Using Population-Size Estimation and Cross-sectional Survey Methods to Evaluate HIV Service Coverage Among Key Populations in Burkina Faso and Togo

Abstract

Objectives:

The objective of our study was to measure progress toward the UNAIDS 90-90-90 HIV care targets among key populations in urban areas of 2 countries in West Africa: Burkina Faso and Togo.

Methods:

We recruited female sex workers (FSWs) and men who have sex with men (MSM) through respondent-driven sampling. From January to July 2013, 2738 participants were enrolled, tested for HIV, and completed interviewer-administered surveys. We used population-size estimation methods to calculate the number of people who were engaged in the HIV continuum of care.

Results:

HIV prevalence ranged from 0.6% (2 of 329) of MSM in Kara, Togo, to 32.9% (115 of 350) of FSWs in Bobo Dioulasso, Burkina Faso. Of those confirmed to be HIV infected, a range of 0.0% (0 of 2) of MSM in Kara to 55.7% (64 of 115) of FSWs in Bobo Dioulasso were using ART. Based on population estimates, the percentage gap between HIV-infected people who should be using ART (per the 90-90-90 targets) and those who reported using ART ranged from 31.5% among FSWs in Bobo Dioulasso to 100.0% among MSM in Kara.

Conclusions:

HIV service coverage among MSM and FSWs in Burkina Faso and Togo was low in 2013. Interventions for improving engagement of these at-risk populations in the HIV continuum of care should include frequent, routine HIV testing and linkage to evidence-based HIV treatment services. Population-size estimates can be used to inform governments, policy makers, and funding agencies about where elements of HIV service coverage are most needed.

Keywords

HIV epidemiology key populations

Men who have sex with men (MSM) and female sex workers (FSWs) are key at-risk populations in the human immunodeficiency virus (HIV) epidemic in low- and middle-income settings.^1,2 Because FSWs and MSM are more likely to be living with HIV infection than other sex- and age-matched adults, the need for HIV prevention and treatment programs in these populations is great.^3,4 However, important barriers to adequate HIV-related care remain for these populations: individual (behavioral, biological), social (network, community), and structural (public policy).^5
–7 Consequently, HIV prevention, diagnosis, and treatment service coverage are limited for both groups in many locations.⁸

An effective HIV continuum of care—from testing for and diagnosing HIV infection to achieving viral suppression among those living with HIV infection—requires frequent and consistent engagement with health services by people who are at risk of HIV acquisition or living with HIV infection.⁹ Connection to and retention in this HIV continuum of care have been shown to improve health outcomes for people living with HIV infection and decrease further transmission of HIV infection to others.^9
–11 However, measuring engagement in the continuum of care among key populations is difficult because of the hidden nature of these groups, limited HIV surveillance methods, incomplete health records, and challenges in measuring HIV viral load suppression.¹² Yet, HIV service coverage estimates can be calculated, optimally by combining multiple population-size estimation methods, and this approach can form the basis for determining whether or not HIV-related service targets for FSWs and MSM are reached.^12,13 UNAIDS and others have contributed to this process by proposing targets for engagement in the HIV continuum of care.^9,14 The UNAIDS 90-90-90 guidelines call for the following by year 2020: 90% of people living with HIV to be diagnosed, 90% of diagnosed people to be using antiretroviral therapy (ART), and 90% of people using ART to have fully suppressed viral loads.¹⁴

The objective of our study was to measure progress toward the UNAIDS 90-90-90 HIV care goals among FSWs and MSM in an HIV epidemic in urban areas of Burkina Faso and Togo. We chose these locations because (1) the HIV epidemic in West Africa is more concentrated in key populations than the HIV epidemics in East Africa and southern Africa; (2) study of these populations in these locations has been limited; and (3) FSWs and MSM in these countries have experienced stigma, discrimination, and policy and legal barriers that limit their access to HIV services.¹⁵ Legal barriers include laws against solicitation of sex work in both countries and laws against same-sex practices in Togo. Our goals were as follows: (1) to measure engagement in the HIV continuum of care (ie, previous HIV testing, previous HIV diagnosis, and ART use) among a cohort of FSWs and MSM in these 2 countries and (2) with use of multiple population-size estimation methods, to contrast our estimates for the broader populations of FSWs and MSM with the UNAIDS 90-90-90 HIV targets.

Methods

Study Population

We conducted a cross-sectional survey based on respondent-driven sampling from January to July 2013 in major cities in Togo (Lomé and Kara) and Burkina Faso (Ouagadougou and Bobo Dioulasso). Participants who were assigned male sex at birth, were at least 18 years of age, and reported anal sex (insertive or receptive) with another man in the 12 months preceding the survey were included in the study and classified as MSM. Participants who were assigned female sex at birth, were at least 18 years of age, and reported having most of their income in the 12 months preceding the survey derived from sex work were included in the study and classified as FSWs.¹³ The Johns Hopkins Bloomberg School of Public Health Institutional Review Board and review boards in each country—the Ethics Committee for Health Research of Burkina Faso and the National Ethics Committee of Togo—granted approval for this study.

