Impact of the 2013 WHO guidelines for screening and treatment of cervical precancerous lesions on women’s screening rates, by HIV-status, in East and Southern Africa: A regression discontinuity design analysis

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Introduction

In 2013, the World Health Organization (WHO) issued guidelines for cervical pre-cancer screening. It recommended screening women aged 30–49, and younger women once they tested HIV-positive. Subsequent WHO guidelines recommended screening women living with HIV (WLHIV) starting at age 25. However, the impact of 2013 guidelines and age to start screening on screening has not been studied.

Methods

We used a regression discontinuity design (RDD) analysis of population-based data to assess the impact of the 2013 WHO guidelines on the screening rates for women according to HIV status and age group in Ethiopia, Malawi, Rwanda, Tanzania, Zambia, and Zimbabwe. The outcome was self-reported ever having been screened for cervical pre-cancer between 2008 and 2018. We compared the screening rates according to HIV status and age group, before and after age 25 years. And before and after 2014, the year countries adopted the 2013 guidelines. We then used a data-driven optimal bandwidth selection procedure to estimate the guidelines’ average treatment effect (ATE), with a local polynomial regression discontinuity and robust bias-corrected confidence intervals. We validated the RDD methodology overall and for women with a significant ATE at the country-level analysis.

Results

We included 73179 women: 6680 (9.1%) living with HIV, 4328 (5.9%) with unknown HIV status, and 62171 (85.0) with a negative HIV status. 5726 (7.9%) reported having ever been screened; 4022 (6.5%) with unknown HIV status, 525 (12.1%) with a negative HIV status and 1179 (17.7%) living with HIV. Adolescent girls and young women living with HIV (AGYWLHIV) aged 15–24 reported screening less often (917 (13.7%)) than their peers with unknown (1677 (38.8%)) or positive HIV status (27278 (43.9%)) (P<0.001), or older women. The ATE of screening was 0 for women with unknown or positive HIV status, for whom the RDD was valid.

Conclusion

We found no evidence the 2013 WHO guidelines increased or reduced women’s cervical pre- cancer screening. However, AGYWLHIV reported screening less often. Policy makers should lower the age to start screening WLHIV from 25 to 15 to screen more AGYWLHIV. Studies are required to examine the impact of the guidelines on cervical pre-cancer screening in more countries.

Article activity feed