may have been so focused on the illnesses of their wives and on the changes in responsibilities that their attention to the functioning of their children may have been affected

73

adolescents regarding the level of problems their children had experienced. It is plausible that mothers may have a better perception of the childs functioning than do fathers, as they are more often responsible for childcare, and therefore spend more time with their children and talk with them more often. The mothers in the ulation, due to the impact that the illness may have had on their children. As a conpresent study may have been even more concerned than mothers in the general popsequence, they may have been even more attentive to problems in their childrens functioning. Future research should focus more on parent-child agreement followin patterns of agreement. ing stressful situations and on the mechanisms that determine agreement or change

further. In contrast with earlier studies.36 mothers seem to have agreed with their

Every study has its strengths and limitations, as did the present study. First,

although the data in this study were derived from a large sample, 57 percent of the families approached for the study declined to participate. Although no differences between respondents and non-participants were found on demographic and illnessrelated variables, a sample bias may exist. Second, this study is based on cross-secthe dynamic interaction of potentially influencing variables. A longitudinal study tional data; such a design gives information at one point in time and does not reflect design may give insight into change over time and causality. Third, 62 percent of the children had siblings who also participated. Although children from the same family share genes and environment, they may react differently to a stressful

event.40 Therefore, no restrictions were made in the inclusion of number of children per family. Future studies may consider the use of multilevel analysis to gain insight into within-family and between-family variation in the functioning of chilresearch should focus on the consequences of parental cancer for sons and daugh-

dren. Fourth, the results of the current and previous studies suggest that future ters, and primary school children and adolescents separately. Specifically, future and resilience factors for children. In this case, studies could focus on the tempera74

studies should pay more extensive attention to the identification of potential risk

ment of the child, the copingstyles, the parent-child relationship, psychological functioning of the parent, family functioning or other stressfull events in a more indepth qualitative manner. In addition to the objective, quantitative methods, qualichildren.

tative research could also contribute to a description of feelings and experiences of The questionnaire used was not designed for the specific purpose of diagnosing emotional and behavioural problems of children of parents diagnosed with cancer. In this context the use of this screening instrument is experimental. Although this school sons and adolescent daughters with this generic questionnaire demonstrates its usefulness with this population. For the lack of other adequately normed instruments we see using it as a viable alternative to receiving no attention at all in clinpsychosocial aspects should receive attention.

is a limitation, the fact that we have discovered the greater vulnerability of primary

ical practice. When children score within the clinical range this is an indication that The current results may heighten the awareness of health care providers that parental cancer may affect children. Parents should be supported in recognizing specific concerns and needs of their children, in particular those of primary school sons and adolescent daughters. It is important to realize that parents may struggle with what they will tell the children and how parenting responsibilities can be comto develop a tailored support program for children and parents.

bined with their illness. Insight into risk and resilience factors of children may help

75

References
1. 2. 3. 4. 5. 6. 7. 8. 9. Rait D, Lederberg M. The family of the cancer patient. In Handbook of Psychooncology, Holland JC, Rowland JH. (eds). Oxford University Press: New York. 1989: 585-597. Christ GH, Siegel K, Sperber D. Impact of parental terminal cancer on adolescents. Am J Orthopsychiatry. 1994; 64: 604-613. Heiney SP, Bryant LH, Walker S, Parrish RS, Provenzano FJ, Kelly KE. Impact of parental anxiety on child emotional adjustment when a parent has cancer. Oncol Nurs Forum. 1997; 24: 655-661. Howes MJ, Hoke L, Winterbottom M, Delafield D. Psychosocial effects of breast cancer on the patient's children. J Psychosoc Oncol. 1994; 12: 1-21. Birenbaum LK, Yancey DZ, Phillips DS, Chand N, Huster G. School-age children's and adolescents' adjustment when a parent has cancer. Oncol Nurs Forum. 1999; 26: 1639-1645. Compas BE, Worsham NL, Epping-Jordan JE, Grant KE, Mireault G, Howell DC, Malcarne VL. When mom or dad has cancer: markers of psychological distress in cancer patients, spouses, and children. Health Psychol. 1994; 13: 507-515. Siegel K, Mesagno FP, Karus D, Christ G, Banks K, Moynihan R. Psychosocial adjustment of children with a terminally ill parent. J Am Acad Child Adolesc Psychiatry. 1992; 31: 327-333. Welch AS, Wadsworth ME, Compas BE. Adjustment of children and adolescents to parental

cancer. Parents' and children's perspectives. Cancer. 1996; 77: 1409-1418. Huizinga GA, Van der Graaf WTA, Visser A, Dijkstra JS, Hoekstra-Weebers JEHM. Psychosocial consequences for children of a parent with cancer: a pilot study. Cancer Nurs. 2003; 26: 195-202. 10. Hoke LA. Psychosocial adjustment in children of mothers with breast cancer. Psychooncology. 2001; 10: 361-369. 11. Verhulst FC, Van der Ende J. Agreement between parents' reports and adolescents' self-reports of

problem behaviour. J Child Psychol Psychiatry. 1992; 33: 1011-1023. 12. Grills AE, Ollendick TH. Issues in parent-child agreement: the case of structured diagnostic interviews. Clin Child Fam Psychol Rev. 2002; 5: 57-83. 13. Achenbach TM. Manual for the Child Behaviour Checklist/4-18 and 1991 Profiles. University of Vermont Department of Psychiatry: Burlington, VT, 1991. 14. Verhulst FC, Van der Ende J, Koot HM. Handleiding voor de Child Behaviour Checklist/4-18 [Manual for the Dutch version of the Child Behaviour checklist/4-18]. Rotterdam: University of Rotterdam, Department of Child Psychiatry, 1996. 15. Achenbach TM. Manual for the Youth Self-Report and 1991 Profiles. Burlington, VT: University of Vermont Department of Psychiatry, 1991. 16. Verhulst FC, Van der Ende J, Koot HM. Handleiding voor de Youth Self-repost [Manual for the 17. 18. 19. 20.

Dutch version of the Youth Self-Report]. Erasmus Rotterdam: University of Rotterdam, Department of Child Psychiatry, 1997. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. The Lancet. 1986; 36: 307-310. Cohen J. Statistical power analysis for the behavioural sciences. Hillsdale, NJ: Erlbaum, 1988. Keenan K, Shaw D. Developmental and social influences on young girls' early problem behaviour. Psychol Bull. 1997; 121: 95-113. Kraemer S. The fragile male. BMJ. 2000; 321: 1609-1612.

76

21. Simmons R, Burgeson R, Carlton-Ford S, Blyth D. The impact of cumulative change in early adolescence. Child Dev. 1987; 58: 1220-1234. 22. Grant KE, Compas BE. Stress and anxious-depressed symptoms among adolescents: searching for mechanisms of risk. J Consult Clin Psychol. 1995; 63: 1015-1021. 23. Lichtman RR, Taylor SE, Wood JV, Bluming AZ, Dosik GM, Leibowitz RL. Relations with children after breast cancer: The mother daughter relationship at risk. J Psychosoc Oncol. 1984; 2: 1-19. 24. Bartels M, Hudziak JJ, Boomsma DI, Rietveld MJH, Van den Oord EJCG. A study of parent ratings of internalizing and externalizing problem behaviour in 12-year-old twins. J Am Acad Child Adolesc Psychiatry. 2003; 42: 1351-1359. 25. Duhig AM, Renk K, Epstein MK, Phares V. Interparental agreement on internalizing, externalizing, and total behaviour problems: A meta analysis. Clin Psychol. 2000; 7: 435-453. 26. Treutler CM, Epkins CC. Are discrepancies among child, mother, and father reports on children's behaviour related to parents' psychological symptoms and aspects of parent-child relationships? J Abnorm Psychol. 2003; 31: 13-27. 27. Hetherington EM, Stanley-Hagan M. The adjustment of children with divorced parents: a risk and resiliency perspective. J Child Psychol Psychiatry. 1999; 40: 129-140. 28. Gore S, Aseltine R, Colten M. Gender, social-relational involvement, and depression. J Res Adolesc. 1993; 3: 101-125. 29. Grant KE, Compas BE, Stuhlmacher AF, Thurm AE, McMahon SD, Halpert JA. Stressors and child and adolescent psychopathology: moving from markers to mechanisms of risk. Psychol Bull. 2003; 129: 447-466.

30. Nolen-Hoeksema S. Gender differences in depression. Curr Dir Psycholog Sci. 2001; 10: 173-176. 31. Nolen-Hoeksema S, Jackson B. Mediators of the gender difference in rumination. Psychol Women Q. 2001; 25: 37-47. 32. Eisenberg N, Cumberland A, Spinrad TL. Parent socialization of emotion. Psychol Inquiry. 1998; 9: 241-273. 33. Garside R, Klimes-Dougan B. Socialization of discrete negative emotions: gender differences and links with psychological distress. Sex Roles. 2002; 47: 115-128. 34. Compas BE, Worsham NL, Ey S, Howell DC. When mom or dad has cancer: II. Coping, cognitive appraisals, and psychological distress in children of cancer patients. Health Psychol. 1996; 15: 167-175. 35. Sourander A, Helstela L, Helenius H. Parent-adolescent agreement on emotional and behavioural problems. Soc Psychiatry Psychiatr Epidemiol. 1999; 34: 657-663. 36. Stanger C, Lewis M. Agreement among parents, teachers, and children on internalizing and externalizing behaviour problems. J Clin Child Psychol. 1993; 22: 107-115. 37. Achenbach TM, McConaughy SH, Howell CT. Child/adolescent behavioural and emotional problems: implications of cross-informant correlations for situational specificity. Psychol Bull. 1987; 101: 213-232. 38. Thompson RJ Jr, Merritt KA, Keith BR, Murphy LB, Johndrow DA. Mother-child agreement on the child assessment schedule with nonreferred children: a research note. J Child Psychol Psychiatry. 1993; 34: 813-820.

77

39. Zukauskiene R, Pilkauskaite-Valickiene R, Malinauskiene O, Krataviciene R. Evaluating behavioural and emotional problems with the Child Behaviour Checklist and Youth Self-Report scales: cross-informant and longitudinal associations. Medicina. 2004; 40: 169-177. 40. Plomin R, Asbury K, Dip PG, Dunn J. Why are children in the same family so different? Nonshared environment a decade later. Can J Psychiatry. 2001; 46: 225-233. 41. Novella JL, Jochum C, Jolly D, Morrone I, Ankri J, Bureau F, Blanchard F. Agreement between patients' and proxies' reports of quality of life in Alzheimer's disease. Qual Life Res. 2001; 10: 443-452.

78

79

80

Parental cancer: characteristics of parents as predictor for child functioning

Harald J Hoekstra, Winette TA van der Graaf, Josette EHM Hoekstra-Weebers

Annemieke Visser, Gea A Huizinga,

Cancer 2006; 106: 1178-1187

81

Abstract

Background. The vulnerability of children when a parent is diagnosed with cancer may depend on a variety of variables. The current study examined the impact of characteristics of 180 parents diagnosed with cancer, along with 145 spouses, on the prevalence of emotional and behavioural problems in children. Methods. Ill par-

ents provided information on socio-demographics and illness-related variables, and (CBCL). Both parents completed the two subscales that measure physical function-

on the prevalence of problems in children by using the Child Behaviour Checklist ing and mental health of the RAND-36. Results. The family situation (single par-

ents, no/few siblings, oldest child) was one of the most important predictors of as having more problems when ill parents experienced treatment complications. A

reported problems in primary school children, whereas adolescents were reported decrease in ill parents physical functioning affected primary school daughters and

adolescents, and adolescents were also affected by the mental health of ill parents.

Problems of ill fathers did not have a different impact on children from those of ill mothers. Spouses physical limitations were indicative for problems in primary school children, whereas a worsening parental mental health was indicative for problems in adolescents. Conclusions. Findings illustrated that parents characteristics must be taken into account when assessing vulnerability of children in this sitdrens ages.

uation. Which variables particularly heighten the risk for problems depend on chil-

82

Introduction
Cancer not only affects the patient, but the experience also encroaches deeply on

family life. Research in recent decades has focused increasingly on the role of the partner in such situations. Researchers have also paid increasing attention to the impact that a diagnosis of cancer in a parent has on the children, although the results more emotional problems in children, whereas others found similar problems in comparable norm groups.1 In addition, a recent study has established that 21% of adolescent sons and 36% of adolescent daughters in such families experience serious stress response symptoms.2 Which children are particularly vulnerable remains unclear.

of these studies are ambiguous. A review study described that some studies reported

First, few studies have investigated the influence that social-demographic charac-

teristics of the parent have on the functioning of children in such situations. The majority of these studies focused on children of mothers who have been diagnosed with cancer. Furthermore, these studies have made no distinction between ill fathers

and ill mothers. Whether the impact of having a mother with cancer differs from the influence of family socioeconomic status (SES) in such situations. Research on the general population has shown, however, that lower SES increases risk for problems in children.3;4 Children from single-parent families and only children general-

that of having a father with cancer is, therefore, unclear. We still know little about

ly tend to be more vulnerable than children from two-parent families or families with more children.5,6 It is possible that this vulnerability increases even further

when a parent has cancer. Children in such families often have more responsibilities and fewer family members with whom they can share their concerns.7-9 Research results concerning the influence of characteristics of the disease are also contradictory. Previous studies argued that the diagnostic phase, period of treatment, and deterioration from the disease were especially difficult times for chil-

ed variables, but rather the effects these variables had on parents functioning

dren.7,10 Another study among chronically ill parents asserted that not illness-relat-

affected the prevalence of problems in children.11 Studies of parents with chronic


83

somatic illnesses have shown that children tend to have more problems when the

physical functioning of the parent is worse.12 The extent to which the physical functhat have considered the psychological functioning of parents with cancer have

tioning of parents with cancer influences children is not known. The few studies yielded inconsistent results. One study found more problems in the parent-child

mental health of parents had no impact on child functioning.14 Differences in since diagnosis, use of different informants) in these studies might have caused these differences. Furthermore, the literature among the general population sug-

relationship when parents experienced mental problems.13 Another study found that

methodology (e.g., a focus on only mothers or on both parents, the type of and time

gests that the functioning of mothers has a different impact on children than the functioning of fathers.15 It has also been suggested that the functioning of parents may have different effects on sons versus daughters.16,17 For example, mothers psydaughters, whereas it was negatively related to educational attainment in sons.18

chological distress was found to be positively related to depressive disorders in

The importance of a well-functioning parent to the well being of the children increases in families in which a parent has cancer. The well functioning parent can tional support for the children. The increased number of tasks and responsibilities ensure that daily life continues to be as normal as possible and can provide emoexpected of the partner combines with the anxiety and concern for the parent who

is ill, placing a particularly heavy burden on the partner. Little research has been

conducted concerning the impact that the health of the partner can have on children in such situations. A meta-analysis of the impact of psychopathology among parchildren further increases when both parents are ill.15 ents on the functioning of their children did show that the risk of problems for the Additional clarity is needed on the influence of various characteristics of parents who have been diagnosed with cancer, as well as the influence of their partners, on the prevalence of problems in children. The goal of the present study was to examine the impact of ill parents demographics, illness-related variables, as well as

physical and mental functioning, on the prevalence of emotional and behavioural


84

problems in children. The current study also examines whether the physical and children over and above ill parents characteristics.

mental functioning of spouses contributed to the prediction of reported problems in

Methods
Sample

The physician or oncology nurse introduced the study by offering written information to all eligible patients consecutively hospitalized or visiting the outpatient clinic for treatment or regular check up at the University Medical Centre Groningen between January 2001 and February 2003. Patients were eligible if diagnosed with cancer between 1-5 years before study entry. Children had to be in the age range of 4-18 and residing with the ill parent at the time of diagnosis. After informed consent was obtained, according to the regulations of the Medical Ethical Committee of the University Medical Centre Groningen, questionnaires and prepaid return envelopes were sent to the families.

The total number of families informed of the study was 476. Of these, 271 families did not agree to participate (response rate, 43%). Nineteen families were excluded from analyses, because the children were > 18 years of age while participating in the study. The sample was 180 parents diagnosed with cancer and 145 spouses. Ill parents reported on 114 primary school and 222 adolescent children. Information on socio-demographic and illness-related variables are displayed in Table 1.

The main reasons for parents declining to participate were that parents or children were too emotionally distressed; they did not want to stir up emotions again; parents expected that the effects of cancer on the children would be small, so consequently children were not informed by them; or parents mentioned that other affairs needed attention (such as school, work or illnesses of others). Parents who particisis from those who declined.

pated in the study did not differ in gender, type of diagnosis and time since diagno-

85

Table I: Socio-demographics and illness-related variables Ill parents characteristics N Fathers 34 Mothers 146 Mean age = 44.3 yrs (SD = 5.1, range = 32.8 to 57.8 yrs) Partner (Yes) Number of children in a family 1 child 2 children 3 children 4 children Highest educational level completed Primary school or lower vocational level Lower general secondary education Secondary vocational education Higher general secondary or pre-university education Professional education or university degree 167 % 19 81 Participating spouses N Fathers 113 Mothers 32 Mean age = 44.8 yrs (SD = 7.3, range = 31.3 to 65.6 yrs) Participating primary school children Sons Daughters Mean age = 8.8 yrs (SD = 2.1, range 4-11 yrs) 59 55 52 48 % 78 22

93

28 90 45 17

16 50 25 9

54 28 43 18 37

30 16 24 10 20

Position of primary school children in the family Only child 8 7 Youngest 59 52 Middle 10 9 Oldest 37 33 Participating adolescent children Sons Daughters Mean age = 15.5 yrs (SD = 2.0, range 12-18 yrs) 105 117 47 53

Illness-related variables Type of cancer Breast Gynaecological Dermatologic Haematological Soft tissue / bone Urologic Gastrointestinal Others Time since diagnosis Mean = 2.7 yrs, SD = 1.2 Treatment modality Non-intensive Intensive Recurrence (Yes) Complications (Yes)

92 22 19 16 10 8 4 9

51 12 11 9 6 4 2 5

Position of adolescent children in the family Only child 20 Youngest 79 Middle 40 Oldest 83

9 36 18 37

30 150 35 33

17 83 19 19

86

Instruments

Socio-demographics. Parents provided information on age, gender, highest educain the family as well as on age and gender of each child.

tional level completed, marital status, number of children, and position of the child

Illness-related variables. Information was obtained from patients on length of time complications of treatment experienced.

since diagnosis, treatment modalities received, experience of recurrent disease, and

pleted the subscales physical functioning (10 items) and mental health (5 items) of tioning scale contains questions about daily activities, such as walking a kilometre, stair climbing, and housekeeping. Mental health was measured with questions prescribed formula and ranged from 0 to 100. Higher scores indicate better health. about depressive feelings and nervousness. Scores were transformed following the Good reliability and validity of the RAND-36 have been replicated for the Dutch (range 0.87 to 0.90). the RAND-36 (a variant of the SF-36 of Ware & Sherbourne).19,20 The physical func-

General health related quality of life (QoL) of parents. Ill parents and spouses com-

translation.20 Cronbach alphas in the present study were high on both subscales

Emotional and behavioural functioning of children. Parents diagnosed with cancer ioural functioning of children over the preceding 6 months.21,22 The CBCL consists

completed the Child Behaviour Checklist (CBCL) to assess emotional and behavof 120 items and has 3 response options (0 = not true, 1 = somewhat or sometimes true, 2 = very true or often true). The internalizing, externalizing, and total behavioural problem scales were used for the present study. The internalizing problem syndromes of withdrawal, somatic complaints, and anxiety-depression. The exter-

scale (32 items) represents the internal mental state of children and measures the nalizing scale (30 items) reflects socially unacceptable behaviour and measures the syndromes of delinquent and aggressive behaviour. The total problem scale repre87

sents the total score derived from the sum of all items and consists of internalizing problems, as well as other problems. Higher scores denote more problems. The reliability and validity of the CBCL has been supported in a wide number of intering, externalizing, and total problem scales ranged from 0.84 to 0.94. Statistical analyses national and national studies. In the present study, Cronbach alphas for internaliz-

and externalizing problems, complemented with thought, social, and attention

Chi-square tests and independent Student t-test were computed to examine differences in demographic and illness-related variables between parents participating in the study and those that did not. Univariate analyses were conducted to determine relations (Pearson correlation analy-

ses) and differences (independent Student t-test) between childrens problems and ill parents socio-demographics, illness-related variables and both parents physical functioning and mental health. Analyses of variance (ANOVAs) were used to examine the

effect of the position of the child in the family on prevalence of problems. Categories used were oldest, middle, youngest, and only child. The Games Howell procedure was used for post hoc pair-wise comparisons because of unequal sample sizes and variances. A dichotomous variable was created for treatment received by parents consist-

ing of surgery only (non-intensive treatment) versus all other treatment (intensive

treatment regimens). This classification was made based on the clinical experience of effects of treatment were the main factors leading to this classification. It might be expected that intensive treatment regimens might have a longer duration and cause more visible side effects in parents than in cases of surgery only and, therefore, would have a more disruptive effect on family life. Complications were defined as: subjecevents that had impact on the subsequent treatment schedule or recovery after thera-

the physicians who participated in this project. Duration of therapy and visible side

tive or objective serious side-effects of treatment or unexpected treatment-related py. Complications experienced were categorized as: fever and infections (36%), seri-

ous side effects of treatment (e.g., serious mucositis, 21%), hemodynamic events
88