Sampling Method

We recruited participants using a separate, simultaneous respondent-driven sampling process at each site.¹⁶ We initiated recruitment with 10 purposely selected, demographically diverse MSM and FSW participants in each country and continued recruitment until we reached the target sample size of approximately 350 for each group in each city. Seed participants were given 3 coupons to recruit up to 3 eligible peers from their social network, and they were given $3 in Burkina Faso and $6 in Togo for each eligible peer recruited into the study. The respondent-driven sampling methods used in this study are described elsewhere.¹⁷

Study Procedures

We screened potential participants for eligibility and asked them to provide informed consent. Then, participants answered structured survey questions about HIV testing history, HIV status, and ART use.

We evaluated participants’ HIV status using sequential blood-testing algorithms.¹⁷ In Burkina Faso, the Determine HIV-1/2 rapid test (Alere, Inc, Waltham, Massachusetts) was used for testing, and the ImmunoComb II HIV 1&2 BiSpot kit (Alere, Inc) was used for confirmation. Discordant results were mediated through use of the ImmunoComb II HIV 1&2 CombFirm kit (Alere, Inc). In Togo, the Determine HIV-1/2 rapid test (Alere, Inc) was also used for testing; the First Response HIV 1-2.0 card test (Premier Medical Corporation Ltd, Nani Daman, India) was used for confirmation; and discordant results were mediated with a Western blot test. Viral load testing was not conducted, because of the infeasibility of appropriate sample collection and storage. All participants were offered their HIV test results. HIV testing and counseling were conducted according to Burkinabe and Togolese national guidelines, which included voluntary post-test counseling. All participants were reimbursed for travel ($4 in Burkina Faso and $10 in Togo) and given male condoms and information on HIV.

Population-Size Estimation Methods and Implementation

We combined several population-size estimation methods to estimate the sizes of the MSM and FSW populations in each city.^12,18,19 We used the unique-object-multiplier method, the social-event-multiplier method, and the wisdom-of-the-masses method for both populations in all 4 cities. We used the capture-recapture method only for the population of FSWs, because they operated out of sex work venues where study staff members could reliably visit them. We evaluated the population-size estimation results for assumption validity. A detailed description of size estimation at all sites is provided in the final study report.²⁰

The unique-object-multiplier method was based on the proportion of participants who had received 1 of up to 350 unique objects distributed by local community peer educators or community group representatives during the 2 weeks before the start of the study. Similarly, the social-event-multiplier method used the proportion of participants who attended a social event during that same period. The multiplier formula used for both the unique-object and social-event estimates at each site was p = [1/(m/n ₂)] × n ₁, where p was the population estimate, n ₁ the number of objects distributed or people who attended the social event, n ₂ the total number of study participants, and m the number of study participants who received the unique object or attended the social event. The wisdom-of-the-masses method utilized the response to the question “How many MSM/FSWs would you guess live in your city?” The medians of the responses at each site were used.

The capture-recapture method, reserved only for FSW populations, involved study staff members visiting 5 sex work venues in each city for the initial “capture” and returning 1 to 3 weeks later for the “recapture.” The formula was p = (n ₁ × n ₂)/r, where p was the population estimate, n ₁ the number of FSWs in the first capture, n ₂ the total number of FSWs in the second visit, and r the number of FSWs in the second visit who were encountered in the first capture, known as the “recapture.”

Population-Size Estimation Analysis

We estimated the proportion of each population (MSM or FSW) in each city by (1) the average of the results from the valid population-size estimation methods in each site as the numerator and (2) the total number of people living in each city who matched the study sample (by sex and age) as the denominator. We determined these denominators through the most recent census data and population projections for Togo (2010 census)²¹ and Burkina Faso (2006 census).²² We used multiple population-size estimation methods for each population in each site to provide a more accurate estimate of MSM and FSW populations and to limit overestimation or underestimation from any one method.¹³

Next, we combined the estimates for the proportion of MSM and FSWs in each city and averaged them for each country to obtain an estimated proportion of MSM and FSWs in all urban settings in each country. We then applied the urban proportion results to the population of reproductive age (ie, males or females aged 15 to 49 years) in each city to determine the final estimates for the MSM and FSW population sizes in the cities of Burkina Faso and Togo. We calculated 95% confidence intervals (CIs) according to the standard deviations of the population-size estimation results at each site.

Service Coverage Analysis

We combined the resulting population estimates with data from our survey to calculate the number of FSWs and MSM in each city who were infected with HIV, previously tested for HIV infection, previously diagnosed with HIV infection, and using ART. We estimated the gap in ART implementation with (1) the number of participants who would be eligible for ART but reported not using it at the time of the survey as the numerator and (2) the number of participants who would be eligible for ART under the UNAIDS 90-90-90 guidelines as the denominator.

Statistical Methods

We conducted statistical analyses with Stata Release 12.²³ Results are presented as crude data because the population at each point in the HIV continuum of care was estimated out of people living with HIV infection, which was a subgroup of the total MSM and FSW sample in each study site, and the subgroup analyses broke the respondent-driven sampling recruitment chains, making such adjustment infeasible.

