Evaluating a community-driven cervical cancer prevention model in western Kenya

Investigator: Megan J. Huchko, MD
Sponsor: NIH National Cancer Institute

Location(s): Kenya


Cervical cancer is the second most common cancer among women, with over 500,000 cases diagnosed annually. This cancer disproportionately impacts women in low-resource settings, where almost 90% of disease mortality occurs. Although effective screening and treatment technologies have been developed for use in low-resource settings, there are complicated, context-specific barriers to their implementation. Successful cervical cancer screening strategies depend on community-wide access to care and linkage to treatment for those women who screen positive. We will identify the factors that inhibit or enable uptake of cervical cancer prevention activities in western Kenya, determine whether screening with HPV reaches more women when offered in community health campaigns compared to clinics, and work with the community to develop and test a strategy to improve linkage to treatment. This work will produce a "toolkit" for scale-up of cervical cancer prevention programs in similar low-resource settings.

Evaluating a community- driven cervical cancer prevention model in western Kenya Cervical cancer impacts over half a million women globally each year. While cervical cancer control in wealthy countries is one of the public health success stories of the past century, the disease remains a significant threat for women in developing countries where almost 90% of cases occur. Although effective screening technologies have been developed for use in low-resource settings, there are complicated, context-specific barriers to their implementation as part of a complete cervical cancer prevention strategy. In addition to employing evidence-based screening and treatment techniques, the overall population impact of cervical cancer screening depends on two main factors: (1) community-wide access to screening and (2) successful linkage to treatment or follow-up for women who screen positive. Strategies to address these factors must also be acceptable, relatively easy to implement, and cost-effective to be sustainable. Our formative work in western Kenya, a country with a high cervical cancer burden, uncovered tangible barriers and facilitators to these key steps in the "cervical cancer prevention cascade" and has led to a strategy that will increase women's uptake of cervical cancer prevention activities. Based on this work, we have developed the following hypotheses: 
(1) cervical cancer screening with self-collected human papillomavirus (HPV) specimens will reach more women when offered through community health campaigns versus government clinics; and 
(2) community-developed strategies will successfully link more women to treatment than the current standard of care (i.e., referral to treatment sites).
To test these hypotheses, we propose a two- phase cluster-randomized trial of implementation strategies for a Ministry of Health and WHO-recommended cervical cancer prevention protocol in western Kenya. During Phase 1, communities will be randomized to HPV-testing in either community-health campaigns or in clinics, with standard referral for treatment of HPV+ women to government facilities. We will use the RE-AIM framework, a framework used to evaluate implementation strategies, (Reach, Efficacy, Adoption, Implementation consistency and costs, and Maintenance) to assess the key outcomes; we will then work in partnership with the community to develop a strategy for enhanced linkage to care. In Phase 2, all communities will offer community-based testing with enhanced linkage to care. Conducting this cluster-randomized trial will enable us to assess the proportion of women in each community who get cervical cancer screening, the gain in treatment access with enhanced linkage to care, and the cost-effectiveness of the two interventions. The RE-AIM framework will allow us to measure and refine the context-specific dimensions of the project to produce a "toolkit" for scale-up within ths region and implementation into similar settings.