Professional Documents
Culture Documents
PII: S0020-7489(19)30261-5
DOI: https://doi.org/10.1016/j.ijnurstu.2019.103454
Reference: NS 103454
Please cite this article as: Peter Johansson , Tiny Jaarsma , Gerhard Andersson , Johan Lundgren ,
The impact of internet-based cognitive behavioral therapy and depressive symptoms on self-care be-
havior in patients with heart failure. A secondary analysis of a randomised controlled trial, International
Journal of Nursing Studies (2019), doi: https://doi.org/10.1016/j.ijnurstu.2019.103454
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
1
Department of Social and Welfare Studies, Linköping University, Norrköping, Sweden.
2
Department of Internal Medicine, Department of Medical and Health Sciences, Linköping
University, Norrköping, Sweden.
3
Department of Behavioural Sciences and Learning, Linköping University, Linköping,
Sweden.
4
Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
Postal address:
Peter Johansson, PhD, Professor, RN
Department of Social and Welfare Studies, Linköping University
Campus Norrköping, SE- 601 74. Norrköping, Sweden
E-mail: peter.b.johansson@liu.se
Phone: +46-11363185
Contribution of the paper
barrier to performing self-care behaviours. This may also be a possible explanation for
why depressive symptoms in heart failure are associated with an impaired quality of
There are studies reporting that psychological interventions have possible positive
short-term impact on self-care in heart failure. However, most of these studies did not
include heart failure patients who also have elevated levels of depressive symptoms.
targeting depressive symptoms in patients with heart failure can improve self-care and
consulting behaviour.
Background: Patients with chronic heart failure may require treatment of depressive
based cognitive behavioural therapy (CBT) on self-care behaviour in heart failure patients,
and to study the association between changes in depressive symptoms and changes in self-
controlled study. Setting: 50 heart failure patients with depressive symptoms were recruited
from four hospitals in Sweden. Methods: Patients were randomized to nine weeks of internet-
based CBT (n=25) or to an active control group participating in an online discussion forum
(n=25). In week two and three, those in the internet-based CBT group worked with
psychoeducation about heart failure and depression, emphasizing heart failure self-care.
During the same weeks those in the on-line discussion forum specifically discussed heart
failure self-care. Patient Health Questionnaire-9 was used to measure depressive symptoms at
baseline and at the nine-week follow-up. The European Heart Failure Self-care Behaviour
Scale-9 was used to measure self-care behaviour (i.e., the summary score and the subscales
autonomous based, provider based and consulting behaviour) at baseline, and at the three-
week and nine-week follow-ups. Results: No significant differences were found in self-care
between the patients in the internet-based CBT and the patients in the online discussion group
at the three- and nine-week follow-up. Within-group analysis of the changes in the European
Heart Failure Self-care Behaviour Scale showed that from baseline to week three, the
summary score increased significantly for the online discussion group (p=0.04), but not for
the internet-based CBT group (p=0.15). At the nine-week follow-up, these scores had
decreased. Similarly, consulting behaviour improved at week three for the online discussion
group (p=0.04), but not for the internet-based CBT group (p=0.22). Provider-based adherence
at the nine-week follow-up had increased from baseline in the internet-based CBT group
(p=0.05) whereas it had decreased in in the on-line discussion group. Improvement in
autonomous-based self-care. ICBT for depression in HF may benefit aspects of self-care that
Depressive symptoms are common in patients with heart failure, and studies suggest a
significantly increased risk of hospital readmissions and mortality (Gathright et al., 2017,
Ghosh et al., 2016) and impairment of quality of life (Muller-Tasch et al., 2007) for these
patients.
Poorer performance of self-care may be an underlying mechanism behind the negative effects
of depressive symptoms in heart failure (Sedlar et al., 2017). For example, studies have
reported that heart failure patients with depressive symptoms have more difficulty adhering to
prescribed medications (Goldstein et al., 2017) and a greater tendency to delay seeking
healthcare despite worsening symptoms (Johansson et al., 2011). This suggests that
interventions targeting depressive symptoms in heart failure patients may improve self-care
(Jiang et al., 2018). However a problem was that only the study by Freedland et al.
