Journal Name
1 2 6 4
Manuscript No. B Dispatch: 4.9.08
Author Received:
Journal: ACP CE: Krishna Sarma
No. of pages: 10 PE: Vijay
Acta Psychiatr Scand 2008: 1–10 Copyright  2008 The Authors
All rights reserved ACTA PSYCHIATRICA
DOI: 10.1111/j.1600-0447.2008.01264.x SCANDINAVICA
Prevalence, onset and comorbidity of
postpartum anxiety and depressive disorders
10 Reck C, Struben K, Backenstrass M, Stefenelli U, Reinig K, Fuchs T, C. Reck1, K. Struben1,
11 Sohn C, Mundt C. Prevalence, onset and comorbidity of postpartum M. Backenstrass1, U. Stefenelli2,
12 anxiety and depressive disorders. K. Reinig1, T. Fuchs1,
C. Sohn3, C. Mundt1
14 Objective: The study presents data on the 3-month prevalences of 1
Department of General Psychiatry, University of
15 postpartum anxiety disorders (PAD) and postpartum depressive Heidelberg, Heidelberg, 2Institute of Statistics,
16 disorders (PDD) and their comorbidity in a German community Wurzburg and 3Gynaecological Clinic, University of
17 sample. Associations with sociodemographic variables and previous Heidelberg, Heidelberg, Germany
18 history of psychopathology were analysed.
19 Method: Data were gathered in a longitudinal study over the first
20 3 months postpartum. In a two-stage screening procedure, a
population-based representative sample of 1024 postpartum women
was assessed for symptoms of anxiety and depression using DSM-IV-
22 based screening instruments.
23 Results: The estimated rates of DSM-IV disorders were 11.1% for Key words: postpartum period; anxiety disorders;
24 depression; epidemiology; womenÕs health
PAD and 6.1% for PDD. Comorbidity was found in 2.1%. The rate
25 for PAD with postpartum onset was 2.2% and for PDD 4.6%. Young Corinna Reck, Klinik fr Allgemeine Psychiatrie, Zentrum
26 fr Psychosoziale Medizin, Vossstr. 2, 69115 Heidelberg,
mothers and mothers with a low education level had a heightened risk
27 of developing depression following delivery. E-mail: corinna_reck@med.uni-heidelberg.de
28 Conclusion: Because of the clinical relevance of PAD, controlled
29 studies and specialized programmes for prevention and treatment are
30 urgently required. Accepted for publication August 15, 2008

32 Significant outcomes
33 • The prevalence of postpartum anxiety disorders was 11.1% and the prevalence of postpartum
34 depressive disorders was 6.1%.
35 • 18.4% of participants with an anxiety disorder (n = 114) were also diagnosed as having a depressive
36 disorder and 33.9% of the women suffering from depression (n = 62) as having an anxiety disorder.
37 • Concerning self-report measures, considerably higher rates of anxiety and depressive disorders were
38 found.
41 Limitations
42 • TheÕ totalÕ prevalence may be underestimated based on our predominantly middle class sample.
43 • Participants were more highly educated than non-participants.
44 • While women were requested to report the onset and history of depression and anxiety prior to
45 delivery, the present assessments were exclusively conducted in the postpartum period and are thus
46 subject to retrospective reporting bias.
49 lished (1–4). In contrast to the well-studied epide-
50 miology and consequences of postpartum
51 Postpartum anxiety disorders (PAD) and postpar- depression on child development, empirical results
52 tum depressive disorders (PDD) are the most concerning postpartum anxiety disorders are scarce
53 frequent maternal psychiatric disorders following (5). Matthey et al. (6) found that 16.2% of mothers
54 delivery. The impact of the motherÕs postpartum were diagnosed with a pure anxiety disorder
55 depression on early interaction experiences and the (phobias, panic, acute adjustment disorder with
56 long-term development of the child are well-estab- anxiety) 6 weeks postpartum. Furthermore, 82%


and v) z = )0. 30 Wenzel et al. 49 and first time onset of PAD and PDD). Measures 19 Socially disadvantaged populations tend to have 20 notably higher postpartum depression prevalence All participants completed a demographic infor- 21 rates than wealthy western industrial nations (13. Despite the high health risks for both phobia and generalized anxiety disorder. There are only two studies (6. participants with SCID (n = 333 with screening positive + n = 171 with screen- 41 ing negative) vs. questionnaire survey. Wenzel Darmstadt who gave birth between December 2003 9 et al. 17 assessment method used to obtain diagnoses and the 18 length of postpartum period under evaluation. The 26 sive disorders with postpartum onset according to Anxiety-SCID-Screening (18) was used as a tele- 27 DSM-IV criteria. valid data on the epidemiology of five anxiety disorder categories.4% for comorbid The Anxiety-SCID-Screening is taken from the 31 generalized anxiety disorder and 0. SCID. approxi.2% and 2. (8) found rates of 1. Sociodemographic characteristics of the participants and comparison of 40 lacking (17).73. Applying Diag. phone screening and the Anxiety Screening Ques- 28 they et al. In his two study samples. according to the with rates reported in other studies (6. Mat.7% for depres. 38 PDD remain scarce in Germany (15. A total of 1464 German- 13 mately 10% of pregnant women develop a post. The sample was mainly middle 10 generalized anxiety disorder. iii) the school leaving certificate z = )0. The prevalence rates of study.2% for and February 2005. participants without SCID (n = 520 with screening negative) 42 Aims of the study U-Test 43 Sociodemographic Participants Without With (SCID ⁄ no 44 The aims of the study were: i) to determine the ÔtotalÕ characteristics (n = 1024) SCID SCID SCID) 45 prevalence of PAD and PDD according to DSM-IV- 46 criteria (including PAD and PDD with onset prior to Mean age (years) 33 € 5 33 € 5 33 € 5 P = 0. structured clinical interview for DSM-IV. z = Age range (years) 15–45 15–45 18–45 0. It contains five screening 33 only very scarce data concerning the comorbidity questions covering the diagnostic categories: panic 34 of postpartum depression and anxiety disorders are disorder. 24 iety disorders. agoraphobia. (Table 1). it should be underscored that disorders (SCID-I) (18). of which 1024 (70%) consented. 2 . Reck et al. as age. A Ôcritical 36 mother and child associated with postpartum scoreÕ resulted from positive screening in one of the 37 disorders.38 Higher level secondary 30 28 31 50 comorbidity of PAD and PDD during the first school leaving certificate 51 3 months following delivery. 