(12%), thrombosis (12%), postoperative bleeding (3%), neurologic events (e.g., ents who experienced treatment-related complications versus those who did not.

epilepsia, 9%), and pneumothorax (6%). A dichotomous variable was created for parHierarchical multiple regression analyses were performed to assess the relative contribution of several characteristics of parents to the prediction of problems reported in children. Only variables that were significant in the univariate analyses were entered into the model. Because of the stable nature of demographic variables, these variables were entered first. The position of the child in the family is a categorical variable and was, therefore, transformed into dummy variable(s). In regres-

sion analysis, there must be a reference category (omitted category), which is children. The following dummy variables were created: oldest child versus the rest; middle child versus the rest; and only child versus the rest. The severity of illness may affect the physical functioning and mental health of parusually the largest subgroup.23 In this study, that was the subgroup of youngest

ents. Therefore, illness-related variables were entered into the second step. The

third step comprised variables measuring physical functioning and mental health of ill parents. The impact of ill parents functioning may depend on parents gender, and it may have a different effect for sons and daughters. Therefore, two-way-inter-

action terms (using standardized scores) were computed between the 2 QoL dimenaction accounted for a unique significant effect it was included in the model. The fourth step contained the physical functioning and mental health of spouses.

sions of ill parents and gender of parents and children. Only when the 2-way inter-

Analyses were conducted separately for primary school (4-11 yrs) age and adolesdifferences between these groups.

cent children (12-18 yrs), because of the cognitive and emotional developmental

89

Results
Univariate analyses

reported in primary school children and the more externalizing problems in adoles-

Socio-demographics. The younger the ill parent, the more total problems were

cent children. Ill fathers and ill mothers did not significantly differ in their perceptions of childrens functioning. Educational level of ill parents was not significantly related to problems reported in primary school age and adolescent children (Table 2). Family characteristics. Single parents and parents from smaller families reported children than those who had a partner and those who had more children. This was not

significantly more internalizing, externalizing, and total problems in primary school found for adolescents. Analyses of variance indicated that the position of the child

Table 2: Relations and differences between ill parents reported problems in children and sociodemographics and illness-related variables Ill parent Family characteristics Illness-related variables Recurrent disease (no/yes) t - 0.0 - 0.9 2.2a 0.9 1.7 Complications (no/yes) t 0.0 0.2 0.3 3.6c 3.2b 3.8c

Educational level

Single /two parents

Time since diagnosis r 0.01 0.05 0.00 0.01 0.03


a,d

Number of children

Age

r Primary school children Internalizing Externalizing Total Problems Adolescents Internalizing Externalizing Total Problems
a d

t - 0.6 0.1

r 0.19 -0.10 0.03

t - 6.1b - 2.4
a

r -0.30c -0.33
c

F 4.8b 2.0 3.3a,d

-0.18 -0.16

2.0b - 0.9 0.5 1.6

-0.20a - 0.3

- 5.2c

-0.37c

-0.05 -0.14
a

- 0.7 -1.4 -1.3

-0.02 0.00 0.00

- 0.3 - 0.3 - 0.6

-0.10 0.00 -0.08

3.2a,d 1.2 3.1

1.2 0.7 0.8

-0.12

0.00

p 0.05; b p 0.01; c p 0.001 pair-wise comparison of groups was not significant

90

Treatment modality t

Gender

Position

within the family affected the prevalence of internalizing and total problems in primary school and adolescent children (Table 2). However, post hoc pair-wise comparisons showed only one significant difference: oldest primary school children were school children (p = 0.017). reported to have significantly more internalizing problems than youngest primary

Illness-related variables. Time since diagnosis was not significantly related to problems reported in children. Parents who had received a more intensive treatment reported more internalizing problems in primary school children than parents who nalizing problems in adolescent children than did parents who did not have recurrent disease. Parents suffering from treatment-related complications (such as infecchildren than parents who did not experience complications (Table 2).

had received only surgery. Parents who had recurrent disease reported more inter-

tions) reported more internalizing, externalizing, and total problems in adolescent

nificantly more internalizing (r = -0.29, p 0.001), externalizing (r = -0.17, p = 0.014), and total problems (r = -0.28, p 0.001) in adolescents. This was not found for primary school children. A decreased mental health in ill parents was significant-

QoL of parents. Ill parents who experienced more physical limitations reported sig-

externalizing (r = -0.23, p = 0.015; r = -0.19, p = 0.005), and total problems (r = respectively. -0.29, p 0.003; r = -0.27, p 0.001) in primary school and adolescent children,

ly associated with more internalizing (r = -0.33, p 0.001; r = -0.27, p 0.001),

Physical limitations experienced by spouses were significantly related to external-

izing problems ill parents reported in primary school children (r = -0.22, p = 0.05).

A decreased mental health in spouses was significantly related to internalizing (r = p 0.001) reported in adolescent children.

-0.39, p 0.001), externalizing (r = -0.19, p = 0.009), and total problems (r = -0.34,

91

Hierarchical regression analyses

Six separate hierarchical multiple regression analyses were performed to examine the contribution of potential predictors in the explanation of internalizing, external(Table 3) and adolescent children (Table 4). izing, and total problems reported by ill parents regarding primary school age

Table 3: Summary of hierarchic regression analyses of variables predicting ill parents reports of primary school childrens problems Internalizing problems Beta Step 1 Gender child Age parent Number of children Single child vs others Middle child vs others Oldest child vs others -0.14 0.00 -0.05 -0.25b R2 0.35 R2Ch FCh 10.8
c

Externalizing problems Bd 0.97 -2.69b -.e R2 0.11 R2Ch FCh 4.6


a

Total problems Beta 0.00 -0.26b -0.43c R2 0.31 R2Ch FCh 15.2c

One/two-parent families -0.48c Step 2 Intensity treatment Step 3 Physical funct Mental health Physical functioning x gender child Step 4 Physical functioning Spouse
a d

0.36 -0.12 0.38 -0.17 -

0.01

2.0 0.22 -2.38 -1.05 3.30a 0.27 0.05 6.0a


a

0.02

3.7

0.12

3.7a -0.16 -

0.33

0.02

3.7a

-1.37a

p 0.05; b p 0.01; c p 0.001 Because of the interaction term entered in this model standardized scores were used and Bs were presented

instead of betas. e Variable is constant. Dashes indicate that variables were not entered into the model, because no significant effects were found in univariate analyses.

92

Internalizing problems

Primary school children. A significant percentage of the variance in reported internalizing problems was explained by demographic variables. Ill parents observed more internalizing problems in primary school children when they were the oldest child and when the ill parents were single parents. Intensity of treatment entered in the second step did not account for a significant increment in the variance explained. The mental health of ill parents entered in the third step also did not account for a significant increment in the variance explained (Table 3). Adolescents. Illness-related variables explained a significant percentage of variance,

but only the experience of complications had a significant independent effect.

Physical functioning and mental health of ill parents and mental health of spouses,

Table 4: Summary of hierarchic regression analyses of variables predicting ill parents reports of adolescents problems Internalizing problems Beta Step 1 Age parent Number of children Step 2 Recurrent disease Complications Step 3 Phys. funct. parent Mental health parent Step 4 Mental health spouse
a

Externalizing problems Beta -0.11 R2 0.01 R2Ch FCh 2.1 0.05 0.18a 0.03 5.7a 0.23b 0.07 -0.10 -0.12 0.03 2.6 -0.18
a

Total problems Beta R2 R2Ch FCh

R2

R2Ch

FCh

0.05 0.10 0.20b 0.14 -0.16


a

4.5a

0.05

9.2b

0.08

8.1c

0.14 -0.19a

0.08

8.0c

-0.21b 0.22 -0.32


c

0.08 17.0c -0.14

0.09

0.02

3.0 -0.25
c

0.19

0.06 11.4c

p 0.05; b p 0.01; c p 0.001

Dashes indicate that variables were not entered into the model, because no significant effects were found in univariate analyses.

93

included in the second and third steps, appeared to account for significant increments in explained variance of internalizing problems in adolescent children (Table 4). Interaction terms failed to contribute significantly to the prediction of internalizing problems in primary school and adolescent children. This indicated that the impact of physical functioning and mental health of the ill parent differed neither for

fathers and mothers nor for sons and daughters. The final model accounted for 38% adolescent children.

of the variance of internalizing problems in primary school children and 22% in

Externalizing problems

centage of variance explained in reported externalizing problems. Only the number of children in a family appeared to have a significant independent negative effect. The next step showed a significant contribution to ill parents QoL; physical func-

Primary school children. Demographic variables accounted for a significant per-

tioning appeared to have a significant unique effect. A significant interaction effect for physical functioning in ill parents and the gender of primary school children was found. This indicated that a worsened physical functioning in ill parents was predictive of more externalizing problems in primary school daughters, whereas this was not the case for sons. The final step showed that physical functioning of spouses accounted for a significant increment in the variance explained (Table 3).

Adolescents. Demographic variables did not account for a significant percentage in two, had a significant and individual effect. The third step showed that physical

the variance explained. The experience of complications in parents, entered in step functioning and mental health of ill parents did not predict a significant percentage of additional variance, nor did mental health of spouses entered in the fourth step. None of the interaction terms accounted for a significant contribution in adolescents externalizing problems (Table 4).

The variables entered into the model explained 27% of variance of externalizing problems in primary school age and 9% in adolescent children.

94

Total problems

Primary school children. A significant percentage of variance in total problems was explained by demographic variables. Number of children in a family and having one or two parents appeared to have significant unique effects. Mental functioning of ill variance explained, but it did not have a significant independent effect (Table 3).

parents, entered into the second step, accounted for a significant increment in the Adolescents. Complications experienced by parents contributed significantly to the mental health of ill parents also appeared to have significant independent effects, as did mental health of spouses (Table 4).

prediction of total problems. The next step showed that physical functioning and

Again, no effects were found for the interaction terms in predicting total problems in primary school age and adolescent children. The overall model accounted for 33% of variance in total problems of primary school and 19% of adolescent children.

Discussion
This study is one of the first to investigate impact of parental characteristics on emotional and behavioural problems in children who have a parent diagnosed with cancer. Attention was paid to ill parents socio-demographic and illness-related

variables, and to physical functioning and mental health of both parents. We used ill parents reports for information on prevalence of problems in children. Analyses were performed for primary school age and adolescent children.

First, the results demonstrated that age, gender, and educational level of ill parents had no noticeable effects on problems reported in children. Information on the impact of parents age and educational level on childrens functioning in this situation is lacking. Studies that paid attention to parents gender found that adolescent in the current study were mothers (81%), which may have prevented us from finding a gender effect. daughters of ill mothers experienced more problems.2,14,24 The majority of ill parents

95

The family setting appeared to be an important predictor for problems in primary lems when living with a single parent, a finding also described previously.25 Also,

school children. Our study shows that children were reported to have more probmore problems in primary school children were reported when they were the old-

est child. Children of this age are dependent on their parents in terms of daily care and attention. If a parent diagnosed with cancer is the sole caregiver or there are only younger siblings to share experiences and feelings with, the life of children

may be more disrupted. Another explanation could be that single parents worry more about their children because they are the sole caregiver, and they are, therefore, more alert to disruptive behaviour. Remarkably, reported problems in adolescents were not associated with family setting. Adolescents may be more capable of from which to receive support.26

managing their own lives, and, in general, they have more extensive peer groups Second, illness-related variables seemed to have an impact on problems parents More specifically, parents who experienced complications during treatment per-

reported in adolescents, whereas this was not found for primary school children. ceived more problems in adolescent children than did parents without complica-

tions, which is in contrast to an earlier study.27 The relatively small sample (n = 32) might have contributed to discrepancies in research findings. The vulnerability of make them aware of the pain and physical discomfort parents may experience.

and the larger age range (several months to 16 years) in the above-mentioned study adolescents in this situation may be caused by their empathic capabilities, which Time since diagnosis, treatment modalities, and recurrent disease did not seem to affect problems reported in children. These findings were in accord with those of others.14,27,28 Previous studies have suggested that the degree to which the child per-

ceives the situation to be serious and stressful, and not objective characteristics of

nation is that characteristics of the illness no longer play an important role 1-5 years following diagnosis. This explanations, however, is in direct contrast to the effect that was found for complications. Relatively few patients received non-intensive
96

the illness itself, determine whether problems will arise.29,30 Another possible expla-

treatment (surgery only) or experienced relapse. It is possible that the groups were too small to reveal any significant differences. Third, the impact on childrens functioning of an ill parents physical and mental functioning was examined. Ill parents who experienced physical limitations perceived more externalizing problems in primary school daughters (not sons). A study in the parent was a predictor for primary school childrens externalizing problems.12,31 However, the last study did not examine effects on sons and daughters separately. In an earlier study, we found a higher percentage of primary school daugh-

among a community sample also demonstrated that worsened physical functioning

ters scoring in the clinical range on externalizing problems than the percentage externalizing problems, especially when the parent experiences physical limitafound in the norm group.24 It may be that daughters at this age are vulnerable to

tions. Physical limitations in ill parents also predicted the internalizing and total tion to the relation between a parents physical functioning and prevalence of problems in children have also considered physical functioning as an important predictor of various problems in children.12,31

problems that were reported in adolescent children. The few studies that paid atten-

Ill parents who had a worsened mental health were more likely to report internalizing and total problems in their adolescent children, which replicates findings of earlier studies.32,33 The current study, however, did not find a relation to externalizing

problems which is in contrast to findings from those earlier studies. One study sugin children exist, whereas depression in combination with a conflicting environ-

gested that a direct relation between parents depression and internalizing problems ment may heighten the risk for externalizing problems.34 Perhaps the diagnosis of

cancer produces a situation in which family members tend to behave in socially

desirable ways, or in which family members become closer to each, rather than a

situation in which conflicts tend to arise. Two factors can play a role in the relation children could actually have more problems as a result of their exposure to depres-

between mental functioning of parents and problems in children. On the one hand, sive behaviour of parents.35 Conversely, mental problems on the part of the parent
97

could also lead to a tendency to over report problems in children.36,37

A further exploration, using interaction-effects, showed that ill parents physical and mental functioning had similar effects on sons and daughters. One exception was found, as described earlier: physical functioning of ill parents affected primary ference between the impact of the functioning of fathers on the prevention of problems in their children and that of mothers. These results were not in line with results

school daughters but not primary school sons. The current study also found no dif-

of previous work that problems in children increased when fathers had a worsened physical functioning, whereas no relation was found for that of mothers.32 parents. Some studies indicated that childrens functioning was equally associated with fathers and mothers mental functioning.38,31,32 Two meta-analyses, however, Inconsistent results were also found in earlier studies concerning mental health of

indicated that problems in children (especially internalizing problems) were more strongly associated with depression in mothers than in fathers.15,39 Perhaps the differ-

ence between fathers and mothers can be explained primarily by the extent to which a child is exposed to the physical and mental limitations of parents. Because, in general, mothers continue to be more involved in care of children than are fathers, the burden on children could be greater when the mother experiences limitations in functioning than when the father experiences similar limitations. Nonetheless, a reorganization of roles can occur in a situation in which one parent has cancer, thus decreas-

ing the differences between mothers and fathers regarding the care of their children. The absence of a gender effect in such situations may also be attributable to the limof functioning of the parent on children could be generated by investigating this relavarious family factors, and the complex relations between these factors, into account. ited participation of fathers with cancer in this study. Further insight into the impact tion in the context within which it occurs. In other words, future studies should take Finally, findings support the notion that spouses functioning had an additional

effect on problems in children. These results are consistent with previous research among children of parents with cancer33 and parents with psychopathology.15,35,40 The impact of spouses functioning seemed to be similar to that of ill parents: worsened

98

physical functioning in spouses predicted externalizing problems in primary school children; spouses mental health was associated with internalizing and total problems in adolescents. Confrontation with depressive behaviour on the part of the

partner can also increase vulnerability of adolescents in such situations. In addition, the partner can have an important role in emotional support of adolescents, even can limit the ability of the partner to fulfil this role. compensating for the role of the ill parent when necessary. Impaired mental health

Results of the study are based on a large and heterogeneous research group. In addi-

tion, it considered not only functioning of the parent who has been diagnosed with cancer but also that of the partner. The influence of the parent variables can be age groups allow detection of such differences. dependent on the developmental phase of the children. Distinctions among various The number of single-parent families participating in the study was small. This may suggest that single parents more often decided not to participate in the study than parents who had a partner. The finding that children in single-parent families were reported to have more problems coupled with the small percentage of single-parpresent in the population of children of cancer patients. ents that participated may indicate that our results show less problems than may be The cross-sectional design of this study does not allow causative conclusions to be

drawn. The use of a longitudinal design will give insight into the direction of sta-

tistical relations and may, therefore, do more justice to bidirectional and dynamic

processes within families in this situation.35 Besides, pre-existing problems in chil-

dren may make an important contribution to current problems in children, but this was not examined. Unfortunately, studies like this do not have the capability of assessing whether problems existed before cancer diagnosis. Furthermore, such stressful situations may cloud the parents perception. Involving other informants, image of the functioning of the children. Potential predictors of childrens functioning and interaction between these factors have not received sufficient attention in

including teachers and the children themselves, could provide a more accurate

99

the literature. Therefore, more research exploring the mechanism that may play a role in the impact parents cancer has on children is needed in order to prevent problems. Such studies should consider not only characteristics of the family and relaparticularly those of adolescents. These characteristics require more attention.

tionships among family members but also the role of social network and age peers, Results of this study show that the impact of cancer on children depends upon family factors and the level of functioning of both parents. The diagnosis and treatment ical and mental problems. These problems can limit the ability of parents to fulfil help to balance the energy the illness demands with the care needed for the chilof cancer is often an enormous burden on the parent, and it can lead to many phystheir parental roles. Social workers could provide support for the parent and, thus, dren. Additional attention to this matter is needed for families in which the parent

experiences treatment complications or physical or emotional problems, as well as problems require particular attention.

for single-parent families. In addition, families in which the partner also has health

100

References
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. Visser A, Huizinga GA, Van der Graaf WTA, Hoekstra HJ, Hoekstra-Weebers JEHM. The impact of parental cancer on children and the family: a review of the literature. Cancer Treat Rev. 2004; 30: 683-694. Huizinga GA, Visser A, Van der Graaf WTA, Hoekstra HJ, Klip EC, Pras E, Hoekstra-Weebers JEHM. Stress response symptoms in adolescent and young adult children of parents diagnosed with cancer. Eur J Cancer. 2005; 41: 288-295. Serbin L, Karp J. The intergenerational transfer of psychosocial risk: mediators of vulnerability and resilience. Annu Rev Psychol. 2004; 55: 333-363. Kahn R, Brandt D, Whitaker R. Combined effect of mothers and fathers mental health symptoms on childrens behavioural and emotional well-being. Arch Pediatr Adolesc Med. 2004; 158: 721-729. Taanila A, Ebeling H, Kotimaa A, Moilanen I, Jrvelin MR. Is a large family a protective factor against behavioural and emotional problems at the age of 8 years? Acta Paediatr. 2004; 93: 508-517. Spruijt E, Goede Md, Vandervalk I. The well-being of youngsters coming from six different family types. Patient Educ Couns. 2001; 45: 285-294. Hilton BA, Elfert H. Childrens experiences with mothers early breast cancer. Cancer Pract. 1996; 4: 96-104. Gates MF, Lackey NR. Youngsters caring for adults with cancer. Image J Nurs Sch. 1998; 30: 11-15.

Vess J, Moreland J, Schwebel A. An empirical assessment of the effects of cancer on family role functioning. J Psychosoc Oncol. 1985; 3: 1-16. Helseth S, Ulfsaet N. Having a parent with cancer: coping and quality of life of children during serious illness in the family. Cancer Nurs. 2003; 26: 355-362. Steele RG, Forehand R, Armistead L. The role of family processes and coping strategies in the relationship between parental chronic illness and childhood internalizing problems. J Abnorm.Child Psychol. 1997; 25: 83-94. Stein A, Newcomb MD. Childrens internalizing and externalizing behaviours and maternal health problems. J Pediatr Psychol. 1994; 19: 571-594. Lewis FM, Hammond MA. The fathers, mothers, and adolescents functioning with breast cancer. Fam Relat. 1996; 45: 456-465. Compas BE, Worsham NL, Epping-Jordan JE, Grant KE, Mireault G, Howell DC, et al. When mom or dad has cancer: markers of psychological distress in cancer patients, spouses, and children. Health Psychol. 1994; 13: 507-515. Connell AM, Goodman SH. The association between psychopathology in fathers versus mothers and childrens internalizing and externalizing behaviour problems: a meta-analysis. Psychol Bull. 2002; 128: 746-773. Davies T, Lindsay LL. Interparental conflict and adolescent adjustment: why does gender moderate early adolescent vulnerability? J Fam Psychol. 2004; 18: 160-170. Leadbeater BJ, Kuperminc GP, Blatt SJ, Hertzog C. A multivariate model of gender differences in adolescents internalizing and externalizing problems. Dev Psychol. 1999; 35: 1268-1282. Ensminger ME, Hanson SG, Riley AW, Juon HS. Maternal psychological distress: adult sons and daughters mental health and educational attainment. J Am Acad Child Adolesc Psychiatry. 2003; 42: 1108-1115.