Results

A total of 2738 eligible people participated in our study. In Burkina Faso, 343 MSM and 349 FSWs consented and participated in Ouagadougou, and 329 MSM and 350 FSWs consented and participated in Bobo Dioulasso. In Togo, 354 MSM and 354 FSWs consented and participated in Lomé, and 329 MSM and 330 FSWs consented and participated in Kara (Table 1).

Table 1.

Study demographics and population estimates for FSWs and MSM, selected cities in Burkina Faso and Togo, January to July 2013

Population	Location	Study Participants, n	Estimated FSW/MSM Proportion,^a % (95% CI)	Female/Male Population^b Aged 15-49 y, n	FSW/MSM Population-Size Estimate, No. (95% CI)
Burkina Faso
FSWs	Ouagadougou	349	1.17 (0.67-1.67)	426 321	4988 (2856-7120)
	Bobo Dioulasso	350		142 579	1680 (962-2398)
MSM	Ouagadougou	343	1.00 (0.88-1.12)	394 383	3944 (3417-4417)
	Bobo Dioulasso	329		129 189	1292 (1137-1447)
Togo
FSWs	Lomé	354	0.82 (0.57-1.07)	358 967	2944 (2046-3841)
	Kara	330		25 102	206 (143-269)
MSM	Lomé	354	1.65 (0.44-2.86)	333 245	5499 (1466-9531)
	Kara	329		26 170	432 (115-748)

Abbreviations: CI, confidence interval; FSW, female sex worker; MSM, men who have sex with men.

^aBased on population-size estimates through the unique-object-multiplier method, social-event-multiplier method, wisdom-of-the-masses method, and (for FSWs only) the capture-recapture method. The unique-object-multiplier method used the proportion of participants who had received 1 of up to 350 unique objects distributed by local community peer educators or community group representatives during the 2 weeks before the start of the study. The social-event-multiplier method used the proportion of participants who attended a social event during that same period. The multiplier formula used for both the unique-object and social-event estimates at each site was p = [1/(m/n ₂)] × n ₁, where p was the population estimate, n ₁ was the number of objects distributed or people invited to the social event, n ₂ was the total number of study participants, and m was the number of study participants who received the unique object or attended the social event. The wisdom-of-the-masses method used the response to the question “How many MSM/FSWs would you guess live in your city?” The medians of the responses at each site were used. The capture-recapture method, used only for FSW populations, involved study staff members visiting 5 sex work venues in each city for the initial “capture” and returning 1 to 3 weeks later for the “recapture.” The formula used was p = (n ₁ × n ₂)/r, where p was the population estimate, n ₁ was the number of FSWs in the first capture, n ₂ was the number of FSWs in the recapture, and r was the total number of FSWs in both captures. Data source for review of population-size estimation methods for key populations: Abdul-Quader et al.¹²

^bEstimate based on population projections from the 2006 Burkina Faso census: L’Institut National de la Statistique et de la Démographie and Ministère de l’Economie et des Finances²² and 2010 Togo census: Direction Générale de la Statistique et de la Comptabilité Nationale.²¹

HIV Seroprevalence

In Burkina Faso, 16 of 339 MSM (4.7%) sampled in Ouagadougou and 16 of 329 MSM (4.9%) sampled in Bobo Dioulasso were infected with HIV. Test results for 4 participants in Ouagadougou were missing. Of 349 FSWs sampled in Ouagadougou, 31 (8.9%) were infected with HIV, as were 115 of 350 FSWs (32.9%) sampled in Bobo Dioulasso (Table 2).

In Togo, 65 of 354 MSM (18.5%) sampled in Lomé and 2 of 329 MSM (0.6%) sampled in Kara were infected with HIV. Test results for 2 participants were missing in Lomé. Of 354 sampled FSWs in Lomé, 96 (27.1%) were infected with HIV, as were 33 of 330 sampled FSWs (10.0%) in Kara (Table 2).

Table 2.

HIV prevalence and engagement in the HIV continuum of care, by HIV-infected FSWs and MSM, selected cities in Burkina Faso and Togo, January to July 2013

		Participants, No. (%)
			HIV Infected
Population	Location	Total	Total	Previous HIV Testing^a	Previous HIV Diagnosis^a	Using ART^a
Burkina Faso
FSWs	Ouagadougou	349 (100.0)	31 (8.9)	25 (80.6)	11 (35.5)	7 (22.6)
	Bobo Dioulasso	350 (100.0)	115 (32.9)	105 (91.3)	67 (58.3)	64 (55.7)
MSM	Ouagadougou	343 (100.0)	16 (4.7)^b	12 (75.0)	5 (31.3)	4 (25.0)
	Bobo Dioulasso	329 (100.0)	16 (4.9)	15 (93.8)	5 (31.3)	1 (6.3)
Togo
FSWs	Lomé	354 (100.0)	96 (27.1)	68 (71.6)^c	19 (20.0)^c	14 (14.7)^c
	Kara	330 (100.0)	33 (10.0)	28 (84.8)	10 (30.3)	6 (18.2)
MSM	Lomé	354 (100.0)	65 (18.5)^d	56 (86.2)	10 (15.4)	4 (6.2)
	Kara	329 (100.0)	2 (0.6)	1 (50.0)	0 (0.0)	0 (0.0)

Abbreviations: ART, antiretroviral therapy; FSW, female sex worker; HIV, human immunodeficiency virus; MSM, men who have sex with men.

