HIV testing, care and viral suppression among men who have sex with men and transgender individuals in Johannesburg, South Africa.

INTRODUCTION: Men who have sex with men and transgender individuals (MSM/TG) carry a disproportionately high burden of HIV, including in South Africa. However, there are few empirical population-representative estimates of viral suppression and the HIV care cascade including HIV testing among this population, nor of factors associated with these outcomes.
METHODS: We conducted a respondent driven sampling (RDS) survey among 301 MSM/TG in Johannesburg in 2017. Participants gave blood samples for HIV testing and viral load. Participants self-completed a survey including sociodemographics, HIV testing history, and engagement in care. We calculated RDS-II weighted estimates of the percentage of HIV-negative MSM/TG reporting HIV testing in the previous 6 months, their testing experience and preferences. Among those HIV-positive, we estimated the percentage status-aware, on ART, and virally suppressed (<50 viral copies/ml plasma). We conducted RDS-weighted robust Poisson regression to obtain weighted prevalence ratios of factors associated with 1) HIV testing among those HIV-negative; and 2) viral suppression among those HIV-positive.
RESULTS: There were 118/300 HIV-positive MSM/TG, (37.5%). Of the HIV-negative MSM/TG, 61.5% reported that they had tested for HIV in the previous 6 months, which was associated with selling sex to men (Prevalence Ratio = 1.67, 95% CI 1.36-2.05). There were 76/118 HIV-positive MSM/TG (56.5%) who reported having previously tested positive for HIV and 39/118 (30.0%) who reported current ART. There were 58/118 HIV-positive MSM/TG with viral loads <50 copies/ml plasma (46.9%). Viral suppression was associated with older age (adjusted PR = 1.03, 95% CI 1.00-1.06 for each year), neighbourhood, and having bought sex from men (adjusted PR = 1.53, 95% CI 1.12-2.08).
CONCLUSIONS: HIV prevalence was very high. Viral suppression among those HIV-positive was similar to the general male population in South Africa, but remains far short of national and international targets. A majority of HIV-negative MSM/TG had HIV tested in the previous 6 months, though there is room for improvement.

Introduction

HIV surveillance and programming targeting gay and bisexual men, other men who have sex with men and transgender women (MSM/TG) have until recently been neglected in countries with generalised HIV epidemics[1]. Over the last ten years, this has begun to change, with increasing funder and policymaker attention paid to the burden of HIV experienced by these populations[2-4], including in South Africa’s National Strategic Plan for HIV, TB and STI’s[5]. Over this same period, coverage of antiretroviral treatment (ART) in sub-Saharan Africa has greatly expanded and AIDS-related mortality in general populations has fallen dramatically[6, 7]. Effective treatment to achieve viral suppression also avoids sexual transmission of the virus, including among MSM[8], and at the population level, ‘Treatment as Prevention’ (TasP) has the potential to reduce HIV incidence[9, 10]. Accordingly, UNAIDS set ‘care cascade’ targets specifying that of all HIV-positive individuals in a given population, 90% should be aware of their status, 90% of those aware should be on antiretroviral treatment (ART) and 90% of those on ART should be virally suppressed by 2020[11].

Targets have not yet been reached in South Africa: among all males aged 15 years and above in Gauteng Province, 85% of those HIV-positive were estimated to be aware of their status, 43% on ART and 34% virally suppressed in 2017[12]. A modelling study for Johannesburg estimated that among people aged 15 and older living with HIV in 2016, 73% were aware of their status, 45% on ART and 26% virally suppressed[13], indicating large gaps between diagnosis and access to treatment. It is possible, however, that the cascade for MSM and transgender populations differs substantially, but to date little empirical population-representative evidence from sub-Saharan Africa has been published to offer insight[14-18].

Experiences of violence, harassment, and stigma routinely inhibit MSM/TG’s access to healthcare[19, 20]. While same-gender sex is legally protected in South Africa, MSM/TG nevertheless face societal discrimination[21]. There is concern that stigma and discrimination toward MSM/TG may pose barriers to prompt diagnosis, linkage to and retention in care, and adherence to ART, leading to worse HIV care outcomes. Stigma might act differentially across the cascade; there could be more pressure to disclose sexual orientation at the point of HIV testing than at the point of treatment initiation or follow-up. A study of MSM and other men attending two ‘Health4Men’ MSM-friendly clinics in Johannesburg found no evidence for a difference in ART retention after 24 months among men reporting same sex behaviour and those not[22], but it is unknown whether the same would be found for clinics not specifically tailoring for the needs of MSM/TG.

Prompt initiation of HIV care necessarily depends upon uptake and frequency of HIV testing, which itself may be inhibited due to stigma. MSM/TG who initially test HIV-negative need to test repeatedly during periods at risk. Men in South Africa test less frequently than women[23], while MSM could be further inhibited due to aforementioned healthcare-related stigma. A qualitative study among urban MSM found that HIV testing was often viewed as a one-off requirement[24]. While there is historical evidence of low testing rates among South African MSM[25], testing appears to have increased between 2008 and 2013, concordant with increasing provision of MSM programming[26] and increases in HIV testing amongst men as a whole during this period[27]. There is a need now for up-to-date estimates and preferences for HIV testing among MSM/TG.

To inform HIV prevention and care programming, the TRANSFORM study set out to provide population-representative empirical estimates and correlates of the HIV care cascade and viral suppression among HIV-positive MSM/TG, and recency of and preferences for HIV testing among HIV-negative MSM/TG in Johannesburg, and to explore characteristics associated with these outcomes and behaviours.

Methods

Population

Eligibility criteria included being aged 18 years or older; male gender assigned at birth or current male identity; sex with a man in the past 12 months; and resident in Johannesburg.

Sampling and recruitment

To obtain representative estimates, we recruited participants using respondent-driven sampling (RDS) [28], informed by formative research conducted within the TRANSFORM study. Recruitment proceeded in ‘waves’ with two coupons valid for two weeks issued to each participant, except when nearing the target sample size when we reduced to one coupon. Participants were reimbursed ZAR 150 for participation in the study and received ZAR 150 for each recruitee.

Data collection

Recruitment began in April 2017 with four seed participants, with five additional seeds added until September 2017. Following eligibility screening and informed consent, participants privately completed a self-administered survey in English or Zulu using a tablet computer (see S3 Survey instrument of S3 Data), had HIV pre-test counselling, physical examination, blood sample collection, gave swabs for STI testing with a research nurse, and received HIV/STI post-test counselling and STI treatment. Participants were given direct referrals for HIV related care or Pre Exposure Prophylaxis (PrEP) access depending on HIV test outcome, and referrals for syndromic STI treatment onsite or at a local government clinic for partners.

HIV testing and care cascade indicators

Participants gave a blood sample for HIV rapid-testing, ELISA-confirmed if positive (Alere Combo and Advanced Quality HIV tests, back-up Alere Determine), CD4 count and viral load-testing (GeneXpert) if HIV-positive. We considered individuals as virally suppressed if they had <50 viral copies/ml plasma, in line with current South African treatment guidelines[29]. There is also evidence that individuals with viral loads greater than 50 copies/ml plasma but less than 1000 copies/ml show increased likelihood of poor treatment outcomes compared to those with viral load measures less than 50 copies/ml[30]. To aid comparison with past studies and those from other settings, we also show the proportion virally suppressed according to cut-offs of <200 copies/ml, <400 copies/ml and <1000 copies/ml in S2 Viral suppression according to different viral load cut-offs of S2 Data.