101

nal dstress. Fam Process. 2001; 40: 163-172. 38. Dierker LC, Merikangas KR, Szatmari P. Influence of parental concordance for psychiatric disorders on psychopathology in offspring. J Am Acad Child Adolesc Psychiatry. 1999; 38: 280-288. 39. Beardslee R, Versage EM, Gladstone TR. Children of affectively ill parents: a review of the past 10 years. J Am Acad Child Adolesc Psychiatry. 1998; 37: 1134-1141. 40. Tannenbaum L, Forehand R. Maternal depressive mood: the role of the father in preventing adolescent problem behaviours. Behav Res Ther. 1994; 32: 321-325.

19. Hays RD, Sherbourne C, Mazel RM. The RAND-36-item Health Survey 1.0. Health Econ. 1993; 2: 217-227. 20. Van der Zee, K, Sanderman R. The measurement of generic health with the RAND-36. A manual. Northern Centre for healthcare Research, University of Groningen, 1993. 21. Achenbach TM. Manual for the Child Behaviour Checklist/4-18 and 1991 Profiles. Burlington, VT: University of Vermont Department of Psychiatry, 1991. 22. Verhulst FC, Van der Ende J, Koot HM. Handleiding voor de Child Behaviour Checklist/4-18 [Manual for the Dutch version of the Child Behaviour Checklist/4-18]. Rotterdam: University of Rotterdam, Department of Child Psychiatry, 1996. 23. Kleinbaum DG, Kupper LL, Muller KE. Applied regression analysis and others multivariable methods. Boston: Duxbury Press, 1986. 24. Visser A, Huizinga GA, Hoekstra HJ, Van der Graaf WTA, Klip EC, Pras E, Hoekstra-Weebers JEHM. Emotional and behavioural functioning of children of a parent diagnosed with cancer: a crossinformant perspective. Psychooncology. 2005; 14: 746-758 25. Lewis FM, Zahlis EH, Shands ME, Sinsheimer JA, Hammond MA. The functioning of single women with breast cancer and their school- aged children. Cancer Pract. 1996; 4: 15-24. 26. Christ GH, Siegel K, Sperber D. Impact of parental terminal cancer on adolescents. Am J Orthopsychiatry. 1994; 64: 604-613. 27. Howes MJ, Hoke L, Winterbottom M, Delafield D. Psychosocial effects of breast cancer on the patients children. J Psychosoc Oncol. 1994; 12: 1-21. 28. Lewis FM, Woods NF, Hough EE, Bensley LS. The familys functioning with chronic illness in the mother: the spouses perspective. Soc Sci Med. 1989; 29: 1261-1269. 29. Lazarus RS, Folkman S. Stress, appraisal and coping. New York: Springer, 1984. 30. Compas BE, Worsham NL, Ey S, Howell DC. When mom or dad has cancer: II. Coping, cognitive appraisals, and psychological distress in children of cancer patients. Health Psychol. 1996; 15: 167-175. 31. Phares V, Compas BE. The role of fathers in child and adolescent psychopathology: make room for daddy. Psychol Bull. 1992; 111: 387-412. 32. Rodrigue JR, Houck GM. Parental health and adolescent behavioural adjustment. Child Health Care. 2001; 30: 79-91. 33. Lewis FM, Darby EL. Adolescent adjustment and maternal breast cancer: a test of the Faucet Hypothesis. J Psychosoc Oncol. 2003; 21: 81-104. 34. Downey G, Coyne JC. Children of depressed parents: an integrative review. Psychol Bull. 1990; 108: 50-76. 35. Goodman H, Gotlib IH. Risk for psychopathology in the children of depressed mothers: a developmental model for understanding mechanisms of transmission. Psychol Rev. 1999; 106: 458-490. 36. Krain L, Kendall P. The role of parental emotional distress in parent report of child anxiety. J Clin Child Psychol. 2000; 29: 328-335. 37. Kinsman AM, Wildman BG. Mother and child perceptions of child functioning: relationship to mater-

102

103

104

Temperament as a predictor of internalizing and externalizing problems in adolescent children of parents diagnosed with cancer

Harald J Hoekstra, Winette TA van der Graaf, Josette EHM. Hoekstra-Weebers

Annemieke Visser, Gea A Huizinga,

Supportive Care in Cancer 2006 August, 30 (Epub ahead of print)

105

Abstract

izing and externalizing problems among children of parents diagnosed with cancer,

Objective. This study examined the relationship between temperament and internal-

beyond the effects of socio-demographics, illness-related variables, and life events. Materials and methods. Three hundred and forty adolescent children and their 212 Youth Self Report and the Child Behaviour Checklist, respectively. Children comDaughters of parents with cancer were reported as having more internalizing probpleted also the Early Adolescent Temperament Questionnaire. Main results.

parents diagnosed with cancer participated. Children and parents completed the

lems than their counterparts did. Prevalence of problems did not depend on chil-

drens and parents age or educational level. Recurrent disease and number of life events experienced by children and parents affected the problems reported. The in children were shyness and fear/worry, to a lesser extent, frustration and percepmost important temperament dimensions in the prediction of internalizing problems tual sensitivity (children only) and lower scores on pleasure intensity (parents only). Externalizing problems were associated with effortful control and in childrens reports with frustration. Temperament seemed to be a more important pre-

dictor of problems reported by children than parents. Conclusion. Findings suggest

that temperament is useful in determining the relative vulnerability of children of parents who have been diagnosed with cancer. Social workers may help parents to recognise individual differences between children and to support children by using techniques that are compatible with the temperament of children.

106

Introduction
All people are confronted with intense experiences during the course of their lifetime. Even children are not spared from them. Adolescent children, especially, are vulnerable to stressful events.1 One stressful incident with which families with ado-

lescent children are increasingly confronted is the diagnosis of cancer in one of the ters particularly, experience more emotional problems than their age peers do.2,3 attention, research into risk and resilience factors is still in its infancy. Recent stud-

parents. Studies have shown that adolescent children in such situations, the daughAlthough the functioning of children of cancer patients has received increasing ies found evidence for the impact of parental characteristics,4 parent-child commu-

to the different effect of parental cancer for boys and girls, and children of different ages, but research into other child characteristics is lacking. Temperament is one of the factors that may have an important influence on how children cope with stressful events, and therefore why some children are more vulnerable to the development of problems than others.7 The primary hypothesis of most models of tem-

nication5 and family functioning6 on child functioning. Increased attention is given

perament is that specific dimensions are related to prevalence of specific problems. Research into the role of temperament can provide insight into which children are at greater risk in such situations and which children are better equipped to handle the situation.

The developmental model of temperament as proposed by Rothbart and Derryberry is one of the few models that offer a measurement tool that is specifically designed proposes that temperament is relatively stable, but that the expression of temperament can change as a result of maturation and (social) environmental influences, and self-regulation. Reactivity is comprised of the physical and emotional differences that exist among individuals in reaction to stress. Children who have a high for use with adolescents, and it was therefore used in the present study. Rothbart

including life events.8,9 The model distinguishes between two concepts: reactivity

degree of reactivity are more easily upset and need more time to recover than other
107

children do. The term self-regulating system refers to such processes as attention, ulation of emotions and behaviour. These children may be able to direct attention away from the stressor, such as parental cancer.

activation, and inhibition. These processes have an important influence on the reg-

To our knowledge, no studies have reported on the effect of different temperament dimensions on the functioning of children of parents who have been diagnosed with cancer. The aim of the current study was to investigate the effects of temperament demographics (age, gender and educational level), illness-related variables (recurchildren experienced during the year before assessment.

on prevalence of problems in these children, beyond the possible effects of sociorent disease and time since diagnosis) and the number of negative life events that

Materials and methods


Procedure

Between January 2001 and February 2003, written information on this study was offered to all cancer patients who were consecutively hospitalised or who visited the outpatients clinic at the University Medical Centre Groningen by their physi-

cians or oncology nurses. In addition, information was sent to patients and their family members who had contacted the researchers in response to media attention had been diagnosed between 1 to 5 years before study entry and if they had chilbecause they wished to participate in the study. Families were eligible if patients dren between 4 to 18 years of age. Participants had to be fluent in Dutch. Patients

discussed study participation with their partners (if present) and children. Informed

consent was obtained according to the regulations of the Medical Ethical Committee of the University Medical Centre Groningen. After informed consent was received, questionnaires and prepaid return envelopes were provided. will be described completely anonymous.
108

Participants were guaranteed that answers were treated absolutely confident and

Participants

The current study is part of a larger study in which 476 families with children aged 4 to 18 years were approached, and information was mailed to 110 families who the first group and 89 families from the second group consented to participate (response: 43 and 81%, respectively). Ill parents who did not participate did not sigand time since diagnosis. had contacted us for information about participation. Of these, 205 families from

nificantly differ from those who participated with respect to gender, tumour type, The current study focused on the responses of children of 11 years of age and older, as they completed the questionnaires themselves. The sample for the present study consisted of 340 adolescent children (149 sons and 191 daughters), between the (80% mothers, mean age = 45.4 years, SD = 4.7). Twenty-one percent of the chil-

ages of 11 and 18 years (mean age = 14.9 years, SD = 2.3) and their 212 ill parents dren were receiving education in primary school, 9% at the lower vocational level, 17% in lower general secondary education, 12% in intermediate vocational education, 38% in high school, and 3% in higher vocational education or university.

Nine percent of the children were from single-parent families. Parents (43 fathers; types of cancer: breast (55%), haematological (9%), skin (9%), gynaecological

169 mothers, mean age = 45.4 years, SD = 4.7) had been diagnosed with various (9%), urological (5%), bone tumours (4%), gastrointestinal (5%) or other cancers, mean time since diagnosis was 2.6 years (SD = 1.2). Twenty-nine percent of the parents had suffered relapses. Children and parents approached in the hospital did not significantly differ in age

such as cancer of the central nervous system or head and neck cancer (6%). The

or gender from children and parents who had volunteered for participation. 5.8, p .001) than that of the first-named group, but this was not found for chil-

Educational level of parents of the last-named group was significantly higher (t = drens educational level. Furthermore, children and parents in both groups reported similar levels of internalizing and externalizing problems, and children did not differ significantly in temperament.

109

Measures

Revised Early Adolescent Temperament Questionnaire (EATQ-R) of Rothbart and

Temperament. Temperament was measured using the adolescent version of the

Derryberry.10-12 The EATQ-R consists of 53 items and includes ten subscales that are control, high intensity pleasure, shyness, fear, frustration, affiliation, perceptual sensitivity, and pleasure sensitivity. Answers were rated on a 5-point Likert-type scale (1 = almost always untrue to 5 = almost always true). Higher values represent a higher availability of the temperamental dimension concerned. The psychometric quality of the EATQ-R was reported to be sufficient among American adolescents with Cronbachs alphas ranging from = 0.55 to = 0.78.13 In the correlations from r = 0.11 to r = 0.37. Five of the ten dimensions had mean inter-

designed to measure temperamental attention control, activation control, inhibitory

present study, alpha scores ranged from = 0.36 to = 0.74, and mean inter-item

item correlations of < 0.20. Therefore, factor analyses were carried out in order to

study the extent to which the temperamental dimensions identified by Rothbart and colleagues emerged from the data of the present Dutch study (see preliminary analyses).

Internalizing and externalizing problems. The Youth Self-Report (YSR) and Child Behaviour Checklist (CBCL) were used to assess, respectively, adolescent childrens self-reported and parents reported behavioural and emotional problems in options (0 = not true, 1 = somewhat or sometimes true, 2 = very true or often true). Higher scores indicated more problems. The Internalizing (TIS) and Externalizing scale (32 items) consists of the syndrome subscales withdrawal, somatic complaints Behaviour Problem Scales (TES) were used in the present study. The internalizing and anxiety/depression. The externalizing scale (30 items) consists of the syndrome subscales delinquent behaviour and aggressive behaviour. The YSR is one of the scores in the present study were high ( > 0.80). The manual provides norm data
110

children.14-16 The YSR/CBCL consists of 102/120 problem items with three response

most commonly used questionnaires in adolescent research. Cronbachs alpha

based on a randomly selected Dutch sample of 560 adolescent boys and 564 adolescent girls.

Life events. The Questionnaire of Recently Experienced Events was used to ask

children and parents about the number of life events experienced during the past developed by Rahe.17 Questions measuring negative events (14 items; e.g., divorce, illness by other family members than the parent with cancer) were used.

year. This questionnaire is based on the Recent Life Change Questionnaire (RLCQ)

Analyses

Factor analyses of the EATQ-R were executed using simultaneous confirmatory analysis (SCA) and exploratory principal component analysis (PCA) to investigate factor validity.

Chi-square and t-tests were performed to compare children and parents who had been recruited in the hospital and those who had volunteered for participation on demographic characteristics and the problems reported.

One-sample t-tests were performed to compare prevalence of internalizing and externalizing problems reported by children and parents diagnosed with cancer with those of the norm group.

Univariate statistics (t-tests and pearson correlation analyses) were performed to investigate effects of study variables (age, gender, educational level, time since diagnosis, recurrence, number of negative life events, and temperament) and problems in children. Hierarchical regression analyses were conducted to examine the contribution of in children. Socio-demographics and illness-related variables (first step), and num-

temperament dimensions to prediction of internalizing and externalizing problems ber of life events (second step) significantly related to temperament were entered into regression analyses to ensure that any effect found for temperament (third step) whether multicollinearity exists between the independent variables Pearson correon childrens problems would not be attributed to these variables. To examine

111

lation analyses and variance inflation factors (VIF) were performed. If the mean VIF is considerably larger than one and the largest VIF is greater than 10, multicollinearity exists.18

There is evidence that sons and daughters may respond differently to the cancer in (using standardised scores) were computed to examine whether the pattern of the relationship between temperament and prevalence of problems differ between sons effect was it included in the model. cance was set at p 0.01. and daughters. Only when the 2-way interaction accounted for a unique significant Owing to the large number of comparisons in relation to the sample size signifithe parent.19 and that they differ in temperament.13,20 Two-way-interaction terms

Results
Preliminary analyses

Simultaneous confirmatory analysis (SCA) was conducted to examine differences

in the percentage of variance explained by the original structure and by the exploratory structure over the same number of factors. The difference in the variance explained by the original structure (41.0%) and by the exploratory structure (46.1%) was considered too large (> 2% rule of thumb21), to continue with the origusing orthogonal rotation followed by varimax procedure. The number of constitut-

inal structure. An exploratory principal component analysis (PCA) was conducted ing factors was determined based on the scree-test. Items that loaded consistently low (< 0.30) or on varying components were excluded. In addition, all items particularly those with loadings between 0.30 and 0.40 were critically examined for the degree to which they formed a good reflection of the dimensions on which they loaded (face validity). Dimensions were assessed as consistent by an alpha above items were excluded, including the entire inhibition-control (five items) and the 0.60 and an inter-item correlation above 0.20. On the basis of these criteria, 17

112

affiliativeness scale (five items). Thirty-six items remained, distinguishing seven consists of all five items from activation control and four of the six items from temperamental dimensions. The first dimension was called effortful control and

attention control. One of the attention control items was excluded and the other was

of several things that are happening around me) that represent the second dimenteria and was excluded. The fear/worry dimension forms the fourth dimensions. sion. The third dimension is pleasure sensitivity, just one item did not meet the cri-

added to the original four perceptual sensitivity items (I am good at keeping track

Three of the six items loaded on this scale. One of the other items was excluded, who likes to speed), and one was added to the original four items of the shyness

one was added to pleasure intensity (I get frightened when I ride with a person scale (Some of the kids at school make me nervous ). Shyness was the fifth

dimension. The sixth dimensions form the pleasure intensity. Three of the six items loaded low or were not consistent, and were therefore excluded. The seventh dimension forms frustration. One of the seven items did not meet the criteria and

was excluded. The seven temperamental dimensions can be defined as follows. Effortful control measures the capacity to start and persist in an action and to focus attention. Pleasure sensitivity is the pleasure related to stimuli involving low intensity, complexity, novelty, and incongruity. Perceptual sensitivity pointed to detec-

tion or perceptual awareness of slight, low-intensity stimulation in the environment. Pleasure intensity represents the pleasure derived from activities involving high intensity. Shyness is the behavioural inhibition to novelty and challenge, especially social. Frustration measures the negative effect related to interruption of ongo-

ing tasks or goal blocking. Fear/worry represents the worry for occurrence of unpleasant situations. The seven factors that were formed of remaining items explained 47.2% of the variation. Alpha coefficients from the other dimensions to r = 0.48 (Table 1). ranged from = 0.61 to = 0.79, and inter-item correlations ranged from r = 0 .21 Cross validity. The same research group conducted a prospective study among children of parents recently diagnosed (1-16 weeks ago) with cancer using a similar

113

procedure in approaching families as was used in the current study. The adapted

structure was tested among 144 adolescent children (54% daughters; mean age = 14.2 years, SD = 2.3 years) that participated in the prospective study (comparison group). Results of PCA among children in the prospective study were similar to those found in the current study. The internal consistency and mean inter-item coravailable on request.

relations of the two groups were comparable (Table 1). Results of the PCA are

Internalizing and externalizing problems in children

T-tests revealed no significant differences in prevalence of internalizing and exter-

nalizing problems between sons and norm group boys as reported by themselves internalizing problems than found in norm group girls (Table 2).

and their ill parents. Daughters and their ill parents reported significantly more

Childrens problems and socio-demographics, illness-related variables, and life events Socio-demographics. Daughters experienced significantly more internalizing probt = 4.8, p 0.001). Neither childrens age and educational level, nor parents age,

lems than sons as was reported by ill parents and self-reports (t = 2.9, p = 0.004;

Table 1: Cronbachs alpha EATQ-R dimensions of children in the current study and a control study Dimensions Current study Cronbachs alpha Effortful control Pleasure sensitivity Shyness Frustration Perceptual sensitivity Pleasure intensity Fear/Worry 0.75 0.79 0.69 0.61 0.62 0.71 0.61 Mean inter-item correlations 0.26 0.36 0.30 0.21 0.25 0.37 0.34 Control Cronbachs Mean inter-item correlations alpha 0.74 0.83 0.63 0.69 0.70 0.74 0.61 0.25 0.56 0.25 0.27 0.31 0.42 0.35 Number of items 9 4 5 6 5 4 3

114

gender or educational level were significantly related to the prevalence of problems reported in children. Illness-related variables. Parents and children reported more internalizing probt = -3.2, p 0.001). Time since diagnosis was not significantly associated with

lems in case of recurrent disease than in case of primary disease (t = -3.3, p 0.001

reports of either internalizing or externalizing problems (even after controlling for recurrent disease). Life events. The number of negative life events children or parents experienced dur-

ing the preceding year was significantly positively correlated with internalizing (r = 0.36, p 0.001, r = 0.25, p 0.001, respectively) and externalizing problems (r = 0.25, p 0.001) reported by or in children.

Childrens temperament and socio-demographics, illness-related variables, and life events Socio-demographics. Sons had significantly higher mean scores on pleasure p = 0.005) than did daughters. Age of children was significantly negatively related to effortful control (r = -0.18, p 0.001). No relationship was found between childs educational level and temperament. intensity (t = 4.7, p 0.001) and significantly lower scores on shyness (t = -2.8,

time since diagnosis and childrens temperament. Children of parents who had

Illness-related variables. No significant relationships were found between length of

Table 2: Descriptive statistics of the YSR and one sample t-tests for comparison of study and norm group Sons Mean SD Internalizing probl YSR Externalizing probl YSR Internalizing probl CBCL Externalizing probl CBCL
a

Norm group boys Mean SD 8.6 11.5 5.7 7.1 5.8 6.7 5.7 7.1 t 1.6 -0.2 0.8 -0.9

Daughters Mean SD 13.9 10.8 8.4 5.9 9.9 6.2 7.2 6.0

Norm group girls Mean SD 10.8 10.0 6.5 5.5 7.1 6.1 6.0 5.8 t 4.3a 1.7 3.5a 0.9

9.6 11.4 6.1 6.6

7.3 6.2 6.2 6.1

p < 0.001

115

recurrent disease differed significantly in fear/worry from children of parents with primary disease (t = 3.1, p = 0.002). Life events. Number of negative life events experienced was significantly positive-

ly correlated with pleasure sensitivity (r = 0.15, p = 0.007), perceptual sensitivity (r = 0.15, p = 0.005) and fear/worry (r = 0.30, p 0.001), and negatively with effort-

ful control (r = -0.15, p = 0.008).