^aPercentage based on total people living with HIV, by row.

^bHIV testing results missing for 4 participants.

^cPercentage is calculated out of 95 because of missing data for 1 participant.

^dHIV testing results missing for 2 participants.

Population-Size Estimates

We estimated that in Burkina Faso, 1.2% of the female population were FSWs and 1.0% of the male population were MSM, whereas in Togo, 0.8% of the female population were FSWs and 1.7% of the male population were MSM. The estimated population of FSWs was largest in Ouagadougou (n = 4988) and smallest in Kara (n = 206). The estimated number of FSWs in Kara was lower than the study sample size (n = 330) of FSWs in Kara because some study participants were recruited from outside the city. The estimated population of MSM was largest in Lomé (n = 5499) and smallest in Kara (n = 432; Table 1).

Service Coverage Estimates

The level of engagement in the HIV continuum of care for those who were HIV infected varied by site. We found that a range of 0.6% (2 of 329) of MSM in Kara to 32.9% (115 of 350) of FSWs in Bobo Dioulasso were HIV infected and that previous HIV testing had been conducted among a range of 1 of 2 HIV-infected MSM in Kara to 15 of 16 HIV-infected MSM in Bobo Dioulasso. None of the 2 HIV-infected MSM in Kara reported using ART. In contrast, 55.7% (64 of 115) HIV-infected FSWs in Bobo Dioulasso reported using ART. Reported ART use was >25.0% only among FSWs in Bobo Dioulasso. We also found that none of 2 HIV-infected MSM in Kara had been previously diagnosed with HIV infection. However, 58.3% (67 of 115) of HIV-infected FSWs in Bobo Dioulasso had been diagnosed with HIV infection (Table 2).

We estimated that the number of HIV-infected FSWs was highest in Lomé (n = 798) and lowest in Kara (n = 21). Similarly, the estimated number of HIV-infected MSM was highest in Lomé (n = 1016) and lowest in Kara (n = 3). For both countries, estimates for the number of FSWs and MSM reporting previous HIV testing ranged from 1 to 876; reporting previous HIV diagnosis, from 0 to 322; and using ART, from 0 to 307 (Tables 3 and 4).

Table 3.

Population-size estimates and engagement in the HIV continuum of care for FSWs and MSM, aged 15 to 49 years, Burkina Faso, January to July 2013^a

Population: City	Population-Size Estimate	HIV Infected^b	Previous HIV Testing^c	Previous HIV Diagnosis	Target for ART^d	Using ART^e	Gap in ART Coverage, %
FSWs
Ouagadougou	4988 (2856-7120)	443 (254-632)	357 (205-510)	157 (90-224)	359 (205-512)	100 (57-143)	72.1
Bobo Dioulasso	1680 (962-2398)	553 (316-789)	505 (289-720)	322 (184-459)	448 (256-639)	307 (176-438)	31.5
MSM
Ouagadougou	3944 (3417-4417)	186 (161-208)	140 (121-156)	57 (50-64)	151 (131-169)	46 (40-52)	69.5
Bobo Dioulasso	1292 (1137-1447)	63 (55-70)	59 (52-66)	20 (17-22)	51 (45-57)	4 (3-4)	92.2

Abbreviations: ART, antiretroviral therapy; FSW, female sex worker; HIV, human immunodeficiency virus; MSM, men who have sex with men.

^aValues are presented as No. (95% confidence interval), unless noted otherwise.

^bEstimated with MSM and FSW HIV seroprevalence in the study sample and with results of population-size estimates.

^cEstimated with the percentage of the study sample that reported testing.

^dEstimated as 90% of people living with diagnosed HIV to be on treatment. Adapted from UNAIDS 90-90-90 guidelines.¹⁴

^eEstimated with the percentage of the study sample that reported using ART and with results of population-size estimates.

Table 4.

Population-size estimates and engagement in the HIV continuum of care for FSWs and MSM, aged 15 to 49 years, Togo, January to July 2013^a

Population: City	Population-Size Estimate	HIV Infected^b	Previous HIV Testing^c	Previous HIV Diagnosis	Target for ART^d	Using ART^e	Gap in ART Coverage, %
FSWs
Lomé	2944 (2046-3841)	798 (555-1042)	568 (395-742)	158 (110-206)	647 (449-844)	116 (81-152)	82.1
Kara	206 (143-269)	21 (14-27)	17 (12-23)	6 (4-8)	17 (12-22)	4 (3-5)	76.5
MSM
Lomé	5499 (1466-9531)	1016 (271-1760)	876 (233-1517)	155 (41-269)	823 (219-1426)	62 (17-108)	92.5
Kara	432 (115-748)	3 (1-5)	1 (1-2)	0 (0-0)	2 (1-4)	0 (0-0)	100.0

Abbreviations: ART, antiretroviral therapy; FSW, female sex worker; HIV, human immunodeficiency virus; MSM, men who have sex with men

^aValues are presented as No. (95% confidence interval), unless noted otherwise.

