Sex Differences in HIV Testing — 20 PEPFAR-Supported Sub-Saharan African Countries, 2019

Despite progress toward controlling the human immunodeficiency virus (HIV) epidemic, testing gaps remain, particularly among men and young persons in sub-Saharan Africa (1). This observational study used routinely collected programmatic data from 20 African countries reported to the U.S. President's Emergency Plan for AIDS Relief (PEPFAR) from October 2018 to September 2019 to assess HIV testing coverage and case finding among adults (defined as persons aged ≥15 years). Indicators included number of HIV tests conducted, number of HIV-positive test results, and percentage positivity rate. Overall, the majority of countries reported higher HIV case finding among women than among men. However, a slightly higher percentage positivity was recorded among men (4.7%) than among women (4.1%). Provider-initiated counseling and testing (PITC) in health facilities identified approximately two thirds of all new cases, but index testing had the highest percentage positivity in all countries among both sexes. Yields from voluntary counseling and testing (VCT) and mobile testing varied by sex and by country. These findings highlight the need to identify and implement the most efficient strategies for HIV case finding in these countries to close coverage gaps. Strategies might need to be tailored for men who remain underrepresented in the majority of HIV testing programs.

Despite progress toward controlling the human immunodeficiency virus (HIV) epidemic, testing gaps remain, particularly among men and young persons in sub-Saharan Africa (1). This observational study used routinely collected programmatic data from 20 African countries reported to the U.S. President's Emergency Plan for AIDS Relief (PEPFAR) from October 2018 to September 2019 to assess HIV testing coverage and case finding among adults (defined as persons aged ≥15 years). Indicators included number of HIV tests conducted, number of HIV-positive test results, and percentage positivity rate. Overall, the majority of countries reported higher HIV case finding among women than among men. However, a slightly higher percentage positivity was recorded among men (4.7%) than among women (4.1%). Provider-initiated counseling and testing (PITC) in health facilities identified approximately two thirds of all new cases, but index testing had the highest percentage positivity in all countries among both sexes. Yields from voluntary counseling and testing (VCT) and mobile testing varied by sex and by country. These findings highlight the need to identify and implement the most efficient strategies for HIV case finding in these countries to close coverage gaps. Strategies might need to be tailored for men who remain underrepresented in the majority of HIV testing programs.
In 2014, the Joint United Nations Programme on AIDS (UNAIDS) launched its 90-90-90 strategy for ending the global HIV pandemic: 90% of all persons living with HIV/AIDS (PLHIV) know their status; of these, 90% are receiving antiretroviral treatment (ART); and of these, 90% are virally suppressed (2). PEPFAR provides guidance on reaching these targets to all its supported countries (3). PEPFAR also collects data on standardized indicators as part of its Monitoring, Evaluation, and Reporting system (4). These data are collected by facility and community sites and are reported quarterly by each country program.
Routine program data reported to PEPFAR from October 2018 to September 2019 from 20 sub-Saharan African countries and modeling data from UNAIDS were used to identify progress toward achieving the first of the three 90-90-90 goals. These countries were selected because they collectively represent the highest HIV prevalence among PEPFAR-supported countries. Indicators used included the number of HIV tests conducted among adults, the number of HIV-positive test results, and yield (or percentage positivity) defined as the number of positive test results divided by the total number of tests reported. Results for each country are presented overall and disaggregated by sex and testing strategy. Testing strategies include index testing (offering an HIV test to the partners and biologic children of PLHIV), PITC (providers recommending an HIV test as part of routine care), VCT (HIV testing at a clinic dedicated to this purpose), and mobile testing (HIV testing offered at an ad hoc location in the community [e.g., van, workplace, or school]).The standard test statistic for the difference in proportions between men and women was computed on the basis of pooled variance formulation. P-values <0.05 were considered statistically significant for each pairwise difference and were computed under the asymptomatic normality assumption (5). PITC was used as the reference strategy for comparisons among strategies because the highest number of new HIV cases are identified by PITC in sub-Saharan Africa (1). The percentage positivity for the three other strategies (index testing, VCT, and mobile testing) was compared with PITC for each country and by sex.
From October 2018 to September 2019, PEPFAR supported 60,945,355 tests that identified 2,603,560 adults with positive HIV test results (5.0% yield; Table 1). Approximately one fifth (19.9%) of all testing occurred in South Africa. More women received tests than men (women, 40,263,510; men, 20,681,845). However, yield was slightly higher among men (970,100; 4.7% yield) than among women (1,633,460; 4.1% yield). Over one half (51.6%) of all HIV-positive results among men were reported by South Africa, Tanzania, and Zambia, and approximately one third (29.2%) were reported by South Africa alone.
Across the 19 countries (excluding Malawi because of limited data), PITC identified the most PLHIV (63.2%) compared with index testing (17.4%), VCT (11.0%), and mobile testing (8.4%). HIV case finding among men followed a similar  (Table 2). Although PITC identified more PLHIV than any other testing strategy, the percentage positivity of index testing was higher than PITC in all countries and for both sexes ( Table 3).The number of PLHIV identified through index testing ranged from 500 in South Sudan (11.0% yield) to 116,500 in Tanzania (21.4% yield) ( Table 2). The five countries that identified the most PLHIV through index testing were Uganda The yield for VCT among men was significantly higher than for PITC in eight countries (Angola, Cameroon, Côte d'Ivoire, Malawi, Mozambique, Uganda, Zambia, and Zimbabwe) but was lower in the Democratic Republic of Congo (DRC) ( Table 3). Similarly, mobile testing had a significantly higher yield than PITC among men in Cameroon, Côte d'Ivoire, DRC, Malawi, Mozambique, Nigeria, Uganda, and Zimbabwe but a lower yield in Botswana, Lesotho, and South Africa. Among women, VCT had a significantly higher yield than PITC in all countries except Botswana, whereas mobile testing had a significantly higher yield than PITC in all but three countries (Botswana, South Africa, and Kenya).
These findings will help guide the Ministries of Health in selecting specific testing strategies to increase HIV-testing coverage. This will help identify persons living with HIV infection.

