Sage Journals: Discover world-class research

Abstract

Objective: To compare existing community treatments for post-partum depression (PPD), treatment efficacy was evaluated for 146 women seeking treatment for PPD in the local community.

Method: Self-report questionnaires were designed to assess clinical depression and anxiety (psychological and physiological), risk factors, treatment satisfaction and the impact of social supports.

Results: Comparison of treated subjects with those on a wait-list demonstrated that depression and the psychological component of anxiety decreased significantly after treatment. Psychological and pharmacological interventions produced similar clinical benefit in the treatment of psychological symptoms, and receiving both treatments in combination was of no added clinical benefit in the immediate or longer term. Individual treatment was associated with more rapid treatment gains initially than group treatment. However, the benefits of group treatment emerged during the 6 months following treatment, leading both interventions to be equally effective in the longer term. Cognitive behavioural therapy was not superior to the combination of non-specific counselling or behavioural strategies, either immediately following treatment or 6 months later.

Conclusions: Clinical efficacy together with other clinical and financial benefits strongly advocate the application of psychological treatment for PPD. Clear parallels between PPD and general depression support the application of knowledge about general depression to extend understanding of PPD and refine clinical management practices.

Keywords

depression management post-partum depression treatment efficacy

Psychotherapeutic and pharmacological treatments both effectively reduce depressive symptomatology [1–4]. In particular, treatment efficacy appears equivalent for cognitive behaviour therapy (CBT) and pharmacotherapy maintenance in the immediate and longer term [5–7]. Although each form of treatment is effective, combining pharmacological and psychological treatments appears to confer little additional advantage [5], [8]; however, combination therapy might be optimal for treating severe disorders [8].

The generalization of these findings to specific subtypes of depression, such as post-partum depression (PPD), is uncertain. Although similar in presentation to other forms of depression, PPD may be influenced by hormonal fluctuations, sleep disturbances, or other adjustments following childbirth. The few outcome studies that have employed a control group support the value of psychotherapeutic and pharmacological treatment for PPD [9–12]. Randomised control trials have demonstrated that fluoxetine and counselling were equally effective in reducing PPD over a 3-month period [13] and research focusing on interpersonal therapy specifically also indicated this to be effective [9], [10], [13], [14]. However, whether pharmacotherapy would enhance treatment effects is unknown. The longer term efficacy of treatments for PPD has not been evaluated.

Little is known about the influence of process variables, such as the treatment setting, on treatment outcome for PPD. Group and individual treatment approaches seem equally effective for general depression immediately following treatment and 6 months later [15–18]. While group CBT was effective for PPD in the study by Meager and Milgrom [19], there is no comparative research examining the efficacy of individual versus group treatment for PPD. General satisfaction with services and therapist–client rapport are thought to be important in the treatment of depressive disorders [3], but this remains unexplored in PPD.

The aim of the present study was to determine the immediate and long-term efficacy of psychological and pharmacological interventions currently available for treating PPD. We examined the role of social support, medication and treatment setting to identify the crucial elements for effective intervention for PPD, and permit comparison with the extensive body of research on general depression.

Method

Setting

The research was conducted across a range of settings within and outside the Perth metropolitan area, Western Australia. Subjects were obtained via clinics and a range of health professionals offering treatment for PPD.

Instruments

Clinical questionnaires

Structured questionnaires were employed to evaluate subjects across three assessment intervals: prior to treatment, immediately following treatment, and at 6 months follow-up.

The pre treatment questionnaire evaluated the following areas.

Information section: obtained demographic information, medication history, social support network, and evaluated risk factors for PPD (e.g. history of depression, stressful life events before, during and after the birth).

Medication status: evaluated the type, duration and current and previous doses of medication for anxiety and depression.

Depression: evaluated using the Edinburgh Postnatal Depression Scale (EPDS) [20]. The scale is brief and easy to administer, and research evidence supports its validity in terms of detection, specificity, sensitivity to change over time [20] and predictive power [21].

Physiological and psychological anxiety: each aspect of anxiety was evaluated independently and with reference to the past 7 days. Psychological anxiety was assessed using the 20-item Trait Form of the State Trait Anxiety Inventory [22] which has proven validity and reliability. The tool is also sensitive to change in anxiety over time [23] and has been used previously to study post-partum anxiety [24]. Physiological anxiety was assessed through two subscales of the General Health Questionnaire (GHQ) [25] which specifically pertained to stress and physiological symptoms of anxiety. The GHQ is considered a valid and reliable instrument, is sensitive to change [23] and is also supported for the assessment of anxiety in the postpartum [26].

