A brief report informing the adaptation of a behavioural activation intervention for delivery by non-mental health specialists for the treatment of perinatal depression

Abstract

Background:

Behavioural activation (BA) is recommended by the National Institute for Health and Care Excellence guidelines for the treatment of perinatal depression; however, there is limited evidence about whether it is effective when delivered by non-mental health specialists (NMHS) in a perinatal setting in the UK.

Aims:

This study aimed to adapt a BA intervention manual and guided self-help booklet intended for delivery by NMHSs for the treatment of perinatal depression.

Method:

Interviews were conducted with 15 women and 19 healthcare professionals (HCP) within the first study element. Four experience-based co-design (EBCD) workshops were held, with the involvement of 14 women and three HCPs, to modify the BA documents for the specific needs of perinatal women. Thematic analysis was used to analyse the data.

Findings:

The findings from the study elements were presented with themes. The co-designers (women and HCPs) pointed out that having sleeping problems, changes in appetite, feeling exhausted and feeling emotional, may be experienced by non-depressed mothers as well during pregnancy or in the postpartum period, especially around the fourth day after giving birth. Therefore, it was important to differentiate these feelings from depression. The women also wanted to see an example for each activity before being asked to do it. Having examples would help them to see the possibilities before creating their own diary sheets or tables of activities.

Conclusions:

Aside of the tool adaptation, the findings of this study provide a foundation to assess the effectiveness of the adapted intervention in a subsequent feasibility trial.

Background

Perinatal depression is a substantial public health concern because of its high prevalence (9–19%), impact on women themselves, their children and partners, and its costs to the family and the wider society (Bauer et al., 2014). Improving the provision of perinatal depression treatments is on the Government agenda and this is one of the action plans of the National Maternity Review (National Health Service England, 2016). In the perinatal period, the mental health needs of women should be met by most effective and low-cost treatments. Behavioural activation (BA) is a structured, brief and effective psychotherapeutic approach that relies on changing behaviour through adaptive activities to treat depression (Uphoff et al., 2020). It also offers the possibility of being more cost-effective because its simplicity means it can be delivered by non-mental health specialists (NMHSs). BA is, therefore, a good candidate for the treatment of perinatal depression that can be delivered by trained NMHSs.

Maternity services, community services and health visiting services consist of a diverse workforce in England; however, there are challenges to capacity in terms of allocating the role of delivering a brief psychological intervention to support perinatal women. The role of maternity support workers (MSWs) includes supporting women and providing complex care. MSWs are an essential part of the maternity team and they can contribute to fulfilling the goals of the National Maternity Review. They can lessen role demands on midwives by working with them. MSWs could also decrease the demand for specialist mental health workers by delivering a brief low-intensity psychological intervention to women with mild to moderate depression during pregnancy or afterwards. A study showed that women prefer traditional individual talking therapies as the first choice of treatment for depression over online psychological therapies (Hantsoo et al., 2017). A previous meta-analysis found beneficial effects of psychological interventions delivered by NMHSs for the treatment of perinatal common mental health disorders (Clarke et al., 2013).

A randomised controlled trial was found which trained a mixed group of specialists and non-specialists to deliver a 10-session BA therapy during pregnancy in the USA (Dimidjian et al., 2017). The study concluded that BA is an effective and feasible intervention for the treatment of depression during pregnancy and 3 months postpartum. There is at present an evidence gap for the effectiveness of low-intensity BA delivered by NMHSs for women experiencing depression in the perinatal period in a UK setting. The prior study (Dimidjian et al., 2017) did not provide information on how the BA treatment was adapted for perinatal women. This study aimed to adapt a BA intervention manual and guided self-help booklet suitable for perinatal women by NMHSs, for example, MSWs. The BA booklet and manual were adapted from documents developed for the CASPER and CHEMIST studies (Gilbody et al., 2017; Littlewood et al., 2019).

Method

Studies 1 and 2 – design and procedure

The Medical Research Council recommends researchers to acknowledge and use appropriate methods in the development of interventions. The experience-based co-design (EBCD) was used in the adaptation of the intervention (Bate and Robert, 2007). It involved a ‘discovery’ phase within the first study element when the women’s and healthcare professionals’ (HCP) experiences were explored, and the ‘co-design’ phase within the second study element, when the BA therapy manual and booklet were adapted.

Women and HCPs were recruited from three areas within Yorkshire and the Humber in 2019. Women were recruited through advertisement posters placed at three NHS Trusts and online advertisements posted on a number of websites and Facebook accounts related to mothers. HCPs were informed about the study by email sent by the gatekeepers in each NHS Trust. A purposive sampling method was used to select women who have experience of perinatal depression, and HCPs who have experience of providing support and care for women experiencing perinatal depression. Women who have experienced perinatal depression are a potentially vulnerable group; therefore, an ‘assessment and management of risk protocol’ was prepared and women who had clinical depression or were in the perinatal period were excluded from the studies. In recognition of the women’s effort in the studies, they were given a One4all gift card worth £20. Travel expenses of women and HCPs were also reimbursed.

Semi-structured individual face-to-face interviews and focus group interviews were used to collect data from women and HCPs within the first study element. Sampling, transcribing and analysing were carried out iteratively until no or few new codes and themes were identified from the data.