54 history on the prevalences of PAD and PDD.7 P = 0. structured clinical interview for DSM-IV. 16) and 39 findings with respect to PAD are completely Table 1. Secondary parameters Vocational A levels 4 5 4 52 were analysed in accordance with the primary target A levels 14 12 16 53 parameters: iv) the impact of previous disorder University degree 46 49 43 Mean number of children 2 € 0. (8) revealed a prevalence rate of 8.35 55 correlations of sociodemographic variables with 56 prevalence rates of PAD and PDD. Study sample 3 Miller et al. The 6 6 weeks to 6 months postpartum.1%. number of children and education level 23 ria in diagnosing postpartum depression and anx. mation sheet covering sociodemographic data such 22 Very few studies have employed DSM-IV-crite. 15). Exclusion criteria for participation in the 11 According to epidemiological studies carried out study included poor command of the spoken and 12 mainly in the English-speaking world. specific 35 available. 1 of diagnosed phobias were found to have occurred Material and methods 2 for the very first time in the postpartum period. speaking mothers were asked to participate in the 14 partum depression (9–12). using two different screening instruments. 14). (7) were able to show in a recent study 4 employing a self-report measure that 10% of The study was carried out in south Germany in two 5 women suffered symptoms of anxiety and stress middle-sized towns and their surroundings. In summary. The partic- 15 postpartum depression have been shown to vary for ipation rate of 70% is acceptable and comparable 16 women from different cultures.35 47 delivery). written German language. ii) rates of PAD and PDD with onset in the Educational level (%) 48 3-month postpartum period (including recurrent Lower level secondary 6 6 7 P = 0.73.7 2 € 0. total sample in this study consisted of female in- 7 nostic and Statistical Manual of Mental Disorders patients of six maternity hospitals in Heidelberg and 8 (DSM-IV) criteria 8 weeks postpartum. social phobia. class. axis I 32 sion.8 2 € 0. (6) documented rates of comorbid tionnaire (ASQ-15) (19) in the context of a 29 depression and anxiety of 4. 8) Screening for anxiety disorders was performed 25 reporting comorbidity rates of anxiety and depres.18.

95 and its specificity screening instruments and conducted the DSM-IV- 13 from 0. .Screening questionnaires: 42 measurement and diagnosis of selected axis I EPDS n = 891 43 mental syndromes and disorders according to the ASQ n = 893 44 criteria of the DSM-IV (25). 1). screening and 56 disorder emerged within the first 12 weeks follow. The full mood and anxiety module 2 comprises 15 items and serves as a syndromatic including the assessment of previous history of 3 screening tool for current anxiety and generalized psychiatric disorders was applied. a cut-off 33 value of 10 or more indicates the presence of a minor 34 and 13 or more the presence of a major depressive Refused to Agreed to 35 disorder. social phobia. Obtained results 4 anxiety disorders. Flowchart depicting sample recruitment. symptoms. 11 sensitivity of the ASQ-15 ranges according to The research assistants who administered the 12 diagnosis group from 0. (23) was Procedure 19 used to screen for major and minor depression 20 according to DSM-IV criteria (24). Therefore. efficient and reliable instrument for the . Postpartum prevalence study 1 The ASQ-15 is a self-report instrument which ing delivery. A German version of the short 18 form of the -D translated by Loewe et al. ÔCritical with informational material and the first set of 25 scoresÕ for minor and major depression are based questionnaires (including a written consent form) 26 on DSM-IV criteria (25).following negative screening: n = 171 54 ment disorder with anxiety (AADA) (6). Contact was initially made with the mothers 1 day 21 prises nine items. 22).96. The SCID is a semistructured. . Structured Clinical Interview (SCID I) 50 ing the criteria for generalized anxiety disorder in n = 504 51 the 3-month postpartum period with a minimum . procedures. According to the German EPDS validation 32 study conducted by Bergant et al. it is impossible for a de novo onset of 48 generalized anxiety disorder to occur in the 49 postpartum period. The would have exceeded the scope of our study. interview had received training to ensure reliability. 3 . agoraphobia. were sub. specific common anxiety screening tools do not cover OCD 7 phobia or generalized anxiety disorder. women meet. Addi- 55 tional Ôpostpartum onsetÕ was diagnosed when the Fig. 17 (EPDS) (21. panic presented in this study because of the fact that 6 disorder. Screening for anxiety and depression 41 nomical. 22). Each of the items in (n = 1464. women obtained a Ôcritical participate in study participate in study 36 scoreÕ given an EPDS score of 10 or more. The -D com. eco. OCD screening instrument would have been 9 IV criteria. an additional and specific 8 sequently interviewed using the SCID and DSM. For this reason. n = 440 n = 1024 37 Participants who reached a Ôcritical scoreÕ in one 38 of the four described screening instruments were 39 additionally interviewed in a second stage using the 40 SCID-I (18).Telephone screening: 45 As the diagnosis of generalized anxiety disorder SCID-Anxiety-Screening and PHQ-D: n = 1014 46 requires a minimum symptom duration of 47 6 months. In this study. 27 The EPDS is an internationally well-established 28 and validated 10-item instrument for the screening Approached 29 of postpartal depression (21. 14 Screening for depression was carried out using Continuous checks were made throughout the 15 the Patient Health Questionnaire-Depression (-D) study for the purpose of ensuring that reliability 16 (20) and the Edinburgh Postnatal Depression Scale was maintained. (22). The women were informed 23 the depression module represents one of the nine about the Mother–Child Project and were provided 24 DSM-IV criteria for major depression. 1.88 to 0. This self-rating n = 1464 30 scale assesses mental state during the previous 31 7 days. Responses are to be made with after delivery in the respective maternity hospitals 22 reference to the past 14 days. Women with critical screening for obsessive–compulsive disorders (OCD) are not 5 scores in one of the diagnostic categories. The ASQ has been validated in terms of required in the screening stage and this in turn 10 its concordance with DSM-IV diagnosis.51 to 0.following positive screening: n = 333 52 symptom duration of 2 weeks within the last 53 4 weeks were diagnosed as having acute adjust. see Fig.

