Psychosocial support interventions to improve treatment outcomes for people living with tuberculosis: a mixed methods systematic review and meta-analysis.

Background: People with tuberculosis (TB) face multi-dimensional barriers when accessing and engaging with care. There is evidence that providing psychosocial support within people-centered models of care can improve TB outcomes, however, there is limited consensus on what works. It remains important for such interventions to be rigorously assessed, and mixed methods systematic reviews are one way of synthesising data for policy makers to be able to access such evidence. Mixed methods reviews take a complexity perspective, with qualitative data being used to contextualise the quantitative findings and giving an insight into how interventions are contingent on variations in design and context. Methods: Five electronic databases were searched from January 1 2015 to 14 January 2023 for randomised controlled trials, quasi-experimental trials, cohort studies and qualitative studies of interventions providing psychosocial support (material and/or psychological-based support) to adults with any clinical form of active TB. Studies with inpatient treatment as the standard of care were excluded. Quantitative studies reporting pre-specified standard TB outcomes were eligible. In line with established mixed methods review methodology, a convergent parallel-results synthesis design was followed: quantitative and qualitative syntheses were distinct and carried out using appropriate methods. A convergent coding matrix was then used to integrate the results. The protocol was registered on PROSPERO (CRD42021235211). Findings: Twenty-three studies of interventions were included (12 quantitative, 10 qualitative, and 1 mixed methods study) were included. Most studies were conducted in low-and middle-income countries with a high-burden of TB. Three explanatory and contextual middle-range theories from the integration of qualitative and quantitative data were developed: effective interventions provide multi-dimensional support; psychological-based support is transformative but there is insufficient evidence that it improves treatment outcomes on its own; intervention delivery shapes a logic of care. Interpretation: This review takes a complexity perspective to provide actionable and timely insight to inform the design and implementation of locally-appropriate and people-centered psychosocial support interventions within national TB programmes. Funding: There was no funding source for this study.