Discussion
PITC identified the largest number of PLHIV, but index testing was more efficient at finding PLHIV as evidenced by a higher positivity rate. Several factors, including the absolute number of PLHIV identified, testing yield, cost per diagnosis, and the ability to reach persons with undiagnosed HIV infection, can help identify the best HIV-testing approach (1).
Although an efficient testing strategy in terms of identifying PLHIV, index testing is resource-intensive (1). For large yields of persons with a new HIV diagnosis to be identified, index testing requires a massive, potentially expensive, scale-up (6). However, Kenya, Tanzania, Mozambique, and Uganda have achieved substantial scale-up, demonstrating the feasibility of index testing in sub-Saharan Africa. Although the majority of PLHIV in sub-Saharan Africa receive their diagnosis through PITC, findings from this analysis show that index testing is identifying new PLHIV in many countries. Screening tools that use a combination of clinical and behavioral questions to identify persons most likely to be HIV-positive can increase the percentage positivity of PITC. However, these screening tools must be validated to ensure high sensitivity and good specificity to avoid missed opportunities for diagnosis (3).
These findings show that men are less likely to be tested for HIV than women and represented 37% of HIV-positive results. Although women were twice as likely to be tested for HIV, approximately one third of new cases identified occurred among men. In all countries, women are routinely tested for HIV as part of antenatal testing, regardless of their clinical or behavioral risk factors. This might help explain why the number of tests conducted was higher among women compared with men, but the percent positivity rate was lower. PITC and index testing identified the most HIV-positive men, but mobile testing remains an important strategy to reach men who are unable or unwilling to attend health care facilities. Interventions that might improve coverage among men include flexible hours, male counselors, and integration of HIV testing into screening for other chronic conditions (7). Although not included in this analysis, HIV self-testing, directly offered within health facilities (8) or distributed by sex or needle-sharing partners as part of index testing (9), might also be an efficient method for reaching men.
The findings in this report are subject to at least three limitations. First, cost data are lacking. Second, data quality varies across countries despite PEPFAR's monitoring and reporting guidance. Finally, PITC data include antenatal testing, which results in more testing among women. Testing antenatal women at lower risk also explains why the yield for PITC is lower for women than men, even though the yield is higher  for women with the other three testing strategies. Strengths include the number of countries involved in the analysis and the volume of testing data collected by implementing partners. As more PLHIV are identified, finding the remaining PLHIV will become more difficult and expensive because new cases will be harder to find. Multiple testing strategies, tailored to the epidemiologic context and targeted to populations with low access to HIV testing, can help reach these persons. Recency testing now allows programs to identify persons infected with HIV in the past year as well as clusters of incident cases. Immediate and intensified index testing efforts targeted to these clusters might reduce transmission and help countries achieve epidemic control (10). What is added by this report?
From October 2018 to September 2019, men in 20 sub-Saharan African countries were half as likely as women to receive an HIV test, and 37% of the HIV-positive results were among men. Similar sex differences were observed across HIV testing strategies based upon percent positivity rates.
What are the implications for public health practice?
These results highlight provider-initiated testing and index testing as strategies that might improve HIV testing coverage and maximize the number of persons newly identified as HIV-positive in 20 African countries.