Social support scales developed by Collins et al. [27] were selected to measure quantity and quality of social support across three dimensions: general support, support received from the baby's father, and support from healthcare providers. Both general and baby's father support were assessed in relation to the past 7 days, while support from healthcare providers was examined since becoming pregnant. While the original healthcare provider scale was designed to evaluate support from doctors and nurses, the scales were adapted to enable other possible healthcare services to be identified and rated. These scales were selected as they represent areas associated most closely with the development of PPD, were developed and standardized on a comparable population sample, were designed for repeated administration, and have psychometric data derived from validity studies [28] which support their application.

The post-treatment and follow-up questionnaire containing sections (ii)–(v) above, re-evaluated depression, anxiety, general support and support received from the baby's father in relation to the past 7 days. Support from healthcare providers was assessed in relation to the past month at post-treatment, and to the past 6 months at follow-up.

The waitlist questionnaire also measuring the above dimensions was administered to subjects presenting for treatment when there was a time gap between initial presentation and commencement of treatment.

Treatment description form

As the nature of treatment was likely to vary across health professionals and treatment settings, a detailed treatment description form was developed to identify specific strategies and ascertain their role in the treatment process. Information was obtained about the professional background of the treatment provider(s), treatment duration, inclusion of partner/significant other in the treatment process, and specific strategies used in the treatment protocol. Ascertaining specific strategies and their proportional use in treatment not only validated assignment of subjects to treatment groups for the study, but also allowed closer analysis of treatment components with respect to clinical outcome.

Satisfaction with services

Since delivery of any intervention may also impact on treatment outcome, satisfaction with services was measured using the Sharp Consumer Satisfaction Questionnaire [27] which was administered by telephone 2 weeks after completion of treatment. The scale measures satisfaction across five subscales: general satisfaction, perceived helpfulness, accessibility, therapist respect and partnership. The instrument is brief, and reported to be valid and reliable in its assessment of satisfaction with services [23].

Procedure

Health professionals in the public and private sectors who were likely to come into contact with women presenting with PPD asked women to complete a pre-treatment questionnaire after the initial PPD diagnosis. Following treatment, women were given or sent a posttreatment questionnaire, and within 2 weeks of receiving this, satisfaction with services was measured via a telephone interview. Meanwhile, healthcare providers were sent the treatment description form to obtain details of the treatment that each subject received. Six months following treatment, each subject was telephoned and reminded of the study prior to posting the follow-up questionnaire, and further reminders were conducted as required to maximize response rates.

Results

Sample description

Initial data was collected from 188 women who had sought or been referred to treatment for PPD. Of these, 42 were excluded from the final sample as: they were not contactable at post-treatment (n = 28); were not considered to have PPD by their healthcare provider (n = 10); refused to take part in the study (n = 2); or stopped treatment prior to completion (n = 2). Thus, the final treatment sample consisted of 146 subjects. Analysis of demographic variables indicated a representative population sample in terms of marital status, religious background, economic and educational status.

Subjects were distributed unevenly across treatment interventions, which can be considered indicative of the relative utilization of particular treatment services for PPD (the actual number of subjects in each treatment condition will be given with reference to each hypothesis). A wait-list condition was derived from a few subjects who had to wait at least 3 weeks to receive group intervention (range 3–10 weeks, mean 4 weeks). On initial presentation for the various treatments, half the subjects were already taking antidepressant medication, highlighting the major role that general practitioners play in the treatment of PPD. However, subjects treated solely with medication are likely to be under-represented because collecting data from general practitioners proved to be extremely difficult.

Hypothesis 1: psychological and pharmacological intervention (CBT, support counselling and pharmacotherapy) reduces clinical symptomatology more than receiving no intervention (wait-list)

Although subjects in treatment (n = 136) greatly outnumbered those on a wait-list (n = 10), comparisons may still be valid because the groups were similar in terms of clinical status and social support received across all scales, and levels of support also remained consistent for both groups across assessment intervals.

The role of treatment in the reduction of symptomatology was investigated in 2 × 2 (group [treatment, wait-list] × time [pretreatment, post-treatment]) analyses of variance. Neither depression nor anxiety changed significantly across the waitlist period. Receiving treatment resulted in significant decreases in depression when comparing pre- (mean = 16.49, SD = 5.29) and post- (mean = 10.06, SD = 6.30) scores (group–time interaction, F_1,137 = 11.89, p < 0.05), whilst the decrease in psychological anxiety approached significance (group–time interaction, F_1,134 = 3.08, p = 0.081). Physiological anxiety scores also decreased (particularly for those receiving treatment), but this could not be attributed to the effect of treatment for PPD specifically because the decrease was not significantly greater than in the waitlist condition.