Mothers n = 171 mothers without positive screening com- 17 were further sent a questionnaire set which pleted a SCID as control for false-negative screen- 18 included the EPDS and the ASQ-15 and which ing results). For the evaluation of confidence 39 tolerated for organizational reasons (e. Mothers with a 49 PHQ PHQ positive screening completed a SCID. 25). 4. 28 which symptom onset occurs 4–8 weeks after The study protocol was approved by the inde- 29 delivery (21. 2. 4 . In the case of a positive decision. 1 given their interest in study participation. pendent ethics committee of the University Med- 30 Telephone screening (Anxiety-SCID-Screening ical Faculty. but sent back the 35 questionnaires. the EPDS and the scores. if a SCID diagnosis was 19 was to be completed and returned 6 weeks post. Heidelberg.g. 10 detailed explanation of study procedures. The SCID was first type of a = 0. EPDS EPDS Afterwards the screening procedure was 50 continued just like for mothers without 51 a critical screening score.e. women were contacted by (20) and the Anxiety-SCID-Screening were con- 6 telephone 14 days postpartum and asked to decide ducted in addition to the EPDS (21. according to comorbitiy were diagnosed contemporaneously). the screening procedure was 20 partum. poor intervals. This 22 selected according to the following considerations: was done to cover for cases in which an additional 23 Ô2 weeks postpartumÕ was selected with the aim of disorder would occur (diagnosis of depression after 24 avoiding a temporal overlap and potential con. 10 and was obtained 2 weeks postpartum following a 33 12 weeks after delivery (see Fig. Temporal deviations of plus 4 days 37 for the first telephone screening and plus or minus Prevalence rates were calculated using simple 38 4 days for the remaining telephone screenings were percentages. Anyhow. A second 52 SCID would have been undertaken in 53 case of mothers with positive screening High score High score High score High score High score High score results for depressive disorders addi- 54 tionally developing anxiety disorders 55 and the other way round. Flowchart of screening occa- 47 Screening: Screening: sions for postpartum anxiety and dep- Anxiety Anxiety Anxiety Anxiety 48 Anxiety PHQ Anxiety PHQ PHQ PHQ ressive symptoms. with these mothers a second SCID would 26 ÔSix weeks postpartumÕ was selected in line with the have been conducted. a diagnosis of anxiety disorder or the other way 25 founding with the occurrence of maternal blues. The SCID was conducted within the 12- 15 ASQ-15. but this did 56 SCID SCID SCID SCID SCID SCID not occur during our study. additionally 16 the stamped addressed envelope provided. 6. the . postpartum preva- 42 43 2 weeks 6 weeks 44 postpartum 4 weeks 8 weeks 10 weeks 12 weeks Delivery 45 46 Screening: Screening: Screening: Screening: Fig. 2. therefore they have been included Statistical analyses 36 in the sample). The fortnightly telephone screenings with the -D 5 partum. Study clinically relevant symptoms (Ôcritical scoreÕ) in the 2 research assistants noted the telephone number and course of screening at any one of the six measure- 3 childÕs date of birth of those women who consented ment occasions. round. 8. a global and two-sided decision error 40 obtainability of the participants). 22) (n = 891) 7 whether they wanted to take part in the study and ASQ-15 (19) (n = 893) questionnaire surveys 8 (n = 1024). This procedure enabled the 11 postpartum. the women were additionally asked to direct arrangement of an appointment for the 12 complete the questionnaire set that had been performance of a more extensive clinical interview 13 distributed in the maternity hospital and which (SCID-I) (18) in the case of critical screening 14 included a written consent form. made (PAD or PDD). EPDS measurement occasions were mothers without a critical screening scores. at 2 and 6 weeks postpartum in light of our goal to 9 D and Anxiety-SCID-Screening were conducted promptly assess the occurrence of depressive 10 with the mothers for the first time. Reck et al. Two weeks symptoms or anxiety. i. n = 1014. A stamped addressed envelope was again continued up to 12 weeks postpartum just like for 21 enclosed. 34 women were not obtainable. As agreed. 4 to being re-contacted by telephone 14 days post. These were to be returned by post using week period of study (n = 333. Written informed consent 32 on six measurement occasions: 2.05 was selected for five major 41 additionally performed given the occurrence of confirmative comparisons. Patient confidentiality 31 and -D) was conducted with participating women was in no way breached. but in our study all cases of 27 definition of postpartum depression.

