Community Singing Interventions for Postnatal Depression: a Hybrid Type II Effectiveness-implementation Trial (SHAPER-PND)

Community Singing Interventions for Postnatal Depression: a Hybrid Type II Effectiveness-implementation Trial

Community Singing Interventions for Postnatal Depression: a Hybrid Type II Effectiveness-implementation Trial

Postnatal depression (PND) affects over 13% of new mothers but there is still not an ideal treatment for all cases. Pharmaceutical and psychotherapy have offered solutions but there are challenges in treatment uptake and adherence and long waiting-lists for psychotherapy. Many mothers attend group activities with their babies, some including music and singing. Community group singing has shown improvement in mental health and singing to babies has shown improvement in mother-infant interaction and reduced infant distress. In this realm, Melodies for Mums (M4M) is a programme based in Lambeth and Southwark providing 10-week singing and music sessions for mothers with postnatal depression (PND) and their babies in community Children's Centres or online, according to government social distancing guidelines. Studies have demonstrated its effectiveness in reducing symptoms of PND faster than usual care or social groups, and preliminary process evaluations have suggested its suitability. It has also been identified as a strong way of engaging mothers from minority backgrounds who are less likely to seek professional support for their mental health. However, the programme is reliant on short-term grants and has not been implemented in clinical care. Therefore, there is a clear need to invest more research into this programme to help it achieve its potential. The investigators aim to conduct M4M in a clinical trial aimed at women experiencing symptoms of postnatal depression in the boroughs of Southwark, Lambeth and Lewisham. The investigators will collect data on the wellbeing of the women through a series of interviews and questionnaires and the investigators will also collect biological samples for stress and immunity markers from mothers and babies. In the long term the investigators intend to establish defined clinical referral pathways for patients from primary (GPs, community, among others) and secondary care (specialist doctor, hospital clinic) settings. In addition, the investigators will collect further evidence of the clinical, implementation and economic effectiveness of the intervention.

