Findings from the pre-implementation phase of the study were reviewed at an annual ASSET meeting . The TDF  and the CICI framework  were used to identify contextual and behavioural determinants that could influence the implementation of HSS interventions and implementation outcomes that may have been missed in the ToC workshop. If there were determinants not identified in the pre-implementation phase, these were accounted for at this stage. The HSS interventions that were selected in the ToC workshop were also labelled according to the EPOC taxonomy [18, 20]. Finally, we ensured that implementation outcomes were aligned with the selected HSS interventions. Table 1 provides on overview of the contextual barriers that were identified, the HSS interventions that were selected and the implementation outcomes that will be assessed.
Developing a training model
To ensure sustainability of the HSS programme, a cascaded training model (Supplementary file 2) was developed, whereby the ASSET team provided master training workshops at centralised locations. Three master training workshops were held with selected DoH staff members: (i) facility-based psychiatric nurses received training on health promotion and awareness, (ii) facility PACK trainers received training on detection and referral, and (iii) community-based services (CBS) trainers from each of the sub-districts (whose primary role was to provide on-going training to community-based healthcare workers on a range of topics) and NPO managers from the supporting NPOs received the counselling and supervision training. Master trainers were tasked with providing training to small groups of facility- and community-based healthcare workers whose role would include implementing the selected HSS interventions.
Addressing poor patient knowledge and health seeking behaviour
Training and health promotion were selected as the HSS interventions to address poor patient knowledge of mental health and domestic violence, poor health seeking behaviour, and the high levels of stigma among patients. The awareness raising component of the HSS programme will consist of health promotion officers or other lay healthcare workers providing daily, 5-7-minute talks to groups of pregnant women in waiting areas at facilities. Information to be covered during the talks will include the signs, symptoms, risk factors and consequences of depression, anxiety and experiences of domestic violence as well as the treatment options available.
Facility-based psychiatric nurses who were the recipients of the master training, will be tasked with providing cascaded training to health promotion officers or other lay healthcare workers at facilities. Master trainers will be provided with a Health Promotion and Awareness of Maternal Mental Health Training Manual  to guide the delivery of the training content. Healthcare workers who will be trained to implement the health promotion talks will be provided with an A3 size, colour flipchart to guide the delivery of the talk.
Addressing low levels of detection
Training, delivery of individual-level care and audit and feedback were selected as the HSS interventions to address the low levels of detection. ANC nurses will be trained to screen all pregnant women for symptoms of CMDs and experiences of domestic violence as part of routine care, using the mental health screening questionnaire  available in the MCR  and the PACK guidelines . The training material was developed in conjunction with the Knowledge Translation Unit (KTU), a clinical research unit at the University of Cape Town, that led the development of PACK. The PACK guide was used as the foundation for the detection and referral process. A PACK Antenatal Women and Mental Health Module  was developed to strengthen the mental health component of routinely provided antenatal care. Four case studies formed the backbone of the module, complemented by HSS discussions, completion of relevant stationery and a focus on effective communication strategies. PACK facility trainers were the recipients of the master training and will be tasked with training ANC nurses at their facilities. The cascaded training will consist of four, weekly, 2-hour training sessions.
During the implementation phase, audit and feedback will be used to assess the screening rates at facilities and provide feedback to the relevant ANC nurses and their managers at bi-monthly meetings.
Addressing poor linkage to care
Referral systems, training and audit, and feedback were selected as the HSS interventions to address the poor linkage to care. Standardised referral pathways were developed. PACK Facility Trainers (facility-based healthcare workers who are PACK trained and assigned to deliver PACK training to healthcare workers in their own facility) were the recipients of the master training and will be tasked with training ANC nurses at their facilities. ANC nurses will be trained to assess the severity of symptoms in pregnant women who screen positive, and to use the standardised referral pathways to link them to care. Pregnant women with mild to moderate symptoms of depression will be referred to CHWs, while women with severe symptoms of depression will be referred to healthcare workers providing specialised care such as Medical Officers, Psychiatric Nurses, and Psychologists. Pregnant women experiencing domestic violence will be referred to a Social Worker for support.
During the implementation phase, audit and feedback will be used to assess the referral rates at facilities and provide feedback to the relevant ANC nurses and their managers at bi-monthly meetings.
Addressing the limited availability of treatment
Several HSS interventions were selected to address the limited availability of treatment, including task-sharing, training, delivery of individual-level care, change to the healthcare environment, audit and feedback, and performance monitoring. A task-sharing psychological counselling programme was developed, to be delivered by CHWs to pregnant women with mild to moderate symptoms of depression or anxiety. It consists of three, 45-minute, structured, individual-level counselling sessions using problem-solving therapy, delivered weekly by CHWs, in patients’ homes . The development of the psychological counselling programme was informed by a manual review of counselling interventions, semi-structured interviews with healthcare workers and pregnant women, and finally through several stakeholder engagement meetings. CHWs will be supervised and supported by Outreach Team Leaders (OTLs) to reinforce the counselling skills, ensure fidelity to the psychological counselling and manage difficult cases.
CBS trainers employed in each sub-district were selected as the recipients of the master training. Master trainers received a 4-day training workshop, where they were provided with a Counselling Skills for Community Health Workers  training manual and a Maternal Mental Health Counselling Support Guide to assist them in delivering cascaded training on (i) counselling, and (ii) supervision and support. The cascaded counselling training will be delivered over three days to CHWs and OTLs and consist of five sections: (i) understanding depression, anxiety and experiences of domestic violence, (ii) basic counselling skills, (iii) patient assessment, (iv) 3-session counselling intervention, and (v) coping skills. The cascaded supervision and support training session will be delivered to OTLs during a one-day training which will include (i) counselling support styles and skills, and (ii) the use of individual and group support. CHWs and OTLs will be provided with a Reference Guide for Community Health Workers, in addition to the Counselling Skills for Community Health Workers training manual to support the delivery of the counselling intervention.
During the implementation phase, audit and feedback will be used to assess the counselling rates and provide feedback to the relevant CHWs, OTLs and their managers at bi-monthly meetings. OTLs will also observe the delivery of the counselling sessions to monitor the performance of the CHWs.