Relationships between temperament and problems in children

All temperament dimensions were significantly related to childrens self-reported

internalizing problems. Shyness, pleasure intensity and fear/worry were significantly related to internalizing problems as reported by ill parents. Effortful control, frusizing problems in children as reported by children and parents (Table 3). Predictors of internalizing and externalizing problems tration and fear/worry were significantly associated with the prevalence of external-

percentage of the variance in the prevalence of internalizing problems in childs

Internalizing problems. Childs gender and recurrent disease explained a significant

Table 3: Descriptive statistics of the temperament dimensions and correlations of these variables with internalizing and externalizing problems Children
Internalizing Externalizing

Ill parents
Internalizing Externalizing

Mean Temperament Effortful control Pleasure sensitivity Shyness Frustration Perceptual sensitivity Pleasure intensity Fear/worry
a

SD 6.2 4.0 3.8 3.8 3.5 3.7 2.7

r -0.25b 0.22b 0.30b 0.29b 0.15a -0.18b 0.46b

r -0.44b 0.03 0.00 0.36b 0.07 -0.02 0.29b

r -0.05 0.13 0.19b 0.10 -0.01 -0.19b 0.23b

r -0.24b -0.04 -0.13 0.14a -0.09 0.10 0.16a

27.8 11.2 12.0 18.4 16.4 14.7 7.4

p < 0.01; b p < 0.001

116

a significant independent effect. Life events accounted for a significant increment in explained variance in childs (R Ch = 0.10) and parents reports (R Ch = 0.07).
2 2

(R Ch = 0.10) and parents reports (R Ch = 0.08). Both variables appeared to have


2 2

The temperament dimensions predicted a significant percentage of additional variance in childrens (R Ch = 0.27) and parents reports (R Ch = 0.04). Shyness, frus2 2

tration, perceptual sensitivity and fear/worry appeared to have a significant positive independent effect in childrens reports. None of the variables appeared to have a significant independent effect in parents reports. Interaction terms between temperament dimensions and childs gender failed to contribute significantly to the pre-

diction of internalizing problems in children and parents reports. The final models accounted for 48 and 19% of the variance in childs and parents models, respec-

Table 4: Regression analyses examining temperament as a predictor of internalizing problems reported by children and ill parents Children Internalizing problems Beta Step 1 Childs gender Recurrent disease Step 2 Negative life events Step 3 Effortful control Pleasure sensitivity Shyness Frustration Perceptual sensitivity Pleasure intensity Fear / worry
a

Ill parents Internalizing problems FCh 17.1


b

R2 0.10

R2Ch

Beta -0.18b 0.22b

R2 0.08

R2Ch

FCh 13.5b

-0.25b 0.20
b

0.20 0.33
b

0.10

40.6b 0.26
b

0.15

0.07

24.4b

0.48 -0.10 0.09 0.31b 0.15b 0.12a 0.02 0.25b

0.27

22.8b 0.10 -0.09 0.13

0.19

0.04

5.3b

p < 0.01; b p < 0.001

Note: The dashes indicate that variables were not entered into the model, because no significant univariate relationship was found

117

tively (Table 4). VIFs ranged between 1.0 - 1.3, suggesting that there was no problem of collinearity. Externalizing problems. Life events accounted for a significant percentage of the accounted for a significant increment in childrens (R Ch = 0.22) and parents
2

explained variance in childrens reports (R = 0.07). The temperament dimensions


2

both children and parent, and frustration also for childrens reported externalizing reported. The variables entered into childrens and parents models explained 29 able, ranging between 1.1 - 1.2.

reports (R Ch = 0.07). Effortful control had a significant unique negative effect for
2

problems. The interaction terms did not contribute significantly to the problems and 7%, respectively, of the variance (Table 5). VIFs in both models were accept-

Table 5: Regression analyses examining temperament as a predictor of externalizing problems as reported by children and ill parents Children Externalizing problems Beta Step 1 Negative life events Step 2 Effortful control Frustration Fear / worry
a

Ill parents Externalizing problems FCh 22.6a Beta R2 R2Ch FCh

R2 0.07

R2Ch

0.26a 0.29 -0.33a 0.23a 0.10 0.22 33.7a

0.07 -0.21a 0.06 0.10

7.9a

p < 0.001

Note: The dash indicates that the variable was not entered into the model, because no significant univariate relationship was found

118

Discussion and conclusion


The present study is the first to examine the contribution of temperament to prevalence of problems among adolescent children of parents diagnosed with cancer. The findings suggest that temperament predicted internalizing and externalizing probables (recurrent disease) and number of negative life-events. The most powerful tem-

lems, beyond the effects of socio-demographics (child gender), illness-related variperament dimensions for internalizing problems were shyness and fear/worry (reactive factor). To a lesser extent, frustration and perceptual sensitivity (childrens anxious children generally have a tendency to withdraw and are hesitant to seek support from their surroundings,22 whereas seeking support might be important, especialreports only) heighten the risk for internalizing problems. It was argued that shy and

ly in situations in which a parent had cancer. High levels of frustration are related to relax and direct themselves toward matters other than the stressor.22 Withdrawn

reduced ability to regulate attention and emotions, whereby children are less able to behaviour, anxiety and depression might be consequences of these tendencies. Internalizing problems reported by the child also increase when they were highly sensitive to things and people around them (perceptual sensitivity). Although,

Rothbart and Bates (1998) hypothesised that children with this kind of sensitivity are more vulnerable for internalizing and externalizing problems, research on these dimensions is lacking.9 The relationships between shyness, anxiety and frustration

and internalizing problems found in children reports have also been found in studies Those who have a temperament characterized by a low level of effortful control (regulative factor) and a high level of frustration are more at risk to develop externalizing problems. The relationship between effortful control and externalizing problems has been found in other studies as well.9,11,26-29 Children who have more control show to focus, and are less easily distracted by circumstances.27 The control that children have over their behaviour in this regard decreases the chance of externalizing probmore initiative in undertaking activities, have more ability to shift their attention and among the general population.9,24-26

119

lems. As mentioned above, frustration decreases emotional regulation and can lead to externalizing problems, in addition to manifesting itself in internalizing problems.25 No effect was found for pleasure intensity, which was in contrast to the results of a recent study among children in the general population. That study found that children who scored high on pleasure intensity experienced more externalizing problems.26

Whether the relationship between temperament and problems is specific for situations in which children are exposed to a parent with cancer, is not clear. It is argued that reactive and regulative temperament factors really are important when the child forth negative emotions, specifically in children high in emotionality. Children with low levels of effortful control may have difficulties to deal with these emotions, and may develop as a consequence emotional or behavioural problems.30 experienced stressful events.30 A stressful event, such as illness in the parent brings

Although the sons and daughters differed in the problems experienced and in their both genders, as was found in a previous study.26

temperament, the relationship between temperament and problems were similar for Similar patterns were found for parents and self reported problems and temperament in the current study, according to a previous study.26 Our results showed, however, a than parents reports of problems, while Oldehinkel and colleagues found the oppostronger relationship between temperament and the problems reported by children site. This inconsistency may be attributed to the informant; Oldehinkel and col-

leagues used parents reports to examine temperament, whereas the current study used self-reports from children. A number of studies found that parents and children The differences in perception may have caused also the differences between parents and children found in the present study. Another interesting result from this study is that, four of the seven temperament dimensions were related to the number of negative life events experienced. This is temperament is influenced by experiences. The influence of stressful environmental consistent with Rothberts theory, which suggests that, despite its biological base, factors on the development of temperament is an interesting phenomenon. Because
120

perceived the level of problems of children under these circumstances different.19,31

the current study uses a cross-sectional design no causal statements can be derived from the results.

The current study is one of few to use self-reports of temperament from a large num-

ber of children to examine the relationship between temperament and internalizing and externalizing problems. It is generally assumed that self-description is an important source of information in the field of personality research. Nonetheless, it was

faced with some difficulties. First, Rothbart and colleagues paid a lot of attention to the development of the theory around temperament. The empirical implementation of the EATQ-R, however, was limited. The original structure of the EATQ-R seemed not applicable for the Dutch children in the current study. A recent study among

Dutch adolescents reported also some problems with the self-reported version of the from the current study with those of studies that used other theories of temperament corresponding dimensions. More research using reliable, validated instruments to

EATQ-R26 and used, therefore, the parent version. Further, the comparison of results

was difficult, due to differences in conceptualizing and in labelling with regard to investigate the temperament of children is necessary. Second, some of the items used

to measure temperament resemble items on the problem scale and can cause itemoverlap. Previous studies show that the association between constructs remains essentially the same when correcting for possible overlapping items,26,32,33 indicating that they are separate concepts. To examine whether patterns were similar among different informants, parents reports were used also. Fourth, the current study is cross-

sectional. Longitudinal designs may provide more insight into the causality of relationships. Fifth, most of the relations between temperament and problems found in the current study were similar to those found in other studies. The use of a control between temperament and problems differed for children of parents with cancer and those of children in the general population. Finally, the current study did not pay attention to the interaction between childrens temperament and other potential pre-

group may have provided more detailed information about whether the relationships

dictors, such as parenting, the parent-child relationship and coping. Thomas and
121

Chess introduced the goodness-of-fit concept, which means that problems in children arise only when temperament and the expectations of the surroundings are not well adjusted to each other.34

The fact that temperament of children can have an impact on the prevention of problems by children is important information. Health care providers can use this account and by means of this to understand their childrens behaviour better. ment of the child and the consequences of having a parent with cancer. knowledge to assist parents to take the individual characteristics of the child into Additionally, parents might be supported to improve the fit between the tempera-

122

References
1. 2. 3. 4. 5. 6. 7. 8. 9. Walker EF, Sabuwalla Z, Huot R. Pubertal neuromaturation, stress sensitivity, and psychopathology. Dev Psychopath. 2004; 16: 807-824. Compas BE, Worsham NL, Epping-Jordan JE, Grant KE, Mireault G, Howell DC, Malcarne VL. When mom or dad has cancer: markers of psychological distress in cancer patients, spouses, and children. Health Psychol. 2004; 13: 507-515. Visser A, Huizinga GA, Van der Graaf WTA, Hoekstra HJ, Hoekstra-Weebers JEHM. The impact of parental cancer on children and the family: a review of the literature. Cancer Treat Rev. 2004; 30: 683-694. Visser A, Huizinga GA, Hoekstra HJ, Van der Graaf WTA, Hoekstra-Weebers JEHM. Parental cancer: characteristics of parents as predictors for child functioning. Cancer. 2006; 106: 1178-1187. Huizinga GA, Visser A, Van der Graaf WTA, Hoekstra HJ, Hoekstra-Weebers JEHM. The quality of communication between parents and adolescent children in the case of parental cancer. Ann Oncol. 2005; 16: 1956-1961. Watson M, St. James-Roberts I, Ashley S, Tilney C, Brougham B, Edwards L, Baldus C, Romer G. Factors associated with emotional and behavioural problems among school age children of breast cancer patients. Br J Cancer. 2006; 94: 43-50. Compas BE, Connor-Smith JK, Saltzman H, Thomsen AH, Wadsworth ME. Coping with stress during childhood and adolescence: problems, progress, and potential in theory and research. Psychol Bull. 2001; 127: 87-127. Bates JEE, Wachs TDE. Temperament: individual differences at the interface of biology and behaviour. xiii ed. Washington, DC, US: American Psychological Association, 1994. Rothbart MK, Bates JE. Temperament. Damon, W. and Eisenberg, W. Handbook of Child

of Vermont Department of Psychiatry, 1991. 15. Verhulst FC, Van der Ende J, Koot HM. Handleiding voor de Youth Self-Report (YSR) [Manual for the Dutch version of the Youth Self-Report]. Erasmus Rotterdam: University of Rotterdam, Department of Child Psychiatry, 1997. 16. Verhulst FC, Van der Ende J, Koot HM (1996). Handleiding voor de Child Behaviour Checklist (CBCL) [Manual for the Dutch version of the Child Behaviour Checklist/4-18]. Rotterdam: University of Rotterdam, Department of Child Psychiatry, 1996.

Psychology, 1998. 5 ed. New York: Wiley p.105-176. 10. Capaldi DM, Rothbart MK. Development and validation of an early adolescent temperament measure. J Early Adolesc. 1992; 12: 153-173. 11. Ellis LK, Rothbart MK. Revision of early adolescent temperament questionnaire. Minneapolis, Minnesota: biennial meeting of the society for research in child development, 2001. 12. Hartman C. De Nederlandse vertaling van de Early Adolescent Temperament Questionnaire, intern rapport [The Dutch translation of the Early Adolescent Temperament Questionnaire, internal report]. Discipline of Psychiatry, University of Groningen, 2001. 13. Ellis LK. Individual differences and adolescent psychosocial development. Dissertation, University of Oregon, 2002. 14. Achenbach TM. Manual for the Youth Self-Report and 1991 Profiles. Burlington, VT: University

123

26. Oldehinkel AJ, Hartman CA, de Winter AF, Veenstra ROJ. Temperament profiles associated with internalizing and externalizing problems in preadolescence. Dev Psychopath. 2004; 16: 421-440. 27. Eisenberg N, Spinrad T, Fabes R, Reiser M, Cumberland A, Shepard S, Valiente C, Losoya S, 28. 29. 30. 31. 32.

17. Van de Willige G, Schreurs P, Tellegen B, Zwart F. Het meten van 'life-events': de Vragenlijst Recent Meegemaakte Gebeurtenissen (VRMG) [Assessment of life events: questionnaire of recently experienced events]. Ned Tijdschr Psychol. 1985; 40: 1-19. 18. Chatterjee S, Hadi AS, Price B. Regression analysis by example. New York: Wiley, 2000. 19. Visser A, Huizinga GA, Hoekstra HJ, Van der Graaf WTA, Klip EC, Pras E, Hoekstra-Weebers JEHM (2005). Emotional and behavioural functioning of children of a parent diagnosed with cancer: a cross-informant perspective. Psychooncology. 2005; 14:746-758. 20. Kim S, Brody GH, Murry VM (2003). Factor structure of the Early Adolescent Temperament Questionnaire and measurement invariance across gender. J Early Adolesc. 2003; 23: 268-294. 21. Tuinstra J. Health in adolescence: an empirical study of social inequality in health, health risk behaviour and decision making styles. Dissertation, University of Groningen, 1998. 22. Eisenberg N, Shepard SA, Fabes RA, Murphy BC, Guthrie IK. Shyness and children's emotionality, regulation, and coping: contemporaneous, longitudinal, and across-context relations. Child Dev. 1998; 69: 767-790. 23. Fox NA, Calkins SD. The development of self-control of emotions: intrinsic and extrinsic influences. Motiv Emot. 2003; 27: 7-26. 24. Eisenberg N, Cumberland A, Spinrad TL, Fabes RA, Shepard SA, Reiser M, Murphy BC, Losoya SH, Guthrie IK. The relations of regulation and emotionality to children's externalizing and internalizing problem behaviour. Child Dev. 2001; 72: 1112-1134. 25. Rothbart MK, Posner MI, Hershey KL (1995). Temperament, attention, and developmental psychopathology. Cicchetti, D. and Cohen, D. J. Developmental psychopathology, Vol. 1: Theory and methods. Wiley series on Personality Processes. Oxford, England: John Wiley & Sons p. 315-340.

33. Lengua L, West S, Sandler I. Temperament as predictor of symptomatology in children: addressing contamination of measures. Child Dev. 1998; 69: 164-181. 34. Thomas A, Chess S. Temperament and development. New York: Brunner/Mazel, 1977.

Guthrie I,Thompson M. The relations of effortful control and impulsivity to children's resiliency and adjustment. Child Dev. 2004; 75: 25-46. Merikangas KR, Swendsen JD, Preisig MA, Chazan RZ . Psychopathology and temperament in parents and offspring: results of a family study. J Affect Disord. 1998; 51: 63-74. Lengua LJ, Long AC. The role of emotionality and self-regulation in the appraisal-coping process: tests of direct and moderating effects. J Appl Dev Psychol. 2002; 23: 471-493. Muris P, Ollendick T. The role of temperament in the etiology of child psychopathology. Clin Child Fam Psychol Rev. 2005; 8: 271-289. Birenbaum LK, Yancey DZ, Phillips DS, Chand N, Huster G. School-age children's and adolescents' adjustment when a parent has cancer. Oncol Nurs Forum. 1999; 26: 1639-1645. Lemery KS, Essex MJ, Smider NA. Revealing the relation between temperament and behaviour problem symptoms by eliminating measurement confounding: expert ratings and factor analyses. Child Dev. 2002; 73: 867-882.

124

125

126

Emotional and behavioural problems in children of parents recently diagnosed with cancer: a longitudinal study

Winette TA van der Graaf, Stacey M Gazendam-Donofrio, Josette EHM Hoekstra-Weebers

Annemieke Visser, Gea A Huizinga, Harald J Hoekstra,

Acta Oncologica 2007; 46: 67-76

127

Abstract

This study examines the prevalence of problems in children within four months wards. Sixty-nine ill parents and 57 spouses completed the Child Behaviour Checklist for 57 primary school (aged 4-11 years) and 66 adolescent children (aged 12-18 years). Adolescents completed the self-report version. Childrens functioning was compared to that of the norm group and a sample of families that were con-

after a parents cancer diagnosis (T1) and six (T2) and twelve months (T3) after-

fronted with parental cancer between one to five years before study participation

(retrospective study). Most children were reported as having a similar level as or fewer problems than was reported in the norm and retrospective studies. Reported problems decreased with time, but children who initially had more problems remained vulnerable during the year. Fathers and mothers highly agree in their perception of childrens behaviour, with the exception of adolescent daughters behavagreement between fathers and adolescent sons and daughters, and mothers and not experience problems shortly after the parents diagnosis and were functioning mants perceptions appear and remain of interest.

iour. Agreement between mothers and adolescent daughters was high, whereas adolescent sons was low to moderate. The outcomes suggest that most children do over time on a level equal to or better than that of their peers. Differences in infor-

128

Introduction
It can be very distressing for a child to have a parent be diagnosed with cancer. emotional problems or become stressed, which may result in their having less atten-

Being confronted with a life-threatening disease may cause a parent to experience tion for their children.1 Additionally, a parent may be away from home more often

while they are admitted to the hospital for treatment. While it has been hypothesized that a distressful event has an enormous impact on children, conflicting results have been found. Primary school children (4-11 years) who have a parent the same age.2-4 Parents, however, report that primary school children have more examine differences between boys and girls in this age group found that parents norm group; reports on daughters showed no difference with the norm.6 emotional problems than children in the norm group.3,5 One of the few studies to

with cancer do not report experiencing any more problems than other children of

with cancer reported that their sons experience more emotional problems than the Studies on adolescents (12-18 years) of a parent with cancer suggest that adolescents report having more problems than other adolescents.3,5,7 Other studies have ents reports of the childrens functioning have found either no difference in prevafound that this mainly applies to adolescent daughters.2,4,6,8 Studies that rely on parlence of problems in comparisons with norm groups.3,4 or that patients reports show not.5,6 That results vary from study to study would seem to be related to the source

that children experience more problems than the norm, but spouses reports do

of the information: the parents or the children themselves.

In many of the studies performed on this subject, the patient was either been diag-

nosed with cancer years before or the amount of time between diagnosis and study participation ranged from a couple of weeks and nine years (see review9). Only a handful of studies have examined the consequences for children in the period emotional problems in the short term.10 Long-term consequences are more often the immediately after diagnosis.2,4,8 despite the finding that distressful events can cause

result of the extent to which a parents cancer leads to a chronic disturbance of nor129

reason, it is important to gain insight in an early stage in the degree to which problems develop. Furthermore, studies for the most part have been cross-sectional in design, while longitudinal research is necessary to study change over time.

mal life and of the demands and burden associated with the disease.11,12 For that

This study examined the prevalence of and change in emotional and behavioural problems in children during the first year after a parent is diagnosed with cancer. diagnosed one to five years ago and with children of the norm group. The second Childrens functioning will be compared with that of children of parents who were issue concerned the predictive power of earlier emotional and behavioural problems in children on later problems. Lastly, agreement among informants in the perception of emotional and behavioural problems will be investigated.

Methods
Procedure

Patients recently diagnosed with cancer that had an expected chance of survival of one year or more, children between 4-18 years of age, and sufficient command of the Dutch language were asked to participate. Written information about the study Gynaecological and Medical Oncology and Haematology at the University Medical Centre Groningen, the Netherlands between January 2001 and February 2003. Each

was provided by the physician or oncology nurse at the Departments of Surgical,

family member was asked for written informed consent according to the regulations of the Medical Ethical Committee. After written informed consent was obtained, each family member. Questionnaires for the second and third measurement were mailed at six (T2) and 12 months (T3) after the first measurement. Family members were instructed to fill in the questionnaires independently of each other. questionnaires and prepaid envelops for the first measurement (T1) were sent to

130

Instruments

Emotional and behavioural functioning. The Child Behaviour Checklist (CBCL) and adolescents self-reported emotional and behavioural problems.15,16 The CBCL includes 120 and the YSR 102 items. Broad-band internalization and externalization scale scores and total problem score were used in the present study. The interand Youth Self-Report (YSR) were used to obtain parents reports of children.13,14

nalization problem scale (32 items) represents the subscales social withdrawal, form the externalization problems scale (30 items). The total problems scale repre-

somatic complaints, and anxietydepression. Aggressive and delinquent behaviour sents the total score derived from the sum of the internalization and externalization

syndrome scales plus thought problems, attention problems, social problems, and three-point scale: 0 = not true, 1 = somewhat or sometimes true, 2 = very true or often true. Higher scores denote more problems. In addition, raw scores were trans-

sex problems and remaining items of the CBCL/YSR. Items can be scored on a

formed into T-scores. T-scores are standardized scores with a mean of 50 and an SD of 10 on the basis of a normative national sample of children who had not been referred for mental health services in the previous year. Using T-scores provides a basis for comparison of, for example, boys and girls taking into account the differual of the CBCL/YSR defines cut-off points to differentiate children considered to function normally from those considered to have clinically elevated problems. The CBCL/YSR are well constructed questionnaires that are widely used nationally and internationally to describe the functioning of children. The CBCL/YSR showed good internal consistency for the three scales at T1, T2 and T3 (range = ences that exist between boys and girls in the norm group.13 Furthermore, the man-

0.84 to = 0.94). The Dutch manual of the CBCL provides norm data on a randomly selected Dutch sample of 1241 primary school (623 boys and 618 girls, aged 4 to 11 years) and 986 adolescent children (493 boys and 493 daughters, aged 12 to ed Dutch sample of 560 boys and 564 girls.16 Response in these Dutch samples was 18 years).14 The manual of the YSR provides norm data based on a randomly select82%. A significantly greater number of parents of primary school children partici131

responders and non-responders in regarding childrens gender.14

pated than of adolescent children (p.001). No difference was found between Additional comparisons were made with a group of children of parents diagnosed with cancer one to five years prior to assessment (mean 2.3 years). These families were approached according to the same procedure, eligibility criteria, during the same time and in the same hospital as described for the present study. Of the 476 cancer patients approached, 205 consented to participate (response 43%). Parents who declined to participate did not differ significantly from participating parents

concerning gender of the ill parent, type of cancer, and time since diagnosis. Ill par-

ents and spouses in that study provided information on emotional and behavioural problems of 114 primary school (59 sons and 55 daughters) and 222 adolescent children (105 sons and 117 daughters). Adolescents reported also on their own functioning. This group will be referred to as the retrospective group. More information on the retrospective group can be found in a previously published study.6 Analyses

Independent t-tests and Chi-squares tests were used to examine differences between

families who participated in this study and those who declined to participate in age, gender of ill parents and type of cancer. Furthermore, independent t-tests were used to examine differences in problems reported at T1 by parents and children who dropped out of the study after T1 and by parents who participated at the three measurements.

