|
The Effect of Peer Support on Postpartum Depression: A Pilot Randomized Controlled Trial
Intervention
Peer support was defined as a specific type of social support that incorporates informational, appraisal (feedback), and emotional assistance. This lay assistance is provided by volunteer individuals who are not part of the mother’s own family or immediate social network but who possess experiential knowledge of the targeted behaviour or stressor (that is, PPD) and similar qualities (such as similar residency, age, socio- economic status, or ethnicity). Because no program previously existed, a mother-to-mother telephone-based support intervention was developed, entitled “Mothers Helping Mothers with Postpartum Depression.”
Recruitment and Training of Peer Volunteers. Potential volunteers were reached through the distribution of flyers, ads in the local newspapers, and by word of mouth. Nineteen experienced mothers volunteered and met the following selection criteria: history of and recovery from PPD, desire to help new mothers, and completion of a 4-hour training session. The training session focused primarily on developing telephone support and referral skills and included role-playing, verifying problem-solving skills, and developing the ability to refer mothers to an appropriate professional service. The author developed a 118-page handbook for distribution to all peer volunteers. This handbook outlined professional services available for referral and was to be used as a reference guide. It also incorporated various topics, including a definition of peer support, potential benefits, how to develop a relationship, skills and techniques for effective telephone support, general PPD information, and the helping process.
During the training sessions, the author described the pilot trial and answered questions. Peer volunteers who wanted to participate in the study were assigned a volunteer code number to promote confidentiality, were requested to complete a demographic form, and were given peer-volunteer activity logs that included postage-paid, addressed envelopes. These activity logs enabled the peer volunteers to document their specific interactions with the trial participants. Most of the peer volunteers who participated in the pilot trial were married (90%) and had some postsecondary education (79%); 84% were multiparous, and 42% were employed outside the home, either full-time or part-time. In addition to the peer volunteers, 2 regional volunteer coordinators were selected. The volunteer-coordinator responsibilities included attending the training sessions, matching trial participants with an appropriate peer volunteer, ensuring support was initiated, and providing assistance to peer volunteers, as required.
Intervention Procedures. The volunteer coordinator paired each new mother with a peer volunteer, based on residency and availability. There were no efforts to standardize the number of new mothers supported by a peer volunteer at any particular time or throughout the pilot trial. Peer volunteers were contacted within 1 to 2 days of trial enrolment and provided with the new mother’s telephone number and address. Peer volunteers were asked to contact the new mother within 48 hours and as frequently thereafter as the individual mother deemed necessary. To individualize the intervention to each mother’s specific needs and to give credibility to the peer volunteers’ experiential knowledge, contact frequency was not standardized, a known effective strategy (63). To enhance understanding of the peer-support intervention and to monitor trial fidelity, the volunteer activity logs were reviewed in relation to the peer-volunteer interactions.
Outcome Measures
Depressive Symptomatology. The primary outcome analyzed in this pilot trial was depressive symptomatology, determined at 8 weeks postrandomization and defined as a score > 12 on the EPDS. The EPDS is a 10-item, self-report instrument developed to assess maternal mood. Items are rated on a 4-point scale to produce a summative score ranging from 0 to 30, with higher scores indicating lower maternal mood. This inter-nationally used instrument does not diagnose PPD, which is possible only through a psychiatric interview, but it is the most frequently used instrument to assess for PPD symptomatology (44). Validated by standardized psychiatric interviews with large samples (18), the EPDS has well-documented reliability and validity in multiple languages (64); a large community study revealed a specificity of 92.5% and a sensitivity of 88% (65). Apart from its widespread use, the EPDS was chosen to identify mothers and measure the primary outcome because it has the following features: ease of administration (including via telephone) (66); uncomplicated interpretation (8); high maternal and health-professional acceptance (18,65,66); good sensitivity, specificity, and predictive power when a cut-off point of > 9 has been used for community-level screening (65,66); and simplicity of incorporation into routine clinical practice (8), should future PPD programs be developed. The respective Cronbach’s alpha coefficients for this scale at baseline and the 4- and 8-week assessments were 0.87, 0.88, and 0.89.
Maternal Self-Esteem. This outcome was measured using the Rosenberg Self-Esteem Scale (SES) (67), a 10-item, self-report instrument developed to assess global feelings of self-worth. Items are rated on a 4-point Likert-type scale to produce a summative score ranging from 10 to 40, with higher scores indicating higher levels of self-esteem. The SES has been psychometrically tested using diverse samples, including new mothers (68,69), and it has demonstrated good reliability and validity. Lower self-esteem has been associated with more depressive symptomatology in postpartum women (4,44,70,71) and is amenable to psychosocial interventions (72,73). The respective Cronbach’s alpha coefficients for this scale at baseline and at the 8-week assessment were 0.93 and 0.87. It was hypothesized that peer support would increase self-esteem.