^bEstimated with MSM and FSW HIV seroprevalence in the study sample and with results of population-size estimates.

^cEstimated using the percentage of the study sample that reported testing.

^dEstimated as 90% of people living with diagnosed HIV to be on treatment. Adapted from UNAIDS 90-90-90 guidelines,¹⁴ which call for the following by year 2020: 90% of people living with HIV to be diagnosed, 90% of diagnosed people to be using ART, and 90% of people using ART to have fully suppressed viral loads.

^eEstimated with the percentage of the study sample that reported using ART and with results of population-size estimates.

The estimated gap between HIV-infected people who should be using ART (per the 90-90-90 targets) and those who were using ART ranged from 31.5% among FSWs in Bobo Dioulasso to 100.0% among MSM in Kara (Tables 3 and 4). We also estimated that the largest number of people eligible for ART (per the 90-90-90 targets) and not currently using it was 761 (92.5%) among MSM in Lomé (Figure).

Figure.

HIV service targets and levels of engagement in the HIV continuum of care among MSM and FSWs living with HIV, January to July 2013: (a) Burkina Faso and (b) Togo. HIV service targets and levels of engagement were calculated through use of population estimates. Diagnosed HIV and treatment targets were based on 2 UNAIDS 90-90-90 targets: 90% of people living with HIV to be diagnosed and 90% (net 81%) of diagnosed people to be using ART.¹⁴ The number of people who reported living with HIV was estimated through the percentage of the study sample that reported being previously diagnosed with HIV. No MSM in Kara reported living with HIV. Error bars indicate 95% confidence intervals. Abbreviations: ART, antiretroviral therapy; FSW, female sex workers; HIV, human immunodeficiency virus; MSM, men who have sex with men.

Discussion

This study found limited engagement in the HIV continuum of care among MSM and FSWs in populations with low and high HIV prevalence. The levels of diagnosed HIV and ART treatment use were much lower than the 90-90-90 UNAIDS guidelines. Although this study found different engagement in the HIV continuum of care in different settings, a trend in all sites was that the largest gaps in the HIV continuum of care were found in diagnosing HIV infections.

The results of this study showed that, by site, a range of 41.7% of FSWs in Bobo Dioulasso to 100.0% of MSM in Kara were living with HIV infection but reported no previous diagnosis, far higher than the 10% goal set by UNAIDS.¹⁴ This trend has been reported in other studies evaluating the HIV continuum of care, even in cities where >90% of the at-risk population reported being tested for HIV infection.^24,25 These findings suggest that infrequent random testing alone may not be sufficient to achieve timely diagnosis of HIV infection, and they support recommendations, such as those from the Centers for Disease Control and Prevention, that call for HIV and sexually transmitted infection testing every 3 to 6 months for at-risk groups (eg, MSM) with multiple or anonymous sex partners.²⁶ The low proportion of previous HIV testing in our study indicates the need to scale up the frequency and regularity of testing in Burkina Faso and Togo.

Timely HIV diagnosis is an important step in the HIV continuum of care. The UNAIDS 2014 Gap Report showed that 86% of people living with HIV infection in sub-Saharan Africa who knew their HIV status were receiving ART and about 76% of these same people had achieved viral suppression.⁸ The report also found that with a different denominator (ie, all diagnosed and undiagnosed people living with HIV infection), the proportion using ART was much lower (39%). Furthermore, according to the report, the proportion of people living with HIV infection who reported receiving ART varied greatly by country. Only 43% of adults living with HIV infection in Burkina Faso and 34% of adults living with HIV infection in Togo were on ART.⁸ Our study confirmed these low percentages of ART use among HIV-infected participants.

Strategies to facilitate frequent regular testing of at-risk populations are still needed to close the gap in HIV diagnosis. Decentralized HIV testing, counseling, care, and treatment may reduce barriers to accessing health services for these at-risk populations.²⁷ Self-testing may also allow at-risk populations to obtain confidential nondiscriminatory testing.²⁸

Although improved access to HIV testing is an important entry point into the HIV continuum of care, strategies focused on improving access to and retention in HIV treatment programs are crucial to improving the health of those living with HIV infection and decreasing HIV transmission. The clinical benefits of starting HIV treatment early were confirmed in the Strategic Timing of Antiretroviral Treatment study, which found a 53% reduction in the risk of AIDS, serious illness, or death in the early treatment group compared with the delayed treatment group.²⁹ Moreover, extended results from the HIV Prevention Trials Network study HPTN052 reinforced the importance of universal access to treatment to prevent HIV transmission: the results showed a 93% decrease in HIV infections after >10 000 person-years of follow-up.³⁰ The move to universal coverage of ART further facilitates reaching 90-90-90 HIV treatment service targets because ART initiation is not limited by CD4 levels and people can start treatment on the same day as diagnosis.^31
–33