Participants were considered aware of their HIV-positive status if they reported in the self-administered survey ever having tested for HIV and receiving a positive result at their most recent HIV test. Participants reported month and year of most recent HIV test. Assuming tests were conducted on the first of each month, together with interview date we created binary indicators for having tested in the previous three, six and twelve months. Individuals reporting that they had previously tested HIV-positive were asked if they had received care, whether they had “ever started taking ART” and whether they were “currently taking ART”.

Other indicators

Participants reported their age in years, previous month’s income, highest educational attainment, employment status, religion, population group (using South African categories), marital status, spouse’s gender, neighbourhood of residence and place of birth, their sexual identity, their sex assigned at birth and their current gender identity. We considered participants ‘cisgender male’ if both sex assigned at birth and current gender were male; ‘transfeminine’ if sex assigned at birth was male and current identity was female or transgender, ‘transmasculine’ if sex assigned at birth was female and current identity was male/transgender and ‘non-binary’ if this was their current gender identity. We used the AUDIT scale[31] to assess alcohol use and asked about frequency of use of a variety of recreational drugs, reporting here on use at least once in the past month, excepting tobacco. We used the PHQ-9 scale to assess depressive symptomatology, with cut-off points at 0, 5, 10, 15 and 20 for none, mild, moderate and moderately severe and severe depression, and asked participants whether they openly talked, tried ‘very hard’ or ‘somewhat hard’ to hide that that they had sex with men with their friends, family and healthcare workers, or neither spoke openly nor tried to hide it. Participants reported aspects of their sexual behaviour in the previous 3 months separately for male and female partners (if applicable), including number of partners, sexual roles (insertive/receptive anal sex or no anal sex, oral sex with men, and vaginal or anal sex with women), frequency of condom use by role (‘always’, ‘most of the time’, ‘some of the time’, ‘rarely’, ‘never’, made binary always/not always), and whether they had given, or had been given money, gifts or favours for sex in the previous 12 months. Participants self-reported experiencing STI symptoms at the time of survey and in the previous 12 months (urethral, genito-urinary or rectal).

For AUDIT and PHQ-9, if participants were missing just one or two responses to the scale of questions, we replaced those missing with the mean of the individual’s other responses (mean of questions 1–9 for PHQ9[32] and questions 2–8 or 9–10 for AUDIT).

RDS recruitment diagnostics

As recommended, recruitment was monitored biweekly and assessed against RDS assumptions[33]. We describe RDS recruitment in detail in S1 Respondent Driven Sampling recruitment diagnostics of S1 Data. Convergence of HIV status, viral suppression and HIV testing in the last 6 months were judged as reasonable when the cumulative RDS-weighted estimate stabilised prior to the full recruitment amongst HIV-positive and negative participants respectively, examined visually.

Statistical analysis

RDS-II weighting[34] was used to account for sampling design, dropping seed participants (who were purposively selected), and weighting by inverse MSM network size, assuming this was proportional to the probability that they would be given a coupon to participate in the study. Confidence intervals were obtained via bootstrapping.

We described the sociodemographic characteristics of MSM/TG by HIV status, as tested in the study. We examined reported HIV testing behaviour among participants testing HIV-negative in the study, and the care cascade amongst those HIV-positive. We described the proportion of HIV-negative MSM/TG who had ever tested, tested in the last three, six and twelve months, the location of and satisfaction with most recent test, and future testing preferences. We examined the associations between these preferences and age, sexual identity and gender identity characteristics using chi-squared tests with a correction for survey data[35]. As the outcome was prevalent, we used RDS-II weighted robust Poisson regression (also dropping seed participants) to obtain weighted prevalence ratios[36, 37] to assess associations between testing in the last 6 months and sociodemographic, sexual behaviour and self-reported STI symptoms, examining first bivariate associations and then adjusting sexual behaviour and STI symptom models for any sociodemographic variables showing evidence for associations with the outcome (Wald test p<0.2).

The HIV care cascade was constructed by using the RDS-II weighted percentage of all MSM/TG testing HIV-positive in the study who knew their status, were on ART and whose viral load was suppressed, and assessing these against the 90-90-90 targets[11]. We described the proportion of HIV-positive MSM/TG who were ‘satisfied’/ ‘very satisfied’ with the privacy and respect they were accorded in their most recent HIV care interaction.

Associations with viral suppression among HIV-positive MSM/TG were also examined using RDS-II weighted robust Poisson regression, as above, first examining sociodemographic characteristics, recreational drug use, depression, disclosure and sexual behaviours adjusted only for age, and then retaining in the adjusted model those characteristics for which the Wald test p value was <0.2.

Analyses were conducted using the RDS: Respondent-Driven Sampling package version 0.9–2[38] and the survey: analysis of complex survey samples package version 3.33–2[39] for R version 3.5.1[40] and Stata version 12[41].

Ethics statement

This study was approved by the ethics committees at the University of the Witwatersrand, and the London School of Hygiene and Tropical Medicine.

Results

Sociodemographic characteristics of HIV-positive and HIV-negative MSM/TG

118/300 (37.5%) MSM/TG were HIV-positive and 182/300 were HIV-negative. One seed participant refused HIV testing. Mean age was 26 years, though HIV-positive MSM/TG were older on average than those HIV-negative, Table 1. The majority were born in Johannesburg, had completed secondary education, were Black African, unemployed, Christian and unmarried. While most MSM/TG identified as cisgender male, a substantial minority were transfeminine (15.7% among those HIV-positive and 11.7% among those HIV-negative). HIV-negative MSM/TG more often identified as bisexual, heterosexual or other compared to HIV-positive MSM/TG (40.5% compared to 11.9%).

Table 1

Sociodemographic characteristics of MSM/TG in Johannesburg by HIV status.

Sociodemographic Characteristics		HIV-positive, n = 118		HIV-negative, n = 182
		n	RDS%	n	RDS%
Age in years	18–21	15	12.8	59	30.6
	20–24	16	13.4	51	30.0
	25–29	36	27.6	36	19.7
	30+	51	46.2	36	19.4
Place of birth	Born in Johannesburg	76	66.9	105	59.0
	Born in South Africa, but not Johannesburg	33	26.6	65	34.6
	Born outside of South Africa	7	6.6	11	6.4
	Missing	2		1
Neighbourhood residence	Soweto	66	58.3	92	53.6
	Braamfontein	9	7.0	18	7.7
	Hillbrow	16	15.6	28	17.1
	Orange Farm	7	5.4	6	3.1
	Other	20	13.7	38	18.6
Religion	Christianity	104	83.8	155	83.7
	Islam	3	2.4	1	1.0
	None	9	13.8	26	15.3
	Other	1	0.0	0	0.0
	Missing	0		1
Education completed	Higher Education	34	31.2	41	18.2
	High School	70	57.4	129	75.6
	Junior High	12	9.7	10	5.7
	Primary	1	1.7	1	0.3
	None	0	0.0	1	0.3
	Missing	1		0
Employment status	Employed full-time	16	13.3	16	6.1
	Employed part-time	13	12.5	29	16.5
	Self employed	14	9.2	19	10.0
	Student	3	2.3	16	7.1
	Unemployed	69	60.1	98	5.9
	Other	2	2.6	4	1.4
	Missing	0		1
Income per month / ZAR	Under 500	33	28.6	49	27.8
	500–999	10	11.1	29	21.6
	1000–1999	22	14.5	35	22.5
	2000–4999	32	27.3	41	22.6
	5000+	16	18.6	14	5.5
	Missing	5		14
Population group	Black African	113	96.4	173	94.1
	Coloured	4	1.9	5	3.4
	White	0	0.0	2	0.5
	Prefer not to say	0	0.0	1	1.0
	Other	1	1.7	1	1.0
Sexual Identity	Gay or homosexual	103	88.1	112	59.6
	Bisexual	12	11.7	59	35.3
	Heterosexual	0	0.0	3	1.3
	Other/Don't know	1	0.2	8	3.9
	Missing	2		0
Gender identity	Cisgender male	95	80.6	138	76.9
	Transfeminine	20	15.7	24	11.7
	Transmasculine	1	0.3	1	0.3
	Non-binary	2	3.4	19	11.2
Marital Status	Not married	103	88.2	163	88.4
	Married to a man or transgender individual	13	11.8	17	10.6
	Married to a woman	0	0.0	2	1.0
	Missing	2		0

n = 1 individual did not consent to HIV testing.