(Freedland et al., 2015) in the meta-analysis included patients who were diagnosed with
depression. In that study, six months of face-to-face cognitive behavioural therapy (CBT),
compared to enhanced standard care, improved depression but surprisingly did not improve
self-care. But self-care has more than one aspect. For example, self-care as measured by the
9-item European Heart Failure Self-care Behaviour Scale-9 (EHFScBS) (Jaarsma et al., 2009)
has been reported to consist of three different aspects of heart failure self-care (Vellone et al.,
2014): autonomy-based adherence (three items: I weigh myself every day, I limit the amount
of fluids, I exercise regularly), provider-based adherence (two items: I eat a low-salt diet, I
breath increases I contact my doctor or nurse, If my legs/feet are more swollen I contact my
doctor or nurse, If I gain weight more than two kg in seven days I contact my doctor or nurse,
If I experience fatigue I contact my doctor or nurse). Thus, instead of analysing self-care as
one concept can be of interest to explore if CBT may have a different impact on the various
consulting behaviour, in heart failure patients with elevated levels of depressive symptom.
Regarding CBT, one problem is the low access to CBT due to a lack of psychologists.
Provision of CBT over the internet (i.e. internet-based CBT) has been proven effective in
patients with depression only (Andrews et al., 2018, Karyotaki et al., 2018). However, there is
a lack of studies that have examined whether internet-based CBT in heart failure aimed at
improving depression can also improve self-care. and different facets of self-care.
Therefore, the objectives of this explorative study including heart failure patients with
elevated levels of depressive symptoms were: (I) to investigate the impact of Internet-based
Cognitive Behaviour Therapy (ICBT) on different aspects of self-care behaviour and (II) to
study the association between changes in depressive symptoms and different aspects of self-
care behaviour.
2. METHODS
A secondary analysis of data collected in a RCT aimed to evaluate the impact of a nine-week
internet-based CBT programme on depressive symptoms in patients with heart failure. The
design of the RCT study has been reported in the main publication (Lundgren et al., 2016). In
brief, a total of 50 heart failure patients with at least mild depressive symptoms (i.e. Patient
Health Questionnaire-9 score >5) were recruited from four hospitals in southeastern Sweden.
active control group participating in a nine-week online discussion group (n=25). The
regional ethical review board of Linköping Sweden approved the study (ref. no. 2011/166-
The internet-based CBT programme (Lundgren et al., 2015, Lundgren et al., 2016). consists
of seven modules: (1) Introduction; (2) Living with heart failure; (3) Depressive symptoms
and heart failure; (4) Behaviour activation – enabling change; (5) Behaviour activation –
implementing change; (6) Problem-solving – a tool for dealing with problems, and (7)
Consummation. At weeks 2 and 3, heart failure self-care was addressed in both groups,
and homework and in the the on-line discussion group by discussions in writing.
2.3 Measurements
The Patient Health Questionnaire-9 (Spitzer et al., 1999) was used to measure depressive
symptom at baseline and at the nine-week follow-up. The Patient Health Questionnaire-9
consists of nine items to be answered on a four-point scale, and summed to a total score in the
range of 0-27, where higher numbers represent a higher level of depressive symptoms
(Kroenke et al., 2001). The Patient Health Questionnaire-9 has been found to be valid and
The 9-item European Heart Failure Self-care Behaviour Scale (Jaarsma et al., 2009) was used
to measure heart failure self-care behaviour and the three sub-scales autonomy-based as
consulting behaviour at baseline, three weeks and at the nine-week follow-up. The items are
rated on a five-point scale between 1 (I completely agree) and 5 (I completely disagree). The
score was standardized to 0-100, and reversed so that a higher score indicated better self-care
Descriptive data are presented as percentages or mean ± SD. A Student’s t-test was used for
continuous variables and the Chi-square test was used for discrete variables. An evaluation of
differences between groups regarding the summary score of the European Heart Failure Self-
care Behaviour Scale and the three subscales scores (i.e. autonomy-based, provider-based and
consulting behaviour) at the three and nine-week follow-up was performed with Analysis of
Covariance (ANCOVA). A paired Student’s t-test was used to analyze if the changes in
European Heart Failure Self-care Behaviour Scale from baseline to three weeks, and from
three weeks to follow-up at nine weeks differed significantly within the internet-based CBT
group and the on-line discussion group. To explore possible associations between changes in
depressive symptoms and changes of European Heart Failure Self-care Behaviour Scale
and consulting behavior we did not separate data from the two groups. Since the change
scores were skewed, analyses were performed with Spearman rank correlations. A positive
correlation between the change in Patient Health Questionnaire-9 and the change in the
depressive symptoms correlates with an improved self-care. P-values < 0.05 were considered
significant. All data were analyzed using standard software (IBM, SPSS version 24).