Two sided t-tests were used to examine differences in the prevalence of emotional parents, spouses and adolescents themselves and children of the norm and retro-

and behavioural problems between children in the current study as reported by ill spective groups. Children who scored above the cut-off points as defined by

could be suggested that differences between parents and childrens reports reflect differences in perspectives. Therefore, conform an earlier study,7 a child was considered to be experiencing clinically-elevated problems when his scores, according

Verhulst and associates14,16 were defined as having clinically elevated scores. It

132

to either the patient, spouse or the child himself, fell above the cut-off points. above and below the cut-off points with the norm group

Chi-square tests were used to compare the frequencies of children who were scored Repeated measures analyses of variance (ANOVAs) were computed to investigate

change over time in emotional and behavioural problems in children and adolesschool and adolescent sons and daughters) to examine between-subject effects. power of earlier emotional and behavioural problems on later problems.

cents during the first year. Subjects were divided into four subgroups (primaryPearsons product-moment correlations were performed to analyze the predictive Lastly, Pearsons product-moment correlations were conducted using CBCL Tan earlier study,6 agreement was examined between fathers, mothers and adolestion coefficient lower than 0.30 indicates weak agreement, a coefficient between indicates strong agreement.17

scores to examine agreement between fathers, mothers and adolescents. In line with cents rather than between ill parents, spouses and adolescents. A Pearson correla0.30 and 0.50 indicates moderate agreement, and a coefficient higher than 0.50 Multiple comparisons were corrected for using a Bonferroni corrected alpha (alpha/number of comparisons) as described by Dunn.18

Results
Participants

Of the 222 families approached, 112 agreed to participate at T1 (response = 50%). Families who declined participation did so because parents were not interested in too aggravating for families (n = 17), parents had a poor (n = 3) or a good prognothe study (n = 27), children did not want to participate (n = 17), participation was sis (n = 6), children were considered too young (n = 5), or were not informed about the cancer diagnosis (n = 3). For the remaining families (n = 32), reasons to decline participation were not given.

133

Of the 112 families who participated at T1, 74 families (66%) participated at T2 and 69 at T3 (62%). In total, 43 families dropped out of the study. The 69 famithe current study. lies who completed questionnaires at the three measurements were the focus of The sample consisted of 69 parents diagnosed with cancer and 57 spouses report-

ing on 57 primary school and 66 adolescent children. More information on sociodemographic variables is listed in Table 1. Breast cancer was most prevalent (n = 24) followed by gynaecological (n = 8), dermatological (n = 8), gastrointesti-

Table 1: Demographic characteristics Parent characteristics Ill parents Mothers Fathers N 69 46 23 67 33 % Spouses Mothers Fathers N 57 22 35 39 61 %

Mean age = 43, SD = 5.1, range = 33 to 55 yrs Highest level of education completed by ill parents Low1 Middle2 High3 Child characteristics Primary school children Daughters Sons Mean age = 8 yrs, SD=2.1 Family characteristics One-parent families Number of children in a family 1 2 3 >4
1 2 3

Mean age = 43, SD = 5.7, range = 27 to 56 yrs Highest level of education completed by spouses Low1 Middle2 High3 16 23 18 N Adolescents 66 35 31 53 47 Daughters Sons Mean age = 15 yrs, SD=2.0 28 40 32 %

11 30 28 N 57 32 25

16 44 41 % 56 44

N 6 13 36 11 9

% 9

Number of children participating in the study per family 19 52 16 13 1 2 3 >4 28 31 7 3 41 45 10 4

Primary school, or lower vocational degree Lower general secondary education or intermediate vocational education or high school degree Higher vocational education or university degree

134

nal (n = 7), urological (n = 7), haematological (n = 5) and other cancers (such as soft tissue, central nervous system or head and neck cancer, n = 10). Mean time seven months at T2 (SD = 1.2 month, range 6 - 10 months) and 13 months at T3 (SD = 1.2 months, range 12 - 18 months). One parent suffered from recurrent illness at T3. since diagnosis was two months at T1 (SD = 1.0 month, range 1 week to 4 months),

Representativeness

Parents diagnosed with cancer who declined participation did not differ significant-

ly in age and gender from parents who did participate. Type of malignancy, how-

ever, affected study participation (2 = 41.7, p 0.001). The percentage of patients group (22%) than in the participating group (8%, 2 = 10.5, p 0.001). In contrast,

with gynaecological malignancies was significantly higher in the non-participating

a higher percentage of patients with urological (12%) and soft tissue malignancies (10%) participated than declined (2.7%, 2 = 18.1, p 0.001; 4.7%, 2 = 6.5, p = 0.011, respectively). Percentages of participants and non-participants with other malignancies were comparable.

Ill parents who dropped out of the study after T1 did not differ in age from those who continued to participate. The percentage of families in which the father was diagnosed with cancer was significantly higher in the group that ended participation (57%) than in the group that continued to participate (34%, 2 = 14.3,

p 0.001). Furthermore, ill parents who dropped out reported significantly more externalizing problems (t = 2.1, p = 0.037) in children at T1 than did ill parents who continued to participate. Spouses who dropped out of the study reported significantly more internalizing problems (t = 3.0, p = 0.001), externalizing (t = 3.6, p 0.001) and more total problems (t = 3.7, p 0.001) in children at T1 than did spouses who

continued to participate. Adolescents who dropped out reported significantly more internalizing problems (t = 2.5, p = 0.014), and tended to report more total problems (t = 2.0, p = 0.051) at T1 than adolescents who continued to participate.

135

Table 2: Descriptive statistics for raw scores on the Child Behaviour Checklist at T1, T2 and T3, t-tests for differences between primary school and adolescent sons

136
Retrospective group T3 M 3.9 (4.3)a 7.1 (6.9) 8.9 (6.6) 24.7 (18.6) 21.7 (25.2) 18.2 (25.3) 16.8 (25.1) 7.5 (7.7) 5.3 (6.9) 5.3 (7.4) 7.6 (6.0) 20.8(16.0) 4.7 (6.6) 5.0 (8.2) 3.9 (7.0) 5.7 (6.5) 6.1 (6.4) 17.4 (13.3) (SD) M (SD) M (SD) M (SD) M (SD) M (SD) T1 T2 T3 M (SD) 4.8 (4.7) 8.6 (6.7) 22.5(15.2) Spouses Retrospective group Norm group 4.8 (6.5) 5.8 (7.3) 15.3 (16.7) 20.7 (14.2) 13.6 (9.1)1 13.6 (10.6) 6.9 (6.6) 3.9 (4.4) 3.6 (4.2) 6.3 (4.9) 4.8 (3.7) 4.6 (4.6) 3.2 (3.6) 3.2 (5.0) 10.1 (10.5)1 5.4 (5.7) 6.3 (7.2) 19.1 (18.0) 5.4 (5.5) 6.2 (5.9) 20.0 (16.1) 3.5 (4.4) 3.8 (4.0) 11.1 (11.0)1a 18.0 (14.2) 15.9 (13.9)
1a

and daughters and children of the retrospective and norm group

Ill parents

T1

T2

M (SD)

(SD)

Primary school sons

Internalizing

5.4 (5.4)

5.1 (5.6)

Externalizing

7.6 (6.6)

7.2 (6.9)

Total Problems

22.0 (18.2)

20.8 (18.3)

Primary school daughters

Internalizing

4.9 (4.7)

4.2 (4.6)

Externalizing

5.0 (5.4)

4.9 (5.2)

Total Problems

15.6(12.4)

14.4 (12.8)

Adolescent sons 5.4 (5.5) 6.4 (5.7) 4.8 (3.9)


1

Internalizing

6.0 (5.4)

5.4 (6.1)

4.9 (5.7)

3.5 (4.0) 4.1 (4.8)


1

2.7 (4.6)1 3.1 (4.2)1 9.4 (12.3)1

4.1 (3.9) 4.7 (4.2) 14.2 (9.7)

5.7 (5.7) 7.1 (7.1) 20.1 (16.5)

Externalizing

5.4 (4.6)

5.2 (6.0)

Total Problems

18.0 (14.0)

16.7 (16.1)

12.2 (11.9)1

Adolescent daughters 5.4 (6.1) 4.2 (3.0) 13.5 (11.1)a 21.5 (17.0) 6.1 (5.9) 8.5 (7.2) 6.5 (5.1) 4.8 (3.0) 15.9 (10.3) 4.9 (5.3) 3.6 (4.8) 12.3 (12.4) 5.7 (5.1) 3.6 (3.2) 12.6 (10.9) 5.3 (5.0) 5.2 (4.9) 15.2 (12.6) 6.5 (6.0) 5.5 (5.8) 18.7 (14.6)

Internalizing

7.4 (6.6)

6.5 (5.1)

Externalizing

5.2 (4.5)

4.6 (4.2)

Total Problems

19.0 (15.2)

15.8 (13.8)

p 0.008 significance of independent t-tests with the norm group

p 0.008 significance of independent t-tests with the retrospective group

Note: Bonferroni-corrected alpha = 0.008

Prevalence of emotional and behavioural problems as perceived by ill parents and spouses Table 2. Ill parents and spouses reports of their childrens functioning are summarized in Primary school sons. No significant differences were found between ill parents and

spouses reports of problems in sons at the three measurement times and reports of problems in sons of the norm group. In comparison to the retrospective group, ill parents in the current study reported significantly less internalizing problems at T3.

Primary school daughters. Ill parents reported that daughters had similar levels of problems to daughters in the retrospective and norm groups. Spouses reported signorm population. nificantly fewer total problems in daughters at T1 and T3 than did parents of the Adolescent sons. Adolescent sons were perceived by ill parents as having significantly fewer externalizing and total problems at T3 compared to sons in the norm problems at T1, T2 and T3, fewer total problems at T2 and T3 and fewer internaland retrospective group. Spouses reported also significantly fewer externalizing izing problems at T3 than spouses in the norm population. According to spouses, the prevalence of problems in sons did not differ significantly from those in sons of the retrospective group. Adolescent daughters. In adolescent daughters, ill parents reported significantly

fewer total problems at T3 than did parents in daughters of the retrospective group. No significant differences were found in comparison to reported problems in daughters of the norm group. Spouses reported a similar level of problems than was found in the retrospective and norm group.

Prevalence of emotional and behavioural problems as reported by the adolescents Adolescents self-reports are summarized in Table 3.

lems at T2 and T3 than peers in the norm group did, but no significant differences
137

Adolescent sons. Adolescent sons reported significantly fewer externalizing prob-

were found in comparison to the retrospective group.

Adolescent daughters. Compared to the retrospective group, significantly fewer internalizing and total problems were reported at T2 and T3 by adolescent daughters in the current study.

Clinically elevated scores

At T1, 24% of the primary school sons were identified as having internalizing prob-

lems above the clinical cut-off point in comparison to 10% in the norm group as having total problems in comparison to 10% and 8%, respectively, in the norm group (no significant differences). Three percent of the primary school daughters were identified as having clinicallyelevated scores on internalizing and total problems, and 12% on externalizing problems at T1, compared to 8 - 10% in the norm group (no significant differences). (2 = 5.4, p = 0.02). Twelve percent was reported as having externalizing and 16%

At T1, 32% of the adolescent sons were reported as having clinically-elevated


Table 3: Descriptive statistics for raw scores on the Youth Self-Report at T1, T2 and T3, t-tests for differences between adolescent sons and daughters and those of the retrospective and norm group Self reports Retrospective group T3 M (SD) 6.9 (6.2) 7.4 (5.1)1 24.2 (17.0) M (SD) 8.7 (6.8) 10.6 (5.7) 32.4 (17.3) M (SD) 8.6 (5.8) 11.5 (6.7) 33.7 (16.8) Norm group

T1 M (SD) Adolescent sons Internalizing Externalizing Total Problems Adolescent daughters Internalizing Externalizing Total Problems
1 a

T2 M (SD) 6.2 (5.4) 7.8 (5.0)1 24.8 (16.9)

9.0 (6.0) 9.6 (5.7) 33.1 (17.9)

9.9 (7.1) 9.8 (6.7) 32.3 (18.6)

9.0 (8.3)a 7.6 (5.2) 28.2 (20.1)a

8.1 (8.3)a 8.4 (6.6) 27.0 (20.4)a

13.7 (9.6) 10.6 (6.1) 39.6 (21.6)

10.8 (7.1) 10.0 (6.1) 34.5 (18.0)

p 0.008 significance of independent t-tests with the norm group p 0.008 significance of independent t-tests with the retrospective group

Note: Bonferroni-corrected alpha = 0.008

138

scores on internalizing problems and 26% on total problems, while 9% was found in the norm population (2 = 20.5, p 0.001; 2 = 10.7, p = 0.001, respectively). Three percent had clinically-elevated scores on externalizing problems.

Among adolescent daughters, at T1 26% were identified as having clinically-elecompared to 8 - 9% of norm group girls (2 = 14.9, p 0.001; 2 = 6.8, p = 0.009,

vated scores on internalizing, 20% on externalizing and 23% on total problems, 2 = 8.2, p = 0.004, respectively).

In general, these percentages decreased over time to 8 - 14%. One exception was nalizing problems (2 = 10.7, p = 0.001).

found: at T3, 20% of adolescent sons still had clinically-elevated scores on inter-

Changes in the prevalence of problems over time

Reports from ill parents, spouses and adolescents revealed a significant decrease in

total problems over time. Spouses and adolescents reported a significant decrease in externalizing and ill parents and adolescents in internalizing problems over time. No significant effects of group or interactive effect of group and time were found (Table 4).

Table 4: Repeated measures analyses of variance Ill parents Test of overall trend Effects Internalizing Time Group1 Interaction Externalizing Time Group1 Interaction Total Problems Time Group1 Interaction
1

Spouses Test of overall trend F 4.4 0.7 1.0 8.4 1.1 0.6 9.3 0.8 0.7 p ns ns ns 0.002 ns ns 0.001 ns ns F

Adolescents Test of overall trend p 0.005 ns ns 0.001 ns ns 0.001 ns ns 6.0 1.0 0.9 6.9 0.2 0.4 10.8 0.2 0.7

F 8.3 1.6 1.3 1.9 3.7 1.6 18.1 0.7 2.1

p 0.001 ns ns ns ns ns 0.001 ns ns

Four groups were distinguished, namely primary school and adolescent sons and daughters

Note: Bonferroni-corrected alpha = 0.005

139

Stability of problems

Relationships between T1 and T2, T2 and T3, and T1 and T3 in internalizing, exterranging from r = 0.65 to r = 0.88. This means that on an individual level children with high scores at T1 also scored high at later time points (Table 5). Agreement among informants at T1 nalizing and total problems were strong (p 0.001) with correlation coefficients

High levels of agreement were found between fathers and mothers regarding primary school sons and daughters, and adolescent sons on internalizing (r = 0.65, r = 0.56, r = 0.67, respectively), externalizing (r = 0.53, r = 0.69, r = 0.51,

respectively), and total problems (r = 0.67, r = 0.70, r = 0.50, respectively). externalizing (r = 020) and total problems (r = 0.19) were low.

However, levels of agreement on adolescent daughters internalizing (r = 0.06), Levels of agreement between adolescent sons and fathers, and between adolescent sons and mothers were low to moderate on internalizing (r = 0.27, r = 0.34, respectively), externalizing (r = 0.39, r = 0.20) and total problems (r = 0.33, r = 0.24). Between adolescent daughters and fathers, low agreement was found on internaliz-

Table 5: Relationships between earlier and later internalizing, externalizing and total problems Ill parents Spouses Adolescents Measurement times Internalizing T1-T2 T2-T3 T1-T3 Externalizing T1-T2 T2-T3 T1-T3 Total Problems T1-T2 T2-T3 T1-T3 r 0.85 0.71 0.69 0.77 0.65 0.68 0.88 0.72 0.74 r 0.72 0.67 0.67 0.67 0.74 0.74 0.78 0.82 0.79 r 0.73 0.79 0.81 0.69 0.76 0.73 0.74 0.80 0.77

All correlation coefficients were significant at a p 0.001 level

140

ing (r = 0.02) and total problems (r = 0.18), while agreement on externalizing was high (r = 0.54). Agreement between adolescent daughters and mothers was high on internalizing (r = 0.67), externalizing (r = 0.66) and total problems (r = 0.79).

Discussion
The current study focused on the prevalence of and change in problems of children during the first year after the parents cancer diagnosis. Primary school and adolescent children were reported to have similar levels of emotional and behavioural problems within four months after the parents cancer diagnosis to children of the group). However, 20 - 25% of the primary school and adolescent sons and adolesinternalizing and/or total problems at the first measurement. norm group and children of parents diagnosed one to five years ago (retrospective cent daughters were reported as having scores in the clinical range, particularly for Our results generally confirm the results of previous studies that have also found comparable to norm group children.4,19 Other studies have also reported that a high

that children of parents recently diagnosed with cancer had an amount of problems percentage of children had clinically-elevated scores on emotional problems in the vulnerable differed. Earlier studies found clinically-elevated scores primarily among adolescent daughters.2,4,8 Interestingly, the current study also found a high percentage of primary school and adolescent sons having clinically-elevated scores. Studies focusing on gender differences have not identified adolescent sons as being first months after the parents diagnosis.20,21 However, the groups identified as being

vulnerable. A possible explanation for this difference may be that the current study focused on the prevalence of internalizing problems and not just anxiety and depression. The internalizing scale measures anxiety and depression, but also measless expressive of emotions and react to their parents illness by withdrawing.22

ures withdrawal and somatic complaints. Earlier studies have shown the sons are Our study suggests, in line with others, that although some children do experience
141

problems, the majority of children do not experience an increase of problems durchildren will try to protect their parents in this stressful period and may, therefore,

ing the first year after the parents diagnosis. A possible explanation is that many mask their true emotions. Children begin to show their true feelings when the fam-

rospective study and other studies that have focused on long term effects of a parent with cancer have been reported to have elevated levels of problems.5,7 Selection bias may be another possible explanation for our findings. Children in families who

ily gets back to normal life again.23,24 This could explain why children in the ret-

dropped out of the study reported having more problems than families who continued participation. This suggests that the results of the current study underestimate the extent of problems in children of parents with cancer.

In the course of the year, the amount of reported emotional and behavioural problems decreased and children were reported to have even fewer problems than children in the norm and retrospective groups. A decrease of problems over time was

also found in an earlier study.4 Although a decline in problems was found for the group as a whole, some childrens problems did not decrease. The level of problems at the first measurement was a relatively strong predictor of problems at later measurements. These results are consistent with a previous study showing high levels of continuity in anxiety/depression in children over time.20 How vulnerable a child is experience fewer problems may have more personal and environmental resources problems, while children who experience more problems may have fewer resources to protect them. depends on biological, psychological and social factors.11 Children who initially

(personality, family functioning or social support) that protect them from long-term

The high levels of agreement between parents found in the present study is conform findings from other studies.6,14,25 We did find one exception: levels of agreement Mothers might judge adolescent daughters behaviour more accurately than fathers between parents on the problems reported in adolescent daughters were low. because daughters are more likely to discuss their worries and problems with their

mothers.6 This suggestion is supported by the finding that levels of agreement


142

between mothers and adolescent daughters appeared high, while levels of agreement between fathers and daughters were low. Levels of agreement between parents and adolescent sons were low. It is believed that parents are less aware of the impact of illness on adolescent sons because parents have less contact with adolescent sons and sons talk less often about their experiences and emotions.26 Limitations

The current study is one of the few prospective studies investigating childrens

functioning during the first year after a parents cancer diagnosis where information has been gathered from various sources. However, this study has a number of limto be daunting task. Families were approached carefully, with a great deal of attenitations. Participating in a research study during this hectic period of time can seem tion. Nonetheless, half of the families approached declined to participate, which

may have led to selection bias. The reasons given by the non-participating families been under- or over-reported. On the one hand, it is possible that families who

varied considerably and we cannot determine whether the presence of problems has decline did so because they were experiencing high levels of problems. On the

other hand, two reasons stated for not participating were a positive prognosis and in those families who declined was small. A second source of bias may come from

not having informed the children, which could mean that the impact of the illness the number of families who dropped out of the study after the first measurement. As discussed earlier, children who continued in the study had fewer problems than children who dropped out.