Postnatal depression (PND) affects at least 12-15% of new mothers, with symptoms including fatigue, anhedonia, insomnia and irritability. However, challenges surround the fact that there is still no complete treatment solution. Although pharmacological treatment has had positive results, these are hampered by low uptake and adherence amongst mothers, while psychotherapy has produced mixed results and has similar challenges around low uptake or delayed treatment. However, many mothers engage in community group activities with their babies, such as attending mother-infant play groups. Such activities have been identified as ways of relaxing mothers, providing good sources of social interaction, decreasing the monotony of each day and also providing a sense of personal fulfillment for mothers. There is also a growing body of evidence demonstrating the effects of community group singing on mental health. Singing to new babies is practised in cultures around the world, and research has demonstrated valuable benefits such as improving mother-infant interaction and reducing distress in babies. Listening to music during pregnancy is also associated with higher levels of wellbeing and reduced symptoms of PND in the first 3 months post-birth, while daily singing to babies is associated with fewer symptoms of PND and higher levels of wellbeing, self-esteem and perceived mother-infant bond. Consequently, there is a strong theoretical background to why singing could support mothers with PND. While small studies have supported the efficacy of this intervention, the present large scale study is needed to measure effectiveness and promote the uptake within the NHS, as well as in order to fully understand the mechanisms underpinning its efficacy. Trial design and aims: These sessions/activities will be delivered for 10-weeks and the women will be followed up at 20- and 36-weeks post randomisation. On week 0 (before the start of the study), 6, 10, 20 and 36 women will either be visited by members of the KCL research team (to be assessed given questionnaires and provide samples) or will complete online questionnaires (see table of measures). The end of the trial for a participant will be after the 36-week follow up, or withdrawal. This is a multi-centre trial that will be run in locations across London (primarily in the boroughs of Lewisham, Lambeth and Southwark), specifically in children's or community venues. These venues will be thoroughly safety- and risk-assessed in light of COVID-19 and adherence to any COVID-secure measures will be closely monitored by Breathe. In the event that one or several women need to self-isolate due to confirmed or suspected exposure to COVID-19, the sessions will be delivered to her/them online via Zoom. This will allow for the sessions to still be conducted in person with the remaining women. In order to enrol a sufficiently large sample of women, there will be 12 blocks of the 10-week singing programme in total: 10 intervention blocks will be for the experimental group, and 2 blocks will be for women in the wait-list control group that wish to take part in the singing sessions. The 2 wait-list blocks will be offered singing sessions after a woman has completed the first 10 weeks of the study and no data will be collected regarding these singing sessions, apart from the control follow-up data that this group is expected to provide. In the event that one or several women need to self-isolate due to confirmed or suspected exposure to COVID-19, the sessions will be delivered to her/them online via Zoom. This will allow for the sessions to still be conducted in person with the remaining women. In the event of the artist or the Breathe officer having to self-isolate, a replacement Breathe-trained artist will be ready to step-in and deliver the sessions. The same will apply to Breathe staff; there will be another staff member available to be present at the sessions. The aims are to explore the clinical effectiveness of the intervention in a larger sample size than previous studies for scalability (to ascertain whether the initial findings can be replicated at a larger scale) and to explore the implementation effectiveness of the intervention, including its uptake, suitability, acceptability, appropriateness and feasibility (to identify not just 'if' but also 'why' the intervention works and support our understanding). It will also explore the cost effectiveness of the intervention, including the cost of delivering the interventions and the balance of benefit for the health sector, in order to be able to develop strong business plan for the intervention. Participants and Recruitment process: This study aims to recruit 400 women (and their babies) that meet criteria on the Structured Clinical Interview for DSM-IV Disorders (SCID) for major depressive disorder and score equal to or above 10 on the Edinburgh Postnatal Depression Scale (EPDS). Participants will be recruited through clinic and the community (including study advertisement on social media, patient groups and self-referral). In order to recruit a sufficient number of women for this study in the current pandemic, most of the signposting will be done through GPs, midwives, health visitors and community mental health teams that remain engaged with the population despite the pandemic. In addition, signposting via social media, especially targeted advertising in local women's groups will be another valuable route for recruitment. The women will be explained the trial, given the PIS and ICF and if consented, will be randomised (2:1) to an intervention group (M4M singing sessions) or a control group (other community-based mother and baby activities). This group will be involved in the clinical and the implementation effectiveness of the study. In addition, circa 30-50 wider stakeholders involved in the intervention will be recruited by research staff. This group includes artists, psychiatrists, GPs, health visitors, commissioners and others involved in the delivery of the intervention. The wider stakeholders and mothers involved in implementation science research will be provided with a specific ICF and a PIS. Measures collected: Measures will be collected from both the intervention and control groups. Participants will be visited at their homes (or online via Zoom if necessary) for baseline/week 0 and week 10 assessments. Week 6, 20 and 36 visits will be completed online due to the self-reporting nature of the measures to be captured. If a mother cannot access a computer/laptop to complete the online questionnaires, these will be printed and posted to them. In order to allow for flexibility in the schedule, a +/-1 week variation in the date of collection of the measures below (apart from week 10, when the window will be weeks 10-12) will be accepted. These measures will be collected from the control and the intervention groups simultaneously. The following measures will be collected: Week 0 (screening/baseline): baseline demographics, repeated demographics, Brief Life Events Scale, Child Experience of Care and Abuse (CECA-Q), Composite Abuse Scale (CAS) - Pregnancy Version, Intrusive Life Events Scale; Structured Clinical Interview for DSM-IV (SCID), Edinburgh Postnatal Depression Scale (EPDS), Hamilton Depression Rating Scale (HDRS), Beck Depression Inventory (BDI), Office for National Statistics Wellbeing Scale (ONS), State Trait Anxiety Inventory (STAI), Perceived Stress Scale (PSS); CARE-index, Maternal Postpartum Attachment Scale (MPAS), Parent Reflective Functioning Questionnaire (PRFQ), UCLA Loneliness Scale, Short General Self-Efficacy Scale (G, SE-6), Multidimensional Scale of Perceived Social Support (MSPSS); Diurnal saliva samples in mothers and babies. Week 1: Pre-post session saliva samples in mothers and babies. Week 6: Repeated demographics, EPDS, BDI, ONS, STAI, PSS, MPAS, UCLA Loneliness Scale, GSE-6, MSPSS, Pre-post session saliva samples in mothers and babies. Week 10: Repeated demographics, EPDS, HRDS, BDI, ONS, STAI, PSS, CARE-Index, MPAS, PRFQ, UCLA Loneliness Scale, GSE-6, MSPSS, Pre-post session saliva samples in mothers and babies, Diurnal saliva samples in mothers and babies, Hair cortisol sample, focus groups (qualitative research). Week 20: Repeated demographics, EPDS, BDI, ONS, STAI, PSS, MPAS, UCLA Loneliness Scale, GSE-6, MSPSS. Week 36: Repeated demographics, EPDS, BDI, ONS, STAI, PSS, CARE-Index, MPAS, PRFQ, UCLA Loneliness Scale, GSE-6, MSPSS. Collection, transportation and storage of samples: Pre-post session saliva samples collected at sessions from mothers and babies will be retrieved using passive drooling (mothers) or salivettes (babies) and transported by one of the researchers in the team and used for cortisol, oxytocin, cytokines and other stress hormones analysis. The samples will be kept on ice and brought to the Maurice Wohl Clinical Neuroscience Institute (King's College London, Denmark Hill) to be stored at -80 degree C until analysis. Six saliva samples from mothers (and two samples from their baby, providing consent) will be collected by mothers at home at awakening, +15, +30+ and +60 min after awakening, at 12 noon and 8 pm, and used for diurnal cortisol and other stress hormones. They will be kept in the fridge at home, and then posted or transported to the Maurice Wohl Clinical Neuroscience Institute (King's College London, Denmark Hill), to be stored at -20⁰C until analysis. Hair samples will be kept at room temperature for short-term storage but will be transferred to -20⁰C for longer-term storage, until analysis (cortisol and other stress hormones) is conducted. Samples will be processed, analysed and remaining material will be kept for 10 years after completion of the study. Results will be identified according to anonymised codes and stored securely. Follow-up: Follow-up will be done at weeks 20 and 36, using the questionnaires mentioned above. Participants' wellbeing will be monitored through the questionnaires and any issues will be flagged by the KCL or UCL team or by Breathe that will do onward reporting if needed. No samples will be collected at follow-up. Wider stakeholders: a series of questionnaires/interviews will be performed to collect implementation science data only.