the first 3 months postpartum was 6. 37 groups (n = 905 and 385 respectively). we found a prevalence 15 participants with and without SCID with respect to of 8.5–10. The effect of covariates such as age.99 = 99 % confi. The ÔtotalÕ prevalence of depressive disorders 27 pants and those refusing to participate showed that (including PDD with onset prior to delivery) 28 participants were on average 3 years older [age of (Table 2) in the entire group of women during 29 refusers: 30 ± 7 (n = 383) years.6% (n = 47). phobiaÕ.0%. educa.2% 11 reference level (26).2) and the remaining 16 also had a received a positive SCID diagnosis (anxiety or 47 depression at the time of measurement (15. age) group as disorders with postpartum onset was 2. 50 1 As a total of five estimators were regarded as primary.52. depression was 2.9% (n = 30) and 2.0%. 18. we combined the clinically 3 (fourth) and their comorbidity (fifth parameter)1. 54 All other coefficients and tests are provided with 95% confi. ii) ÔAADAÕ and iii) 6 previous anxiety and depression on disorders in the Ôspecific and social phobiasÕ to form single categories. which included panic disorder.8% (n = 18) was 20 found and 0. group of women during the first three postpartum 9 tion or number of children was also estimated months was 11.8–23.55.6% (n = 19) of which had a first 12 parameters age and education was analysed using a onset of anxiety disorders (1. 95% CI 0.4% (n = 4) for Ôpanic disorder ⁄ ago- 21 raphobiaÕ with postpartum onset. 7 postpartum period and disorders with postpartum The rate of anxiety disorders across the entire 8 onset. 26 n = 899) (Table 1). 45 sion without current symptoms (85.1% (n = 21). These statistical analyses onset. 48 95% CI 8. P < 0. 43 history of depressive disorders (n = 107).001.2% (n = 23) 39 for minor depression with postpartum onset – all 40 of which lacked a history of depression.9%.15.1% (95% CI 6. prevalence (third) and onset of depression diagnostic category. 30 z = )6. two-sided]. n = 11 of 1024). 13 logistic regression model.9% (n = 30) was 33 44% (n = 374) of non-participants.5–1. An analysis of the frequency of comorbid 55 dence intervals for descriptive purposes only.4% of the women reported a positive Prevalence of postpartum scid-anxiety disorders. For 41 Prevalences dysthymia. P < 0.4% of 5 .1% (n = 114). the 51 global a was adjusted according to BonferroniÕs method resulting in a comparison-wise level of decision error of Comorbidities of SCID-postpartum depressive and anxiety 52 4 a = 0. sion.9. a prevalence rate of 2. Mann–Whitney test. A rate of 2. Mann–Whitney test.9. however.15. no differences with respect to the which had a first onset and no previous history of 36 number of children [an average of two in both depressive disorders (1. depression) at the second stage of measurement.1–2.1%.01 (26). A prevalence 22 Results rate of 2. There major depression with postpartum onset.5 and r in version Regarding the subgroup Ôpanic disorder ⁄ agora- 19 1 2.3 % (95% CI 0.1).1–0. 95% CI 1. a prevalence rate of 1.3% (n = 24) was observed for 34 z = )10. 42 10.g.2.9) (n = 3) for postpartum 17 tests were conducted.9–91. relevant subgroups i) Ôpanic disorder ⁄ agoraphobiaÕ 4 Odds ratios with confidence intervals were cal. 46% of 35 were. panic disorder with 5 culated as risk measures to determine the impact of agoraphobia and agoraphobia. Differences between refusers and participants 24 25 Participants had a mean age of 33 years (SD = 5. 95% CI Prevalence rates include only those women who 46 76. found. Regarding major depres- 32 participants (n = 845) with higher education vs.5% (n = 5) was revealed. The rate of anxiety 10 using odds ratios with one (e.001. Postpartum prevalence study 1 lence (first parameter) and onset rates (second) of Because of the small number of cases in each 2 anxiety. bidity rate of 2.3% (n = 24) was found for AADA and 23 1. 82.1% (n = 62). two-sided] and The rate of depressive disorders with postpartum 31 that they were more highly educated [64% of onset was 4.0) (n = 83) and a rate of 16 sociodemographic characteristics. With regard to the subgroup of 14 and refusers. a rate of 0.1 (26). An interaction effect of the (n = 23). Mann–Whitney test. P < 0. disorders 53 dences were thus calculated for these five primary coefficients. Analyses comparing partici. The prevalence rate of minor 38 z = )0. Two-sided 1 )a ⁄ 5 = 0. To compare participants n = 19 of 1024). participants and drop-outs as well as specific and social phobias. All two-sided occurrence of PAD und PDD revealed a comor- 56 confidence intervals were calculated for the single proportion. 18 were performed using spss 11. Mann–Whitney 0.5% (n = 15) for AADA with postpartum onset. Prevalence of postpartum SCID -depressive disorders. All 16 recurrent sufferers had 49 a major depression. two-sided]. Table 2 presents the number of patients meeting 44 Ninety-one of these had a history of depres- criteria for the diagnosis of an anxiety disorder.

