Psychological interventions for maternal depression among women of African and Caribbean origin: a systematic review

The review protocol was developed and registered with the International Prospective Register of Systematic Reviews (PROSPERO; CRD42019149392). The review used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) to ensure methodological rigour and global best practice for systematic review [20]. Utilising Boland, Cherry and Dickson’s [21] model, a protocol outlining the inclusion criteria and specific methods of analyses postulated in advance of undertaking this review [21]. All records were collected on RefWorks, a bibliographic tool used in research to develop personalised databases.

The review inclusion criteria were: (1) study population of women and mothers, self-identified as Black or having African or Caribbean origin; (2) providing intervention in the form of psychological therapy, educational programs or health visitor interventions; (3) studies which demonstrate maternal depression and well-being outcomes and/or child development/ behaviour/well-being outcomes; and (4) studies using either qualitative or quantitative methods with pre- and post-intervention measures or randomised controlled trials.

Papers were excluded if: (1) the study population had a majority of Caucasian participants or contained ethnic minority groups that were not inclusive of participants with Black, African or Caribbean origin; (2) studies where interventions did not focus on maternal well-being and depression; (3) studies that only measured the outcomes of child development/behaviour/well-being; or (4) studies with correlational design, with an emphasis on the prevalence of depression only without any intervention tested or provided.

The strategy adopted to formulate the review was the PICo model (Population [Black, African and Caribbean mother/child], the phenomenon of Interest [depression], Context [maternal mental health and child well-being]) [22]. The Boolean operators (AND/OR/NOT) were also used for strategic search [23]. The search terms include “postpartum depression”, “maternal depression”, “postnatal depression”, “intervention”, “treatment”, “program*”, “black Caribbean”, “black African”, “black”, “Afro-Caribbean”, “African” and “Caribbean” “mothers” and “women” to search five databases (Scopus, PsycINFO, ProQuest Central, Applied Social Science Index and Abstracts (ASSIA) and Web of Science) for published studies between 2000 and July 2020.

The search yielded a total of 553 records, of which duplicates removed were n = 259. Two reviewers screened the available titles and abstracts by applying the inclusion–exclusion criteria. After full texts were screened, 13 studies were selected for the systematic literature review. Any discrepancies between the two reviewers were resolved through discussion with a third reviewer.

The present review used the Standard Quality Assessment Criteria (QualSyst Tool) for the risk of bias evaluation [24] (see the Additional file 1 table showing the risk assessment of each study). All studies included in the review showed a low risk of bias. Majority of the studies used control groups (collaborative care or treatment as usual). Due to mothers’ vulnerable nature with maternal depression, none of the studies withheld treatment as usual from participants. Most studies included in the review used randomisation to allocate participants to the different arms of the trial. The majority of the studies provided limited information on the process of blinding or whether this occurred.

However, one study reported proximal outcomes and related this to the small sample size [25], and another study only supported the conclusions made due to lack of significance in the results [26]. The majority of research were of small sample size pilot studies testing the feasibility of maternal interventions. Attrition is a challenge in these types of studies where the sample populations are under-treated. For example, one study had only 3% attendance in the control group compared to 83% attendance in the intervention group [27]. The majority of studies recruited non-treatment seeking participants which may reflect the challenges with engagement and one internet-based intervention study examined mothers’ intention to seek treatment for maternal depression.

The findings from the various studies were thematically synthesised into a qualitative narrative addressing the research question. Therefore, the conceptual framework of synthesis was adopted [28]. Each source was critically analysed and evaluated by major themes in terms of strengths, weaknesses and critical gaps using the synthesis matrix. The synthesis was conducted in four stages: (1) implementation of a relevant literature search; (2) identified key ideas and elements; (3) organised the key ideas and elements; (4) synthesised data and built a case for new research intervention. An ‘index card’ like method was employed to identify and organise the key ideas and elements from stages 2 and 3. Each pile or category of organised ideas were re-arranged into (a) a logical flow of information; (b) compare and contrast; (c) critique; and the (d) postulation of an alternative model for future interventions research aimed at African and Caribbean mothers/women.