Hypothesis 2: the reduction of clinical symptoms of PPD for subjects receiving cognitive behavioural therapy will be comparable with those who receive pharmacological treatment (medication)

Subjects who received CBT with medication had more contact time (mean = 10.56, SD = 5.37) than those who received CBT without medication (mean = 8.31, SD = 3.88) or who received medication alone (5.00, SD = 5.29) (F_2,86 = 32.74, p < 0.05). However, subjects were similar in terms of initial clinical status on the EPDS and both anxiety scales, social support and ratings of satisfaction with treatment received.

Immediate treatment gains

Outcome data (Table 1) revealed that medication was no more effective than CBT for the treatment of depression, psychological anxiety or physiological anxiety. However, subjects treated with CBT (alone or in combination with medication) had greater decreases in depression (group–time interaction, F_1,82 = 11.08, p < 0.05) and psychological anxiety (group–time interaction, F_1,79 = 5.98 p < 0.05) following treatment than those who received medication alone. Combining CBT with medication was of no additional benefit to receiving CBT alone on any clinical scales.

Table 1.

Pre- post and follow up ratings of depression and anxiety across comparative groups

Long-term treatment gains

Since only three subjects received medication without psychological treatment at 6 months follow-up, a full analysis across the three conditions was not possible. Subjects who were still taking medication at follow-up (n = 18) had significantly more access to healthcare provider support during follow-up than unmedicated subjects (F_1,31 = 10.25, p < 0.05], while general support and support received from the baby's father did not differ between groups.

Although depression continued to decline during the 6 months following intervention (main effect for time, F_1,32 = 4.31, p < 0.05), non-significant interaction effects indicates that receiving medication in conjunction with CBT did not influence levels of depression, psychological anxiety or physiological anxiety at 6 months follow-up.

Hypothesis 3: subjects receiving individual treatment will have a greater reduction in PPD symptoms than those receiving group treatments

To evaluate the role of treatment setting, subjects who received CBT in group or individual settings were selected. Cognitive behavioural therapy was chosen over other treatments because it involved standard techniques and strategies, resulting in three separate conditions (Table 1).

Subjects treated in groups had more contact time (17 h) than those treated on a purely individual basis (7 h) (F_2,86 = 32.74, p < 0.05). Partner involvement across the three conditions was negligible, with no involvement for those receiving group treatment alone, and minimal involvement for those treated individually (mean = 0.56 h, SD = 1.2) or in the combination setting (mean = 0.66 h, SD = 1.00). Almost half the subjects were taking antidepressant medication on initial presentation, and this did not differ between the three treatment conditions at any of the assessment intervals. On initial presentation subjects had similarly high depression scores (averaging in the moderate clinical range) and anxiety. Satisfaction ratings were generally high, but differed across conditions (F_2,67 = 3.72, p < 0.05) as those receiving individual treatment alone indicated greater levels of satisfaction than those receiving treatment in group settings (p < 0.05, Student Newman-Keuls test).

Subjects reported similar levels of social support on initial presentation, and support ratings on all scales did not differ during treatment.

Treatment outcome (immediate and long-term)

To determine the specific role of treatment setting, planned contrast analyses were performed between: (i) subjects treated in groups (alone and in conjunction with individual treatment) versus those treated individually; and (ii) subjects treated in combined settings versus those treated only in groups. Depression decreased significantly following treatment (main effect for time, F_1,83 = 128.74, p < 0.05) and these treatment gains were maintained at follow-up. Comparison of subjects treated in groups (alone and in conjunction with individual treatment) versus those treated individually revealed a significant group × time interaction (F_1,83 = 16.98, p < 0.05). In particular, depression was significantly lower at post treatment in subjects treated individually as opposed to those who received group or combined intervention (t₈₄ = 3.9, p < 0.05), despite similar depression scores on initial presentation.

At follow-up there was also a significant decrease in depression (main effect for time, F_1,63 = 11.36, p < 0.05), particularly in those treated in both group and individual settings (group–time interaction, F_1,63 = 5.95, p < 0.05). Specifically, depression continued to decline for those who had been treated in the combined setting (t₃₄ = 5.26, p < 0.05), while there was no change for those treated in groups only.