and ninety-three women completed an ASQ-15 at 31 weeks 2 or 6.5% of which proved critical (95% 32 CI 29.5) 20 21 Target parameters are printed in bold.2)* 7 Postpartum onset 2.4) 15 Previous occurrence of disorder 1. With regard to the screening of anxiety symptoms.1–0.6 (3.08) Increased 53 Depression Anxiety 2. anxiety symptoms were 33 Impact of previous SCID-depressive and anxiety disorders revealed for 42. Prevalences and confidence intervals of depressive and anxiety disorders (total prevalence. CI 20.2–3. 32. Eight hundred 30 13.1 (6.4 (2. 36 anxiety disorder developed an anxiety disorder An analysis of available data rates based on 37 with postpartum onset.1 (0.1 (8. n = 1024 (n = 504 with SCID.0–4.9% of the women suffering from depres.6)* 6.9) 12 Specific and social phobia Minor depression SCID 13 14 Total prevalence 8.5).1.5–4. panic and agoraphobia Major depression SCID 9 10 Total prevalence 1. Reck et al.3 (4.1–6.2) 11 Postpartum onset 0.9–32.1) 6. n = 1024 (n = 504 with SCID.4–4.3 (0. According to the -D (weeks 2–12).8–26.6% (95% CI 25.5 (0. 95% CI Screening carried out at weeks 4–12. prevalence of disorder with postpartum onset and prevalence of previous disorder in patient history) 2 3 Anxiety disorders Depressive disorders 4 Anxiety disorder Depression SCID (minor or major depression.1% (95% CI 3.4 (0.4–3.7–13.1%.8) 10.45 (0.54) Increased 56 6 .37–9.2) 3.5–10.7) 16 Acute adjustment disorder with anxiety1 Dysthymia 17 Prevalence 2.5) of the 1024 women 28 hundred and fifty-five women had an anxiety showed symptoms of anxiety in the Anxiety-SCID- 29 disorder or a depression or both (15.4–3. n = 520 without 48 SCID due to negative screening results) 49 50 Previous occurrence of disorder Postpartum onset (recurrent and first time onset) Odds ratio (95% CI) Evaluation of relative risk 51 Anxiety Anxiety 6. critical depression scores 45 46 47 Table 3.3 (1.4) – 19 Previous occurrence of disorder 0.0) of partic- 34 11. In total. One 28.78) Increased 55 Depression 3.4–35. 27 sion (n = 62) also had an anxiety disorder.05–10. n = 520 without SCID due to negative screening results). 9% 40 postpartum onset. n = 4 of 36) of the ipating women using either the SCID-Screening or 35 women who had previously suffered from an the ASQ-15.7).4 (8.27) Increased Depression or anxiety Anxiety 3.5) Panic disorder.62 (1.2–1.68) Increased 52 Depression 3. Risk of developing a disorder with postpartum onset given previous history of disorder assessed using the SCID.9% (95% CI 39.3 (1.9) 2.1 (4.5) 0. 1 Table 2.9) of participating women screened 41 risk for depressive and anxiety disorders with positive for a major or minor depressive disorder.2 (1.1–26.6% (95% 44 (Table 3).5) 2.0–0.9. Altogether.0–2.9 (2.3–8.02) Unchanged 54 Depression 3.7–2.5).68–7. 22 23 24 participants with an anxiety disorder (n = 114) 25 Screening results were also diagnosed with a depressive disorder 26 and 33.8 (1.5–3.5–3.2 (1. dysthymia) 5 6 Total prevalence 11.49–20.69–7.0–0.2) Postpartum onset 0.7)* 4.0–4.1 (0.9 (2.8–7.3–10. an increased (95% CI 7.86 (1.0) 2. Using odds ratios.1) 18 Postpartum onset 1.91–7.58 (1.1–1. depression screening measures revealed that the data 38 n = 13 of 107) of women reporting a previous of 1014 women were available for the -D and of 891 39 depressive episode developed a depression with for the EPDS.35 (1.5) 0.6–12.8–2.6–19.2 (1.3 (0. agoraphobia.5 (0. 42 postpartum onset was found for those mothers The rate of women with depressive symptoms 43 with a history of depressive or anxiety disorders according to EPDS (week 2 or 6) was 23.0–17.6)* 8 Previous occurrence of disorder 3. 12.0) 2.1% (95% CI 6.5 (2.5) Previous occurrence of disorder 2.80 (1.8–46.

With phobias constituting 8.13.19). EPDS. first investigation of the prevalence of postpartum 49 The influence of education was. 32 developing depression following the birth of their )0. the comorbidity of PDD and PAD. P < 0. of women. all phobias and panic Drop-out analyses 54 disorder) in this study was 11. number of children).8) of the women sociodemographic variables (age.0–12. The risk of suffering from anxiety having been screened negative for anxiety (four 26 disorders following delivery was not affected by the false negatives of 585 negatively screened anxiety 27 motherÕs age or education. P > 0.9) corresponding to a 23 cases odds ratio = 1. prevalences of PAD and PDD as well as their 43 Using the EPDS and a cut-off score of 13 or more. 7 . 2 according to either the -D or EPDS.1) 6 weeks postpartum.1%.2) of women in this between drop-outs (n = 101) after week 12 and 4 study reached the cut-off for a minor depression those continuing in the study (n = 923) (Mann– 5 (>9) 2 weeks postpartum and 15. mothers 34 above 35 years had a decreased risk (odds Discussion 35 ratio = 0. P > 0.7%. 9.4) 6 weeks postpartum. v2 = 11. 95% CI 0.7.1%.4–27. ASQ-15 and 13 lence rates proved persistent across measurement SCID-Screening) and were randomly selected to 14 occassions without eminent deviations at the upper take part in the clinical interview with the aim of 15 and lower ends.9%. 6 12. 95% CI 0. 95% CI 0. however. participants fulfilling 17 criteria for depression or anxiety who were not Sociodemographic correlates 18 identified in the screening process. comorbidity based on DSM-IV-criteria within the 44 we were able to confirm the influence of age on the first 3 months following birth.9%. Four individuals 24 childÕs gender (in both cases odds ratio = 1. SCID subjects (n = 504) did not differ from non- 29 icantly impacted the risk of developing a postpartum SCID subjects (n = 520) in demographical param- 30 depression: age and education of the mother. preva.3.0. P = 0.9%. 