Mothers will start a block of 10-week classes and continue with their group for the duration of the course. Classes will take place in Children's Centres (or online). Mothers will attend with their babies and will sit in a socially-distanced circle on the floor surrounded by soft play cushions and mats. Classes will start with welcome songs, introducing everyone to one another, and involve a range of singing and music activities. Mothers will be required to respect social distancing guidelines. Music activities will include learning songs from around the world and will be accompanied by instruments that the mothers and babies can play together. Instruments will not be shared and will be disinfected before and after the singing sessions. Mothers will also work to write some of their own songs over the weeks. Recordings of the group singing the songs together will be made for the mothers to listen to at home. Classes will be led by professional workshop leaders trained by Breathe.

Our control group will be a 'active' control. During the first 10 weeks (during the study period), mothers in the control group will receive details of other non-music classes available to them in the community (or online if necessary, depending on the programs available at the time and government guidelines) and will receive the same schedule of texts and phone calls to encourage them to join these activities. They will still be seen by the researchers to collect clinical measures and biological samples (including the pre-post saliva samples) and to monitor engagement in other activities. Following the first 10 weeks, the mothers in the control group will be offered a place on the singing programme, but these data will not be part of the study, and they will not join groups with women who are in the study

Melodies for Mums (M4M) is an intervention that was developed and tested as part of a collaboration between the Royal College of Music, Imperial College London and University College London from 2015-2017. The programme involved weekly singing classes for mothers and their babies delivered in groups of 8-12 in Children's Centres for 10 weeks. A trained artist delivers 1-hour long singing sessions to mothers with postnatal depression (and their babies) for a period of 10-weeks. The sessions are aimed at the mothers, with songs ranging from sound baths to folk songs and songs created by the mothers themselves. Mothers randomised to the intervention will be prompted via text message, each week before the session, to attend these singing sessions.