3. RESULTS
3.1 Population
The population has been described in detail in the main publication (Lundgren et al., 2016). In
brief, the mean age of the study population was 63 years (SD 12.8), 59% were male and 4%
were smokers. About three-quarters of the population were in New York Heart Association
Class II (40%) and III (36 %). Approximately 90 % took beta-blockers and angiotensin-
converting enzyme inhibitors or angiotensin receptor blockers. The mean score in the Patient
Health Questionnaire-9 before the start of the study did not vary between the groups (internet-
No significant differences were found in self-care between the patients in the internet-based
CBT and the patients in the online discussion group at the three- and nine-week follow-ups
(Table 1). As can be seen in Table 1, self-care behaviours increased from baseline to week
three in both groups. From week three to the end of the study at week nine, a slightly different
pattern was seen. The mean scores of the patients in the on-line discussion group decreased in
all four scales, whereas in the internet-based CBT group the scores remained stable or
increased in two of the four scales (i.e. autonomy-based and provider-based adherence).
Within-group analysis (Table 1) of the changes in the European Heart Failure Self-care
Behaviour Scale shows that from baseline to week three, the summary score increased
significantly for the online discussion group (p=0.04), but not for the internet-based CBT
group (p=0.15). At the nine-week follow-up, these scores had decreased. Similarly, consulting
behaviour improved at week three for the online discussion group (p=0.04), but not for the
internet-based CBT group (p=0.22). At the nine-week follow-up, these scores had also
decreased to approximately the level found at baseline. For autonomy-based adherence, only
minor increases were found at three weeks. However, at nine weeks, the mean score in the on-
line discussion group had decreased to baseline level, whereas the score of the internet-based
CBT group was sustained. Provider-based adherence increased in the internet-based CBT
group significantly from baseline to the nine-week follow-up (p=0.05) whereas in the on-line
(mean change score 0.9 SD 4.0) and improvement in autonomy-based adherence (mean
change score 1.2 SD 22.4) was found (r=0.34, p=0.03). The correlation between the
improvement in depressive symptoms and the summary score of European Heart Failure Self-
care Behaviour Scale was of the same magnitude (mean change score 2.8 SD 14.2), but not
significant (r=0.28, p=0.07). For subscales consulting behaviour (mean change score 1.8 SD
19.7), and provider-based adherence (mean change score 7.1 SD 18.3), the correlations were
weaker and not significant (r=0.18, p=0.27 and r=0.07, p=0.66). Correlational analysis in
4. Discussion
discussion group in the improvement of self-care behaviour in heart failure patients with
depressive symptoms. We could not find any beneficial effect of internet-based CBT on self-
care in depressed heart failure patients. Freedland et al. (Freedland et al., 2015) also reported
no improvements from CBT for self-care in heart failure patients. An explanation may be that
CBT interventions may not impact all aspects of self-care. We therefore undertook further
analyses to explore whether internet-based CBT could have different impact on the three
subscales in the European Heart Failure Self-care Behaviour Scale, but no beneficial effects
could be found. Another possible explanation for the lack of effects on self-care, is that in our
study and that of Freedland et al (Freedland et al., 2015), poor self-care was not a criterion for
improvement in the summary score at week 3, but not at week 9. This may be the result of
increased attention by way of participation in the study, thus indicating a possible digital-
placebo effect (Torous and Firth, 2016). On the other hand, the score for provider-based
adherence was significantly different than baseline (p=0.05) in the internet-based CBT group
at the nine-week follow-up. It is possible that the internet-based CBT programme was helpful
module and the homework assignments were intended to increase the understanding of, and
the benefit of, medication and exercise. Furthermore, in the behavioural activation module of
the internet-based CBT programme, the participants could, in a structured way, plan and
the data from the two groups to study how change in depression was related to changes in
self-care, regardless of group assignment. We realize that from this analysis, one cannot
determine whether internet-based CBT has a direct effect on self-care, or if it affects self-care
indirectly by improving depression. However, in our pilot RCT study, only patients in the
internet-based CBT group had a significant decrease in depressive symptoms (Lundgren et al.,
These results illustrate the difficulty of finding the optimal combination of programmes in the
CBT on top of exercise may be a more effective option than exercise or CBT alone. A
previous study by Gary et al. (Gary et al., 2010) showed that CBT combined with home-based
exercise had the best effect in reducing depression, increasing physical function, and
improving health-related quality of life. These data also suggest a new hypothesis regarding
physical well-being and symptom relief are reported as important motivational goals for heart
failure patients (Jaarsma et al., 2017). These goals may be more associated with intrinsic
motivations, i.e. the patient performs self-care activities due to internal satisfaction, and thus
based self-care behaviours such as exercise, choosing healthy food, increasing control by
daily weighing, and experiencing fewer symptoms by taking heart failure medications. In our
internet-based CBT programme, participants established their own personal goals (i.e.