An additional limitation is that a general questionnaire was used to measure the childrens functioning. It is possible that this instrument is not sensitive enough to measure consequences related to having a parent diagnosed with cancer. Questionnaires specifically developed for this situation or qualitative research methods could provide more insight into problems that may be prevalent in this situation or in this phase of the illness.

143

Future research and implications

Future researchers may need to pay more attention to how patients are approached for participation so that a representative sample group is procured and the consequences of having a parent with cancer can be more fully mapped out. In order to

facilitate early recognition of children at risk, more longitudinal studies are needed to investigate protective and risk factors, such as temperament, coping and family functioning.

Results of the current study may heighten awareness of health care providers that some children may have difficulty adapting to the parents illness. Health care providers may assist parents in supporting their children in the overwhelming early stage of cancer and to be specifically sensitive to children at risk. Results also showed that different informants may paint a different picture of how a child functions. Therefore, health care providers should provide parents with information to share their emotions with their parents under these circumstances.

heighten their awareness of their childrens feelings and give children the chance to

144

References
1. Armsden GC, Lewis FM. Behavioural adjustment and self-esteem of school-age children of women with breast cancer. Oncol Nurs Forum. 1994; 21: 39-45. 2. Compas BE, Worsham NL, Epping-Jordan JE, Grant KE, Mireault G, Howel DC, et al. When mom or dad has cancer: markers of psychological distress in cancer patients, spouses, and children. Health Psychol. 1994; 13: 507-515. 3. Heiney SP, Bryant LH, Walker S, Parrish RS, Provenzano FJ, Kelly KE. Impact of parental anxiety on child emotional adjustment when a parent has cancer. Oncol Nurs Forum. 1997; 24:655-661. 4. Welch AS, Wadsworth ME, Compas BE. Adjustment of children and adolescents to parental cancer. Parents' and children's perspectives. Cancer. 1996; 77: 1409-1418. 5. Birenbaum LK, Yancey DZ, Phillips DS, Chand N, Huster G. School-age children's and adolescents' adjustment when a parent has cancer. Oncol Nurs Forum. 1999; 26: 1639-1645. 6. Visser A, Huizinga GA, Hoekstra HJ, Van der Graaf WTA, Klip EC, Pras E, et al. Emotional and behavioural functioning of children of a parent diagnosed with cancer: a cross-informant perspective. Psychooncology. 2005; 14: 746-758. 7. Watson M, St. James-Roberts I, Ashley S, Tilney C, Brougham B, Edwards L, et al. Factors associated with emotional and behavioural problems among school age children of breast cancer patients. Br J Cancer. 2006; 94: 43-50. 8. Grant KE, Compas BE. Stress and anxious-depressed symptoms among adolescents: searching for mechanisms of risk. J Consult Clin Psychol. 1995; 63: 1015-1021. 9. Visser A, Huizinga GA, Van der Graaf WTA, Hoekstra HJ,Hoekstra-Weebers JEHM. The impact of parental cancer on children and the family: a review of the literature. Cancer Treat Rev. 2004; 30: 683-694. 10. Rutter M, Silberg J, O'Connor T, Simonoff E. Genetics and child psychiatry: II. Empirical research findings. J Child Psychol Psychiatry. 1999; 40: 19-55. 11. Compas BE, Hinden BR, Gerhardt CA. Adolescent development: pathways and processes of risk and resilience. Annu Rev Psychol. 1995; 46: 265-293. 12. Hetherington EM, Stanley-Hagan M. The adjustment of children with divorced parents: a risk and resiliency perspective. J Child Psychol Psychiatry. 1999; 40: 129-140. 13. Achenbach TM. Manual for the Child Behaviour Checklist/4-18 and 1991 Profiles. Burlington, VT: University of Vermont Department of Psychiatry; 1991. 14. Verhulst FC, Van der Ende J, Koot HM. Handleiding voor de Child Behaviour Checklist [Manual for the Dutch version of the Child Behaviour Checklist/4-18]. Rotterdam: University of Rotterdam, Department of Child Psychiatry, 1996. 15. Achenbach TM. Manual for the Youth Self-Report and 1991 Profiles. Burlington, VT: University of Vermont Department of Psychiatry: 1991. 16. Verhulst FC, Van der Ende J, Koot HM. Handleiding voor de Youth Self-Report [Manual for the Dutch version of the Youth Self-Report]. Erasmus Rotterdam: University of Rotterdam, Department of Child Psychiatry, 1997. 17. Cohen J. Statistical power analysis for the behavioural sciences. Hillsdale, NJ: Erlbaum: 1988. 18. Dunn OJ. Multiple comparisons among means. J Amer Statist Ass. 1961; 56: 52-64.

145

19. Hoke LA. Psychosocial adjustment in children of mothers with breast cancer. Psychooncology. 2001; 10: 361-369. 20. Mireault GC, Compas BE. A prospective study of coping and adjustment before and after a parent's death from cancer. J Psychosoc Oncol. 1996; 14: 1-18. 21. Nelson E, While D. Children's adjustment during the first year of a parent's cancer diagnosis. J Psychosoc Oncol. 2002; 20: 15-36. 22. Davey M, Gulish L, Askew J, Godette K, Childs N. Adolescents coping with mom's breast cancer: developing family intervention programs. J Marital Fam Ther. 2005; April. 23. Helseth S, Ulfsaet N. Having a parent with cancer: coping and quality of life of children during serious illness in the family. Cancer Nurs. 2003; 26: 355-362. 24. Davey MP, Askew J, Godette K. Parent and adolescent responses to non-terminal parental cancer: a retrospective multiple-case pilot study. Fam Syst Health. 2003; 21: 245-259. 25. Duhig AM, Renk K, Epstein MK, Phares V. Interparental agreement on internalizing, externalizing, and total behaviour problems: a meta analysis. Clin Psychol. 2000; 7: 435-453. 26. Seiffge-Krenke I, Stemmler M. Factors contributing to gender differences in depressive symptoms: a test of three developmental models. J Youth Adolesc. 2002; 31: 405-417.

146

147

148

General discussion

149

Positive and negative experiences during childhood may influence a childs func-

tioning. A parent with cancer is an event that causes a lot of stress and uncertainty

and may change the life of children radically and for a long time. In the United States in particular a number of studies have been conducted lately paying attention to the impact of such an event on children. The results of these studies however are conflicting, for which several reasons can be argued. Often no distinction is made

in gender and age of the children. Informants about the childrens functioning vary, sometimes the parents report and sometimes the children themselves. The quality validity and reliability of the results.

of some studies is limited (e.g. study design, sample size), which may influence the In the Netherlands so far no research has been performed into the consequences of the diagnosis cancer in a parent for children. Whether children have more problems than children who did not experience such a situation is unknown. This thesis

focuses on the prevalence of problems in children under such circumstances and on factors that may interrelate with the degree in which these problems occur. The results are presented separately for boys and girls and for primary school children and adolescents. Childrens functioning is described from three perspectives, the factors this study focused upon can be seen in Figure 1 (Introduction, page 24). parents with cancer, the spouse (if present) and the adolescent children. Further

Emotional and behavioural functioning


Childrens functioning was studied in two ways: in a prospective, longitudinal study after diagnosis and in a cross-sectional study on the functioning of children 1 to 5

on the functioning of children at three measurement moments during the first year years after diagnosis. Children of parents with cancer who had their diagnosis four months prior to measurement appeared to experience a degree of emotional and behavioural problems comparable to children of parents with cancer diagnosed 1 to 5 years prior to measurement and to children of the norm group. Six and 12 months
150

later these children on average seemed to experience fewer problems than the two comparison groups. However, sons in the primary school age group and adolescent daughters appeared to experience more emotional problems 1 to 5 years after diagvidual level one of four or five children, depending on age and gender, both during nosis than their peers, when looking at the mean scores of the total group. On indithe four-months period and 1 to 5 years after diagnosis, appeared to experience problems so serious that professional help was needed. A number of children with serious problems shortly after diagnosis recovered in the course of the year.

Cancer is a high impact and life threatening illness that in many cases disrupts a the lapse of time after diagnosis. Remarkably, children do not appear to experience more emotional and behavioural problems during the first year after diagnosis than urement moment (6 or 12 months after diagnosis), experienced even fewer problems may be very stressful for the parents due to the confrontation with the disease and dren may put aside their own needs and may not express their own problems when

childs life. The impact of the illness on the child could be expected to decrease with

children 1 to 5 years after diagnosis and some, depending on age, gender and measthan children in the retrospective and the norm group. The first year after diagnosis the subsequent medical examinations and treatments. Particularly in this phase chilseeing their parent having a difficult time.1 There is also the possibility that the con-

sequences of the parents illness do not become manifest until later. Previous stud-

ies show indications of a delayed effect which means that high impact events do not effect is that children in the early phase of the parents illness have enough resources

become noticeable at once but a number of years later. The explanation for a delayed to be able to cope. These resources often disappear over time.2 Sample bias howev-

er might be another plausible explanation. The children reported about are the chil-

dren who have participated in all three study measurements. Drop out children after all measurements.

first measurement seemed to have more problems than children who participated in Children of parents with cancer appeared to experience not so much behavioural problems such as aggressive or delinquent behaviour as primarily emotional prob-

151

lems such as fear and depression, withdrawal behaviour and physical complaints.

studies also report behavioural problems such as anger as a reaction of children of parents with cancer. 4 Behavioural problems are also found in children of a terminally ill parent.5,6

This corresponds with previous quantitative research,3 but qualitatively oriented

Predictors of childrens functioning


Child characteristics

Gender, age and temperament of children appeared to be related to the prevalence of emotional and behavioural problems. The results of our study show that adolescent daughters have more problems 1 to 5 years after diagnosis than adolescent girls in the norm group. Development theories suggest that adolescents tend to con-

form to traditional masculine and feminine role patterns. For adolescent boys this protection and autonomy. Girls are supposed to focus more on their social sur-

means that they focus more on matters concerning themselves, independence, self roundings and they learn to care for the well-being of others.7,8 This difference in

socialization and the subsequent feelings of responsibility of daughters for taking

on care tasks may explain the vulnerability of adolescent daughters of parents with cancer. True enough, the increasing demands may also stimulate the daughter into acquiring new competences and so contribute to her feeling being able to do someher, development problems may follow.

thing for the parent. However, when the demands exceed her capacities and block The present study also shows that sons in the primary school-age have more probthe socio-emotional areas may be primarily an explanation. This staying behind

lems. The less rapid development of boys in comparison with girls in language and may interfere with the coping with a stressful event such as cancer.9 The literature

about the consequences of a divorce for children shows that boys in this age are more vulnerable, but that varying mechanisms underlie boys and girls vulnerabil152

ity.10,11 Girls e.g. appear to experience problems primarily when feeling responsible for the conflict between the parents, whereas boys experience problems when not feeling able to cope with the situation. Similar processes may very well play a role in children with a parent with cancer. Further research is needed to be better able to explain the vulnerability of sons in the primary school age group. Results suggest as well that the childrens temperament correlates significantly with the prevalence of problem. Shy and fearful children primarily show a high prevalence of emotional problems; children with a low frustration tolerance and children not so capable of effortful control seem to have primarily behavioural problems. Although various studies show that these temperamental dimensions correlate as well with the functioning of children without an ill parent, it is suggested that this correlation is mainly found when children are confronted with a stressful event.12 Parents characteristics

Children appeared to experience more problems when the father had a cancer diag-

nosis than when the mother did. This result is at right angles to those of previous studies that showed that children of an ill mother almost always had more probBesides it is not clear to what extent the illness characteristics of the fathers were different from those of the mothers. Further studies with a larger number of particinterpretable. lems. The number of participating fathers in the present study is relatively low.

ipating fathers will make the effect of the gender of parents more researchable and Neither age nor educational level appeared to be related to the degree of problems in children. The general literature about this is rather consistent and describes that a higher educational level of parents correlates with a better functioning of chil-

dren.13,14 These studies state that not the educational level in itself correlates with a higher problem risk in children but the often allied worse financial situation and tors have played a part in the present study. worse working and living circumstances.13 It is not clear to what extent these fac-

Attention has been paid to the physical and mental functioning of the parent with
153

cancer and the spouse as well. As the parent and the spouse experience more limiproblems in children depend on the age of the child. Primary school children primarily appeared to experience problems when both parents had physical limitations. Children in this age group depend on their day-to-day life on the care provided by their parents. More physical limitations in a parent will lead to a larger disruption of the life of these children. The prevalence of problems in adolescents seemed only to correlate with the physical limitations of the ill parent. A possible

tations the problems in the children increase. The limitations that correlate with the

explanation for this may be that because of the physical limitations the realization lems in adolescents might be a consequence of the increasing number of tasks that will have to be taken over from the parent who has physical limitations.

dawns that the illness is serious indeed. Another explanation might be that the prob-

Another correlation was found between the mental functioning of both parent with cancer and spouse and the prevalence of problems in adolescents, whereas this was not found for primary school children. Adolescents are in a phase of dissociation of Apart from the decreased accessibility of a parent as a consequence of this, the parent may also in coping with the event be less capable of attending to and supporting adolescents.

the parent and so dependent on the psychological stability and flexibility of parents.

Family characteristics

Primary school children of single-parent families and children with fewer siblings appeared to have more problems than children of two-parent families and with several siblings. Primary school children of single-parent families having more problems has been shown in previous studies among children of a parent with cancer15 situation cannot of course be supported by a second, often healthy, parent. Besides, and children in the general population.16 Children of single-parent families in this

the threatening passing away of the parent in a single-parent family can bring with be the case in two-parent families. This is notably found among primary school
154

it more fear because the consequences for children have a higher impact than will

children and not among adolescents. From developmental psychology however it is known that adolescents have more contacts outside the family. It may very well be port, receive this support from peers or adults outside the family. that adolescents of small or single-parent families experiencing insufficient supThe more negative life events (such as illness in one of the other members of the family) a family had gone through the more problems a child of a parent with cancer experienced. This finding in accordance with other studies that report an increasing number of life events, despite the presence or absence of other risk factors, enlarges the risk of problems.2 Illness-related variables

Intensive treatment, post-treatment complications and recurrence are related to more problems in certain sub-groups of children, probably because the perception of the illness is more threatening in these situations. Time since diagnosis and

length of treatment showed no correlation with problem prevalence in children.

Although it seemed not surprising that the seriousness of the illness and the impact of treatment is associated with the functioning of children, previous studies lack family members (spouse, adult children).19 proof of this correlation.17,18 Only recurrence showed an increased problem risk in In the present study parents with divergent diagnoses were approached for participation. Thus a heterogeneous group of cancer patients participated with divergent types of cancer and treatment stages, with subsequent various physical and psycho-

logical limitations. The diversity in illness-related variables in this study makes the results difficult to interpret and further research necessary in order to obtain more insight into the impact of these variables. Possibly it is not as much the specific illexperience, but the confrontation with cancer and the possible long-term consequences for the physical and/or mental functioning of the parents. ness characteristics themselves that give indications for the problems children may

155

Informant agreement

The agreement between parents about the functioning of children appeared to be first months and the years after diagnosis. The type of problem reported about (emotional or behavioural), age (primary school children or adolescents) and the gender of the child seemed not to have any effect on the degree of agreement. Mothers reported notably often a comparable level of problems to that of the adolescents themselves, whereas fathers often reported fewer problems. This suggests problems of their children than fathers. This may be due to the fact that children

high, whereas between parents and children was low to moderate both during the

that mothers, despite the possible illness, are generally better able to assess the tend to discuss their problems sooner with the mother than with the father. have for their children.

Moreover, mothers might have more empathy for the impact such an event might

Methodological remarks
The design of this study contains a number of strong characteristics. The results are first of all based on information obtained from a relatively large group of children and their ill and healthy parent. Family members provided at various moments in mainly focused on children of just mothers with cancer, both mothers and fathers

time information about the functioning of children. Contrary to earlier studies that diagnosed with cancer participated in our study group. In addition the problems of children were inventoried by means of a questionnaire of which the validity and reliability have been shown nationally and internationally.

Despite these positive study characteristics a number of remarks can be made. First, Although two third of the families that did not participate has reported the reason for non-participation in the study (from participating is too large an emotional bur-

as a consequence of the relatively low response sample bias may have taken place.

den to we have left the illness behind us), it is difficult to conclude whether this
156

has led to an over- or an underreporting of problems. On the one side the problemfore participated, which may have led to over-reporting. On the other it may well be that families experiencing problems or not informing their children and/or havparticipating, which may have led to underreporting. This representativity moreover is a matter of discussion in the longitudinal study because in addition a num-

experiencing families may have comprehended the necessity of the study and there-

ing trouble communicating about the illness and its consequences refrained from

ber of families dropped out before second and third measurement. Children having

dropped out after first measurement appeared to experience more problems than underreporting of problems. Another marginal note about the representativity of the

children having participated in all measurement moments, which may have caused results is the overrepresentation of the number of participating mothers diagnosed with cancer in the cross-sectional study. The percentage of women in the age range of 35-55 who have cancer is higher (65%) than that of men.20 This percentage agrees with the number of participating ill mothers in the prospective, longitudinal study (67%), but is lower than the percentage in the cross-sectional study (81%).

Second, sometimes several children from the same family participated in the study, for which no correction has taken place. Every child is unique and reacts in its own way on the illness of the parent. There is no denying however that the functioning of children in one family is interrelated, they are exposed to the same family and surrounding characteristics such as parents functioning, manner of upbringing,

family functioning and life events. No multilevel analyses were performed in the been taken into account which may have caused a distortion of the results.

present study and so the interdependence of children from the same family has not

Future research
Based on this study a number of recommendations can be made for further research. Future studies should pay even more attention to the non-response. The risk of selec157

tivity generated by non-response could not be eliminated in this kind of research.

