A multi-methods implementation research evaluation of a program scale cluster randomized trial to improve integration of HIV prevention and treatment services in family planning clinics in Kenya

Abstract

Funding cuts to the US President’s Emergency Plan for AIDS Relief (PEPFAR) undermine the global HIV response and threaten harm reduction efforts across low-and middle-income countries (LMICs). In the current geopolitical and funding landscape, the urgency of integrating HIV services within broader health systems is greater than ever to sustain hard-won gains. Integrating HIV prevention and treatment services into family planning (FP) clinics offers a promising strategy to address HIV and unwanted pregnancy simultaneously while supporting the UNAIDS 95-95-95 targets for testing, treatment, and prevention. However, evidence on how best to integrate these two services in LMICs remains limited. The FP HIV SCALE study is a hybrid type II effectiveness-implementation trial that evaluates the effectiveness of the Systems Analysis and Improvement Approach (SAIA) as an implementation strategy to increase the integration of HIV prevention and treatment services at program scale in FP clinics when implemented by the Mombasa County public health workforce in Kenya. The work described in this dissertation is nested within the FP HIV SCALE study. We applied a combination of multi-method analytical approaches, including linear regression models, mixed-methods studies, and a rigorous configurational comparative method called coincidence analysis (CNA) to conduct an implementation evaluation for the FP HIV SCALE study. The primary aims were to: 1) evaluate associations between organizational readiness for change, organizational climate, and successful integration of HIV counseling, HIV testing, screening and linkage for PrEP, and linkage to HIV care in FP clinics; 2) evaluate implementation outcomes of acceptability, appropriateness, and feasibility from healthcare workers’ perspectives of using a modified national register (research record) to document programmatic integrated FP/HIV performance data; and 3) define the difference-makers of SAIA micro-interventions that are necessary or sufficient for the successful integration of HIV counseling and testing in FP clinics. In the first aim, we did not observe associations between organizational readiness, organizational climate metrics, and clinical outcomes for integrated HIV services. We identified organizational climate metrics that are important predictors of readiness for change. For both FP clinic staff and clinic managers, upward communication and innovation/flexibility metrics were predictive of organizational readiness. Furthermore, clinic managers’ perceptions of management support and commitment to the facility were strongly associated with organizational readiness. The second aim used a convergent mixed-methods approach to evaluate the research record, applying the Consolidated Framework for Implementation Research (CFIR) to assess outcomes of acceptability, appropriateness, and feasibility from Proctor’s Implementation Outcomes Framework. Family planning clinic staff and managers in intervention and control clinics found the research record highly acceptable, appropriate, and feasible for documenting integrated outcomes in a single place. Finally, the third aim on CNA identified difference-makers of SAIA micro-intervention categories that resulted in achieving success of integrated HIV counseling and testing in FP clinics. For HIV counseling, three candidate models each with three pathways were identified. A sufficiency model with three pathways was identified as the best because of its high coverage scores compared to the other two models. The patient education category of micro-interventions was a sufficient condition for achieving the outcome. The other two pathways represented the data quality category bundled with either the training category or the external support category of micro-interventions. For HIV testing, we identified a single model consisting of two pathways. The training category of micro-interventions alone was sufficient for producing the outcome, and a second pathway was a conjunct of patient education and data quality micro-interventions. This dissertation offers novel insights that are broadly relevant in the field of implementation research, as well as a roadmap for integrating HIV prevention and treatment services into FP clinics. By applying implementation science to real-world scale-up efforts, this work helps bridge the “know-do gap” in integrating health services. These findings deepen our understanding of how we and others have hypothesized the relationship between readiness and clinical outcomes while also highlighting potential gaps in measurement, the practical tools essential for integration, and the optimal combination of factors that drive success of integrated outcomes. Collectively, these findings offer important lessons for LMICs and provide actionable insights for researchers and policymakers working to advance the integration of HIV and FP services globally.