Child-Care Stress. This outcome was measured using the Child-Care Stress Checklist (CCSC), a 20-item, self-report instrument developed to assess stress related to the birth of a new baby. Items are rated on a yes–no scale to produce a summative score ranging from 0 to 20, with higher scores indicating higher levels of child-care stress. The CCSC was developed by the author based on a literature review of common maternal stressors and in response to the lack of a valid measure. The CCSC was psychometrically tested with 645 new mothers and demonstrates good reliability and validity (unpublished data). For this study, the respective Cronbach’s alpha coefficients at 4 and 8 weeks postpartum were 0.81 and 0.81; the retest correlation between weeks 4 and 8 was 0.72. Because higher child-care stress has been associated with more depressive symptomatology in postpartum women (44,74–76), construct validity was supported through significant correlations between the CCSC and corresponding EPDS at 4 weeks (r = 0.61) and 8 weeks (r = 0.60) postpartum. The respective Cronbach’s alpha coefficients for this scale at baseline and at the 8-week assessment were 0.80 and 0.78. It was hypothesized that peer support would decrease child-care stress.
Maternal Loneliness. This outcome was measured using the short version of the UCLA Loneliness Scale (LS) (77), a 10-item, self-report instrument designed to measure the extent to which an individual feels emotionally and socially lonely. Items are rated on a 4-point Likert-type scale to produce a summative score ranging from 10 to 40, with higher scores indicating higher degrees of loneliness. The LS has been shown to have good psychometric characteristics with diverse populations, including new mothers (77). Higher perceived loneliness has been associated with more depressive symptomatology in postpartum women (7,78), and with depression in general (77). The respective Cronbach’s alpha coefficients for this scale at baseline and at the 8-week assessment were 0.90 and 0.91. It was hypothesized that peer support would decrease maternal loneliness.
Maternal Perceptions of Peer Support. This outcome was measured using the Peer Support Evaluation Inventory (PSEI), a 4-subscale self-report instrument developed to measure a mother’s perception of the support received from her peer volunteer. The subscales assess supportive interactions (for example, emotional, appraisal, and informational support), relationship qualities (for example, perceived peer responsiveness, extent of interdependence, and peer qualities), perceived benefits (for example, potential health outcomes related to the 3 theoretical perspectives of social integration, stress and coping, and social constructionism), and satisfaction with support (for example, access, convenience, and perceived quality). This self-report instrument is based on extensive theoretical work completed by the author during a postdoctoral research fellowship. Content validity was assessed by 1 Canadian and 2 US social-support experts. The Cronbach’s alpha coefficients for the subscales were as follows: supportive functions = 0.95; relationship qualities = 0.96; perceived benefits = 0.92; and satisfaction = 0.96. This outcome was assessed via mailed questionnaires sent between 12 and 14 weeks postrandomization.
Peer-Volunteer Perceptions of Peer Support. This outcome was measured using the Peer Volunteer Experience Questionnaire (PVEQ) (78). Questions are related to program training and expectations, interactional characteristics, volunteer roles, intrapersonal effect, and recruitment and retention. This outcome was assessed via mailed questionnaires sent at the end of the pilot trial.
Peer-Volunteer Activities. All intervention activities, including telephone discussions, left messages, and face-to-face contacts, were documented by the peer volunteers using the Peer Volunteer Activity Log (63). Peer volunteers were asked to return all activity logs 8 to 12 weeks after being matched with a new mother.
Data Management and Analysis
Data were entered into a data management system by a research assistant blind to group allocation, and logic and range checks were used to verify the accuracy of the data. Any discrepancies were compared with the original data forms. Discontinuation of the intervention (n = 1) did not entail the participant’s exclusion from the study, and an “intention to treat” approach was used to analyze these data. The data are presented using descriptive statistics (means, SDs, or proportions). For categorical data, Pearson’s chi-square test was used to examine differences between the 2 study groups. Independent 2-sample t-tests were conducted for data at the interval level of measurement. Pearson’s correlations were used to examine the relation between the frequency of peer-volunteer contacts and maternal mood. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were estimated.