Group mother-baby interventions in the community (i.e. baby play, baby reading sessions, baby massage, etc). A list of these activities will be collected from the activities on offer in the community. All the mothers randomised to the control group will be pointed towards these activities, prompted by text messages in line with the prompts received by the mothers in the intervention group. Mothers will be encouraged to attend these non-music based community activities once a week with their babies.

To assess the effectiveness of group singing interventions on symptoms of postnatal depression using the Edinburgh Postnatal Depression Scale (EPDS)

To assess the effectiveness of group singing interventions on symptoms of postnatal depression using the Edinburgh Postnatal Depression Scale (EPDS) The EPDS was developed to assist health professionals in detecting mothers suffering from Postnatal depression. The scale consists of 10 short statements. A mother checks off one of four possible answers that is closest to how she has felt during the past week. The EPDS is measured on a scale of 0-30, where a higher score indicates more severe depression. Mothers scoring above 12 or 13 are likely to be suffering from depression.

The primary outcome measure is change in EPDS total score between baseline and Week 10 (end of treatment).

To assess the acceptability of the intervention using the Acceptability of Intervention Measure (AIM)

To assess the acceptability of the singing intervention, using the Acceptability of Intervention Measure (AIM). A 4-item measure of perceived intervention acceptability. Items are measured on a 5-point Likert scale (Completely Disagree-Completely Agree). Score is calculated mean.

To assess whether singing improves (changes) further aspects of mental health, including depression using the Hamilton Depression Rating Scale (HDRS).

To assess whether singing improves (changes) further aspects of mental health, including depression, using: Hamilton Depression Rating Scale (HDRS): is a semi-structured clinician-administered depression assessment scale. Scoring is based on the 17-item scale and scores of 0-7 are considered as being normal, 8-16 suggest mild depression, 17-23 moderate depression and scores over 24 are indicative of severe depression; the maximum score being 52 on the 17-point scale.

To assess whether singing improves (changes) further aspects of mental health, including depression using the Beck Depression Inventory (BDI).

To assess whether singing improves (changes) further aspects of mental health, including depression, using: Beck Depression Inventory (BDI): is a 21-item, self-rated scale that evaluates key symptoms of depression. Each item is rated 0-3, whereby a higher score indicates more severe depression. 1-10 is considered normal; 11-16 is mild mood disturbance; 17-20 is borderline clinical depression; 21-30 is moderate depression; 31-40 is severe depression; 40 and above is extreme depression.

To assess whether singing improves (changes) further aspects of mental health, including stress using the Perceived Stress Scale (PSS)

To assess whether singing improves (changes) further aspects of mental health, including stress. Stress will be evaluated using the following scale: Perceived Stress Scale (PSS): is an 11-item questionnaire assessing one's perceived stress levels. Each question must be answered on a likert scale, ranging from 'never' (0) to 'very often', (5), whereby a higher score indicates greater stress.

To assess whether singing improves (changes) further aspects of mental health, including wellbeing using the Office for National Statistics Wellbeing Scale (ONS):

Wellbeing will be evaluated using the following scale: Office for National Statistics Wellbeing Scale (ONS): is a four-item questionnaire assessing life satisfaction, worthwhile, happiness, and anxiety. People are asked to respond to the questions on a scale from 0 to 10 where for life satisfaction/worthwhile/happiness a higher score indicates greater wellbeing, and for anxiety a greater score indicates greater anxiety.

To assess whether singing improves (changes) further aspects of mental health, including anxiety, using the State-Trait Anxiety Scale (STAI)

Anxiety will be evaluated using the following scale: State-Trait Anxiety Scale (STAI): is a self-rated questionnaire which assesses intensity or frequency of anxiety. It is divided into two sections, one measuring 'state' and the other 'trait' of anxiety. The range of possible scores for each section is 20 to 80, whereby a higher score indicates greater anxiety. A cut-off score of at least 40 has been considered to be clinically meaningful.