95% CI 0.01. Seven individ- 19 Sociodemographic correlate analyses were con. cases = 0.5 in each case. 2 screening results (PHQ. obtained a positive anxiety SCID result despite 25 95% CI 0.8% (95% CI Whitney test. education and number of children) 31 young mothers (<25 years) had a heightened risk of (Mann–Whitney test. not anxiety disorders in a German community sample 50 confirmed (chi-squared test. 28 Two major factors were discovered which signif.3–0. prevalences of postpartum anxiety disorders and 38 mothers with a low education level (lower secondary depression as well as their comorbidity in Germany 39 school leaving certificate) had an increased risk of and how these prevalences compare with those 40 developing depression (odds ratio = 2.4% (95% CI 17.e.7).6–1. 171 subjects had negative 12 Anxiety-SCID-Screening and the ASQ-15.35) (Table 1). 95% CI found in studies of different countries.38 < z < 0.1) of the women had a major depression 8 (>12) 2 weeks postpartum and 8.4–1.09 < z < )0. as well as in the these 504 women. no differences were found 3 20. df = 4.05 in each case.9). There was no having been screened negative for depression 21 effect of number of children (one vs. with the specific 55 By week 12.0). education. 52 The prevalence of PAD (acute adjustment 53 disorder with anxiety. two-sided. the attrition rate was 9.2–1. The results further provide new data on 51 v2 = 7. Furthermore. df = 2.3–1. 7. Furthermore.5).8–7. The influence of 45 development of depressive symptoms (chi-squared previous disorder histories and sociodemographic 46 test. 33 child (odds ratio = 3.5– Representativeness of the SCID-sample 9 11. uals were tested positive by the SCID despite 20 ducted based on total prevalences.6). This rate corresponds 56 respect to screening measures (EPDS.6–1. 95% CI 1.7% (95% CI 6. Ages ranged between 36 15 and 45 years.7–23.04). Younger correlates on the risk of developing a PAD or PDD 47 women (below the age of 25) were significantly was further evaluated.0.55). 8).7).5.5. 95% CI: 0. nor of false-negative rate of 0. ASQ-15) and with the findings of other studies (5. controlling for possible estimation errors in terms 16 of Ôfalse negativesÕ – i. Postpartum prevalence study 1 emerged for 25% (95% CI 22. The Hei- 41 1–5) whereas mothers with a university degree were delberg Postpartum Study aimed to examine the 42 less at risk (odds ratio = 0. 7 8. more than (seven false negatives of 766 screened nega- 22 one) on anxiety or depression disorders (in both tive = 0.9% (95% CI )1. 10 Regarding the number of positively screened SCID results were obtained for 504 subjects. While eters (age. There were no exclusion criteria These findings provide new insights concerning the 37 relating to the age of the mother. This study represents the 48 more likely to obtain an EPDS score or 13 or more. Of 11 women in the -D and EPDS.8–19.

(30) showed that may be underestimated. Dimensional were thus identifiable. when applying the same cut-off value at necessary to initiate preventive measures for suf- 29 the same measurement occasion. 23). a rate of 2. While women were 50 postpartum depression. the results show a heightened risk of 7 findings for depressive disorders. 15. a considerably higher rate of anxiety disorders on the onset of these disorders following 6 symptoms (32. Concerning studies in postpartum research (6. lower than prevalenc. rates of anxiety developing postpartum anxiety disorders and 8 disorders closely correspond to those reported for depression in the case of a previous history (6. 41 models might be preferable. Several limitations of the study should also be 42 chandani et al. a number of studies 17 this study revealed a PDD rate of 4.Õs (30) results as an indication depression and anxiety prior to delivery. women who gave birth in the number of women 8 .6% 36). Different results surement occasions. With respect to to be faced. The strength of this study is to be found in its 32 es found in less economically developed countries longitudinal design with multiple fortnightly mea- 33 such as India (29) or Turkey (13). it is remarkable that the prevalence found in interventions for postpartum anxiety disorders (35. Younger mothers were found to be especially 10 The prevalence of PDD (minor and major exposed to a heightened risk of postpartum 11 depression) in the German predominantly middle depression.7% a cut-off value above 12. participants were more 48 and carrying out testing at the same measurement highly educated than non-participants (64% vs. The influence of education 45 depression diagnosed using the EPDS (cut-off level on rates of refusal underscore the bias in the 46 value above 12) also had a significant effect on the sample selection and the associated risk of preva- 47 childÕs development.8% of the women had a cut-off vitally important to preventive medicine.6%. During the last few years. differences in prevalence 39 describe the acute clinical syndromes of postpartum rates resulting from the assessment method used 40 depression and anxiety disorder. The 26 value above 9 and 8. SKID). very similar to ferers and facilitate an untroubled postpartum 30 those found in economically developed western period and healthy development of their children. 22. studies exist focusing on appropriate preventive 20 many. 8).5%) was found.2% was observed. We found higher prevalences for self-reported der within the first 3 months following delivery. 