Psychological anxiety declined at post-treatment (main effect for time, F_1,80 = 50.72, p < 0.05) and during the 6 months follow-up (main effect for time, F_1,61 = 6.30, p < 0.05), particularly in those who received individual treatment only (group–time interaction, F_1,80 = 13.43, p < 0.05). Specifically, anxiety decreased more for those treated only on an individual basis than for subjects treated in groups (t₈₃ = 3.99, p < 0.05), despite similar percentile scores prior to treatment. Physiological anxiety also decreased significantly during treatment (main effect for time, F_1,79 = 60.22, p < 0.05) and remained low at follow-up, but this could not be attributed to the treatment setting per se.

Hypothesis 4: subjects receiving group CBT will have a greater reduction of PPD symptoms than those receiving group behaviour therapy, particularly in the longer term

To examine the role of cognitive therapy relative to behavioural therapy, those who received CBT in a group setting were compared with those who received behavioural therapy. Since only a few subjects received cognitive therapy only (especially at follow-up), these subjects were not included in the analysis.

First, the therapeutic techniques employed for the selected subjects (derived from the treatment description form) were assessed. Although both groups had a high frequency of general support counselling in the treatment process, those in the CBT group received more behavioural intervention techniques than those in the behavioural support group, and may be considered representative of community treatment interventions.

Contact time was significantly greater for those who received group behaviour therapy than those who received CBT therapy (25 ± 17 h vs. 16 ± 7 h, t₃₂ = 2.74, p < 0.05), whereas partner involvement in treatment was negligible in both groups. At post-treatment, a higher proportion of subjects were on medication in the behavioural therapy group than in the CBT group [67% vs. 19%, Fishers exact test, p < 05]. Both groups were similar in terms of clinical presentation and satisfaction with treatment services. Although those in the CBT group had received more general support before treatment (F_1,31 = 6.50, p < 0.05), social support during and following treatment was similar across groups.

Treatment outcome (Immediate and long-term)

Following treatment, there were significant reductions in depression (main effect for time, F_1,31 = 30.54, p < 0.05), psychological anxiety (main effect for time, F_1,30 = 6.82, p < 0.05) and physiological anxiety (main effect for time, F_1,29 = 19.51, p < 0.05). Non-significant changes over time at follow-up indicated that these treatment gains were maintained 6 months later in both treatment conditions. While CBT was no more effective than behavioural-based supportive counselling, confounding effects of greater medication use and greater treatment duration for those in the latter group may result in underestimation of the efficacy and efficiency of CBT for this sample.

Discussion

The findings indicate the importance of receiving intervention for PPD. When compared with subjects on a wait-list, depression and psychological symptoms of anxiety decreased in treated subjects. It is important to consider however, that the wait-list sample was small (and thus possibly unrepresentative), and that the waitlist period was shorter than the mean duration of treatment. While care must be taken not to overstate the importance of treatment based on this study, nevertheless the findings are consistent with the few existing treatment studies on PPD [9–11] as well as with the general depression literature [1–4] indicating the clinical benefit of receiving treatment when compared with wait-list conditions.

One important aim of the present study was to evaluate the role of medication in treating PPD. Indicative of community treatment interventions, subjects treated with medication alone had significantly less contact time with healthcare professionals during the treatment period than did those receiving CBT. Overall, both treatments were comparable in reducing symptomatology immediately following treatment, and thus were consistent with other single PPD studies demonstrating the efficacy of CBT [19] and medication [11] in reducing depressive symptoms. These results also parallel the literature for general depression [1]. There was no advantage to receiving medication in conjunction with CBT in the immediate term, a finding replicated with reference to PPD [11] as well as general depression [8].

Although insufficient sampling at follow-up prevented a direct comparison between medication and CBT in the longer term, some analysis of the role of medication was made through comparison of subjects who received CBT with and without medication 6 months following treatment. Since over half the subjects initially taking medication were still on medication 6 months later, these subjects made greater demands on health services during follow-up than those who did not receive medication. Importantly, very few subjects who had not received medication initially had started taking medication following psychological treatment.

Treatment gains were maintained for both medicated and non-medicated subjects at follow-up; this is consistent with the literature for general depression [5]. However, there was no clinical advantage to receiving medication in conjunction with CBT. Since medication was of no added benefit during or following treatment, the need to use medication in conjunction with CBT is questionable, at least for subjects who are not severely depressed.

