VMMC Coverage Proposal (CDC HQ)

[jb3436]

VMMC Request

Status:
Exploratory discussions to determine next steps on request from Carlos Toledo.
ICPI/Engagements#65
Meeting with CDC team Tuesday July 16, 2019 to discuss next steps
Will determine if it is a denominators project or other

Proposal for creating an ongoing measure of male circumcision coverage

Problem statement

PEPFAR has supported over 18 million voluntary medical male circumcisions (VMMC) since the program's inception. However, accurate, reliable estimates of the proportion of men circumcised nationally, or at sub-national levels which are crucial for program planning, are missing.  Various estimate sources exist that currently offer that information, however, discrepancies have been noted between sources. In particular, existing estimates are not consistent with cumulative MER VMMC_CIRC indicator values. As countries continue to reach more and more men, accurate, reliable estimates of VMMC coverage will become increasingly important as plans are made to transition from the scale up phase of VMMC to a longer-term approach to maintaining high MC coverage to sustain epidemic control.

At the same time, other PEPFAR indicators and tools have overlapping functions with these coverage estimates, but are not cross-harmonized and involve parallel efforts. An accurate COP planning process would require that agreed-on MC coverage estimates be used to populate the datapack. Meanwhile, the VMMC_CIRC_NAT and VMMC_TOTALCIRC_NAT MER indicators face the same technical challenges as the coverage estimates (e.g. accurately capturing past medical MCs done outside PEPFAR) but use different approaches and, as currently specified, cannot be directly used for their ultimate purpose of clarifying what unmet need remains. All of these entities need to be formulated to be usable for planning, based on best-available data, and harmonized with each other.

Objective

The major goal of this activity is to develop a tool that summarizes national and SNU-level coverage of medical circumcision (e.g., proportion of men medically circumcised) across 15 priority countries for VMMC, triangulated and validated through an agreed-on process that incorporates the full range of available data streams.  This will include circumcision coverage estimates in priority age bands (i.e., 15-29) and other age groups. This tool will also aim to assist countries in longer-term planning for VMMC by establishing a way to plan for ongoing circumcision of rising cohorts of males as they age into the eligible population (at least 10 years of age).

This tool should be interagency, housed in OGAC, and intrinsically harmonized with related entities like MER VMMC indicators and datapack fields. Every effort should also be made to ensure output harmonization with the VMMC achievement data annually released worldwide by WHO, which is the global standard reference.

Proposed approach

We propose that the current VMMC_TOTALCIRC_NAT indicator be redefined as a coverage indicator, calculated with inputs that will include the most recent VMMC_CIRC_NAT value to ensure it is update with the most recent achievements. The outputs of the proposed tool will be the values of this redefined indicator, and will directly populate the datapack. This will ensure the necessary harmonization across PEPFAR indicators, and ensure that the process is interagency and OGAC-led and that coverage outputs are universally available.

Agreement must still be reached on how to address the technical issues that have faced all previous attempts to estimate coverage. We proposed the following steps be conducted by an interagency team, with ICPI leading the process and creating the final tool.

• Review currently available primary data sources (e.g., PEPFAR indicators, PHIA, MoH reporting, and others) and existing approaches/estimates (datapack, DMPPT2 tool, and others) to identify limitations and gaps in data, and inconsistencies in approaches.

• Identify standardized data elements and a process for verification to produce reliable national and SNU-level coverage of medical circumcision. Any approach will include selecting a source of baseline coverage data and sources of updated data for each year. Some of these may vary between countries.

  • We anticipate that the most appropriate source of baseline coverage for countries with PHIA data is the PHIA. Technical issues around how to classify circumcisions as medical or nonmedical must still be resolved, and approaches for countries without PHIAs must be developed.
  • For fiscal years following a country's PHIA but before the VMMC_TOTALCIRC_NAT indicator was developed, a process must be developed for collecting MoH achievement data and ensuring it is consistent with PEPFAR achievements for the year. Notably, a process like this already exists involving WHO and PEPFAR: MoHs report annual MC achievements to WHO and these are then harmonized with PEPFAR before dissemination. The final process should build on this practice.
  • For fiscal years after development of the VMMC_TOTALCIRC_NAT indicator, this value should be used. It may be necessary to do a landscape review of how this data is currently collected, and specify minimum verification steps to ensure data quality and basic methodologic consistency between countries.
  • To convert VMMC achievement data to age-specific coverage, population data and a process for incorporating aging are needed. Agreement is needed on population data sources. Some existing coverage tools incorporate aging, and this aspect could possibly be adapted into the new tool.
  • Coverage age bands must be selected. The 5-year bands used for VMMC achievement reporting are an option, but this level of precision sacrifices accuracy, increases complexity, increases standard errors for baseline values due to smaller sample sizes, and may not be programmatically relevant. Broader coverage bands like 10-14, 15-29, and 45 and above could be considered.

• Develop the tool. It must summarize national and SNU-level coverage of medical circumcision (e.g., proportion of men medically circumcised) across 15 priority countries for VMMC, within agreed-on age bands.

• Propose changes to the datapack process and PEPFAR MER Indicators in time to inform the COP20 process and future reporting. 

Deliverables:

• Comparison table of data sources, variables collected, definitions used, process for verification
• Tool summarizing national and SNU-level coverage of medical circumcision (e.g., proportion of men medically circumcised) across 15 priority countries for VMMC
• Agreed-on process and roles for incorporating available data streams to update tool at least annually
• Recommendations to changes to datapack process and PEPFAR MER Indicators to inform COP20 and future reporting