Implementation process and outcomes of a mental health programme integrated in primary care clinics in rural Mexico: a mixed-methods study

BACKGROUND: Policies and programmes in Mexico promote the integration of mental health services into primary health care (PHC), however these services remain largely unavailable in the country. Since 2014 a non-governmental organisation has delivered a mental health programme at PHC clinics in the state of Chiapas, in partnership with the local Ministry of Health (MoH). The programme provides mental health services based on the mhGAP guidelines through multiple implementation strategies, including programme financing, infrastructure strengthening, high-intensity training, and supervision. This study aimed to examine the implementation process and outcomes of this mental health programme to understand the extent to which mental health care integration has been achieved and to identify the successes and remaining challenges in order to inform the development and implementation of similar programmes. METHODS: We used a mixed-methods convergent design. Quantitative data for the period between December 2016 and December 2017 were extracted from the organisation’s health information system to capture process indicators, including the amount (dose) and quality (fidelity) of services delivered. We conducted two focus groups and 24 semi-structured interviews with health providers and managers to ascertain implementation outcome data: penetration, fidelity, acceptability, appropriateness and feasibility. Quantitative and qualitative data were analysed using descriptive and framework analyses, respectively. RESULTS: During the study period, health providers delivered mental health consultations to 486 adults diagnosed with a mood or anxiety disorder. Programme fidelity was limited given that talk-based interventions, which are required in all consultations according to programme guidelines, were only provided in 24% of consultations. Only 42% of service users attended more than 50% of scheduled mental health follow-up consultations, which also hindered fidelity. Low attendance is partially attributed to limited programme appropriateness, given that interventions to address social risk factors are not available. High levels of acceptability and feasibility enabled through strong support from the organisation were key programme strengths. CONCLUSIONS: Mental health programmes at PHC can be implemented when adequate support and supervision structures are in place, and key resources are available. There is an urgent need for health systems strengthening to support efforts to provide mental health care, and to link PHC with locally-relevant social interventions.