HIV testing behaviours and preferences amongst HIV-negative MSM/TG

24/171 HIV-negative participants reported previous HIV testing but had a missing or partially completed month and year of last HIV test. We did not find statistical evidence for systematic differences between those with and without last HIV test dates (chi-squared tests p>0.1 for all characteristics shown in Table 2 and social network size, with the exception of paying a man in money, gifts or favours in exchange for sex in the last 12 months (‘buying sex’)). Participants reporting buying sex were more likely to have missing dates for last HIV test, 6/17 compared to 18/153, p = 0.008.

Table 2

Associations with testing for HIV in the previous 6 months among MSM/TG testing HIV-negative in the study.

Characteristics of the HIV-negative MSM/TG population		n	RDS%	Crude PR	95% CI		p value
Sociodemographic Characteristics
Age in years	mean not tested / tested	24.5	24.3	1.00	0.97	1.02	0.795
Born in Johannesburg	Born in Johannesburg	55/93	58.9	1.00			0.573
	Born in SA, not Johannesburg	37/53	69.1	1.17	0.87	1.58
	Born outside of SA	8/10	61.6	1.05	0.54	2.01
Religion	None	12/24	61.0	1.00			0.965
	Christianity/Islam/Other	88/133	61.6	1.01	0.68	1.50
Education completed	Junior High School	6/10	65.0	1.06	0.61	1.87	0.973
	High School	70/111	61.0	1.00
	College/University/Higher Education	24/36	61.3	1.02	0.71	1.46
Employment status	Unemployed	59/88	65.3	1.00			0.325
	Employed or student	41/69	55.8	0.85	0.63	1.17
Income in the last month	0–499	26/41	55.8	1.00			0.596
(ZAR)	500–999	13/21	68.8	1.23	0.78	1.94
	1000–1999	19/31	51.4	0.92	0.55	1.56
	2000–4999	27/37	65.9	1.18	0.77	1.82
	5000+	4/14	41.3	0.74	0.32	1.71
Sexual Identity	Bisexual, heterosexual, other	36/65	60.8	1.00			0.846
	Gay or homosexual	64/92	62.0	0.97	0.72	1.31
Gender identity	Cisgender male	81/120	64.4	1.00			0.222
	Transfeminine	9/19	36.9	0.57	0.30	1.09
	Transmasculine	0/1	0.0	-	-	-
	Non-binary	10/17	67.1	1.04	0.68	1.59
Marital Status	Single/divorced/widowed	90/141	60.0	1.00			0.274
	Married to a man or transgender individual	8/12	74.0	1.23	0.85	1.80
	Married to a woman	2/2	100.0	-	-	-
Neighbourhood	Soweto	51/83	63.1	1.00			0.524
	Hillbrow	15/19	74.6	1.18	0.81	1.72
	Braamfontein	14/18	66.9	1.06	0.65	1.73
	Orange Farm	4/5	65.1	1.03	0.45	2.35
	Other*	16/32	44.7	0.71	0.43	1.17
Sexual behaviours
Sex with a man (3months)	No	19/41	47.0	1.00			0.083
	Yes	81/116	67.6	1.44	0.96	2.16
Sex with a woman (3 months)	No	61/94	56.5	1.00			0.239
	Yes	39/63	67.3	1.19	0.89	1.59
2+ male partners (3 months)	No	83/131	63.0	1.00			0.433
	Yes	17/26	51.4	0.82	0.49	1.35
Sell sex to men (12 months)	No	78/129	55.1	1.00			<0.001
	Yes	22/27	92.0	1.67	1.36	2.05
Buy sex from a man (12 months)	No	91/144	61.1	1.00			0.313
	Yes	8/11	76.1	1.25	0.81	1.91
Condomless Anal Intercourse with a man (3 months)	No	61/104	58.8	1.00			0.383
	Yes	39/53	67.1	1.14	0.85	1.54
Sexual role with men during anal sex, (3 months)	No anal sex with men	23/45	53.0	0.68	0.45	1.03	0.262
	Receptive anal intercourse	27/39	56.4	0.73	0.47	1.13
	Insertive anal intercourse	35/53	67.2	0.86	0.61	1.22
	Versatile	15/20	77.8	1.00
Self-reported STI symptoms (12 months)	No	82/123	66.2	1.00			0.094
	Yes	18/34	44.5	0.67	0.42	1.07

n = 157 HIV-negative with responses to ever tested and date of last test.

n's do not add to total where there are some missing responses.

All models exclude seed participants and include weighting for inverse network size (RDS-II).

There was not evidence for associations (Wald test p<0.2) between sociodemographic characteristics and testing, so we show only univariate associations.

*Other neighbourhoods include those within Johannesburg with fewer than 10 participants in the study sample.

Few HIV-negative MSM/TG reported never previously HIV testing (10/181, 5.5%), while 100/157, 61.5%, reported having tested in the previous 6 months and 118/157, 73.0% tested in the last 12 months, Table 3.

Table 3

HIV testing history and experiences amongst MSM/TG testing HIV-negative in the study.

Among all testing HIV-negative, n = 182
Testing history	n/N	RDS %
Ever had an HIV test	171/181	94.5
Had an HIV test in the previous 12 months	118/157*	73.0
Had an HIV test in the past 6 months	100/157*	61.5
Had an HIV test in the past 3 months	77/157*	49.8
			'Satisfied' or 'Very satisfied' with privacy		'Satisfied' or 'Very satisfied' with respect shown
Testing preferences	n/N	RDS %	n	RDS %	n	RDS %
Where last tested for HIV, of those ever testing**
Public hospital or clinic	69/167	47.1	68	97.7	66	98.7
Private hospital or clinic	20/167	9.7	19	98.1	19	98.1
Community HIV testing service for the public	41/167	24.9	37	92.7	40	99.1
Community HIV testing service for MSM only	25/167	13.2	24	95.8	24	95.8
At home	3/167	2.4	3	100	3	100
University/school (write-in response)	7/167	2.4	7	100	7	100
Research study (write-in response)	2/167	0.4	2	100	2	100
Amongst those HIV negative and reporting that their last test was negative, n = 163***:
Where would you choose to test in future?
Public hospital or clinic	69/162	46.9
Private hospital or clinic	27/162	18.3
Community HIV testing service for the public	17/162	8.8
Community HIV testing service for MSM only	39/162	19.8
At home	10/162	6.2
Who would you prefer to perform an HIV test in future?
Doctor or clinical officer	62/162	42.1
Nurse	27/162	17.0
Counsellor	19/162	12.4
MSM community worker	44/162	22.6
Me (i.e., self-test)	10/162	6.0

*n = 24 responses to last HIV test date were missing, n = 1 missing ever tested response.