have served as an aid to achieving self-care. This hypothesis requires testing in additional
research that explores the effects of CBT or internet-based CBT on self-care, and on different
A major limitation is the fact that this was a secondary-analysis of data collected in a pilot-
RCT. Thus, the study is underpowered to detect associations and interactions between
internet-based CBT, self-care behaviours and depressive symptoms. Therefore, the results
from this study should be interpreted with caution. Another limitation is that the analysis of
self-care behaviours was self-reported. It could have been valuable if objective data regarding
self-care behaviours such as physical activity were also collected. Since this study is one of
the first internet-based CBT studies performed with heart failure patients with elevated
depressive symptoms, we believe that the results of this study, despite its limitations, are of
interest.
In conclusion, this study found improvement in depressive symptoms that were associated
with improved autonomous-based self-care adherence. More studies that evaluate whether
internet-based CBT for depression in heart failure has beneficial effects on self-care are
needed.
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
Acknowledgements
This study was funded by grants from the Swedish Heart and Lung Association (grant number
E08/14), the Medical Research Council of Southeast Sweden (grant number FORSS-470121)
and the Region of Östergötland (grant number LIO-470271). The funding source had no
influence on the design, procedure, analysis or interpretation of the results in this study.
References
Andrews, G., Basu, A., Cuijpers, P., Craske, M.G., McEvoy, P., English, C.L., Newby, J.M., 2018.
Computer therapy for the anxiety and depression disorders is effective, acceptable and
practical health care: An updated meta-analysis. J Anxiety Disord 55, 70-78.
Freedland, K.E., Carney, R.M., Rich, M.W., Steinmeyer, B.C., Rubin, E.H., 2015. Cognitive Behavior
Therapy for Depression and Self-Care in Heart Failure Patients: A Randomized Clinical Trial.
JAMA Intern Med 175 (11), 1773-1782.
Gary, R.A., Dunbar, S.B., Higgins, M.K., Musselman, D.L., Smith, A.L., 2010. Combined exercise and
cognitive behavioral therapy improves outcomes in patients with heart failure. J Psychosom
Res 69 (2), 119-131.
Gathright, E.C., Goldstein, C.M., Josephson, R.A., Hughes, J.W., 2017. Depression increases the risk of
mortality in patients with heart failure: A meta-analysis. J Psychosom Res 94, 82-89.
Ghosh, R.K., Ball, S., Prasad, V., Gupta, A., 2016. Depression in heart failure: Intricate relationship,
pathophysiology and most updated evidence of interventions from recent clinical studies. Int
J Cardiol 224, 170-177.
Goldstein, C.M., Gathright, E.C., Gunstad, J., M, A.D., Redle, J.D., Josephson, R., Moore, S.M., Hughes,
J.W., 2017. Depressive symptoms moderate the relationship between medication regimen
complexity and objectively measured medication adherence in adults with heart failure. J
Behav Med 40 (4), 602-611.
Hammash, M.H., Hall, L.A., Lennie, T.A., Heo, S., Chung, M.L., Lee, K.S., Moser, D.K., 2013.
Psychometrics of the PHQ-9 as a measure of depressive symptoms in patients with heart
failure. Eur J Cardiovasc Nurs 12 (5), 446-453.
Jaarsma, T., Arestedt, K.F., Martensson, J., Dracup, K., Stromberg, A., 2009. The European Heart
Failure Self-care Behaviour scale revised into a nine-item scale (EHFScB-9): a reliable and
valid international instrument. Eur J Heart Fail 11 (1), 99-105.
Jaarsma, T., Cameron, J., Riegel, B., Stromberg, A., 2017. Factors Related to Self-Care in Heart Failure
Patients According to the Middle-Range Theory of Self-Care of Chronic Illness: a Literature
Update. Curr Heart Fail Rep 14 (2), 71-77.
Jiang, Y., Shorey, S., Seah, B., Chan, W.X., Tam, W.W.S., Wang, W., 2018. The effectiveness of
psychological interventions on self-care, psychological and health outcomes in patients with
chronic heart failure-A systematic review and meta-analysis. Int J Nurs Stud 78, 16-25.
Johansson, P., Nieuwenhuis, M., Lesman-Leegte, I., van Veldhuisen, D.J., Jaarsma, T., 2011.
Depression and the delay between symptom onset and hospitalization in heart failure
patients. Eur J Heart Fail 13 (2), 214-219.