Families who experienced study participation as too much of a burden can not be

forced. In order to generate more clarity about non-response bias reasons for nonis necessary to provide more insight into the characteristics (such as socio-demographic illness-related variables) of responders in comparison to the population.

participating should be more accurate obtained. Furthermore, to get more clarity it

A relatively large sample and quantitative research methods were used so consequently analyses are mostly executed on a group level. A limitation of this method is that no insight is obtained in the experiences and the degree of adaptation of children on individual level. As a supplement to the large-scale quantitative studies individual experiences of and consequences for children.21

qualitative research methods can be used so as to provide better insight into the Internationally accepted, well-validated and age and gender standardized questionnaires were used to obtain information about the emotional and behavioural problems in children. These questionnaires however focused on general problems in

children instead of on the specific consequences the illness of the parent may have for children. Earlier studies have shown that children may experience shame or the illness of the parent also brings restrictions for their own life or loneliness due to not wanting to ask for attention for their own problems.5 The use of questionfeelings of guilt because they shy away from their parent, anger about the fact that

naires aimed at the specific consequences of this situation for the child may provide additional information. Parent and adolescent reports were used to map out problems in children. Although lescents themselves, it is known that every informants report has its limitations.

the results indicate that the most reliable information comes from mothers and adoParents do not always provide an accurate picture of the problems. First because it the childs behaviour takes place outside the parents range of vision (at school for

is difficult for an outsider to perceive emotional problems, more so because part of example). Adolescents in their turn will not very soon report socially unacceptable behaviour.22 For an overall view on the functioning of children it is advisable to
158

involve informants who have the opportunity to observe children in various circumstances. A teacher for instance or a childs mentor, for a more complete picture. Longitudinal research in which children are followed for more than a year after diagnosis may provide more insight about the increase of problems in a child as a enlarge of diminish the risks of problems in the long term. consequence of the diagnosis cancer in a parent and about the factors that may Next to the factors discussed in this thesis there are many other factors that may

relate to the way a child copes with the event. Another publication of our research stress response symptoms. 23 Other factors needing attention in future research are mediating and moderating factors such as appraisal, coping and parenting, for chiland social contacts outside the family.24,25 drens functioning when confronted with cancer in a parent, and family functioning

group found that the child-parent communication was related to the prevalence of

The present study focused on the negative consequences the diagnosis cancer in a have positive consequences for a child, such as increased sensitivity, empathy, per-

parent may have for children. The confrontation with cancer may possibly also sonal growth, increased feeling of competence and self-respect.26 Future research to the positive consequences of the cancer experience for children.

should therefore not only focus on the negative consequences but also pay attention

Practical implications
Although the majority of the children appears to be resilient in coping with the illness of the parent the remaining minority experiences serious problems. There is no structural focus on the care for children of a parent with cancer in the Netherlands. This study may contribute to improving the skills level of caregivers who are condren. This can be done in several ways. fronted with patients with young children and so incite better care for these chilFirst it is important that parents, where needed, are supported in helping their chil159

dren to cope with this life event. Practice shows that parents often have questions about how best to do so. Parents should be better informed about the possible reactions of sons and daughters in various ages, and what is needed for children to cope. They can possibly observe problems sooner and seek adequate resources for solvthe demands of the illness and the treatment and the care and attention for the chilmoments in the illness process, as e.g. shortly after diagnosis, after recurrence, durwhen the parent experienced serious physical or psychological complaints. ing them. In addition, parents should receive support in finding a balance between dren. This support should especially focus on the consequences of crucial ing heavy treatments or when experiencing complications, as well as on moments It is also important to pay attention to the spouse. As shown the consequences

for children appeared to depend on the functioning of the spouse, who often has an important role in offering consistent and stable family life and in supporting the child. Children under these circumstances are supposed to behave socially acceptable and might be restricted in expressing their feelings openly. By organising meetings children can be informed about the illness and its consequences for the family life in a more free to ask questions and express their fears and uncertainties. Moreover, they

way suitable for their age. Without the presence of their parents, children may feel can exchange experiences with other children in a similar situation. They can learn that their own needs are important too and can be offered tools that satisfy their needs. Chances of a good coping process will be larger when children are offered a out of sight. surrounding in which they are well informed and learn to not lose their own needs A protocol for the way in which children and families are best supported is not available. When offering support the uniqueness of a child and the family surroundfound that suits the needs and customs of that child within the family. ings will have to be taken into account in all situations and a way will have to be

160

References
1. 2. 3. 4. 5. 6. 7. 8. 9. Yahav R, Vosburgh J, Miller A. Emotional responses of children and adolescents to parents with multiple sclerosis. Mult Scler. 2005; 11: 464-468. Forehand R, Biggar H, Kotchick BA. Cumulative risk across family stressors: short- and longterm effects for adolescents. J Abnorm Psychol. 1998; 26: 119-128. Birenbaum LK, Yancey DZ, Phillips DS, Chand N, Huster G. School-age children's and adolescents' adjustment when a parent has cancer. Oncol Nurs Forum. 1999; 26: 1639-1645. Fitch MI, Bunston T, Elliot M. When mom's sick: changes in a mother's role and in the family after her diagnosis of cancer. Cancer Nurs. 1999; 22: 58-63. Christ GH, Siegel K, Sperber D. Impact of parental terminal cancer on adolescents. Am J Orthopsychiatry. 1994; 64: 604-613. Siegel K, Mesagno FP, Karus D, Christ G, Banks K, Moynihan R. Psychosocial adjustment of children with a terminally ill parent. J Am Acad Child Adolesc Psychiatry. 1992; 31: 327-333. Davies T, Lindsay LL. Interparental conflict and adolescent adjustment: why does gender moderate early adolescent vulnerability? J Fam Psychol 2004; 18:160-170. Garside R, Klimes-Dougan B. Socialization of discrete negative emotions: gender differences and links with psychological distress. Sex Roles. 2002; 47: 115-128. Kraemer S. The fragile male. BMJ. 2000; 321: 1609-1612.

ment. J Abnorm Child Psychol. 1998; 26: 199-212. 12. Muris P, Ollendick T. The role of temperament in the etiology of child psychopathology. Clin Child Fam Psychol Rev. 2005; 8: 271-289. 13. Bradley RH, Corwyn RF. Socioeconomic status and child development. Annu Rev Psychol. 2002; 53: 371-399. 14. Von Rueden U, Gosch A, Rajmil L, Bisegger C, Ravens-Sieberer U, the European KIDSCREEN 15. 16. 17. 18. 19. group. Socioeconomic determinants of health related quality of life in childhood and adolescence: results from a European study. J Epidemiol Community Health. 2006; 60: 130-135. Lewis FM, Zahlis EH, Shands ME, Sinsheimer JA, Hammond MA. The functioning of single women with breast cancer and their school- aged children. Cancer Pract. 1996; 4: 15-24. Lipman EL, Boyle MH, Dooley MD, Offord DR. Child well-being in single-mother families. J Am Acad Child Adolesc Psychiatry. 2002; 41: 75-82. Compas BE, Worsham NL, Epping-Jordan JE, Grant KE, Mireault G, Howell DC et al. When mom or dad has cancer: markers of psychological distress in cancer patients, spouses, and children. Health Psychol. 1994; 13: 507-515. Bardwell W, Natarajan L, Dimsdale J, Rock C, Mortimer J, Hollenbach K et al. Objective cancerrelated variables are not associated with depressive symptoms in women treated for early-stage breast cancer. J Clin Oncol. 2006; 24: 2420-2427. Northouse L, Mood D, Kershaw T, Schafenacker A, Mellon S, Walker J et al. Quality of life of women with recurrent breast cancer and their family members. J Clin Oncol. 2002; 20: 40504064.

10. Cumming EM, Davies PT, Simpson KS. Marital conflict, gender and children's appraisal and coping efficacy as mediators of child adjustment. J Fam Psychol. 1994; 8: 141-149. 11. Kerig PK. Moderators and mediators of the effects of interparental conflict on children's adjust-

161

20. Vereniging van Integrale Kankercentra (VIKC). Incidentie nieuw gediagnosticeerde patinten in een jaar [Incidence of newly diagnosed cancer patients]. Cited August 10, 2006. Available on http://www.ikcnet.nl. 2006. 21. Davies PT, Cicchetti D. Toward an integration of family systems and developmental psychopathology approaches. Dev Psychopath 2004; 16: 477-481. 22. Seiffge-Krenke I, Kollmar F. Discrepancies between mothers' and fathers' perceptions of sons' and daughters' problem behaviour: a longitudinal analysis of parent-adolescent agreement on internalizing and externalizing problem behaviour. J Child Psychol psychiatry 1998; 39: 687-397. 23. Huizinga GA, Visser A, van der Graaf WTA, Hoekstra HJ, Hoekstra-Weebers JEHM. The quality of communication between parents and adolescent children in the case of parental cancer. Ann Oncol. 2005; 16: 1956-1961. 24. Compas BE, Hinden BR, Gerhardt CA. Adolescent development: pathways and processes of risk and resilience. Annu Rev Psychol. 1995; 46: 265-293. 25. Marshak L, Seligman M, Prezant F. Disability and the family life cycle: recognizing and treating developmental challenges. xiii ed. New York, NY, US: Basic Books, Inc., 1999. 26. Lewandowski LA. A parent has cancer: needs and responses of children. Pediatr Nurs. 1996; 22: 518-521.

162

163

164

Summary

165

166

Summary

The prevalence rate of cancer is high. In the Netherlands in 2003 more than 72,000 dren living at home. The diagnosis cancer in a parent is an event that affects all fam-

people were diagnosed with cancer. About 9,000 of them are estimated to have chilily members and confronts them with the fragility of life. The months following are

often dominated by diagnostic examinations and one or more treatment modalities. In this period the strain and the uncertainty about the effectivity of the treatment and the survival chances for the patient will prevail. The patient, but the spouse/partner as well, faces the challenge of having to adapt to the menace of the illness and to the Moreover, parents who have children living at home must find a manner to combine the consequences of the illness with housekeeping and parenting tasks.

physical and emotional consequences that go with the illness and the treatments.

This dissertation explores the impact of parental cancer on the functioning of children. The research consists of two parts. Firstly, a cross-sectional study, in which families 1 to 5 years after the cancer diagnosis were approached to participate. Two hundred and five families approached in the hospital and 89 families who had contacted the researchers in response to media attention participated in the study.

Secondly, a longitudinal, prospective study aimed at families during the first year after diagnosis. These families were asked to participate at three moments; within four months after the diagnosis, and 6 and 12 months afterwards. One hundred and

twelve families participated in this study. The family system theory was used to place event touching one family member will impact other family members as well. The child, the parents and the family.

the parental cancer within a larger context. The system theory hypothesizes that an impact the event has for children may depend on characteristics of the illness, the

167

The main purposes of this dissertation were:

1. To examine the prevalence of emotional and behavioural problems in primary school

(aged 4-11 years) and adolescent (aged 12-18 years) sons and daughters of a parent parent.

diagnosed with cancer, by comparing them to children not confronted with cancer in a 2. To investigate risk factors for the prevalence of problems in children, namely the childs socio-demographic variables and temperament; ill parents and spouses demographics and quality of life; illness-related variables, and experienced life events.

Chapter 2 contains a literature review examining the functioning of children whose par-

ent was diagnosed with cancer and the relationship between the prevalence of problems and characteristics of the child, the parent and the family. The search was aimed at English language studies published between January 1980 and March 2004 and result-

ed in fifty-two studies addressing the psychosocial functioning of children aged 0-20

years. The studies varied in design: fourteen studies used a quantitative design, eighteen intervention study. The results of almost all quantitative studies showed emotional probmothers. Evidence of emotional problems in primary school children (4-11 years) was

a qualitative design and thirteen a combination of the two. Seven studies involved an lems in adolescent children (12 years), in particular in adolescent daughters of ill

lacking in almost all quantitative studies. Qualitative studies however found emotional

problems in children of all ages. Quantitative studies generally showed no behavioural

and social problems, whereas qualitative studies found behaviour problems in primary school children and social and cognitive problems in children of all ages. A few of these studies paid attention to the impact of characteristics of the child, the parent or the fampsychological functioning, marital satisfaction and family communication. A better

ily. The most consistent variables related to child functioning appeared to be parental parental functioning, more marital satisfaction and a better communication within the

family decreased the problems in children. The limited number of intervention studies anxiety and better communication. However, because the studies are not randomized
168

directed to the needs of children and their families reported positive effects, such as less

clinical trials results may not be considered as evidence-based. The conclusion therefore their parent is diagnosed with cancer.

can be that high quality research is needed to examine the problems in children when Chapter 3 describes the prevalence of emotional and behavioural problems in 336 chil-

dren 4 to 18 years of age, from 186 families in which a parent had been diagnosed with functioning parents with cancer and their spouses were asked to complete a question-

cancer between 1 to 5 years prior to study entry. To get a comprehensive picture of child naire on the functioning of their children and children from the age of 11 years and older

were asked to complete a self-report questionnaire. The prevalence of problems in chilresults showed that primary school sons and adolescent daughters experienced more problems than their peers. Although the adolescent daughters were also in earlier stud-

dren was compared with that of children in a randomly selected Dutch sample. The

ies identified as vulnerable, this was not earlier found for sons of the primary school age. important as well. Previous studies identified children of ill mothers in particular as vulnerable. In the current study however children of ill fathers appeared to have more prob-

Except for the effects of age and gender of children, the gender of the parent appeared

lems than those of ill mothers. Adolescent sons and daughters seemed to have more problems when the parent had experienced a recurrence of the illness or had received a more intensive treatment, whereas this was not found for time since diagnosis and duration of treatment. No relationships were found for illness-related variables and the spondence was found between the reports of fathers and mothers on their childrens

prevalence of problems in primary school children. Finally, a moderate to high correfunctioning, whereas this was low to moderate between parents and children. Nevertheless mothers reported a level of problems in their children comparable to what the children reported themselves, whereas fathers reported a lower level of problems than their children. Results suggest that subgroups of children of a parent diagnosed and potential influencing variables varied according to informants.

with cancer are at risk for emotional problems. The perception of childs functioning Chapter 4 discusses the extent to which the characteristics of parents diagnosed with cancer and those of the partner were related to the prevalence of emotional and behav169

ioural problems in children. Attention is paid to the effects of socio-demographic

variables, characteristics of the illness, physical and mental functioning of the parents on the prevalence of problems in children. The study is based on the reports of 180 parents 1 to 5 years after the cancer diagnosis and those of their 145 partners. These parents reported on 114 primary school children and 222 adolescents. Results show that

primary school children experienced more emotional and behavioural problems when living in a single parent family, had no or fewer siblings or were the oldest child, whereas this was not found for adolescents. Adolescents functioning was more affected when ill parents experienced treatment complications. Moreover, a decrease in ill parents physical functioning affected primary school daughters and adolescents, and adolesnot have a different impact on children than those of ill mothers. Spouses physical lim-

cents were also affected by the mental health of ill parents. Problems of ill fathers did itations were indicative for problems in primary school children, whereas a worsening parental mental health was indicative for problems in adolescents. These results emphasize that family- and illness-related variables and the functioning of parents may affect the problems children experience. Which characteristics have the higher impact depends on the age of children.

Chapter 5 pays attention to the relationships between temperament and the prevalence of emotional and behavioural problems in adolescents, 1 to 5 years after the cancer diagcer. To get an accurate picture of the relationships between temperament and the prevanosis. Results are based on 340 adolescents and their 212 parents diagnosed with canlence of problems, we controlled for socio-demographic and illness-related variables and the number of life-events that were experienced in the year before study participation. The extent of problems appeared not related to the age and educational level of the increasing number of negative life events experienced by children and parents. Over and above the effect of these factors, temperament affected also the prevalence of problems in children, particularly when problems were reported by children themselves. Shyness problems were related to frustration and perceptual sensitivity in the reports of children
170

child or parent. More problems were reported in case of recurrent disease and an

and fear/worry were strongest related to emotional problems. Furthermore, emotional

and to pleasure intensity in the reports of parents. Behavioural problems were associated with effortful control, and only in childrens reports with frustration. The results of lems in children. this study suggest that temperament is an important predictor of the prevalence of probThe results of chapter 6 are based on the information obtained from the prospective, lon-

gitudinal study. This study was performed to gain insight in the extent of emotional and

behavioural problems shortly after the cancer diagnosis and in the changes of these problems during the first year. Results are based only on those families who participatfour months after the cancer diagnosis and 6 and 12 months later. Sixty nine parents ed at the three measurements, namely on those who completed questionnaires within with cancer and 57 spouses reported on the prevalence of emotional and behavioural problems in 57 primary school children and 66 adolescents. Adolescents completed questionnaires on their own functioning. The prevalence of problems was compared According to mean scores, children were reported to have a similar or even a lower level

with the problems reported in the cross-sectional study (Chapter 3) and the norm group. of problems than the two reference groups. Scores on the individual level however

showed that one of four or five children appeared to have clinically-elevated scores. The found between fathers and mothers on the level of problems in primary school children

number of these children decreased during the year. High levels of agreement were and adolescent sons, but the agreement was low for adolescent daughters. Levels of

agreement between fathers and adolescents were low, but was higher between mothers lems during the first year after the parents diagnosis than children in the retrospective drop out of children during the study, may have led to a selection bias. This suggests that children in this situation.

and daughters. The outcomes suggest that most children do not experience more probstudy or the norm group. However, the non-response at the start of the study and the the results of the current study give an underrepresentation of the extent of problems in Chapter 7 describes and discusses the main findings of this dissertation. Methodological

limitations are discussed. Recommendations are made for future research and for health care workers who are working with families in which a parent is diagnosed with cancer.

171

172

Samenvatting

173

174

Samenvatting

Kanker is een veel voorkomende ziekte. In 2003 werd in Nederland bij meer dan

72.000 mensen kanker vastgesteld. Geschat wordt dat het hier in 9.000 gevallen om mensen met thuiswonende kinderen gaat. De diagnose kanker bij de ouder is een kwetsbaarheid van het leven. Vaak staan de maanden daaropvolgend in het teken heid of de behandeling zal aanslaan en de patint zal overleven. De patint, maar meebrengt en aan de fysieke en emotionele gevolgen van de ziekte en de behandeden om de gevolgen van de ziekte te combineren met huishoudelijke en opvoedingstaken. gebeurtenis die alle gezinsleden treft. Gezinnen worden geconfronteerd met de van onderzoeken en behandelingen. In deze periode heerst de spanning en onzekerook de partner zullen zich moeten aanpassen aan de dreiging die de ziekte met zich ling. Als er thuiswonende kinderen zijn moet daarnaast een manier worden gevon-

In dit proefschrift wordt aandacht besteed aan de gevolgen die de diagnose kanker

bij de ouder voor kinderen kan hebben. Het onderzoek bestaat uit twee delen.

Allereerst een cross-sectionele studie, waarin gezinnen n tot vijf jaar na diagnose voor deelname zijn benaderd. Aan deze studie hebben 205 gezinnen deelgenomen die zijn benaderd in het ziekenhuis en 89 gezinnen die op eigen initiatief contact hebben opgenomen om aan het onderzoek deel te nemen. De tweede studie jaar nadat de diagnose kanker is gesteld. Deze gezinnen werd gevraagd vier maanomvat een longitudinale, prospectieve studie, gericht op gezinnen tijdens het eerste den nadat de diagnose was gesteld en zes en twaalf maanden daarna een vragenlijst in te vullen. Aan deze studie hebben 112 gezinnen deelgenomen. De resultaten van dit proefschrift zijn geplaatst binnen het kader van de systeemtheorie. In de systeemtheorie wordt verondersteld dat een gebeurtenis in het leven van een gezinslid niet alleen dat gezinslid, maar ook het leven van de andere gezinsleden treft. De mate waarin kinderen reageren op deze gebeurtenis kan samenhangen met kenmerken van de ziekte, kenmerken van het kind en de ouders en met kenmerken van het gezin.

175

De belangrijkste doelen van dit proefschrift waren:

1. Vaststellen of thuiswonende kinderen (4-18 jaar) van een ouder gediagnosticeerd niet met een dergelijke stressor zijn geconfronteerd.

met kanker meer emotionele en/of gedragsproblemen ervaren dan kinderen die

2. Onderzoeken in welke mate sociaal-demografische kenmerken en het tempera-

ment van kinderen, sociaal-demografische kenmerken en kwaliteit van leven van de ouder met kanker en de partner, ziektegerelateerde factoren, het meemaken van andere negatieve gebeurtenissen samenhangen met het vrkomen van problemen bij kinderen.

Hoofdstuk 2 omvat een review studie naar het functioneren van kinderen van een ouder gediagnosticeerd met kanker. Vervolgens wordt beschreven in hoeverre in de kind en de kenmerken van het kind, de ouder en het gezin. De zoekactie was gericht Deze zoekactie resulteerde in 52 studies met aandacht voor het psychosociaal funcgereviewde studies aanwijzingen bestaan voor een relatie tussen problemen bij het op alle Engelstalige onderzoeken, gepubliceerd tussen januari 1980 en maart 2004. tioneren van kinderen in de leeftijd van 0-20 jaar. Studies varieerden in design:

14 maakten gebruik van een kwantitatief design, 18 van een kwalitatief design en 13 van een combinatie van deze methoden. Zeven studies beschreven een intervenproblemen bij adolescente kinderen (12-18 jaar), vooral bij adolescente dochters van een moeder met kanker. Bewijs voor emotionele problemen bij basisschoolkintiestudie. De resultaten van vrijwel alle kwantitatieve studies wezen op emotionele

deren (4-11 jaar) ontbrak in vrijwel alle kwantitatieve studies. Kwalitatieve studies

lieten echter zien dat emotionele problemen bij kinderen in alle leeftijdgroepen voorkomen. Tevens vonden de kwantitatieve studies geen bewijs voor het vrkomen bij basisschoolkinderen en sociale en cognitieve problemen bij kinderen van alle leeftijden vonden. In een aantal van de tot dan toe uitgevoerde studies werd aandacht besteed aan het effect van karakteristieken van het kind, de ouder en het gezin op het vrkomen van problemen. De meest consistente resultaten betroffen
176

men van gedrags- en sociale problemen, terwijl kwalitatieve studies gedragsproble-

het psychisch functioneren van de ouder, de tevredenheid van de ouders met de (huwelijks)relatie en de gezinscommunicatie. Een beter functioneren van de ouder, meer tevredenheid met de huwelijksrelatie en een betere gezinscommunicatie leid-

den tot minder problemen bij de kinderen. Het beperkte aantal interventiestudies

dat zich heeft gericht op de behoeftes van kinderen en hun families in deze situatie vond positieve effecten, zoals minder angst en een verbeterde communicatie. Echter, omdat deze studies geen gerandomiseerde of gecontroleerde trials waren is de effectiviteit van deze interventies niet empirisch onderzocht. Geconcludeerd werd dat meer kwalitatief hoogwaardig onderzoek nodig is om problemen van kinbrengen en na te gaan welke kinderen meer risico lopen. deren die geconfronteerd worden met de diagnose kanker bij de ouder in kaart te In hoofdstuk 3 wordt aandacht besteed aan het vrkomen van emotionele en gedragsproblemen bij 336 kinderen in de leeftijd van 4-18 jaar, afkomstig uit 186 gezinnen, waarvan bij de ouder n tot vijf jaar geleden kanker is vastgesteld. Om een zo volledig mogelijk beeld te krijgen van het functioneren van de kinderen werd de ouder gediagnosticeerd met kanker en de partner gevraagd hierover een vragenlijst in te vullen. Kinderen vanaf 11 jaar werden daarnaast gevraagd een vra-

genlijst over zichzelf in te vullen. Het vrkomen van problemen bij kinderen in samengestelde normgroep. De resultaten lieten zien dat zonen in de basisschoolgenoten. Hoewel adolescente dochters in meerdere studies als kwetsbaar werden gedentificeerd, was dit voor zonen in de basisschoolleeftijd niet eerder aangetoond. Behalve de leeftijd en het geslacht van het kind, bleek ook het geslacht van

onze studie werd vergeleken met de problemen bij kinderen in een at random leeftijd en adolescente dochters meer emotionele problemen ervaren dan leeftijds-

de ouder van belang. Eerdere studies wezen voornamelijk kinderen van zieke moemet kanker meer problemen te hebben. Bovendien bleek een recidief of een inten-

ders als kwetsbaar aan, maar in de huidige studie bleken kinderen van een vader sieve behandeling samen te hangen met meer problemen bij adolescente zonen en dochters. Geen effect werd echter gevonden voor de tijd verstreken sinds de diagnose en voor de duur van de behandeling. Geen van de ziektegerelateerde vari177

abelen bleek gerelateerd aan het functioneren van kinderen in de basisschoolleef-

tijd. Tot slot, het vergelijken van de informatie over het functioneren van de kinde-

ren verkregen van verschillende informanten wees uit dat het functioneren verschil-

lend werd gepercipieerd. Hoewel een redelijke overeenstemming tussen ouders

bestaat, is de overeenstemming tussen ouders en kinderen laag tot gemiddeld.