**n = 5 responses to where last tested were missing.

***n = 1 response to where would like to test in future missing.

The most common location for last HIV test was a public clinic or hospital (69/166, 47.1%), followed by a community HIV testing centre for the general public (41/166, 24.8%), while 25/166 (13.2%) had most recently tested at an MSM-specific service. Across all HIV testing locations, MSM/TG reported high levels of satisfaction with the privacy and respect they were shown. The distribution of preferences for future testing venues was diverse and similar to that of last testing location, except for a higher preference for an MSM-specific community testing facility and a private clinic or doctor compared with the percentage who had actually tested in one (19.8% compared to 13.2%, and 18.3% compared to 9.7% respectively). Just under half of participants preferred to test in a public clinic or hospital (46.9%) and preferred to have the test conducted by a doctor/clinician (42.0%), while very few expressed a preference for self-testing (6.0%).

There was little statistical evidence that testing preferences varied by sociodemographic characteristics (full results not shown), with the exception of gender identity (p = 0.006). More transfeminine individuals preferred to test in a public clinic/hospital (18/29, 69.3%) compared to cisgender (62/142, 50.3%) and non-binary (5/18, 22.9%) MSM/TG, while fewer preferred to test at a private clinic (2/29, 10.2%) compared to cisgender (28/142, 19.0%) or non-binary individuals (4/18, 30.5%). There was little difference in the percentages preferring to test in clinics targeted towards MSM by gender identity, though there was no response option given for transgender-tailored clinics.

Associations with having HIV tested in the previous 6 months among HIV-negative MSM/TG

There was little evidence for associations between sociodemographic characteristics and having HIV tested in the previous 6 months among HIV-negative MSM/TG. Compared to HIV-negative cisgender males, transfeminine individuals were 0.57 times as likely to have tested for HIV in the previous 6 months (95% CI 0.30–1.09), but the statistical evidence for an association between testing and gender identity was not significant overall (p = 0.222).

There was strong evidence that MSM/TG who had exchanged sex for money, gifts or favours in the previous 12 months, ‘sold sex’, were more likely to have HIV tested in the last 6 months (PR = 1.67, 95% CI 1.36–2.05). There was weak statistical evidence that MSM/TG who had had sex with a man in the previous 3 months were more likely to have tested in the last 6 months compared to those who had not (PR = 1.44, 95% CI 0.96–2.16), and that those who reported experiencing STI symptoms in the past year were less likely to have tested for HIV in the previous 6 months, (PR = 0.67, 95% CI 0.42–1.07).

Viral suppression and engagement in treatment and care among HIV-positive MSM/TG

The percentage of HIV-positive MSM/TG who were virally suppressed (<50 copies/ml) was estimated to be 46.9% (95% CI 31.5–62.3%, n = 58/118), Fig 1, Table 4. Self-reported knowledge of HIV-positive status was estimated to be 56.5% (95% CI 40.4–72.6%, n = 76/118) and current ART coverage 30.0% (95% CI 17.1–43.0%, n = 39/118) of all those HIV-positive. Of those who reported currently taking ART, there were 10/39 who were not virally suppressed. There were 29 participants who tested positive for HIV and were virally suppressed but who either did not self-report ever testing positive for HIV, or did not self-report current ART use.

Fig 1

The HIV care cascade among MSM/TG in Johannesburg, n = 300.

Table 4

The HIV care cascade among MSM/TG in Johannesburg, n = 300.

		n	Unweighted sample %	RDS-II weighted %	95% CI
Of All HIV-Positive, n = 118
Knows HIV-Positive Status	no	42
	yes	76	64.4	56.5	40.4,	72.6
Currently on ART of all HIV-Positive	no	79
	yes	39	33.1	30.0	17.1,	43.0
Virally suppressed of all HIV-positive	no: 50+ copies/ml	60
	yes: <50 copies/ml	58	49.2	46.9	31.5,	62.3
Expressed as 90-90-90
Knows HIV-Positive Status	no	42
	yes	76	64.4	56.5	40.4,	72.6
Currently on ART of those who know they are positive	no	37
	yes	39	51.3	53.2	32.7,	73.7
Virally suppressed of those on ART	no: 50+ copies/ml	10
	yes: <50 copies/ml	29	74.4	77.3	51.4,	100.0

Among HIV-positive MSM/TG, 75.6% had last attended HIV care at public clinics and 18.1% at MSM-specific clinics, with the remaining at private clinics. There were 89.3% (n = 63/90) who were satisfied with their last clinic’s privacy and 91.5% (n = 65/70) who were satisfied with the respect they were shown; these figures were 100% among those visiting MSM-specific clinics.

Factors associated with viral suppression amongst HIV-positive MSM/TG

Being virally suppressed was associated in crude analyses with increasing age, PR = 1.03 for each year, 95% CI 0.99–1.06, Table 5. HIV-positive MSM/TG living in Braamfontein, a central neighbourhood, as compared to Soweto, were less likely to be virally suppressed, though the numbers were small (1/9 compared to 39/66, PR = 0.04, 95% CI 0.00–0.33). MSM/TG who were not open with their family about the fact that they had sex with men were less likely to be virally suppressed, though the evidence for the association was weak (PR = 0.72, 95% CI 0.46–1.14). There was evidence that MSM/TG reporting recreational drug use in the last month were more likely to be virally suppressed (PR = 1.56, 95% CI 1.00–2.44). Having paid a man in money, gifts or favours in exchange for sex in the past 12 months was strongly associated with being virally suppressed compared to not buying sex in the last 12 months (PR = 2.16, 95% 1.61–2.89).

Table 5

Associations with viral suppression among HIV-positive MSM/TG.

