Appraising the HIV prevention cascade methodology to improve HIV prevention targets: Lessons learned from a general population pilot study in east Zimbabwe

Multiple HIV Prevention Cascades (HPC) formulations have been proposed to assist advocacy, monitoring of HIV prevention implementation and research to identify ways to increase use of HIV prevention methods. Schaefer and colleagues proposed a unifying formulation suitable for widespread use across different populations which could be used for routine monitoring or advocacy. Robust methods for defining and interpreting this HPC formulation using real world data are necessary to aid promotion and utilisation of this framework to address necessary gaps in primary HIV prevention method use. We used 2018-19 data from the Manicaland Pilot HIV Prevention Cascades Study in eastern Zimbabwe to validate the HPC framework for pre-exposure prophylaxis (PrEP), voluntary medical male circumcision (VMMC), female and male condoms, and combination prevention. Validation involved: (1) testing feasibility of populating the HPC; (2) comparing simple vs. complex HPC measures using 2-sample proportion tests; and (3) using logistic regression to assess whether HPC cascade bars predicted prevention use and whether sub-bars explained loss from the HPC. It was possible to populate the HPC for individual and combined prevention methods using pilot survey data. Most steps were associated with prevention method usage outcomes, except for VMMC. There were significant overlaps between individuals reporting positive responses for the main motivation bar and those citing barriers to motivation. To refine the HPC's access bar definition, it is suggested to consider individuals who report access barriers. While the HPC framework identifies barriers to individual prevention methods, challenges arise in identifying those for combined prevention. Our study successfully utilised questionnaires from the Manicaland HPC pilot survey to measure the HPC for individual and combined prevention methods. In conclusion, it is feasible to populate this framework using general population survey data and designated questionnaire modules. We propose a final formulation of the HPC, and questionnaire modules and methods to create it. With proper promotion, such as through measurement within national to local population surveys, the HPC framework can allow comparison of prevention method use, identification of intervention targets and enhance prevention services, aiding in the crucial reduction of HIV incidence.