To ascertain whether singing improves the observed mother-infant interaction using the Crittenden CARE-Index (CCI)

To assess whether the singing intervention improves upon aspects of the mother-infant relationship. The observed mother-infant interaction will be assessed using: Crittenden CARE-Index (CCI): The CCI is a validated tool to clinically assess the mother-infant interaction. Videos are filmed of a mother and her baby playing, and are rated from 0-14 for aspects of maternal behaviour, aspects of infant behaviour, and dyadic synchrony, whereby a higher score indicates greater amounts of that behaviour present.

To ascertain whether singing improves the perceived mother-infant relationship using the Maternal Postpartum Attachment Scale (MPAS)

To assess whether the singing intervention improves upon aspects of the mother-infant relationship. The perceived mother-infant relationship will be assessed using: Maternal Postpartum Attachment Scale (MPAS): The MPAS is a 19-item questionnaire assessing maternal feelings of attachment and bonding towards her baby. Items are rated from 1 to 5, whereby a lower score indicates less bonding/attachment with her baby.

To ascertain whether singing improves the perceived mother-infant relationship using the Parent Reflective Functioning Questionnaire (PRFQ)

To assess whether the singing intervention improves upon aspects of the mother-infant relationship. The perceived mother-infant relationship will be assessed using: Parent Reflective Functioning Questionnaire (PRFQ): The PRFQ is an 18-item questionnaire that asks mothers to reflect on their relationship with their infant and how attuned they perceive themselves to be. It assesses a caregiver's capability to reflect upon her own internal mental experiences as well as those of the baby. Each item is rated on a likert scale from 1 (strongly disagree) to 7 (strongly agree) with a total possible score ranging from 18-126.

To ascertain whether singing improves social support and reduces loneliness using the UCLA Loneliness Scale

To assess whether the singing intervention improves aspects of loneliness and perceived support. UCLA Loneliness Scale: is a 20-item questionnaire that assesses one's perceived sense of loneliness. Questions address relationships and loneliness. Items are rated as 'often,' 'sometimes,' 'rarely,' or 'never.' Scores can range from 20-80, whereby a higher score indicates a greater sense of loneliness.

To ascertain whether singing improves social support and reduces loneliness using the Multidimensional Scale of Perceived Social Support (MSPSS)

To assess whether the singing intervention improves aspects of loneliness and perceived support. Multidimensional Scale of Perceived Social Support (MSPSS): is a 12-item questionnaire that assesses perceived support (practical and emotional) from peers, family, and friends. Each item is rated from 1 (very strongly disagree) to 7 (very strongly agree) and is scored from 12-84, where a higher score indicates a greater sense of perceived social support.

To identify whether there are biological mechanisms underpinning the psychological outcomes assessed using changes in measurements in stress hormones, including hair cortisol, diurnal cortisol and salivary cytokines

Stress hormones, including hair cortisol, diurnal cortisol and salivary cytokines. These measures will be analysed using an array of techniques including enzyme-linked immunosorbent assay (ELISA). Higher levels of cortisol and cytokines may indicate higher levels of stress.

To identify whether there are biological mechanisms underpinning the psychological outcomes assessed using changes in measurements in salivary oxytocin

Levels of salivary oxytocin measured. These measures will be analysed using an array of techniques including enzyme-linked immunosorbent assay (ELISA). Higher levels of oxytocin may indicate positive interactions between mothers and babies.

To identify how the singing sessions affect the lived experience of mothers with PND using focus groups

Focus groups: focus groups will take place immediately following session 10 (if logistically possible) for all mothers focusing on their lived experience of the intervention and their reported mechanisms of effect

To explore the phenomenology of PND and how singing intersects with PND among women with particular risk factors for PND (traumatic birth, adverse childhood experiences, and social isolation/loneliness) using semi-structured interviews

Semi structured interviews: individual or small-group interviews with three sub-groups of women self-reporting particular risk factors for PND: traumatic birth, adverse childhood experiences, and social isolation/loneliness. These interviews will focus in-depth on the phenomenology of PND and how singing intersects with the specific context of PND among the sub-groups.