However, these linkage-to-treatment strategies that are focused on ART initiation among people who are living with HIV infection should be matched by effective retention strategies. The challenges of achieving sustained engagement in treatment and viral suppression^11,24,34 were affirmed by the results of a large cluster randomized controlled trial conducted in KwaZulu-Natal, South Africa, from 2012 to 2016. This trial found no change in HIV incidence at the community level when universal access to ART was provided and >90% of people living with HIV were aware of their HIV status.³⁵ Engagement in treatment may actually support retention in care. In a recent study on universal access to ART among MSM in Nigeria, about 80% of HIV-infected MSM achieved viral suppression after 6 months. In this study, MSM in programs that provided universal access to ART had better rates of retention in care than those who were not engaged in treatment.³⁶

In our study, we found that the percentages of previous HIV diagnosis and ART use among FSWs in Bobo Dioulasso were higher than among FSWs elsewhere or MSM within the same city. These higher percentages may be the result of active research projects and programs with FSWs in Bobo Dioulasso during the past 2 decades. The Yerelon cohort project, which evaluates combined prevention and treatment services for FSWs in Burkina Faso, has been active since 1998, which likely supported engagement in the HIV continuum of care in this city.³⁷ Ultimately, to improve engagement and retention in the HIV continuum of care, interventions are needed at every service coverage level, including HIV testing, linkage to care, program retention, medication adherence, and viral load testing.³⁸ To assist with engagement and retention in the HIV continuum of care, new technology and mobile methods (eg, smartphone and Internet-based interventions) should be explored to increase the uptake of HIV services among hard-to-reach groups.^39,40

As consideration is given to various strategies to improve entry into and continued engagement in the HIV continuum of care, governments and policy makers also need data that help inform their efforts to scale up resources for HIV services. Estimates of at-risk populations, the degree to which they are involved in the HIV continuum of care, and HIV service coverage needs are critical to this process. We demonstrated how combining population-size estimation methods to estimate the size of at-risk populations could help demonstrate geographic variability and identify gaps in HIV testing, diagnosis, and ART use. Through such estimates, it may be possible to identify geographic areas with lower involvement in the HIV continuum of care and higher HIV service coverage needs so that focused interventions can be integrated into national strategic planning or policy making. Context-specific health service coverage data combined with the UNAIDS 90-90-90 targets can help identify gaps in access to and use of HIV services in the HIV continuum of care among key populations.

Limitations

This study had several limitations. First, we did not include viral load testing results, so we could not determine the proportion of those using ART who achieved viral suppression nor compare our findings with the third component of the UNAIDS 90-90-90 targets. Second, no effective system of unique identifier codes was in place at the time of our study. Without unique identifier codes and viral load data, we could not confirm that those diagnosed with HIV infection who reported being unaware of their infection were truly new diagnoses. This lack of viral load data is noteworthy because in another setting, a substantial proportion of people who reported being unaware of their HIV infection had evidence of previous antiretroviral blood level testing and viral suppression.⁴¹ Third, the sample size of people living with HIV infection was small in some of our sites, notably MSM in Kara. However, our population-size estimates were within the expected ranges from the region: previous studies estimated 0.5% of the total adult male population in Ghana to be MSM and 3.5% of the adult population in Nigeria to be MSM or FSWs.^42,43

Finally, our population estimates may have been affected by several variables, potentially resulting in overestimation or underestimation. For example, we observed a challenge in engaging older people, particularly MSM (who are more likely than younger MSM to be HIV infected), in the respondent-driven sampling portion of our study. The challenge of recruiting older MSM is consistent with previous findings that older MSM have smaller social networks and are less likely to be recruited into respondent-driven sampling chains than younger MSM.⁴⁴ Other factors that may have affected our population-size estimates included that study participants may have been more likely than nonparticipants to be linked to HIV care, those more likely to receive an object or participate in a social event may also have been more likely to be recruited into the study than people who did not, and some MSM and FSWs were likely to hide their sexual orientations or occupations.

Conclusion

Engagement in the HIV continuum of care among MSM and FSWs in major cities in Togo and Burkina Faso was shown to be below the standard of the UNAIDS 90-90-90 targets, including low proportions of previous HIV testing, previous HIV diagnosis, and ART use. Priority interventions should include routine and frequent HIV testing and linkage into evidence-based and human rights–affirming HIV treatment services. Population-size estimates can be used to inform governments, policy makers, and funding agencies about elements of HIV service coverage that are most needed.

Footnotes

Acknowledgments

We thank the ministries of health of Burkina Faso and Togo for approving and collaborating on this study. We also acknowledge the team at USAID and USAID West Africa for their continued dedication to and important role in the success of this study and similar studies across the region. Finally, we thank the women and men who participated in this study for being engaged and making this analysis possible.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The USAID–Project SEARCH, Task Order No. 2, was funded by the US Agency for International Development under contract GHH-I-00-07-00032-00, beginning September 30, 2008, and supported by the President’s Emergency Plan for AIDS Relief. The Research to Prevention Project was led by the Johns Hopkins Center for Global Health and managed by the Johns Hopkins Bloomberg School of Public Health Center for Communication Programs. Additional support was provided by the Measurement & Surveillance of HIV Epidemics Consortium, funded by the Bill & Melinda Gates Foundation. Stefan Baral’s effort was funded in part by support from the Johns Hopkins University Center for AIDS Research, a National Institutes of Health (NIH)–funded program (P30AI094189), which is supported by the following NIH cofunding and participating institutes and centers: Fogarty International Center, National Cancer Institute, National Heart, Lung, and Blood Institute, National Institute of Allergy and Infectious Diseases, National Institute of Child Health and Human Development, National Institute of Diabetes and Digestive and Kidney Diseases, National Institute of General Medical Sciences, National Institute of Mental Health, National Institute on Aging, National Institute on Drug Abuse, and Office of AIDS Research. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

References

Baral

Beyrer

Muessig

. Burden of HIV among female sex workers in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Infect Dis. 2012;12(7):538–549.