The co-design workshops were used to meet the study objectives of the second study element (see extended report in Supplementary material). The typical content of a co-design workshop included: (1) giving information about the study; (2) going through the BA booklet and then the proposed BA manual, page by page while explaining the findings from the first phase and asking probing questions to identify, agree and define key points to change; and (3) setting priorities for the adaptation of the BA booklet and manual. The decision on whether to amend the documents or not was a consensus decision-making process between women, HCPs and the researcher in the workshops by asking co-designers whether everyone is agreeing on the amendment or not. The adaptation was made according to the majority’s opinion.

The interviews and workshops were conducted and audio-recorded by the first author using a password-protected audio recorder. The recordings were transcribed by the same author or a professional transcriber. Thematic analysis was used to analyse the data.

Results

Study 1 – findings

The duration of individual interviews with 15 women ranged between 25 and 65 minutes, with an average of 40 minutes and the individual interviews with 13 HCPs ranged between 15 and 60 minutes, with an average of 30 minutes. One focus group interview with two HCPs lasted 40 minutes and another focus group with four HCPs lasted 52 minutes. Eleven of 19 HCPs were midwives; three were health visitors; two were obstetricians; two were maternity support workers; and one was a GP. All the women were white British, and aged 28–41 years. Over two-thirds of them were married or living with a partner and under one-third were single. Over two-thirds of them had completed graduate study and nearly all of them were working in a part-time job. Eight women reported experiencing perinatal depression and seven women reported experiencing only postpartum depression.

The themes are summarised in Table 1 identified from the data and were shared with co-designers in the second study and used in the adaptation process of the BA manual and booklet.

Table 1. Summary of qualitative interviews findings that were used in the adaptation process of the BA manual and booklet

Study 2 – findings

In total, 14 women and three HCPs took part in four co-design workshops across the three research sites (see extended report in Supplementary material). The duration of workshops ranged between 110 and 145 minutes, with an average of 123 minutes. Eleven women were white British, three were other Asian backgrounds, and aged 31–45 years. All the women were married or living with a partner. Eight had completed graduate study and seven were working in a part-time job. Eight women reported experiencing perinatal depression. Five women reported experiencing only postpartum depression.

The following five themes were identified from the data: (1) to differentiate what is common and not common to feel in the perinatal period (risk factors and signs of perinatal low mood and depression) and other adaptations to the content; (2) illustration of mood cycles and suggested activities with examples (breaking the low mood cycle, activities for women, planning manageable activities, example action plan, example diary); (3) using narrative quotations from the first phase; (4) highlighting that ‘BA may not work for you’ (other help sources are available); and (5) minor modifications on format or text (using bullet points, highlighting important texts, choosing brighter and natural pictures, reordering some sections, removing repetitions and unnecessary information, improving the clarity of the texts and scale to use for mood, and deciding the size of the manual and booklet).

Although BA aims to increase women’s engagement with activities, the co-designers also pointed out that a considerable number of women continue their daily routines and attend activities, including mother and baby groups, but do not get enjoyment from them and feel depressed. The consensus among the co-designers was that a statement should be added for these women stating that BA is just one of the options that can be tried, and that other help is available if this booklet does not feel right for them. This statement is critically important as it encourages women not to give up and that there are always other treatment options if one does not work.

Discussion

The aim of this study was to inform the adaptation of the BA manual and booklet, intended for delivery by MSWs for the treatment of perinatal depression. The documents were modified through co-design workshops with the involvement of women who have experience of perinatal depression and HCPs who have experience of providing support and care for those women. This study comprised the ‘co-design phase’ of the EBCD approach. Previous study ‘discovery phase’ included exploring women and HCPs’ experiences of perinatal depression which informed the co-design phase of the EBCD study.

A previous qualitative study conducted in the USA aimed at modifying cognitive behavioural therapy (CBT) for perinatal depression (O’Mahen et al., 2012). The authors conducted qualitative interviews with 23 perinatal women, and they modified the CBT manual according to the themes generated from the interviews. The decrease in women’s activities, their struggles in managing their daily routines and isolating themselves from social activities were found related to the behavioural component of the CBT (O’Mahen et al., 2012) and these findings were consistent with the first element of the study that published elsewhere. This study differs from the study of O’Mahen et al. methodologically. Uniquely, in this study co-design workshops were conducted with the involvement of women and HCPs who shared equal power with the researcher in the adaptation process for the BA manual and booklet instead of CBT as in the O’Mahen et al. (2012) study.

All the women participating in the first study were older than 28 years, and all were white British, thereby limiting the transferability of the findings to younger women or ethnic minorities. Women and HCPs who were involved in the studies responded to advertisements placed by the researcher. Their experiences, therefore, cannot be assumed to be identical to those women who were not involved in the studies.

Although co-design has strengths in the modification of the BA manual and booklet, it has constraints of being tokenistic by researchers who continue holding power and not valuing co-designers’ experiences in the development process (Boylan et al., 2019). During the workshops, the co-designers were encouraged to share their ideas about how to improve the BA manual and booklet, in order to develop them and make them useful for other women who may share similar experiences to theirs. The researcher conveyed that this was an opportunity for them to help other women by using their experiences, allowing everyone to share their thoughts, and was mindful not to take over the conversation by talking more than others, thereby balancing the researcher’s power in the process.

The fundamental contribution of this research is firstly that it is a rigorous adaptation of an evidence-based psychological intervention, behavioural activation, recommended by national, evidence-based guidelines for the treatment of perinatal depression. Although the effectiveness of the adapted BA manual and booklet was not evaluated in this thesis, these documents could be used in a future feasibility study and a related randomised controlled trial.