Findings from the reviewed studies showed that maternal depression and other mental health conditions could be best treated with interventions designed to enhance parenting confidence and self-care [25, 26, 31‐36]. This notion of the intervention was found to be effective across eight studies. For example, Crockett et al. [26] examined the feasibility, effectiveness and acceptability of an intervention program called Reach Out, Stand Strong: Essentials for New Moms (ROSE) for rural, low-income African American pregnant women who were at risk of postpartum depression. The program entailed interpersonal therapy that enhanced familial communication, social support, managing the transition, and focusing on factors associated with major depressive symptoms and perinatal depression. Women in the ROSE condition showed significant improvement in their depressive symptoms and parenting adjustment at 3 months postpartum than women in the treatment as usual condition.

In a randomised controlled trial (RCT) of skill-based intervention that empowers women at risk of postpartum mental health distress, Mendelson et al. [25] measured how Mother-Baby (MB) course could be adopted as a preventive measure for depression and the impact of perinatal depression on coping, social support and mood regulation among African American women. An MB course is a manualised skill-development training material blended with core elements of cognitive behavioural therapy to prevent the risk of perinatal depression among ethnically diverse and low-income mothers [37]. This was delivered by trained facilitators who taught mothers how to harness and utilise enjoyable and supportive contacts, modulate their negative automatic thoughts and increase pleasant activities. Findings showed 16% greater prevention of depression over time and more effective mood regulation from baseline to 6 months follow up in the intervention group compared to the control group. These findings have been supported in previous research on the effectiveness of the MB course and its ability to prevent depressive symptoms and other risks of mental health problems in mothers [37, 38]. Crockett et al. [26] and Mendelson et al. [25] illustrated maternal programmes that encapsulate skills-oriented intervention, enhancing parenting confidence and self-care in postpartum periods and beyond.

Mother–child relationship has significant implications for maternal mental health, and therefore, the current review showed recurrent emergence of interventions targeting effective mother–child relation [25, 31, 32]. For instance, Holditch-Davis et al. [32] examined the mother-infant relationship and its effects on maternal mental distress using a maternally administered intervention designed explicitly for preterm infants in the neonatal intensive care unit (NICU). Two interventions were compared; (1) Kangaroo Care and (2) Auditory, Tactile, Visual, and Vestibular (ATVV) interventions. The effectiveness of these two interventions on the mother-infant relationship and maternal mental distress were evaluated and compared with an attention control group. Findings showed a rapid decline in maternal depressive symptoms and lower parenting stress across all intervention groups with higher scores on the Home Observation for Measurement of the Environment (HOME) inventory. HOME is a scale that measures the quality and quantity of support available in the child's home environment [39]. ATVV is an intervention that involves a series of activities with an infant such as moderate stroking, talking, rocking and keeping eye-contact [40]. Whereas, Kangaroo intervention entailed a process of holding an infant to ensure skin-to-skin contact around the mother’s breasts [32].

Mendelson et al. [25] also focused on the mother–child interpersonal relationship in their study, in which the Mother and baby (MB) Course was adopted as the basis for preventing maternal depression. In this intervention, "the mother-baby relationship is emphasised throughout the course, including ways to enhance parenting practices using course skills" [25  p. 212]. The MB course involves core elements of cognitive behavioural intervention that translates into quality mother–child relationship and has consistently shown to reduce maternal depression, especially among low-income women [37, 38].

Previous studies have shown low engagement and high dropout rates of psychological treatment among ethnic minority populations in HICs compared to the native population which suggests that current treatment options lack cultural relevance or appropriateness [14, 16]. Therefore, cultural beliefs, values and traditions are essential in intervention planning and execution for addressing maternal depression [25, 26, 33, 35, 41‐44]. In the Grote et al.’s [41] study using an RCT design, culturally relevant and enhanced brief Interpersonal Psychotherapy (IPT-B) had a greater effect on treating depression in low-income pregnant women compared to those receiving enhanced usual care. Enhanced IPT-B consists of engagement sessions and later followed by eight IPT-B sessions covering up to 6 months postpartum period.