21 this study is comparable with the rate of 5. Reck et al. the rate poor compatibility of pursuing a career and raising 14 found in this study is comparable with some children with which women in Germany continue 15 reported results (21. The findings of Ram. these findings are in accordance with other 3 PAD). determination of the onset of a postpartum disor- 35 sion. The prevalence rates of postpartum anxiety 22 reported for a subgroup of women between the disorders presented in this study indicate that the 23 ages of 14 and 35 years (28). In contrast.7% of retrospective reporting bias. 27). then according to our results. 36 symptoms (ASQ-15 and EPDS) than in the clinical The study further applied both DSM-IV criteria 37 interview (DSM-IV. Taking into consideration this particular group could be explained by the 13 studies which have used DSM-IV-criteria. It is possible that and common screening tools such as the EPDS for 38 categorical diagnosis (DSM-IV) may fail to the diagnosis of disorders. Concerning 4 self-report measures (ASQ-15) completed at weeks the impact of previous anxiety and depressive 5 2 or 6. however. These are. 9 the general population. we obtained a prevalence rate of 8. It should also be 55 children are at risk of developmentally suffering noted that it was not possible for us to screen all 56 under the depressive mood of their mothers. partum depression (31–34). based on our predominantly 43 support of the clinical relevance of dimensional middle class low-risk sample. development of appropriate screening instruments 24 Regarding clinical cases of EPDS-scores 6 weeks for anxiety disorders in the perinatal period is 25 postpartum. the 52 that subsyndromal postpartal depressive disorders present assessments were exclusively conducted in 53 can have a detrimental effect on childhood devel. 16 those women with a disorder-onset after delivery. 11). 1 Regarding women with a disorder-onset following Regarding the comorbidity rate of PAD and 2 delivery (including recurrent and first time onset of PDD. which is in have examined preventive programmes for post- 18 line with rates obtained in other studies (15. Using the same cut-off value lence-underestimation. Ramchandani et al.7% for 44% higher education). This enabled an accurate 34 were obtained using different measures of depres. 31 nations. The increased risk associated with 12 class sample was 6. If we are to understand requested to report the onset and history of 51 Ramchandani et al. 49 occasion.1%. 8. First. only few 19 Concerning the frequency of depression in Ger. the prevalence results based on the applied for postpartum anxiety disorders seems 28 EPDS are. (30) present one argument in considered. In line with the delivery. the ÔtotalÕ prevalence 44 models. In implementation of a screening instrument routinely 27 conclusion. the postpartum period and are thus subject to 54 opment.

the broad confidence depression – a meta-analysis. Screening psychi- 48 scher Störungen mit dem ‘‘Gesundheitsfragebogen für 49 References Patienten (-D) . 47 34):10–17.2:659–668. 6. [Screening for psychiatric disorders with the ‘‘Pa- 51 development. Pediatrics 2006. 10 tigator. Neale MC.67:1285–1298. Ross LE.74:139–147. Kurstjens S. Swain AM. Meltzer-Brody S.-Biol. Nonetheless. Structured clinical interview for DSM-IV.316:1884–1886. Diagnostica 2004. 42 18. J Clin 7 might bias results. axis I. 16. Lohr KN. Hunt A. A cross-sectional prev- 26 testing. Anxiety disorders during pregnancy 6 it must be critically discussed whether this procedure and the postpartum period: a systematic review. Prince M. Paulson JF. 15 ening effect). 56 Am J Psychiatry 2001. Wenzel A. soziodemographischen Acknowledgements sowie psychosozialen Faktoren. 35 commence in the early postpartum period. partum Depression Scale. Anxiety and stress in the 11 symptoms (strengthening effect) and in turn an postpartum: is there more to postpartum distress than 12 underestimation of prevalence rates. 9 . Haugen EN. Negri LM. Detection of postpartum 53 an overview. Reck C. Postpartum prevalence study 1 screened positive in the respective maternity hos.33:1161–1167. OÕhara MW. Holden JM. 20 most part insensitive to such disturbances.13:63– 68. Gar- tlehner G. The 19. 12. Leiferman JA. Interactive regulation of validation study]. Zaudig M. Gavin NI. depression. Barnett B. Kavanagh DJ. Wolke D. hypothesis. Postpartum depression. Alspac Study 2 pitals and that those women who did not give birth Team. 2005. Heidelberg (funding period: 2003–2004). Postpartum depression and child studie. With 11. Jackson LC. The SCID.g. Kendler KS. Howie P. Murray L. Maternal antenatal anxiety and behavioural ⁄ emo- tional problems in children: a test of a programming 3 in a clinic were automatically excluded. 39 The authors thank the women who participated in the study 17. Obstet Gynecol 19 appears to be methodologically justified and for the 2005. Fuchs T et al. Psychopathology 2004. 10.150:782– 55 analysis of the genetic epidemiology of anxiety disorders. happened to anxiety? J Affect Disord 2003.Ergebnisse der deutschen Validierungs- 50 1. Postpartulme und später auftretende 36 Depressionen bei Müttern: Prävalenz und Zu- 37 sammenhänge mit obstretischen. Wittchen HU. Cooper PJ. 4. bined effects of postpartum depression in mothers and 24 Furthermore. Sagovsky R. Rates and risks of postpartum 21 respect to statistical analyses. community in the developing world. Br J Psychiatry Suppl 1998. 5. 14. Z Klin Psychol Psychother 38 2001. Kächele H. A review and meta.158:1568–1578. Salem Hospital and the 41 Gynaecological Clinic at the University of Heidelberg) and in India Psychol Med 2001. Hospital. Psychol Med 31 Clinicians should be aware that anxiety and 2003. It is conceivable that the mothers Psychiatry 2006. Glover V. Ageing and mental health in a devel- 40 and the staff with the maternity clinics in Heidelberg (St. Mclean LM. Miller RL. OÕconnor TG. 17 Anxiety-SCID-Screening and ASQ-15 across mea. Golding J. Swinson T. Life events. 