Echter, moeders en kinderen bleken een overeenkomstig niveau van problemen te suggereren dat kinderen een verhoogd risico lopen op emotionele problemen als een ouder gediagnosticeerd wordt met kanker. Informanten verschillen in de mate waarin en welke factoren het risico op problemen vergroten. In hoofdstuk 4 wordt beschreven in hoeverre kenmerken van de ouder gediagnosdie bij kinderen werden gerapporteerd. Aandacht werd besteed aan het effect van

rapporteren, terwijl vaders beduidend minder problemen rapporteerden. Resultaten

ticeerd met kanker en van de partner samenhangen met de mate van problemen sociaal-demografische variabelen, ziektekenmerken, het fysiek en psychisch functioneren van ouders op het vrkomen van problemen bij kinderen. De studie baseerde zich op de rapportage van 180 ouders die n tot vijf jaar terug zijn gediag-

nosticeerd met kanker en op die van hun 145 partners. Deze ouders rapporteerden over 114 basisschoolkinderen en 222 adolescenten. Resultaten lieten zien dat kinderen in de basisschoolleeftijd meer problemen hadden als ze uit een eenoudergezin

kwamen, minder broertjes en/of zusjes hadden of het oudste kind in het gezin waren, terwijl dit niet werd gevonden voor adolescente kinderen. Adolescenten blegevolg van de behandeling had ervaren. Het fysiek functioneren van de ouder met ken meer emotionele en gedragsproblemen te ervaren als de ouder complicaties als kanker bleek tevens gerelateerd aan meer problemen bij dochters in de basisschool-

leeftijd en adolescenten, terwijl het psychisch functioneren alleen gerelateerd was aan het vrkomen van problemen bij adolescenten. Een verminderd functioneren slechter fysiek functioneren van de partner was gerelateerd aan het vrkomen van van de zieke vader had een overeenkomstige impact als dat van de moeder. Een problemen bij kinderen in de basisschoolleeftijd, terwijl een slechter psychisch functioneren gerelateerd was aan het vrkomen van problemen bij adolescenten.
178

Deze resultaten wijzen erop dat gezins- en ziektekenmerken en het functioneren van de ouders een bijdrage kunnen leveren aan de mate waarin kinderen problemen leeftijd van het kind. ervaren. Welke kenmerken de meeste impact hebben hangt voornamelijk af van de In hoofdstuk 5 wordt aandacht besteed aan de relatie tussen temperament en het

vrkomen van emotionele en gedragsproblemen bij adolescente kinderen, n tot vijf jaar na diagnose. Resultaten zijn gebaseerd op de rapportage van 340 adolescente kinderen en hun 212 ouders gediagnosticeerd met kanker. Om de relatie tussen temperament en het vrkomen van problemen goed te kunnen vaststellen, werd gecontroleerd voor sociaal-demografisch en ziektegerelateerde variabelen, en het aantal life-events dat in het jaar voorafgaand aan de deelname aan het onderzoek zijn voorgevallen. De mate van problemen die werden gerapporteerd bleken werden meer problemen gerapporteerd als de ouder een recidief had en wanneer

niet gerelateerd aan leeftijd en opleidingsniveau van het kind en/of de ouder. Wel meer negatieve life-events hadden plaatsgevonden. Ondanks de bijdrage van deze factoren, was ook temperament gerelateerd aan het vrkomen van problemen bij kinderen. Verlegenheid en angst waren het sterkst gerelateerd aan emotionele problemen. Daarnaast waren emotionele problemen gerelateerd aan frustratie, percep-

tuele gevoeligheid en plezierintensiteit. De gedragsproblemen werden geassocieerd met effortful control en frustratie. De uitkomsten van deze studie veronderstellen blemen bij kinderen voorkomen. dat temperament een belangrijke voorspellende factor is voor de mate waarin proDe resultaten van hoofdstuk 6 zijn gebaseerd op informatie verkregen uit de prospectieve, longitudinale studie. In dit hoofdstuk wordt de mate van problemen bij kinderen kort na de diagnose en de verandering daarvan in de loop van het eerste

jaar beschreven. Resultaten zijn gebaseerd op die gezinnen die op alledrie de meetmomenten, namelijk binnen vier maanden na diagnose en zes en twaalf maanden en 57 partners rapporteerden over het vrkomen van emotionele en gedragsprodaarna, aan het onderzoek hebben deelgenomen. Negenenzestig ouders met kanker blemen bij 57 kinderen in de basisschoolleeftijd en 66 adolescente kinderen.
179

Adolescente kinderen vulden daarnaast vragenlijsten in over het eigen functione-

ren. Het vrkomen van problemen bij deze groep kinderen werd vergeleken met groep. Op groepsniveau bleken kinderen op alle drie de meetmomenten evenveel of Uitgaande van de individuele scores bleek een op de vier of vijf kinderen klinisch verhoogde scores te hebben. Dit aantal nam echter in de loop van het jaar af. Er

de mate van problemen in de cross-sectionele studie (hoofdstuk 3) en de normzelfs minder problemen te ervaren dan in beide referentiegroepen werd gevonden.

bleek veel overeenstemming tussen vaders en moeders over de mate van problemen bij basisschoolkinderen en adolescente zonen. De overeenstemming tussen ouders bleek echter laag voor adolescente dochters. De overeenstemming tussen vaders en de overeenstemming tussen moeders en adolescente dochters hoog.

adolescente kinderen en moeders en adolescente zonen was laag; hierentegen was Het lijkt wel of kinderen het eerste jaar na diagnose niet meer problemen hebben dan kinderen in de retrospectieve studie en de normgroep. Echter, als gevolg van de non-response bij de start van de studie en de uitval van kinderen in de loop van de dat een onderrapportage is ontstaan van de problemen die bij kinderen in deze situatie voorkomen. In hoofdstuk 7 worden de belangrijkste bevindingen van dit proefschrift beschre-

studie, kan zich een selectiebias hebben voorgedaan. Dit kan ertoe hebben geleid

ven en bediscussieerd. De methodologische beperkingen van deze studie worden beschreven. Tot slot worden aanbevelingen gedaan voor toekomstig onderzoek en ouder met kanker is gediagnosticeerd. voor hulpverleners die in hun werk te maken hebben met gezinnen waarin een

180

181

182

Dankwoord

183

184

Dankwoord
Jaren geleden stapte ik als groentje in een voor mij volledig onbekende onderzoekswereld. En dat was niet altijd even gemakkelijk. Het schrijven van een proefschrift is bij tijd en wijle een worsteling waarvoor enig doorzettingsvermogen een absolute vereiste is. Desondanks kijk ik terug op een periode waarin ik op diverse terreinen verschuldigd. veel heb geleerd en me persoonlijk heb ontwikkeld. Ik ben hierbij veel mensen dank Allereerst gaat dank uit naar de gezinnen die aan het onderzoek hebben deelgenomen. Aan het begin van het traject heb ik veel gesproken met ouders met kanker en hun kinderen. Ik heb erg veel bewondering voor de manier waarop deze gezinnen omgaan met

zon ingrijpende gebeurtenis. Ondanks de hectische tijd die veel gezinnen doormaken hebben ouders en kinderen tijd vrijgemaakt om de uitgebreide vragenlijsten in te vullen. De jongere kinderen hebben de betekenis van de gebeurtenis in een tekening of een verhaal weergegeven. Een aantal van deze creaties is opgenomen in dit proefschrift. De begeleiders dr. J.E.H.M. Hoekstra-Weebers, Prof. dr. H.J. Hoekstra en Prof. dr. eigen manier aan dit proefschrift hebben geleverd. Zonder de inzet en bijdrage van de gezinnen was dit onderzoek niet mogelijk geweest. W.T.A. van der Graaf wil ik hartelijk bedanken voor de bijdrage die ze elk op hun Mijn eerste begeleidster en co-promotor. Josette, we hebben intensieve jaren met het schrijven van artikelen. Je bent ontzettend kritisch en het is een lange weg voor-

elkaar meegemaakt. Ik heb veel van je geleerd op het gebied van onderzoek doen en dat iets aan jouw eisen voldoet. Je hebt veel tijd en energie in dit project gestoken, het

is dan ook mede dankzij jouw inzet dat dit proefschrift tot een goed einde is gekomen. Mijn eerste promotor Harald, jouw kennis als clinicus was belangrijk bij de omgang met de medische gegevens in het onderzoek. Je hebt een actieve rol gespeeld bij de

organisatie van het onderzoek en mede dankzij jouw inzet hebben we een groot aantal patinten kunnen includeren. Jouw pragmatische en no-nonsense aanpak tijdens
185

het hele onderzoek heb ik erg gewaardeerd.

Mijn tweede promotor Winette, jouw deur stond altijd open om me een weg te

wijzen in de complexe wereld van de oncologie. Daarnaast heb je me gedurende het Je hebt me altijd gestimuleerd om kort en bondig te formuleren, en niet te vervallen in wollig taalgebruik. Volgens mij is er progressie?! Prof. dr. E.C. Klip, was onze eerste promotor. Helaas is het promotietraject niet op tijd afgerond. Ed, we hebben vooral de eerste jaren van het project contact gehad. Het leven

onderzoek aangemoedigd de betekenis voor de praktijk niet uit het oog te verliezen.

van een onderzoeker is zo slecht nog niet met een begeleider die je op een zonnige midover een artikel. Jammer dat je niet meer als eerste promotor betrokken kon zijn.

dag uitnodigt in de tuin, om onder het genot van een heerlijke lunch te discussiren Ineke Bakker was onze onderzoeksassistent. Beste Ineke, het was vast voor jou wel even schrikken om met twee van die chaoten te moeten gaan samenwerken. Maar wat hadden wij zonder jou gemoeten. De logistiek van het onderzoek vereiste een strakke organisatie, dat was zonder jouw orde en accurate manier van werken vast niet gelukt. Daarnaast hebben we ook een gezellige tijd gehad, bedankt daarvoor. Ook de (oncologie)verpleegkundigen en artsen van de afdelingen chirurgische onco-

logie, interne oncologie, radiotherapie en gynaecologie van het Universitair Medisch Centrum Groningen, Medisch Centrum Leeuwarden, Scheper Ziekenhuis te Emmen, Martiniziekenhuis te Groningen, Rpcke-Zweers ziekenhuis te Hardenberg en het getoond bij het benaderen en informeren van patinten.

IJsselmeerziekenhuis te Emmeloord wil ik bedanken voor de inzet die zij hebben Collegas van de afdeling Medische Psychologie, de DDQ en ons eigen leesclubje die zich tijdens de verschillende onderzoeksbijeenkomsten hebben voorgedaan.

(Joke, Marrit en Ellen) wil ik bedanken voor de betrokkenheid en de leermomenten Willem Lok, Roy Stewart en Eric van Sonderen wil ik bedanken voor de ondersteu-

ning die ze hebben geboden bij het maken van invoerprogrammas en/of het discussiren over de meeste geschikte statistische methode bij het analyseren van gegevens. Ria Molanus, ook jou wil ik bedanken. Ondanks het enthousiasme over de net geboren tweeling, was je bereid me te helpen de laatste teksten om te zetten in begrijpe186

lijk en leesbaar engels.

De leden van de leescommissie, Prof. dr. J.W. Groothoff, Prof. dr. J.L.N. Roodenburg, het proefschrift.

Prof. dr. P.H.B. Willemse wil ik bedanken voor het lezen en becommentariren van

Mijn huidige collegas van de sectie Toegepast Onderzoek en Public Health wil ik bedanken. Ik ben erg blij in zon stimulerende en gezellige omgeving te kunnen werken. Danielle, hoeveel lunchen ben ik je verschuldigd na al de afgesproken deadlines

die ik weer niet haalde. En Johan, het ging jou allemaal veel te traag, dus bood jij je Bedankt voor je enthousiasme en stimulans. Geke, ik heb het enorm naar mijn zin bij de sectie Toegepast Onderzoek en met jou als hoofd. Bedankt voor de ruimte die je en overal te willen helpen.

maar aan voor de leescommissie zodat je af en toe nog enigszins druk kon uitoefenen.

me de afgelopen tijd hebt gegeven mijn proefschrift af te ronden en je bereidheid altijd Tot slot, sommige mensen hebben een speciale bijdrage geleverd aan het realiseren van dit proefschrift. Allereerst natuurlijk Gea. We hebben samen intensieve jaren meeje gehad en van je geleerd. Je bent erg bijzonder! gemaakt, waarin we een bijzondere vriendschap hebben opgebouwd. Ik heb veel aan Rahul, ook jij hebt de afgelopen jaren veel voor me betekend. Ga nu eindelijk jouw wijze raad eens opvolgen en wat meer van mijn vrije tijd genieten!! Papa en mama jullie waren er altijd en zijn altijd in me blijven geloven. Het doorzettingsvermogen dat me er de afgelopen jaren doorheen heeft geholpen heb ik aan jullie te danken.

Familie en vrienden, jullie wil ik bedanken voor de eeuwige steun, het geduld om al zich hebben voorgedaan.

die jaren die promotieverhalen aan te horen en de leuke ontspannende momenten die En natuurlijk jij, Wiebe. Ondanks alle veranderingen heb je mij de ruimte gegeven om mijn ding te doen. Je bent heel erg belangrijk voor mij. Ik hoop dat we de komenbent nog zo klein, jij laat me nu al zien waar het in het leven om draait. de jaren wat minder life-events en wat meer tijd voor elkaar hebben. Lieve Thijn, je

187

188

Articles related to the project


Gea A Huizinga, Winette TA van der Graaf, Annemieke Visser, Jos S Dijkstra, Josette EHM Hoekstra-Weebers. Psychosocial consequences for children of a parent with cancer: a pilot study. Cancer Nursing. 2003; 26: 195-202 Gea A Huizinga, Annemieke Visser, Winette TA van der Graaf, Harald J Hoekstra, toms in adolescent and young adult children of parents diagnosed with cancer. European Journal of Cancer. 2005; 41: 288-295

Ed C Klip, Elisabeth Pras, Josette EHM Hoekstra-Weebers. Stress response symp-

Gea A Huizinga, Annemieke Visser, Winette TA van der Graaf, Harald J Hoekstra, Josette EHM Hoekstra-Weebers. The quality of communication between parents and adolescent children in the case of parental cancer. Annals of Oncology. 2005; 16: 1956-1961

Gea A Huizinga, Annemieke Visser, Winette TA van der Graaf, Harald J Hoekstra, Roy E Stewart, Josette EHM Hoekstra-Weebers. Predictors of the psychological Submitted functioning of adolescents who have a parent with cancer: a multilevel approach.

Gea A Huizinga, Annemieke Visser, Winette TA van der Graaf, Harald J Hoekstra, Ed C. Klip, Stacey Gazendam-Donofrio, Josette EHM Hoekstra-Weebers. cancer diagnosis. Submitted Posttraumatic stress symptoms in adolescents during the first year after a parents

189

190

NORTHERN CENTRE for HEALTHCARE RESEARCH (NCH)


This thesis is published within the research program Disorder, Disability and Quality of Life of the Northern Centre for Healthcare Research. More information regarding the institute and its research can be obtained from our internetsite: www.med.rug.nl/nch. Previous dissertations from the program Disorder, Disability and Quality of Life Huizinga GA (2006) The impact of parental cancer on children. PROMOTOR: prof dr HJ Hoekstra COPROMOTORES: dr JEHM HoekstraWeebers, dr WTA van der Graaf Fleer J (2006) Quality of life of testicular cancer survivors. PROMOTORES: prof dr HJ Hoekstra, prof dr DTh Sleijfer, prof dr EC Klip COPROMOTOR: dr JEHM Hoekstra-Weebers Arnold R (2004) Quality of life in chronic obstructive pulmonary disease and chronic heart failure. PROMOTORES: prof dr R Sanderman, prof dr GH Koter COPROMOTORES: dr AV Ranchor, dr MJL De Jongste Keers JC (2004) Diabetes rehabilitation; effects and utilisation of a multidisciplinary education programme. PROMOTORES: prof dr R Sanderman, prof dr ROB Gans COPROMOTORES: dr J Bouma, dr TP Links Pool G (2003) Surviving testicular cancer; sexuality and other existential issues. PROMOTORES: prof dr HBM van de Wiel, prof dr DTh Sleijfer, prof dr HJ Hoekstra COPROMOTOR: dr MF van Driel Stiegelis HE (2003) A life less ordinary; cognitive adaptation and psychological functioning among cancer patients treated with radiotherapy. PROMOTORES: prof dr R Sanderman, prof dr AP Buunk COPROMOTOR: dr M Hagedoorn Heuvel ETP van den (2002) Supporting caregivers of stroke patients; an intervention study. PROMOTORES : prof dr B Meyboom-de Jong, prof dr R Sanderman COPROMOTOR: dr LP de Witte REFERENT: dr LM Schure Schroevers MJ (2002) Short- and long-term adaptation to cancer; a comparison of patients with the general population. PROMOTOR: prof dr R Sanderman REFERENT: dr AV Ranchor Hoekstra-Weebers JEHM (2000) Parental adaptation to pediatric cancer. PROMOTORES: prof dr EC Klip, prof dr WA Kamps REFERENT: dr JPC Jaspers Doeglas DM (2000) Functional ability, social support and quality of life: a longitudinal study in patients with early rheumatoid arthritis. PROMOTORES: prof dr WJA van den Heuvel, prof dr R Sanderman COPROMOTOR: dr ThPBM Suurmeijer Nijboer C (2000) Caregiving to patients with colorectal cancer: a longitudinal study on caregiving by partners. PROMOTORES: prof dr GAM van den Bos, prof dr R Sanderman CO-PROMOTOR: dr AHM Triemstra Tiesinga LJ (1999) Fatigue and Exertion Fatigue: from description through validation to application of the Dutch Fatigue Scale (DUFS) and the Dutch Exertion Fatigue Scale (DEFS). PROMOTORES: prof dr WJA van den Heuvel, prof dr ThWN Dassen CO-PROMOTOR: dr RJG Halfens Jong GM de (1999) Stress, stress management and issues regarding implementation. PROMOTORES: prof dr PMG Emmelkamp, prof dr JL Peschar REFERENT: dr R Sanderman Alberts JF (1998) The professionalized patient: sociocultural determinants of health services utilization. PROMOTOR: prof dr WJA van den Heuvel REFERENT: Dr R Sanderman

191


192

Citaten ouders Ik heb mijn dochter zelf een paar keer meegenomen naar bestraling en onderzoek. Daardoor kreeg ze uitleg en dat maakt dat allerlei wilde beelden bij haar realistischer werden. Het werkte goed, maar het initiatief lag bij mij en ik moest om toestemming vragen. Tijdens mijn ziekte had Ameline (nu 14) heel veel behoefte om met lotgenoten te praten/schrijven. Was heel moeilijk om hier iemand voor te vinden. Dat was jammer De dreiging is voorbij en ik ben toch wel enigszins verrast dat zij haar leventje weer heeft opgepakt .Ik heb professionele ondersteuning vooral tijdens de behandeling gemist. Ik had graag gewild dat iemand had aangeboden om af en toe met haar te praten

Door de professionele hulpverleners is er nooit aandacht aan de kinderen besteed, het maakt eenzaam daarin. Gewoon eens er naar vragen of aandacht aan besteden. Veel hoeft niet, maar wel eens alleen er naar vragen. De kinderen zijn soms wat teruggetrokken. Nadenkender. Maar meestal gewoon doorgaand. Tijdens de ziekte van mijn man is mn zoontje een keer op school erg agressief geweest, wat heel ongewoon was. Leerkracht was gelukkig op de hoogte.

Moeilijkste voor de kinderen: de angst voor kanker. Het verdriet mij te moeten missen tijdens de ZHS opname. Ook tijdens de chemokuren liep alles anders, 1e week bij hun vader, elke dag na school en tussen de middag wel even bij mij. De haaruitval was n van de moeilijkste dingen, dat vonden de kinderen heel erg!

Ik maak me zorgen over hoe zij mijn overlijden, wat te vroeg zal komen, zullen verwerken en mee om zullen gaan. Het gevoel van onveiligheid was voor de kinderen het moeilijkste. De angst om weer te verliezen en de machteloosheid van toezien hoe ziek ik was van de chemokuren. Ook een stuk ontregeling van het leven dat weer opgepakt was. Maar dat is wat ik denk, ze hebben dat niet laten zien/geuit.

De oudste was 14,5. Hij werd een spijbelaar. Zei dan dat hij met mij naar de arts moest enzon hele dwarse periode (ging roken, blowen, t had toch geen zin meer). Ze waren geschokt toen ze het hoorden. Dat hun supermamma dit had, waarom dan? Heel troostend ook, je bent nog hartstikke mooi en lief met n borst. Eigenlijk om te huilen.

You might also like