Characteristics of the HIV-positive MSM/TG population		n	RDS%	Crude PR	95% CI		p value	*aPR	95% CI		p value
Sociodemographic characteristics, sexual and gender identity
Age in years	Mean unsuppressed / suppressed	28.1	30.2	1.03	0.99	1.06	0.130	1.03	1.00	1.06	0.051
Income per month / ZAR	Under 500	17/33	42.5	1.00			0.869
	500–999	6/10	48.9	1.15	0.48	2.73
	1000–1999	10/22	41.0	0.96	0.43	2.14
	2000–4999	17/32	56.8	1.34	0.72	2.47
	5000+	7/16	51.2	1.20	0.58	2.51
Born in Johannesburg	Born in Johannesburg	43/76	47.6	1.00			0.631
	Born in SA, not Johannesburg	12/33	43.8	0.92	0.53	1.61
	Born outside of SA	3/7	65.2	1.37	0.66	2.87
Religion	None	3/10	24.3	1.00			0.226
	Christianity/Islam/Other	55/108	50.7	2.09	0.64	6.81
Education completed	Up to Junior High School	8/13	50.0	1.07	0.52	2.20	0.985
	High School	32/70	46.9	1.00
	College/University/HE	18/34	47.0	1.00	0.59	1.71
Employment status	Unemployed	33/69	46.0	1.00			0.768
	Employed full-time, part-time, student	25/48	49.4	1.07	0.67	1.72
Sexual Identity	Gay or homosexual	50/103	52.0	1.00			0.738
	Bisexual, heterosexual or other	7/13	46.3	1.12	0.57	2.20
Gender Identity	Cisgender male	47/95	48.8	1.00			0.477
	Transfeminine	10/20	46.1	0.94	0.52	1.72
	Transmasculine	1/1	100.0	2.05	1.58	2.64
	Non-binary	0/2	0.0
Marital Status	Single/divorced/widowed	50/103	45.8	1.00			0.515
	Married/civil union/legal to a man or transgender individual	7/13	56.8	1.24	0.65	2.35
	Married to a woman	0/0	0	-	-	-
Neighbourhood	Soweto	39/66	55.8	1.00			0.032	1.00			0.070
	Hillbrow	6/16	43.2	0.77	0.38	1.58		0.67	0.32	1.39
	Braamfontein	1/9	2.2	0.04	0.00	0.33		0.05	0.01	0.48
	Orange Farm	5/7	58.2	1.04	0.44	2.46		0.90	0.47	1.71
	Other	7/20	31.5	0.56	0.26	1.22		0.62	0.29	1.32
Mental health, alcohol and substance use, disclosure of sexuality
PHQ9	None	29/57	47.4	1.00			0.924
	Mild depression	10/29	39.1	0.82	0.42	1.61
	Moderate depression	11/18	56.7	1.20	0.65	2.20
	Moderately severe or severe depression	4/8	56.4	1.02	0.38	2.75
	Missing >2 responses to scale questions	4/6	45.1	0.95	0.33	2.71
AUDIT	Low risk drinking	28/53	50.2	1.00			0.632
	In excess of low risk drinking guidelines	19/41	44.7	0.89	0.52	1.54
	Harmful or hazardous drinking	4/10	36.0	0.72	0.24	2.11
	Alcohol dependence	3/8	24.3	0.48	0.15	1.60
	Missing >2 responses to scale questions	4/6	64.3	1.28	0.62	2.65
Use of at least one substance in last month	No	37/81	39.9	1.00			0.052	1.00			0.155
	Yes	21/37	62.4	1.56	1.00	2.44		1.36	0.89	2.06
Openly talk about having sex with men with friends	Yes	38/77	50.5	1.00			0.429
	No	19/40	41.0	0.81	0.48	1.36
Openly talk about having sex with men with family	Yes	26/43	58.5	1.00			0.169	1.00			0.203
	No	30/72	42.4	0.72	0.46	1.14		0.75	0.49	1.16
Openly talk about having sex with men with healthcare workers	Yes	39/77	45.2	1.00			0.704
	No	18/40	49.6	1.10	0.68	1.79
Sexual Behaviours
Sex with a man (3months)	No	5/22	23.8	1.00			0.119	1.00			0.203
	Yes	53/96	52.4	2.20	0.82	5.86		1.53	0.61	3.86
Sex with a woman (3 months)	No	53/107	47.1	1.00			0.941
	Yes	5/11	45.7	0.97	0.44	2.14
2+ male partners (3 months)	No	37/78	47.3	1.00			0.910
	Yes	23/40	45.9	0.97	0.59	1.61
Sell sex to men (12 months)	No	37/78	46.8	1.00			0.860
	Yes	20/37	48.9	1.05	0.63	1.73
Buy sex from a man (12 months)	No	47/102	43.0	1.00			<0.001	1.00			0.008
	Yes	11/14	92.8	2.16	1.61	2.89		1.53	1.12	2.08
Condomless Anal Intercourse with a man (3 months)	No	35/68	51.2	1.00			0.350
	Yes	23/50	40.3	0.79	0.48	1.30
Sexual role with men during anal sex, (3 months)	No anal sex with men last 3 months	5/23	22.8	0.36	0.13	1.01	0.239
	Receptive anal intercourse	26/46	45.7	0.72	0.42	1.24
	Insertive anal intercourse	11/20	53.0	0.84	0.46	1.54
	Versatile	13/23	63.2	1.00

All models exclude seed participants and include weighting for inverse network size (RDS-II).

Adjusted models were adjusted for age and those variables found to be p value <0.2 in the crude analysis. p values are from Wald tests.

n's do not add to total where there are some missing responses.

*Other neighbourhoods include those within Johannesburg with fewer than 10 participants in the study sample.

In adjusted analyses, the evidence that older MSM/TG were more likely to be virally suppressed remained (aPR = 1.03, 96% CI 1.00–1.06), while the statistical evidence for neighbourhood associations weakened somewhat (p = 0.070). The association between viral suppression and recreational drug use also weakened (aPR = 1.36, 95% CI 0.89–2.06, p = 0.155). Strong evidence that MSM who had bought sex were more likely to virally suppressed remained, though the strength of the association reduced somewhat (aPR = 1.53, 95% CI 1.12–2.08).

Discussion

We estimated a very high HIV prevalence among MSM/TG in Johannesburg, and found half of those HIV-positive to be virally unsuppressed. These findings indicate gaps in the HIV care cascade that represent missed opportunities to improve the health of HIV-positive MSM/TG and to prevent ongoing HIV transmission. Younger MSM/TG who were HIV-positive were less likely to be virally suppressed, emphasising the need for regular and accessible testing and support to engage in HIV care for this group.

We found a higher proportion of MSM/TG to be virally suppressed at 47% (32–62%) than the 34% estimated among all adult males living with HIV in Gauteng in 2017 [12], the 26% estimated in Johannesburg in 2016 [13] and similar to the 47% (95% CI 43–52%) estimated to be virally suppressed among the HIV-positive adult population in South Africa as a whole in 2017[42]. This was the case even with our lower viral load cut-off of 50 copies/ml plasma as compared to 400/ml plasma. Our estimate is also higher than modelled estimates for virological suppression among South African MSM (22%, data from 2013–2015)[43]. It is plausible that in addition to challenges related to having sex with men, HIV-positive MSM/TG experience some of the same barriers to achieving viral suppression that men as whole do, as estimates for viral suppression among women are higher than those of men overall (43% compared to 58% nationally among adults[42]). Knowledge of status and current ART use among HIV-positive MSM/TG were much higher than that reported amongst MSM in two districts of neighbouring Mpumalanga province in 2012–2013, (56.5% compared to 28.1% and 14.1% aware of status, 30% versus 14% and 10% on ART)[44]. Our higher estimates could be related to having an urban sample of MSM/TG and to our more recent data collection. Nonetheless, the percentage of HIV-positive MSM/TG whom we found to be virally suppressed remains well below the UNAIDS target of 73% of all those HIV-positive, indicating gaps in the HIV care cascade. It is encouraging however that HIV-positive MSM/TG receiving HIV care, 76% of whom were attending public clinics and hospitals, reported high levels of satisfaction with the service they received.

We found that HIV-positive MSM/TG who had bought sex from a man in the past 12 months were more likely to be virally suppressed. Our findings could possibly reflect a higher recognition of risk and thus increased care engagement among those buying sex, but we are not able to determine causality. Studies among other MSM/TG populations have found that substance use, alcohol use and mental health are associated with worse care cascade outcomes[45], and the fact that we did not find statistically significant associations with these characteristics should be treated with caution. Our sub-sample was small, statistical error remains a possibility and we have not disaggregated different types of drug use.

