Men missing from the HIV care continuum in sub-Saharan Africa: a meta-analysis and meta-synthesis

Abstract

Introduction: Men are missing along the HIV care continuum. However, the estimated proportions of men in sub-Saharan Africa meeting the UNAIDS 95-95-95 goals vary substantially between studies. We sought to estimate proportions of men meeting each of the 95-95-95 goals across studies in sub-Saharan Africa, describe heterogeneity, and summarize qualitative evidence on factors influencing care engagement.

Methods: We systematically searched PubMed and Embase for peer-reviewed articles published between 1 January 2014 and 16 October 2020. We included studies involving men ≥15 years old, with data from 2009 onward, reporting on at least one 95-95-95 goal in sub-Saharan Africa. We estimated pooled proportions of men meeting these goals using DerSimonion-Laird random effects models, stratifying by study population (e.g. studies focusing exclusively on men who have sex with men vs. studies that did not), facility setting (healthcare vs. community site), region (eastern/southern Africa vs. western/central Africa), outcome measurement (e.g. threshold for viral load suppression), median year of data collection (before vs. during or after 2017) and quality criteria. Data from qualitative studies exploring barriers to men's HIV care engagement were summarized using meta-synthesis.

Results and discussion: We screened 14,896 studies and included 129 studies in the meta-analysis, compiling data over the data collection period. Forty-seven studies reported data on knowledge of serostatus, 43 studies reported on antiretroviral therapy use and 74 studies reported on viral suppression. Approximately half of men with HIV reported not knowing their status (0.49 [95% CI, 0.41-0.58; range, 0.09-0.97]) or not being on treatment (0.58 [95% CI, 0.51-0.65; range, 0.07-0.97]), while over three-quarters of men achieved viral suppression on treatment (0.79 [95% CI, 0.77-0.81; range, 0.39-0.97]. Heterogeneity was high, with variation in estimates across study populations, settings and outcomes. The meta-synthesis of 40 studies identified three primary domains in which men described risks associated with engagement in HIV care: perceived social norms, health system challenges and poverty.

Conclusions: Psychosocial and systems-level interventions that change men's perceptions of social norms, improve trust in and accessibility of the health system, and address costs of accessing care are needed to better engage men, especially in HIV testing and treatment.

Keywords: Africa South of the Sahara; HIV infections; HIV testing; continuity of patient care; men; qualitative research.

Figure 1

Study selection. Search process for selected studies in meta‐analysis and meta‐synthesis.

Figure 2

Forest plot of studies reporting data on proportion of men with HIV with knowledge of their status, listed in ascending order of year of publication.

Figure 3

Forest plot of studies reporting data on proportion of men with HIV on ART out of all men with HIV who know their status, listed in ascending order of year of publication.

Figure 4

Forest plot of studies reporting data on proportion of men on ART who were virally suppressed out of all men with HIV on ART, listed in ascending order of year of publication.

Figure 5

Conceptual model of men's engagement in the HIV care continuum.

Figure A1

Forest plot comparing proportions of men and women with awareness of positive HIV status, including only studies reporting these data for both men and women (N=34 studies).

Figure A2

Forest plot comparing proportions of men and women on ART of those who are aware of their positive status, including only studies reporting these data for both men and women (N=35 studies).

Figure A3

(a) Forest plot comparing proportions of men and women who are virally suppressed of those on ART, including only studies reporting these data for both men and women (N=70 studies). Data for men shown here and data for women with overall comparisons shown in Figure 3b. (b) Forest plot comparing proportions of men and women who are virally suppressed of those on ART, including only studies reporting these data for both men and women (N=70 studies). Data for women and overall comparisons shown here and data for men shown in Figure 3a.

Figure A3

(a) Forest plot comparing proportions of men and women who are virally suppressed of those on ART, including only studies reporting these data for both men and women (N=70 studies). Data for men shown here and data for women with overall comparisons shown in Figure 3b. (b) Forest plot comparing proportions of men and women who are virally suppressed of those on ART, including only studies reporting these data for both men and women (N=70 studies). Data for women and overall comparisons shown here and data for men shown in Figure 3a.

Figure A4

Forest plot comparing proportions of men with awareness of positive HIV status in studies focusing exclusively on MSM (N=10 studies) versus studies that did not focus exclusively on MSM (N=37 studies).

Figure A5

Forest plot comparing proportions of men on ART who are virally suppressed in studies focusing exclusively on MSM (N=5 studies) versus studies that did not focus exclusively on MSM (N=69 studies).

Figure A6

Forest plot comparing proportions of men with awareness of positive HIV status in studies in which knowledge of positive HIV status was measured by self‐report prior to testing (N=33 studies) versus studies in which knowledge was measured by other methods, such as awareness after testing in the study (N=7 studies).

Figure A7

Forest plot comparing proportions of men with known HIV on ART in studies in which ART status was ascertained in a healthcare facility (N=16 studies) versus studies in which ART status was ascertained in a community setting (N=27 studies).

Figure A8

Forest plot comparing proportions of men with awareness of positive HIV status in studies conducted in eastern and southern Africa (N=44 studies) versus studies conducted in western and central Africa (N=3 studies).

Figure A9

Forest plot comparing proportions of men with known HIV on ART in studies conducted in eastern and southern Africa (N=36 studies) versus studies conducted in western and central Africa (N=6 studies).

Figure A10

Forest plot comparing proportions of men on ART who are virally suppressed in studies conducted in eastern and southern Africa (N=64 studies) versus studies conducted in western and central Africa (N=8 studies).

Figure A11

Forest plot comparing proportions of men with awareness of positive HIV status in studies in which all quality criteria are met (N=24 studies) versus studies in which quality criteria are partially met (N=23 studies).

Figure A12

Forest plot comparing proportions of men with HIV on ART in studies in which all quality criteria are met (N=26 studies) versus studies in which quality criteria are partially met (N=17 studies).

Figure A13

Forest plot comparing proportions of men on ART who are virally suppressed in studies in which all quality criteria are met (N=43 studies) versus studies in which quality criteria are partially met (N=31 studies).

Figure A14

Forest plot comparing proportions of men with awareness of positive HIV status in studies in which the median year of data collection was between 2009 and 2016 (N=39 studies) versus in studies in which the median year of data collection was between 2017 and 2019 (N=8 studies).

Figure A15

Forest plot comparing proportions of men with HIV on ART in studies in which the median year of data collection was between 2009 and 2016 (N=33 studies) versus studies in which the median year of data collection was between 2017 and 2020 (N=10 studies).

Figure A16

Forest plot comparing proportions of men on ART who are virally suppressed in studies in which the median year of data collection was between 2009 and 2016 (N=52 studies) versus studies in which the median year of data collection was between 2017 and 2020 (N=22 studies).