44 Dawn Girlich for their help in manuscript preparation. 32 depression represent serious health care problems 15.19:295–311. Boyer P. Heron J. 1996. Matthey S. Br Med J 1998. Cooper PJ.27:253–260. Patel V. JosefÕs oping country: who cares? Qualitative studies from Goa. Loewe B. 25 consequence of an a-inflation caused by multiple 13. alence study of depression at various times after delivery in 27 Controlled studies comparing the prevalence of Mersin province in Turkey.44:1025– 4 With regard to the repeated application of the 1036.8:37–54.50:171–181.37:272–280. Saint ElisabethÕs Hospital. 20. 43 authors also thank Eva Stehle. Graefe K. Development of the 10-item Edinburgh Post- 54 3. Clinical re- 16 atic changes using the PHQ and EPDS as well as the view. Herzog W. Diagnosing 8 under investigation felt that they were Ôin safe handsÕ postpartum depression in mothers and fathers: whatever 9 and supported by the regular contact to the inves. Gruschwitz S. 8. 9. many exploratory results carry the fathers on parenting behaviour. Gaynes BN. Cox JL.6:12. Wunderlich U. Strauß M. 786. Tezcan H. Murray L. Int Rev Psychiatry 22 intervals arising from the small sample sizes (e. On the other depression? BMC Psychiatry 2006. Psychol Med 1997.8:29–35.30:33–41. the repetitive completion of questionnaires disorders at eight weeks postpartum. tionnaire (ASQ-15). Bugdayci R. Zipfel S. implications for the utilization of treatment. Dauber S. and Dipl. 28 postpartum anxiety disorders and their comorbid. J Womens Health 2004. Hettema JM.31:29–38. Goettingen: Hogrefe. Br J Psychiatry 1987. given the lack of system. Individual and com- 23 comorbidity tables) should also be critically noted. 1997.173(suppl. and that this may have lead to a reduction in 7. Contribution to the epidemiology of postpartum depression in Germany – 33 in the first few weeks postpartum and that appro. Screening for anxiety disorders: 45 Postpartum Prevalence Study was supported by a grant from sensitivity and specificity of the anxiety screening ques- the Program of Research Support at the University Medical 46 Faculty. Harrington R. Anxiety symptoms and 13 hand. Rahman A. Dipl.106:1071–1083. Ballestrem CL. Wittchen HU. social sup- 29 ity with depressive disorders between postpartum port and depression in childbirth: perspectives from a rural 30 and non-postpartum women are urgently required. Pallant JF. affect in postpartum depressed mothers and their infants: 21.-Psych. the applied study procedure review of prevalence and incidence. J Child Psychol Psychiatry 2003. Sasmaz CT. Perinatal depression: a systematic 18 surement occasions. Arch Womens 34 priate preventive programmes are required which Ment Health 2005. tient Health Questionaire-Depression’’ – German 52 2. J Anxiety Disord 14 may have resulted in a loss of concentration (weak. 5 screening tools over a relatively short period of time. Darmstadt (Alice-Hospital and Darmstadt Hospital). Iqbal Z.

Miller IW. Can midwifes reduce post- 17 clinic system. Caraveo-Anduaga JJ. Rodrigues M. Dtsch Med 2005. Reck et al. Howie P. 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 10 . assistance at risk for postpartum depression. Westley D. Spitzer RL. Barnett B. Yonkers KA.79:113–126. 1990. Prevention of postnatal distress or depression: an evalua- 8 24. Kavanagh DJ. Andrade L. Arch Womens Ment Health 6 fragebogen für Patienten (2. Ramin SM. Ramchandani P. Frilingos M. Gender.365:2201–2205. OÕconnor TG. 23 Am J Psychiatry 2002. partum psychological morbidity? A randomized trial. 30. mised controlled trial.163:1443–1445. postnatal depression. Am J Psy- 15 27. Evans J.25:215–219. Loewe B. Matthey S.282:1737–1744. Sweeney P. Austin MP. Zipfel S. India.158:1856–1863. American Psychiatric Association. 16 postpartum depression in an innercity maternal health 35. Charles M. 2nd edn. Gesundheits.1:176–173. Ulmer H. Int J Methods Psychiatr Res 2003. Karlsruhe: Pfizer. JAMA 1999. Nguyen T. J Affect Disord 2004. 31. poverty.7:???–??? (Epub 22 29. Rush AJ. 1994. Stein A. The epide. Psychosocial and psychological interventions 11 25. postpartum depression: a study of mothers in Goa. 4 Wochenschr 1998. Bergant AM. Hill C. Desouza N. 10 study. Brief antenatal 19 International Consortium of Psychiatric Epidemiology cognitive behaviour therapy group intervention for the 20 (ICPE) Surveys. Zlotnick C. 33. Howard M. Patel V. Diagnostic and statisti. Dapunt O. 18 miology of major depressive episodes: results from the 36. tingen: Hogrefe. Dennis CL. Br Med J 12 cal manual of mental disorders DSM-IV. Spitzer RL. 7 Questionaire-Depression]. Antenatal prevention of 5 23. Auflage) [Patient Health 1999. depression in the postpartum period and child develop- burgh postpartum depression scale’’ [German validation of ment: a prospective population study. 4th edn. New York: John A preventive intervention for pregnant women on public 14 Wiley & Sons. Agresti A. Goet. Herzog W-D. Paternal 2 Deutschsprachige Fassung und Validierung der ‘‘Edin. 13 26. 1998.123:35–40. Lancet 3 the Edinburgh Postpartum Depression Scale]. 32. 2002. 34. Onset and persistence of chiatry 2006.331:15–23. Walkingshaw SA. Validation and utility tion of an intervention at preparation for parenthood 9 of a self-report version of PRIME-MD: the primary care classes. Heim K. Birth 28. 2005. 1 22. Williams JB. Lavender T. Pearlstein T.159:43–47. Am J Psychiatry 2001. Buist A. Kroenke K. and 3 ahead of print). for prevention of postnatal depression.12:3– prevention of postnatal depression and anxiety: a rando- 21 21. Categorical data analysis. Berglund P. Lumley J et al. J Affect Disord 2007.