Baral

Sifakis

Cleghorn

Beyrer

. Elevated risk for HIV infection among men who have sex with men in low- and middle-income countries 2000-2006: a systematic review. PLoS Med. 2007;4(12):e339.

Beyrer

Baral

Kerrigan

El-Bassel

Bekker

Celentano

. Expanding the space: inclusion of most-at-risk populations in HIV prevention, treatment, and care services. J Acquir Immune Defic Syndr. 2011;57(suppl 2):S96–S99.

Chersich

Luchters

Ntaganira

. Priority interventions to reduce HIV transmission in sex work settings in sub-Saharan Africa and delivery of these services. J Int AIDS Soc. 2012;16:17980.

Baral

Holland

Shannon

. Enhancing benefits or increasing harms: community responses for HIV among men who have sex with men, transgender women, female sex workers, and people who inject drugs. J Acquir Immune Defic Syndr. 2014;66(suppl 3):S319–S328.

Beyrer

. Global prevention of HIV infection for neglected populations: men who have sex with men. Clin Infect Dis. 2010;50(suppl 3):S108–S113.

Baral

Logie

Grosso

Wirtz

Beyrer

. Modified social ecological model: a tool to guide the assessment of the risks and risk contexts of HIV epidemics. BMC Public Health. 2013;13:482.

Joint United Nations Programme on HIV/AIDS. The Gap Report. Geneva, Switzerland: UNAIDS; 2014.

Gardner

McLees

Steiner

del Rio

Burman

. The spectrum of engagement in HIV care and its relevance to test-and-treat strategies for prevention of HIV infection. Clin Infect Dis. 2011;52(6):793–800.

10.

Tanser

Bärnighausen

Grapsa

Zaidi

Newell

. High coverage of ART associated with decline in risk of HIV acquisition in rural KwaZulu-Natal, South Africa. Science. 2013;339(6122):966–971.

11.

Hull

Montaner

. Optimizing the engagement of care cascade: a critical step to maximize the impact of HIV treatment as prevention. Curr Opin HIV AIDS. 2012;7(6):579–586.

12.

Abdul-Quader

Baughman

Hladik

. Estimating the size of key populations: current status and future possibilities. Curr Opin HIV AIDS. 2014;9(2):107–114.

13.

Holland

Papworth

Billong

. Antiretroviral treatment coverage for men who have sex with men and female sex workers living with HIV in Cameroon. J Acquir Immune Defic Syndr. 2015;68(suppl 2):S232–S240.

14.

Joint United Nations Programme on HIV/AIDS. Ambitious Treatment Targets: Writing the Final Chapter of the AIDS Epidemic. Geneva, Switzerland: UNAIDS; 2014.

15.

Duvall

Irani

Compaoré

. Assessment of policy and access to HIV prevention, care, and treatment services for men who have sex with men and for sex workers in Burkina Faso and Togo. J Acquir Immune Defic Syndr. 2015;68(suppl 2):S189–S197.

16.

Schonlau

Liebau

. Respondent-driven sampling. Stata J. 2012;12(1):72–93.

17.

Wirtz

Schwartz

Ketende

. Sexual violence, condom negotiation, and condom use in the context of sex work: results from two West African countries. J Acquir Immune Defic Syndr. 2015;68(suppl 2):S171–S179.

18.

Johnston

Prybylski

Raymond

Mirzazadeh

Manopaiboon

McFarland

. Incorporating the service multiplier method in respondent-driven sampling surveys to estimate the size of hidden and hard-to-reach populations: case studies from around the world. Sex Transm Dis. 2013;40(4):304–310.

19.

Okal

Geibel

Muraguri

. Estimates of the size of key populations at risk for HIV infection: men who have sex with men, female sex workers and injecting drug users in Nairobi, Kenya. Sex Transm Infect. 2013;89(5):71.

20.

Papworth

Grosso

Ketende

. Examining risk factors for HIV and access to services among female sex workers (FSW) and men who have sex with men (MSM) in Burkina Faso, Togo and Cameroon. http://www.jhsph.edu/research/centers-and-institutes/research-to-prevention/publications/WA_FinalReport.pdf. Published 2014. Accessed June 8, 2015.

21.

Direction Générale de la Statistique et de la Comptabilité Nationale. Togo: recensement général de la population et de l’habitat (2009-2011), quatrième recensement générale au Togo. http://www.stat-togo.org/nada/index.php/catalog/2. Published 2011. Accessed August 8, 2016.

22.