Karyotaki, E., Ebert, D.D., Donkin, L., Riper, H., Twisk, J., Burger, S., Rozental, A., Lange, A., Williams,
A.D., Zarski, A.C., Geraedts, A., van Straten, A., Kleiboer, A., Meyer, B., Unlu Ince, B.B.,
Buntrock, C., Lehr, D., Snoek, F.J., Andrews, G., Andersson, G., Choi, I., Ruwaard, J., Klein, J.P.,
Newby, J.M., Schroder, J., Laferton, J.A.C., Van Bastelaar, K., Imamura, K., Vernmark, K., Boss,
L., Sheeber, L.B., Kivi, M., Berking, M., Titov, N., Carlbring, P., Johansson, R., Kenter, R., Perini,
S., Moritz, S., Nobis, S., Berger, T., Kaldo, V., Forsell, Y., Lindefors, N., Kraepelien, M.,
Bjorkelund, C., Kawakami, N., Cuijpers, P., 2018. Do guided internet-based interventions
result in clinically relevant changes for patients with depression? An individual participant
data meta-analysis. Clin Psychol Rev 63, 80-92.
Kroenke, K., Spitzer, R.L., Williams, J.B., 2001. The PHQ-9: validity of a brief depression severity
measure. J Gen Intern Med 16 (9), 606-613.
Lundgren, J., Andersson, G., Dahlstrom, O., Jaarsma, T., Kohler, A.K., Johansson, P., 2015. Internet-
based cognitive behavior therapy for patients with heart failure and depressive symptoms: A
proof of concept study. Patient Educ Couns 98 (8), 935-942.
Lundgren, J.G., Dahlstrom, O., Andersson, G., Jaarsma, T., Karner Kohler, A., Johansson, P., 2016. The
Effect of Guided Web-Based Cognitive Behavioral Therapy on Patients With Depressive
Symptoms and Heart Failure: A Pilot Randomized Controlled Trial. J Med Internet Res 18 (8),
e194.
Muller-Tasch, T., Peters-Klimm, F., Schellberg, D., Holzapfel, N., Barth, A., Junger, J., Szecsenyi, J.,
Herzog, W., 2007. Depression is a major determinant of quality of life in patients with chronic
systolic heart failure in general practice. J Card Fail 13 (10), 818-824.
Sedlar, N., Lainscak, M., Martensson, J., Stromberg, A., Jaarsma, T., Farkas, J., 2017. Factors related to
self-care behaviours in heart failure: A systematic review of European Heart Failure Self-Care
Behaviour Scale studies. Eur J Cardiovasc Nurs 16 (4), 272-282.
Spitzer, R.L., Kroenke, K., Williams, J.B.W., Group, a.t.P.H.Q.P.C.S., 1999. Validation and Utility of a
Self-report Version of PRIME-MD. JAMA: The Journal of the American Medical Association
282 (18), 1737-1744.
Torous, J., Firth, J., 2016. The digital placebo effect: mobile mental health meets clinical psychiatry.
Lancet Psychiatry 3 (2), 100-102.
Vellone, E., Jaarsma, T., Stromberg, A., Fida, R., Arestedt, K., Rocco, G., Cocchieri, A., Alvaro, R., 2014.
The European Heart Failure Self-care Behaviour Scale: new insights into factorial structure,
reliability, precision and scoring procedure. Patient Educ Couns 94 (1), 97-102.
Table 1
Table 1. Mean values at the three measurement points and analysis of changes between the internet-based CBT and the discussion group regarding the
summary score, autonomy-based adherence provider-based adherence and consulting behaviors of the European Heart-Failure Self-Care Behavior Scale. The
table also provides analysis of changes within the internet-based CBT group and the discussion group.
Discussion- 40.5 (21) 48.1 (18) 41.8 (18) 42.5 (23) 45.8 (26) 42.2 (20) 67.4 (22) 75 (15) 69.4 (20) 25.7 (31) 36.4 (27) 27.7 (34)
group
Between-
group F(1,41)=0.20, p=0.88 F(1,41)=0.36, p=0.85 F (1,41)=0.69, p=0.79 F(1,37)=0.15, p=0.70
analysis
ANCOVAA
Between-
group F(1,37)=0.08, p=0.78 F(1,37)=0.15, p=0.69 F (1,37)=2.2, p=0.14 F(1,37)=0.28, p=0.60
analysis
ANCOVAB
Within- p=0.151 p=0.352 p=0.711 p=0.712 p=0.201 p=0.052 p=0.221 p=0.922
group
analysis
ICBT-group