Total score of the acceptability of Intervention Measure (AIM). A 4-item measure of perceived intervention acceptability. Items are measured on a 5-point Likert scale (Completely Disagree-Completely Agree).

Total score of the Intervention Appropriateness Measure (IAM): A 4-item measure of perceived intervention appropriateness. Items are measured on a 5-point Likert scale (Completely Disagree-Completely Agree). Score is calculated mean.

Total score of the Feasibility of Intervention Measure (FIM): A 4-item instrument to assess perceived intervention feasibility. Items are measured on a 5-point Likert scale (Completely Disagree-Completely Agree). Score is calculated mean.

To assess intervention adherence and attrition rates using attendance data (rates and drop out reasons)

Data on the overall adherence to the intervention, number of drops-outs each week and reasons why. This data will be compiled to assess the percentage of attendance and collect the reasons why mothers did not attend sessions.

The number of individuals signing mothers up to the intervention, the number of individuals delivering the intervention and the number of individuals supporting the intervention (and continuing to do so)

EQ5D-5L (quality of life measure): The EQ-5D-5L essentially consists of 2 pages: the EQ-5D descriptive system and the EQ visual analogue scale (EQ VAS). The descriptive system comprises five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has 5 levels: no problems, slight problems, moderate problems, severe problems and extreme problems. The patient is asked to indicate his/her health state by ticking the box next to the most appropriate statement in each of the five dimensions. This decision results in a 1-digit number that expresses the level selected for that dimension. The digits for the five dimensions can be combined into a 5-digit number that describes the patient's health state.

To assess the cost effectiveness of the intervention using the Adult Service Use Schedule (AD-SUS): a questionnaire that evaluates healthcare service use by adults.

To assess the cost effectiveness of the intervention using implementation activity logs (to estimate implementation costs).

To assess factors affecting the sustainability and scalability of the intervention using the NOMAD Scale:

To assess factors affecting the sustainability and scalability of the intervention using the NOMAD Scale.

To assess factors affecting the sustainability and scalability of the intervention using semi-structured interviews

To assess factors affecting the sustainability and scalability of the intervention using semi-structured interviews

Inclusion Criteria: Women aged 18 or older Satisfactory understanding of English Women who have a child between 0 and up to 9 months old Women with postnatal depression diagnosed using symptoms of PND at a minimum score of 10 on the Edinburgh Postnatal Depression Scale (EPDS) AND meeting diagnostic criteria for major depressive episode on the Structured Clinical Interview for DSM-IV Disorders (SCID) Access to an internet-connected device (mobile phone, tablet, computer or laptop) to allow completion of assessments and participation in the singing sessions. Exclusion Criteria: Child outside of the age-range specified Unable to give informed consent

Anonymised data sets will be made available on the KCL Research Data Management System. Protocol will be published on a peer-reveiwed journal.

Soukup T, Davis RE, Baldellou Lopez M, Healey A, Estevao C, Fancourt D, Dazzan P, Pariante C, Dye H, Osborn T, Bind R, Sawyer K, Rebecchini L, Hazelgrove K, Burton A, Manoharan M, Perkins R, Podlewska A, Chaudhuri R, Derbyshire-Fox F, Hartley A, Woods A, Crane N, Bakolis I, Sevdalis N. Study protocol: randomised controlled hybrid type 2 trial evaluating the scale-up of two arts interventions for postnatal depression and Parkinson's disease. BMJ Open. 2022 Feb 1;12(2):e055691. doi: 10.1136/bmjopen-2021-055691.

Estevao C, Bind R, Fancourt D, Sawyer K, Dazzan P, Sevdalis N, Woods A, Crane N, Rebecchini L, Hazelgrove K, Manoharan M, Burton A, Dye H, Osborn T, Davis RE, Soukup T, Arias de la Torre J, Bakolis I, Healey A, Perkins R, Pariante C. SHAPER-PND trial: clinical effectiveness protocol of a community singing intervention for postnatal depression. BMJ Open. 2021 Nov 17;11(11):e052133. doi: 10.1136/bmjopen-2021-052133.