When examining the role of treatment setting on outcome with subjects who received CBT, those treated in group settings had almost double the contact time than those treated individually. Satisfaction ratings were also greater for those treated on an individual basis, and this, together with changes in support received from the baby's father, was considered with respect to treatment outcome.

Results clearly demonstrate the efficacy of CBT, regardless of treatment setting, in the reduction of depression and psychological anxiety following treatment and the maintenance of treatment gains 6 months later. The greater efficiency of individual treatment immediately following the treatment may be attributable to subjects being unable to benefit initially from treatment in a group setting if lacking in social skills and energy. Since combination treatment was more effective during the 6 months following treatment than during the intervention, decreases in depression and psychological anxiety may help subjects to benefit from social interactions. Despite more rapid treatment gains initially by those treated individually, all treatment settings were equally effective in reducing depression and psychological anxiety in the longer term, replicating studies for general depression [15–18].

Physiological anxiety decreased significantly following treatment, but this could not be attributed to receiving treatment itself. In other PPD research [19], CBT treatment resulted in decreases in tension as measured on the Profile of Mood States Questionnaire which measures a combination of general psychological and physiological symptoms of anxiety. However, the present results suggest that treatment may be more effective for psychological than physiological aspects of anxiety, a point that may be worthy of consideration for future treatment outcome research.

Our next aim was to determine whether specific components of CBT (behaviour therapy, cognitive therapy or combination CBT) influenced treatment outcome. Non-specific support counselling in combination with behavioural interventions was as effective as CBT in reducing depression, psychological anxiety and physiological anxiety immediately following intervention, and these treatment gains were maintained 6 months later. However, it is important to recognize the potential impact of treatment differences on outcome. In particular, the greater treatment duration and greater use of medication for those in the behavioural intervention may have enhanced treatment outcome for this group. In sum, CBT appears to be at least as effective and probably more efficient than behaviour therapy for PPD.

The role of cognitive versus behavioural treatments has yet to be directly compared for PPD, but support does exist for CBT [11], [19] and supportive counselling [9], [10]. However, a social support group intervention where mothers exercised with their infants and were instructed in infant massage was ineffective in the treatment of PPD [29]. Behavioural intervention in the present study encompassed mastery events, pleasant event scheduling and relaxation, and replicated the efficacy of this form of treatment for general depression [1].

Implications/limitations

There are obvious limitations to conducting research in the community, such as the absence of a true control group and reliance on self-report data. However, the present uncontrolled study approach has provided some insight into the resources required for community treatment of PPD, and suggests the need for a randomised controlled trial to validate these important findings. Psychological intervention was comparable to pharmacological treatment in terms of clinical efficacy. However, the absence of unwanted side-effects associated with medication, the impact on breast-feeding and risk of overdose render psychological treatment as clearly the optimal treatment for PPD. Furthermore, shorter treatment duration, an absence of medication costs, and the ability to effectively implement treatment in group settings, render psychological treatments more timeand cost-effective, and hence the most efficient approach to treating PPD. This highlights the need to channel financial and community resources to further develop and expand psychological treatment services for the treatment of mild to moderately depressed women with PPD.

Finally, the results from the current study parallel those for general depression. Hence the integration of knowledge and research practices for general depression should be applied to extend understanding of PPD and to refine clinical management practices.

References

1. Elkin

Shea

Watkins

J T

. National Institute of Mental Health treatment of depression collaborative research program. Archives of General Psychiatry 1989; 46: 971–982.

2. US Department of Health and Human Services Depression Guideline Panel . Depression in primary care: treatment of major depression. AHCPR, Rockville 1993; 71–73.

3. Scott

. Psychological treatments of depression: an update. British Journal of Psychiatry 1995; 167: 289–292.

4. Oei

TP S

Shuttlewood

G J

. Specific and nonspecific factors in psychotherapy: a case of cognitive therapy for depression. Clinical Psychology Review 1996; 16: 83–103.

5. Syravynski

Verreault

M D

Gaudette

. The treatment of depression with group behavioural cognitive therapy and Imipamine. Canadian Journal of Psychiatry 1994; 39: 387–390.

6. Shea

Pilkonis

Beckham

Collins

Sotsky

Elkin

. Course of depressive symptoms over follow-up: findings from the NIMH treatment of depression collaborative research program. Archives of General Psychiatry 1992; 49: 782–787.