Results
Of the 501 mothers who completed the screening EPDS at 8 weeks postpartum, 96 scored > 9 (19.2%). Of these mothers, a further 33 were ineligible. The most common reasons for ineligibility were current use of antidepressant medication (n = 14), chronic psychiatric condition (n = 9), and limited English (n = 8). It is important to note that the ineligibility of 14 mothers owing to current antidepressant medication use clearly suggests that peer support is not an adequate treatment option for all cases of postpartum depression. Of the 63 eligible mothers, 21 (33%) declined enrolment, most frequently citing a sufficient support network (n = 9). Thus, the acceptance rate for enrolment in the trial was 67%, indicating that, while most mothers were receptive to the offer of peer support, a single intervention is not acceptable to all mothers. No significant baseline differences in age, education, income, and EPDS score were found between mothers who participated in the trial and mothers who were eligible to participate but refused. Of the 42 participants enrolled, only 1 mother (in the control group) did not complete the 8-week assessment. Table 1 presents characteristics of the trial participants. There were no statistically significant differences between the 2 groups.
|
Table 1 Baseline characteristics of randomized participants
|
|
Variable
|
Level
|
Experimental group
(n = 20)
|
Control group
(n = 22)
|
|
|
|
n
|
%
|
n
|
%
|
|
Age
|
18–24 years
25–34 years
³ 35 years
|
2
16
2
|
10
80
10
|
4
16
2
|
18
73
9
|
|
Marital status
|
Married or common-law
|
20
|
100
|
22
|
100
|
|
Education
|
High school
College or university
|
4
16
|
20
80
|
7
15
|
32
68
|
|
Born in Canada
|
Yes
|
17
|
85
|
18
|
82
|
|
Annual household income
|
$0–39 999
$40 000–79 999
$80 000 +
|
9
8
1
|
50
44
6
|
12
6
4
|
55
27
18
|
|
Parity
|
Primiparous
|
7
|
35
|
7
|
32
|
|
Mode of delivery
|
Caesarean section
|
3
|
15
|
4
|
18
|
|
History of postpartum depression
|
Yes
|
4
|
20
|
4
|
18
|
Significantly more mothers in the experimental group exhibited decreased depressive symptomatology at the 4-week (c2 = 5.18, df = 1, P = 0.02) and 8-week (c2 = 6.37, df = 1, P = 0.01) assessment (Table 2). Specifically, at the 4-week assessment, 40.9% (n = 9) of mothers in the control group had EPDS scores > 12, compared with only 10% (n = 2) in the experimental group. Comparable findings were found at the 8-week assessment, at which time 52.4% (n = 11) of mothers in the control group and 15% (n = 3) of mothers in the experimental group continued to score > 12 on the EPDS. After the baseline characteristics evident in Table 1 were controlled, logistic regression was conducted to assess the effect of the peer-support intervention on depressive symptomatology. The results indicated that the peer-support intervention significantly decreased depressive symptomatology at the 8-week assessment (OR = 4.7; 95% CI, 0.91 to 25.46). Specifically, mothers who received the peer-support intervention were over 4 times more likely to have decreased depressive symptomatology, compared with mothers who did not receive the supportive intervention.
|
Table 2 EPDS scores between groups
|
|
EPDS score
|
Time
|
Experimental group
(n = 20)
n (%)
|
Control group
(n = 22)
n (%)
|
OR
|
95% CI
|
P
|
|
> 9
|
4 weeks
8 weeks
|
9 (45)
7 (35)
|
16 (72.7)
16 (76.2)a
|
3.26
5.94
|
0.90 to 11.81
1.52 to 23.18
|
0.06
0.008
|
|
> 12
|
4 weeks
8 weeks
|
2 (10)
3 (15)
|
9 (40.9)
11 (52.4)a
|
6.23
6.23
|
1.15 to 33.77
1.40 to 27.84
|
0.02
0.01
|
|
an = 21; EPDS = Edinburgh Postnatal Depression Scale
|
When EPDS mean scores were assessed at 4 weeks, a significant difference was found between mothers in the control (mean 12.1, SD 4.6) and experimental (mean 8.5, SD 3.7) groups (t [40] = 2.8, P = 0.008). Similar group differences were found at the 8-week assessment (t [39] = 2.9, P = 0.006). While not statistically significant, positive trends favouring the experimental group were found in mean scores related to maternal self-esteem, child-care stress, and loneliness (Table 3).
|
Table 3 Secondary outcomes between groups
|
|
Variable
|
Time
|
Experimental group
(n = 20)
Mean (SD)
|
Control group
(n = 22)
Mean (SD)
|
Mean difference
|
|
Maternal self-esteem
|
Baseline
|
28.25 (4.19)
|
27.82 (3.92)
|
0.43
|
|
|
8 weeks
|
30.00 (4.21)
|
28.57 (3.83)a
|
1.43
|
|
Child-care stress
|
Baseline
|
7.10 (3.24)
|
7.40 (3.44)
|
0.30
|
|
|
8 weeks
|
4.95 (2.68)
|
6.48 (3.63)a
|
1.53
|
|
Maternal loneliness
|
Baseline
|
24.75 (4.88)
|
25.18 (5.50)
|
0.43
|
|
|
8 weeks
|
20.37 (5.23)
|
23.91 (6.07)a
|
3.54
|
|
an = 21
|
1 | 2 | 3 | 4
|