Almost two thirds of HIV-negative MSM/TG reported having HIV tested within the previous 6 months, higher than among MSM in Soweto in 2008 (28%)[24, 25] and MSM in Mpumalanga in 2014–2015 prior to a testing intervention (38%)[46]. Encouragingly, testing was associated with some indicators of higher risk, including selling sex to men, though not self-reported STI symptoms. There remains a need for sustained HIV/STI testing messages. Transfeminine individuals were less likely than cisgender male individuals to have HIV tested, similar to findings from Cape Town[47], though the statistical evidence for an association between testing and gender identity was weak; further investigation is needed. Of MSM/TG who did attend HIV testing, it is encouraging that high levels of privacy and respect at last testing experience were reported. That there was not a clear majority preference for testing location and person performing the tests underscores the need for a high volume and variety of accessible and acceptable services for a diverse MSM/TG population. While few MSM/TG reported a preference for HIV self-testing in our study, it is possible that this stems from unfamiliarity: a recent study of MSM in Mpumalanga found high levels of satisfaction with HIV self-testing and increased testing frequency amongst those who had tried it[46], though studies from elsewhere have also found that men are reluctant to receive a diagnosis on their own[48].

Limitations

Our study set out to obtain population-representative estimate of HIV testing and the care cascade among MSM/TG in Johannesburg, rather than relying upon modelled estimates. While we found that study outcomes of HIV status, viral load and testing and many sociodemographic characteristics showed evidence of converging well, there were some challenges in achieving a representative sample of MSM/TG (S1 Data), as in other MSM surveys from the continent[49, 50]. Compared to another study of MSM in Soweto from 2008, we found a lower proportion of MSM/TG who identified as heterosexual or ‘straight’[25], and our estimates of sexual identity had not converged, S1 Fig A2e of S1 Data, likely indicating a difficulty in sampling this group. A higher proportion of participants described themselves as Black African compared to the population of Johannesburg overall; recruitment chains originating from White seeds did not achieve more than two sample waves and White and Asian individuals were not recruited by others. Most participants were recruited from four of the nine seeds.

Following stratification by HIV status, we might have had limited power for investigating factors that are associated with viral suppression among HIV-positive MSM/TG or recent HIV testing among HIV-Negative MSM/TG.

A further limitation was our reliance upon self-reported measures of HIV testing/status awareness and ART use. That half of MSM/TG verified to be virally suppressed did not report currently taking ART and/or did not report being aware of their status suggests that diagnosis and current ART use were under-reported among men testing HIV-positive in our study, discrepancies that we are investigating further[51]. Other surveys and studies using self-reported care cascade indicators have also pointed out this problem[52-56], with social desirability biases and survey misunderstandings posited as possible explanations. Confusion from care interactions have also been reported in South Africa[57]. Misinterpretations about the distinction between ‘undetectable’ viral load and ‘negative’ HIV status in provider interactions may represent an increasingly importance source of misclassification that should be investigated further. While our estimate of viral suppression is biologically ascertained, these reporting biases limit our ability to guide programming in identifying whether the gaps in diagnosis versus initiation and retention on ART are most significant.

Conclusions

It is encouraging that MSM/TG do not appear to be lagging in the HIV care cascade compared to the general male population in Johannesburg and are reporting positive care experiences. However, given their higher risks as a group and very high HIV burden, there is a need to continue to focus resources towards further improvement, particularly among younger MSM/TG. A better understanding of sexual mixing and transmission patterns could aid in targeting, development of a differentiated care approach to serve the diversity of MSM would be appropriate, and a dedicated study to understand the needs of the transgender population is warranted.

Respondent driven sampling recruitment diagnostics.

(DOCX)

Click here for additional data file.

Viral suppression according to different viral load cut-offs.

(DOCX)

Click here for additional data file.

Survey instrument.

(PDF)

Click here for additional data file.

29 Nov 2019

PONE-D-19-20875

HIV testing, care and viral suppression among men who have sex with men and transgender individuals in Johannesburg, South Africa

PLOS ONE

Dear Dr Smith,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

I invite you to revise and resubmit your manuscript. Please attend to the reviewer's comments, which I believe may help to better strengthen the contributions of this manuscript to the field and public health practice. In addition, please attend to the following issues I noted in my review of the manuscript:

If citation 31 is published then please include a brief statement that reviews the RDS recruitment process (methods) and challenges reaching a representative sample using RDS (discussion). It is not enough to just point the reader to a subsequent manuscript. If this paper remains unpublished by the time of your re-submission, please provide sufficient detail on the RDS method and appropriate data on the representativeness of the sample recruited.

Please indicate the variables missing data in Table 1 as done in Table 2 in the foot note

Please include the test statistics for comparison of recent HIV tests with and without missing date information in text. I t is OK that these data are not presented in any of the tables.

join the reviewers in their enthusiasm for the data you present on an important, but understudied population; highlighting both areas where some successes have been achieved and areas in need of targeted intervention efforts. Please provide a point by point response to each reviewers' response, and indicate in the response how it was addressed. This will help to facilitate a more timely review.

==============================

We would appreciate receiving your revised manuscript by Jan 13 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Laramie Smith, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

1. Please include a copy of the interview guide used in the study, in both the original language and English, as Supporting Information, or include a citation if it has been published previously.

2. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

3. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

4. One of the noted authors is a group or consortium TRANSFORM Study Group. In addition to naming the author group, please list the individual authors and affiliations within this group in the acknowledgments section of your manuscript. Please also indicate clearly a lead author for this group along with a contact email address.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study uses RDS methodology to obtain a representative sample of MSM/TG in Johannesburg, in order to investigate: HIV prevalence, HIV testing behaviour in HIV negative individuals and viral suppression in HIV positive individuals. The study found a high prevalence of HIV, that the majority of negative MSM/TG had tested within the last 6 months, and poor viral suppression in PLHIV, although comparable to that of men in South Africa in general.

This is an important study adding to the evidence base attempting to understand HIV in this key population which has been historically neglected in Sub-Saharan Africa.

I believe the manuscript could be improved by addressing the following comments:

1) Why was a cut-off of 200 used for viral suppression? The cut-off in South African treatment guidelines at the time of the study, as well as National Department of Health programme monitoring, is 400. The other commonly used cut-off at this time (2017) was 1000 (PEPFAR routine reporting; 2017 South African national HIV prevalence, incidence, behaviour and communication survey). In the new guidelines, 50 will be used. It would be more comparable to the other evidence and applicable to the South African setting to use 400 as a cut-off instead.

2) In South Africa it is known that there is a large group of MSM who identify as straight (example Lane et al. High HIV Prevalence Among Men who have Sex with Men in Soweto South Africa: Results from the Soweto's Men's Study AIDS Behav (2011) 15:626–634). The RDS methodology does not appear to have penetrated into this group well (not unexpected). I think it is important to mention this in the discussion.

3) Although the viral suppression rates are similar to those of other men, the rate of viral suppression in men in South Africa is well below that of women. e.g. 2017 South African national HIV prevalence, incidence, behaviour and communication survey: viral suppression in men- 43%; viral suppression in women- 58%. Is it possible that the finding of similar viral suppression in MSM/TG is related to the fact that men in general experience such significant access barriers?

4) Reference 31: has this manuscript been accepted for publication yet? I do not believe it can be referenced as it stands. From the journal website:

"Do not cite the following sources in the reference list: Unavailable and unpublished work, including manuscripts that have been submitted but not yet accepted (e.g., “unpublished work,” “data not shown”). Instead, include those data as supplementary material or deposit the data in a publicly available database."