L’Institut National de la Statistique et de la Démographie and Ministère de l’Economie et des Finances. Recensement general de la population et de l’habitation de 2006: resultats definitifs. http://www.insd.bf/documents/publications/insd/publications/resultats_enquetes/autres%20enq/Resultats_definitifs_RGPH_2006.pdf. Published 2008. Accessed August 8, 2016.

23.

StataCorp. Stata: Release 12. College Station, TX: StataCorp; 2011.

24.

Kranzer

Govindasamy

Ford

Johnston

Lawn

. Quantifying and addressing losses along the continuum of care for people living with HIV infection in sub-Saharan Africa: a systematic review. J Int AIDS Soc. 2012;15(2):17383.

25.

Lourenco

Colley

Nosyk

Shopin

Montaner

Lima

. High levels of heterogeneity in the HIV cascade of care across different population subgroups in British Columbia, Canada. PLoS One. 2014;9(12):e11527 7.

26.

Workowski

Berman

. Sexually transmitted diseases treatment guidelines, 2010. MMWR Recomm Rep. 2010;59(RR-12):1–110.

27.

Fujita

Poudel

Thi

. A new analytical framework of “continuum of prevention and care” to maximize HIV case detection and retention in care in Vietnam. BMC Health Serv Res. 2012;12:483.

28.

Krause

Subklew-Sehume

Kenyon

Colebunders

. Acceptability of HIV self-testing: a systematic literature review. BMC Public Health. 2013;13:735.

29.

US National Institutes of Health. Starting antiretroviral treatment early improves outcomes for HIV-infected individuals. http://www.nih.gov/news/health/may2015/niaid-27.htm. Published May 27, 2015. Accessed June 8, 2015.

30.

Cohen

Chen

McCauley

. Antiretroviral therapy for the prevention of HIV-1 transmission. N Engl J Med. 2016;375(9):830–839.

31.

Layer

Kennedy

Beckham

. Multi-level factors affecting entry into and engagement in the HIV continuum of care in Iringa, Tanzania. PLoS One. 2014;9(8):e10496 1.

32.

Parchure

Kulkarni

Gangakhedkar

. Pattern of linkage and retention in HIV care continuum among patients attending referral HIV care clinic in private sector in India. AIDS Care. 2015;27(6):716–722.

33.

Orne-Gliemann

Larmarange

Boyer

. Addressing social issues in a universal HIV test and treat intervention trial (ANRS 12249 TasP) in South Africa: methods for appraisal. BMC Public Health. 2015;15:209.

34.

Nachega

Uthman

del Rio

Mugavero

Rees

Mills

. Addressing the Achilles’ heel in the HIV care continuum for the success of a test-and-treat strategy to achieve an AIDS-free generation. Clin Infect Dis. 2014;59(suppl 1):S21–S27.

35.

Dabis

. The impact of Universal Test and Treat on HIV incidence in a rural South African population: ANRS 12249 TasP trial, 2012-2016. Presented at: AIDS 2016; July 18-22, 2016; Durban, South Africa.

36.

Charurat

Emmanuel

Akolo

. Uptake of treatment as prevention for HIV and continuum of care among HIV-positive men who have sex with men in Nigeria. J Acquir Immune Defic Syndr. 2015;68(suppl 2):S114–S123.

37.

Konate

Traore

Ouedraogo

. Linking HIV prevention and care for community interventions among high-risk women in Burkina Faso—the ARNS 1222 “Yerelon” cohort. J Acquir Immune Defic Syndr. 2011;57(suppl 1):S50–S54.

38.

Birger

Hallett

Sinha

Grenfell

Hodder

. Modeling the impact of interventions along the HIV continuum of care in Newark, New Jersey. Clin Infect Dis. 2014;58(2):274–284.

39.

Elul

Lahuerta

Abacassamo

. A combination strategy for enhancing linkage to and retention in HIV care among adults newly diagnosed with HIV in Mozambique: study protocol for a site-randomized implementation science study. BMC Infect Dis. 2014;14:549.

40.

Muessig

Nekkanti

Bauermeister

Bull

Hightow-Weidman

. A systematic review of recent smartphone, Internet and Web 2.0 interventions to address the HIV continuum of care. Curr HIV/AIDS Rep. 2015;12(1):173–190.

41.

Sanchez

Kelley

Rosenberg

. Lack of awareness of human immunodeficiency virus (HIV) infection: problems and solutions with self-reported HIV serostatus of men who have sex with men. Open Forum Infect Dis. 2014;1(2):ofu084.

42.

Quaye

Fisher Raymond

Atuahene

. Critique and lessons learned from using multiple methods to estimate population size of men who have sex with men in Ghana. AIDS Behav. 2015;19(suppl 1):S16–S23.

43.

Makofane

Gueboguo

Lyons

Sandfort

. Men who have sex with men inadequately addressed in African AIDS national strategic plans. Glob Public Health. 2013;8(2):129–143.

44.

Ministère de la Planification, du Développement et de l’Aménagement du Territoire, Ministère de la Santé. [Demographic and health survey in Togo, 2013-2014]. https://dhsprogram.com/pubs/pdf/FR301/FR301.pdf. Published 2015. Accessed June 8, 2015.