Most importantly, the culturally relevant materials were embedded to augment the intervention. For example, the engagement session was based on ethnographic and motivational interviewing designed to build rapport and trust to address the practical aspects of lived experiences of cultural and psychological barriers to mother and childcare among low-income depressed women. Therefore, the practitioners approached the mothers with the core elements of culturally sensitive principles of ethnographic interviewing in which the interviewer employs the role of a learner to understand the women from their cultural perspectives without any bias, focusing on; unconditional appreciation of participants’ cultural values, beliefs and traditions in terms of their views about depression, health beliefs, coping practices or health-seeking behaviours. Women were asked what they would like to achieve during the intervention phase. Issues such as the importance of race, ethnicity and spirituality were put forward in the engagement sessions [41].

Considering the low-income and economically disadvantaged predicaments of the depressed women in this minority group, a pragmatic approach was integrated with the Enhanced IPT-B intervention to cater for basic needs (e.g. access to food and free bus passes and bay supplies) and to facilitate access to sustainable social services (e.g. effective childcare, job training and access to housing) which were included as core aspects of the intervention delivery. Findings from the Enhanced IPT-B intervention highlights its effectiveness in reducing antenatal depression, and further improving social functioning and preventing relapse in the postpartum period of up to 6 months.

Furthermore, two studies implemented reinforcement strategies, weekly from their home visitor [25], and the other weekly calls and/or texts about the upcoming session [27]. A combination approach is essential to be considered given the numerous obstacles that may hinder African and Caribbean women from engaging in treatment, especially when they are from low-income backgrounds. Translated materials and topics such as spiritual-related resources were added to the culturally adapted intervention [43].

Modern technological innovation now plays a significant role in the mental health healthcare provision. One of the overarching themes in the present review is internet-mediated care for maternal depression [27, 31]. For example, Logsdon et al. [31] examined the effectiveness of the internet-based approach to treating depression among African American adolescent mothers. The intervention included vignettes, resources, questions and answers developed based on the theory of planned behaviours (TPB). Immediately after the intervention and 2 weeks post-intervention, participants (n = 151) responded to questions on depressive symptoms based on TPB variables. They completed the intervention from any computer device of their choice. This was compared with the control group (n = 138) which entailed visits to homes and community organisations.

The intervention led to significant improvements in perceived control on maternally related self-care, positive attitude change, increased intention to seek help, and active behaviours for seeking actual treatments. Other advantages accrued to the internet-based intervention were about the cost of delivering it, which was least expensive with reduced costs of logistics and time, which may be especially beneficial for mothers with limited resources. A similar study, Boyd et al. [27] conducted a pilot RCT using social media for postpartum intervention for depressed low-income mothers. Groups were compared using social media (Facebook) and in-person interventions for African American mothers who screened positive with postpartum depression from paediatric clinics. Both interventions employed the parents interacting with infants (PIWI) which involved eight weekly sessions including behaviour activation and depression psychoeducation, parent–child interaction such as play, laughter and safety. Findings showed that women in the social media intervention group had lower depressive symptoms than women in the in-person.

Both Logsdon et al. [31] and Boyd et al. [27] studies showed the potential effectiveness of using online facilities for reducing maternal depression. Their findings have been supported in the previous literature. For example, online intervention to improve positive parenting behaviours and ameliorate child-related behavioural problems [45], treatment of depression, improve attitudes about treatment, relapse prevention among young mothers [46, 47], and improved parenting skills and infant feeding [48].

This study was conducted shortly before the Covid-19 pandemic era; however, the unequal distribution of infection and death rates of the Covid-19 pandemic among African and Caribbean populations in HICs has exposed longstanding systemic issues of racial inequalities [19]. For example, in the UK, those who experienced higher societal disadvantages faced more significant health risks [49]. Reports have shown that there have been between a 10% and 50% higher risk of death among African and Caribbean populations compared to the White British population [49]. Considering the current pandemic, which may further alienate already marginalised people, there is an urgency to understand the effectiveness and engagement of culturally adapted online help or ‘tele-intervention’ delivered remotely for these ethnic groups.