5) There are a few minor editing comments:

- line 103: please change "18 years plus" to "18 years or older"

- Table 1, employment status, unemployed- there appears to be an error in the % column

Reviewer #2: Review: HIV testing, care and viral suppression among men who have sex with men and transgender individuals in Johannesburg, South Africa

A very well written paper of major interest in the field. The authors estimated a very high HIV prevalence among MSM/TG in Johannesburg, and half of those HIV-positive were virally unsuppressed. These findings indicate gaps in the HIV care cascade that represent missed opportunities to improve the health of HIV-positive MSM/TG and to prevent ongoing HIV transmission.

I only have one major concern that requires the author’s revision before publication. The authors have conducted univariate and multivariate logistic regression to identify factors associated with HIV-testing in the previous 6 months amongst those HIV-negative; and to identify factors associated with viral suppression amongst those HIV-positive.

While the authors have correctly interpreted the odds ratios (OR). In this population, these binary outcomes (HIV testing, access to care and viral suppression) are common (>10%), it is often more desirable to estimate a relative risk (RR) or prevalence ratio since there is an increasing differential between the RR and odds ratio with increasing prevalence rates, and there is a tendency for some to interpret ORs as if they are RRs. The authors can estimate the RR in Stata with a log-binomial regression model or should there be convergence issues, apply the Poisson regression model with a robust error variance. A major consideration in public health research is need to communicate study results to a wider non-statistical community i.e. the medical community, general public, and policy makers. Poisson regression with robust variance and log-binomial regression provide correct estimates and are a better alternative for the analysis of this kind of study design with binary outcomes than logistic regression, as the relative risk or prevalence ratio is more interpretable and easier to communicate than the odds ratio.

Please see references below.

1. McNutt LA, Wu C, Xue X, Hafner JP. Estimating the Relative Risk in Cohort Studies and Clinical Trials of Common Outcomes. Am J Epidemiol 2003; 157(10):940-3.

2. Zou G. A Modified Poisson Regression Approach to Prospective Studies with Binary Data. Am J Epidemiol 2004; 159(7):702-6.

3. Sander Greenland , Model-based Estimation of Relative Risks and Other Epidemiologic Measures in Studies of Common Outcomes and in Case-Control Studies, American Journal of Epidemiology 2004;160:301-305

Simba Takuva, MBChB, MSc, DTM&H.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Simbarashe Takuva

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

26 Mar 2020

February 28, 2020

Dear Laramie,

Please find attached our revised manuscript submission.

We thank the editor and reviewers for taking the time to read and make suggestions to improve our manuscript. Below, we respond to editor and reviewer comments, and explain why it is not possible to make data publicly available but give guidance on how researchers can obtain the data upon request.

Yours sincerely,

Elizabeth Fearon

Editor Comments

I invite you to revise and resubmit your manuscript. Please attend to the reviewer's comments, which I believe may help to better strengthen the contributions of this manuscript to the field and public health practice. In addition, please attend to the following issues I noted in my review of the manuscript:

1. If citation 31 is published then please include a brief statement that reviews the RDS recruitment process (methods) and challenges reaching a representative sample using RDS (discussion). It is not enough to just point the reader to a subsequent manuscript. If this paper remains unpublished by the time of your re-submission, please provide sufficient detail on the RDS method and appropriate data on the representativeness of the sample recruited.

RESPONSE: We have now included the more extensive set of RDS diagnostics in Appendix 1 and more detail as to the representativeness of the survey to the discussion of study limitations, p.24.

2. Please indicate the variables missing data in Table 1 as done in Table 2 in the foot note

RESPONSE: We had added the numbers missing responses for each variable to Table 1.

3. Please include the test statistics for comparison of recent HIV tests with and without missing date information in text. It is OK that these data are not presented in any of the tables.

RESPONSE: We have added the test statistics to the text, p.12.

I join the reviewers in their enthusiasm for the data you present on an important, but understudied population; highlighting both areas where some successes have been achieved and areas in need of targeted intervention efforts. Please provide a point by point response to each reviewers' response, and indicate in the response how it was addressed. This will help to facilitate a more timely review.

RESPONSE: Thank you, we respond to each comment and indicate where we have made changes to the manuscript.

Reviewer #1: This study uses RDS methodology to obtain a representative sample of MSM/TG in Johannesburg, in order to investigate: HIV prevalence, HIV testing behaviour in HIV negative individuals and viral suppression in HIV positive individuals. The study found a high prevalence of HIV, that the majority of negative MSM/TG had tested within the last 6 months, and poor viral suppression in PLHIV, although comparable to that of men in South Africa in general.

This is an important study adding to the evidence base attempting to understand HIV in this key population which has been historically neglected in Sub-Saharan Africa.

RESPONSE: Thank you.

I believe the manuscript could be improved by addressing the following comments:

1) Why was a cut-off of 200 used for viral suppression? The cut-off in South African treatment guidelines at the time of the study, as well as National Department of Health programme monitoring, is 400. The other commonly used cut-off at this time (2017) was 1000 (PEPFAR routine reporting; 2017 South African national HIV prevalence, incidence, behaviour and communication survey). In the new guidelines, 50 will be used. It would be more comparable to the other evidence and applicable to the South African setting to use 400 as a cut-off instead.

RESPONSE: In the absence of consistent viral load cut-points across studies and contexts, we chose to use 200 viral copies/ml as the cut-off, following CDC recommendations given in:

Centers for Disease Control and Prevention. Questions and Answers for the General Public: Revised Recommendations for HIV Testing of Adults, Adolescents, and Pregnant Women in Health Care Settings. However, we agree with the reviewer that it would be helpful to present cascade findings using the South African guidelines. We think it makes most sense to use the cut-off of 50 viral copies/ml plasma to indicate viral suppression (our cut-off for detection was 40 copies/ml) in line with the 2019 South African guidelines, and because there is evidence that those with viral loads >50 copies/ml but less < 1000 copies/ml have poorer treatment outcomes compared to those with viral load <50 copies/ml (Hermans LE, Moorhouse M, Carmona S, Grobbee DE, Hofstra LM, Richman DD, et al. Effect of HIV-1 low-level viraemia during antiretroviral therapy on treatment outcomes in WHO-guided South African treatment programmes: a multicentre cohort study. Lancet Infect Dis. 2018;18(2):188-97).

We have also included in Appendix 2 the viral suppression n’s and RDS-weighted %’s for other cut-offs (200, 400 and 1000 copies/ml) used in the past and elsewhere in the literature, in order to aid comparison.

2) In South Africa it is known that there is a large group of MSM who identify as straight (example Lane et al. High HIV Prevalence Among Men who have Sex with Men in Soweto South Africa: Results from the Soweto's Men's Study AIDS Behav (2011) 15:626–634). The RDS methodology does not appear to have penetrated into this group well (not unexpected). I think it is important to mention this in the discussion.

RESPONSE: We have added a note to the discussion of study limitations as follows, line 365: “Compared to another study of MSM in Soweto from 2008, we found a lower proportion of MSM who identified as heterosexual or ‘straight’(24), possibly indicating a difficulty in sampling this group.”

3) Although the viral suppression rates are similar to those of other men, the rate of viral suppression in men in South Africa is well below that of women. e.g. 2017 South African national HIV prevalence, incidence, behaviour and communication survey: viral suppression in men- 43%; viral suppression in women- 58%. Is it possible that the finding of similar viral suppression in MSM/TG is related to the fact that men in general experience such significant access barriers?

RESPONSE: Yes, we agree this is a possibility and we have now mentioned this in the discussion (see line 315):

“It is plausible that in addition to challenges related to having sex with men, HIV-positive MSM/TG experience some of the same barriers to achieving viral suppression that men as whole do, as estimates for viral suppression among women are higher at than those of men overall (43% compare to 58% nationally among adults(40)).”