7. Evans

Hollon

DeRubeis

. Differential relapse following cognitive therapy and pharmacotherapy of depression. Archives of General Psychiatry 1992; 49: 802–808.

8. Hollon

S D

Shelton

R C

Loosen

P T

. Cognitive therapy and pharmacotherapy for depression. Journal of Consulting and Clinical Psychology 1991; 59: 88–99.

9. Holden

J M

Sagovsky

Cox

J L

. Counselling in a general practice setting: controlled study of health visitor intervention in treatment of postnatal depression. British Medical Journal 1989; 298: 223–226.

10.

10. Wickberg

Hwang

C P

. Counselling of postnatal depression: a controlled study on a population based Swedish sample. Journal of Affective Disorders 1996; 39: 209–216.

11.

11. Appleby

Warner

Whitton

Faragher

. A controlled study of fluoxitine and cognitive-behavioural counselling in the treatment of postnatal depression. British Medical Journal 1997; 314: 932–936.

12.

12. Cooper

Murray

. Prediction, detection, and treatment of postnatal depression. Archives of Disease of Childhood 1997; 77: 97–101.

13.

13. O'Hara

M W

Scott

Gorman

Wenzel

. Efficacy of interpersonal psychotherapy for postpartum depression. Archives of General Psychiatry 2000; 57: 1039–1045.

14.

14. Stuart

O'Hara

M W

. Interpersonal therapy for postpartum depression: a treatment program. Journal of Psychotherapy Practice and Research 1995; 4: 18–29.

15.

15. Shaffer

Shapiro

Sank

Coghlan

. Positive changes in depression, anxiety and assertion following individual and group cognitive behaviour therapy intervention. Cognitive Therapy and Research 1981; 5: 149–157.

16.

16. Brown

R A

Lewinsohn

P M

. A psychoeducational approach to the treatment of depression: comparison of group, individual and minimal contact procedures. Journal of Consulting and Clinical Psychology 1984; 52: 774–783.

17.

17. Teri

Lewinsohn

P M

. Individual and group treatment of unipolar depression: comparison of treatment outcome and identification of predictors of successful treatment outcome. Behaviour Therapy 1985; 17: 215–228.

18.

18. Toseland

R W

Siporin

. When to recommend group treatment: a review of the clinical and research literature. International Journal of Group Psychotherapy 1986; 36: 171–201.

19.

19. Meager

Milgrom

. Group treatment for postpartum depression: a pilot study. Australian and New Zealand Journal of Psychiatry 1996; 30: 852–860.

20.

20. Cox

J L

Holden

Sagovsky

. Detection of postnatal depression: development of the 10-item Edinburgh postnatal depression scale. British Journal of Psychiatry 1989; 150: 782–786.

21.

21. Murray

Carothers

A D

. The validation of the Edinburgh Post-natal Depression Scale on a community sample. British Journal of Psychiatry 1990; 157: 288–290.

22.

22. Spielberger

C D

Gorsuch

R L

Lushene

Vagg

P R

Jacobs

G A

. Manual for the State-Trait Anxiety Inventory (Form Y) (Self-Evaluation Questionnaire). Consulting Psychologists Press, Palo Alto, CA 1983.

23.

23. Andrews

Peters

Teeson

. The measurement of consumer outcome in mental health: a report to the National Mental Health Information Strategy Committee. Clinical Research Unit for Anxiety Disorders, Sydney 1994.

24.

24. Barnett

Schaafsma

M F

Guzman

A M

Parker

G B

. Maternal Anxiety: a five year review of an intervention study. Journal of Child Psychology and Psychiatry 1991 1981; 32: 423–438.

25.

25. Goldberg

Williams

. A user's guide to the General Health Questionnaire. NFER-Nelson, Windsor, UK 1988.

26.

26. Collins

N L

Dunkel-Schetter

Lobel

Scrimshaw

S C

. Social support in pregnancy: psychosocial correlates of birth outcomes and postpartum depression. Journal of Personality and Social Psychology 1993; 65: 1243–1258.

27.

27. Tanner

B A

. A multi-dimensional client satisfaction instrument. Evaluation and Program Planning 1982; 5: 161–167.

28.

28. Nott

Cutts

. Validation of the 30-item general health questionnaire in postpartum women. Psychological Medicine 1982; 12: 409–413.

29.

29. Fleming

A S

Klein

Corter

. The Effects of a social support group on depression, maternal attitudes and behaviour in new mothers. Journal of Child Psychology and Psychiatry 1992; 33: 685–698.