4) Reference 31: has this manuscript been accepted for publication yet? I do not believe it can be referenced as it stands. From the journal website:

"Do not cite the following sources in the reference list: Unavailable and unpublished work, including manuscripts that have been submitted but not yet accepted (e.g., “unpublished work,” “data not shown”). Instead, include those data as supplementary material or deposit the data in a publicly available database."

RESPONSE: We have now added more detailed RDS diagnostics to Appendix 1 instead of referencing this unpublished manuscript, and have added further detail to the Discussion.

5) There are a few minor editing comments:

- line 103: please change "18 years plus" to "18 years or older"

- Table 1, employment status, unemployed- there appears to be an error in the % column

RESPONSE: Thank you for bringing these issues to our attention. We have rectified them.

Reviewer #2: Review: HIV testing, care and viral suppression among men who have sex with men and transgender individuals in Johannesburg, South Africa

A very well written paper of major interest in the field. The authors estimated a very high HIV prevalence among MSM/TG in Johannesburg, and half of those HIV-positive were virally unsuppressed. These findings indicate gaps in the HIV care cascade that represent missed opportunities to improve the health of HIV-positive MSM/TG and to prevent ongoing HIV transmission.

I only have one major concern that requires the author’s revision before publication. The authors have conducted univariate and multivariate logistic regression to identify factors associated with HIV-testing in the previous 6 months amongst those HIV-negative; and to identify factors associated with viral suppression amongst those HIV-positive.

While the authors have correctly interpreted the odds ratios (OR). In this population, these binary outcomes (HIV testing, access to care and viral suppression) are common (>10%), it is often more desirable to estimate a relative risk (RR) or prevalence ratio since there is an increasing differential between the RR and odds ratio with increasing prevalence rates, and there is a tendency for some to interpret ORs as if they are RRs. The authors can estimate the RR in Stata with a log-binomial regression model or should there be convergence issues, apply the Poisson regression model with a robust error variance. A major consideration in public health research is need to communicate study results to a wider non-statistical community i.e. the medical community, general public, and policy makers. Poisson regression with robust variance and log-binomial regression provide correct estimates and are a better alternative for the analysis of this kind of study design with binary outcomes than logistic regression, as the relative risk or prevalence ratio is more interpretable and easier to communicate than the odds ratio.

Please see references below.

1. McNutt LA, Wu C, Xue X, Hafner JP. Estimating the Relative Risk in Cohort Studies and Clinical Trials of Common Outcomes. Am J Epidemiol 2003; 157(10):940-3.

2. Zou G. A Modified Poisson Regression Approach to Prospective Studies with Binary Data. Am J Epidemiol 2004; 159(7):702-6.

3. Sander Greenland , Model-based Estimation of Relative Risks and Other Epidemiologic Measures in Studies of Common Outcomes and in Case-Control Studies, American Journal of Epidemiology 2004;160:301-305

RESPONSE: Thank you for this suggestion. We have re-done our analyses using robust poisson regression to obtain weighted prevalence ratios instead of odds ratios for measure of association (still dropping seed participants and probability weighting using the inverse network weights in line with the RDS design).

Interview Guides

We have added the survey instruments in English and Zulu as supplementary material (S3).

Data Access

Data from this study cannot be deposited publicly because of a high risk of deductive disclosure when reporting all of the information needed to validly analyse the data. This risk is made worse by the vulnerability of the population, which faces social stigma and has reported high levels of harassment, blackmail and abuse in other studies (eg Zahn 2016 (1)) and in our survey.

The data for this study were collected using respondent driven sampling (RDS), a type of peer-referral or ‘snowball sampling’. Provided that assumptions are met and appropriate weighting is used, the data can be treated as a probability sample and used to obtain representative estimates of the population, eg HIV status. In practice, these assumptions are hard to meet and it is recommended to examine a set of RDS recruitment dynamics to assess the extent to which the sampling might deviate from assumptions, (see guidance on RDS diagnostics (2)). This requires information about participants’ social networks, information about their recruiter and those they recruit, and their position within sampling chains (eg whether one of only nine seed participants). We have given an overview description of the RDS recruitment as an appendix to this manuscript, and one that is specific to this study’s primary outcomes. We have taken care to do this while maintaining confidentiality, eg not giving HIV status of seed participants away via showing bottle net plots or recruitment trees by HIV status, but subsequent studies would need the full RDS information to assess the representativeness of the sample with respect to their specific study primary outcomes, eg to produce and examine convergence and bottleneck plots. We are concerned that there is a high risk of deductive disclosure to making the individual-level RDS data public, in combination with data about other characteristics of participants, including highly sensitive HIV and STI testing results and personal characteristics. Even if we removed individual pieces of identifiable information, we are not confident that this would truly anonymise the data.

Although we cannot make study data publicly accessible at the time of publication, all authors commit to make the data underlying the findings of the study available, in compliance with the PLOS Data Availability Policy. Data can be requested via the London School of Hygiene and Tropical Medicine Research Operations Office Data Management Lead: alex.hollander@lshtm.ac.uk and the first author (elizabeth.fearon@lshtm.ac.uk) and Principal Investigator (tpalanee@wrhi.ac.za). Individuals requesting data should give their research objectives and indicate the list of requested variables, see S3 Survey Instrument. For data involving personally identifiable information or other sensitive data, data sharing is contingent on the data being handled appropriately by the data requester and in accordance with all applicable local requirements.

(1) Zahn R, Grosso A, Scheibe A, Bekker LG, Ketende S, Dausab F, et al. Human Rights Violations among Men Who Have Sex with Men in Southern Africa: Comparisons between Legal Contexts. PLoS One. 2016;11(1):e0147156

(2) Gile KJ, Johnston LG, Salganik MJ. Diagnostics for Respondent-driven Sampling. J R Stat Soc Ser A Stat Soc. 2015;178(1):241-69

Added 26 March, following discussion with editor: for brevity, we propose the following data availability statement:

The data underlying this study cannot be made publicly accessible due to concerns about confidentiality when combining sensitive data about individuals and the respondent driven sampling (RDS) recruitment information needed to reflect the sampling design in analyses.  Although we cannot make study data publicly accessible at the time of publication, all authors commit to make the data underlying the findings of the study available` in compliance with the PLOS Data Availability Policy. Data can be requested via the London School of Hygiene and Tropical Medicine Research Operations Office Data Management Lead: alex.hollander@lshtm.ac.uk and the first author (elizabeth.fearon@lshtm.ac.uk) and Principal Investigator (tpalanee@wrhi.ac.za). Individuals requesting data should give their research objectives and indicate the list of requested variables, see S3 Survey Instrument. To protect the confidentiality of participants, data sharing is contingent on the data being handled appropriately by the data requester and in accordance with all applicable local requirements.

27 May 2020

HIV testing, care and viral suppression among men who have sex with men and transgender individuals in Johannesburg, South Africa

PONE-D-19-20875R1

Dear Dr. Fearon,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Zixin Wang, PhD.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: (No Response)

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: (No Response)

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: (No Response)

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: (No Response)

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: I have no additional comments or edits to add. The authors have attended to all concerns in a satisfactory way. Thank you.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Simbarashe Takuva

4 Jun 2020

PONE-D-19-20875R1

HIV testing, care and viral suppression among men who have sex with men and transgender individuals in Johannesburg, South Africa

Dear Dr. Fearon:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Zixin Wang

Academic Editor

PLOS ONE