Correlates of Water-Based Lubricant Use Among Men Who Have Sex with Men in Blantyre,Malawi

Abstract

Condom compatible lubricants (CCLs), including water-based lubricants (WBL) represent one strategy to prevent the breakage of latex condoms and thus decrease the risk of HIV transmission during anal intercourse. The analyses presented here characterize the correlates of WBL use during anal sex among men who have sex with men (MSM) in Blantyre, Malawi enrolled from April 2011 to March 2012 using respondent-driven sampling (RDS). Bivariate and multivariable logistic regression analyses with RDS-weighting were conducted on a total sample of 338 MSM. With RDS-weighting, 25.4% [95% confidence interval (CI): 20.3–31.4] of MSM (106/329) reported primarily using WBL during anal sex. In multivariable analysis, higher income [adjusted odds ratio (aOR): 5.9; 95% CI: 2.48–14.19], family being aware of their sexual practices (aOR: 2.52; 95% CI: 1.29–4.92), and reporting consistent condom use in the last 6 months (aOR: 1.27; 95% CI: 1.06–1.52) were positively associated with WBL use. Increasing age (per 1 year increase in age; aOR: 0.89; 95% CI: 0.83–0.95) was negatively associated with WBL. Taken together, these data highlight the limited uptake of WBL among MSM in Blantyre, Malawi, especially among older men and those belonging with lower income. Older MSM in Malawi are known to have a higher prevalence of HIV and lower reported use of WBL, suggesting significant risks of onward HIV transmission. Separately, the limited use among those with lower incomes suggests the need for free or subsidized distribution of CCL together with condoms and counseling about their use specifically for MSM in Malawi.

Introduction

Men who have sex with men (MSM) are disproportionately affected by HIV across the world, including in more broadly generalized epidemics observed in Southern and Eastern Africa.^1,2 These disparities in HIV burden are due primarily to the high HIV acquisition and transmission risks associated with condomless anal intercourse (CAI) with serodiscordant and viremic partners.³ A meta-analysis exploring probability of HIV transmission observed nearly twenty times higher likelihood of HIV transmission in the context of anal compared with vaginal intercourse.⁴ However, these biological risks are contextualized by structural risk determinants such as stigma, violence, and discrimination, leading to limited provision and uptake of HIV services for these men.^1,5

–10

Thus, as part of a comprehensive package for HIV prevention among MSM, the World Health Organization recommends the use of condom compatible lubricants (CCLs), which include silicon and water-based lubricants (WBL).¹¹ WBL are recommended rather than petroleum or glycerin-based products, which have been shown to degrade latex condoms,^11,12 and consequently are associated with both higher rates of condom breakage^13,14 and increased porousness.¹⁵ Not using WBL can not only increase the risk of condom rupture but also increase the risk of rectal lesions and physical trauma.⁵ Greater condom use with WBL therefore can afford greater protection against HIV, although use of lubricants without condoms can increase the risk of other sexually transmitted infections (STIs).⁵

In 2011, the Office of the U.S. Global AIDS Coordinator (OGAC) which administers the President's Emergency Plan for AIDS Relief (PEPFAR) released the Technical Guidance on Combination HIV Prevention for MSM,¹⁶ in which it recommended both distribution of condoms and of CCL as a part of a comprehensive HIV prevention strategy for MSM. This guidance built on the 2008 Central Contraceptive Project from PEPFAR, which began procuring and shipping lubricants to PEPFAR countries.¹⁷ However, CCLs were not added to the Contraceptive and Condoms Product Catalog until 2011.¹⁸ Also, despite the guidance and the addition of CCL to the commodity list within the contraceptive and condom product catalog, there remain limited HIV prevention programs which effectively integrate information about CCL usage.

Beyond this, there is also limited distribution of this core commodity for the prevention of HIV acquisition and transmission together with condom distribution.¹⁹ In neighboring Tanzania, in an attempt to discourage same-sex practices, the government first enforced a ban on the import and sale of lubricants in 2016 before fully banning HIV/AIDS outreach projects aimed at gay men and other MSM several months later.²⁰ More recently, persecution and arrests of those believed to be gay in Dar es Salaam have only intensified.

Gay men and other MSM in higher income countries have been reported to be several times more likely to be able to access CCL than their counterparts in low and middle income countries, although this comparison did not note within country disparities.¹ Of a global online study conducted in 2010 among 5,066 MSM and their providers, only 29% reported that lubricant was easily accessible.²¹ Another study conducted by the same team in 2012 again found a low percentage of participants reporting easy access to lubricant. Specifically, only 8% of participants in low income countries reported easy access compared to 14% in lower-middle income countries and 34% of participants in high income countries.²² However, these estimates did not specify the type of lubricant accessed, and other studies have found that MSM often use condom incompatible lubricant, such as body cream and cooking oil, when they cannot access CCL.^19,23

–26

Other factors associated with higher access to lubricants were more community engagement and more comfort with providers.²² Barriers to lubricant access were identified, including homophobia and the level to which the respondent's sexual orientation was known to peers (“outness”).²²

Across sub-Saharan Africa, several studies have explored the correlates and uptake of CCL use among both MSM and heterosexual men. One study conducted in Lagos State, Nigeria in 2011 found that 85.5% of MSM reported lubricant use; however, only 55.4% (138/249) reported use of a CCL, in this case KY jelly.²⁴ In addition, 58.2% (145/249) reported using lubricant use “all the time” and 67.9% (169/249) reported always using lubricant with a condom. Another study conducted in Zambia between 2003 and 2006 among 155 males reporting heterosexual serodiscordant relationship status found that at baseline, none of the men had previous exposure to lubricant products.²⁷ A more recent study in Tanzania conducted between 2012 and 2013 found relatively high levels of CCL (two-thirds of men self-reporting use).²⁸ The most common reasons for not using lubricants were lack of availability, heat of the moment, and knowing their partner's HIV status.²⁸

In a 2008 study in Malawi, the prevalence of self-reported WBL use among MSM was 11.3%,²⁹ although there has been limited research evaluating determinants of CCL use among MSM in the country. In Malawi, same-sex sexual practices are illegal, and punishment for men can carry an up to 14-year prison sentence³⁰; however, in 2014, the Malawian government announced the suspension of enforcement of these laws.³¹ Through the few studies that have been conducted in Malawi, the prevalence of HIV among Malawi MSM is estimated to be between 15.4% and 21.0% compared to ∼8.1% among men 15–49 in the broader population.^32
–34

Despite higher HIV prevalence among MSM, in a qualitative assessment conducted in Malawi, health care workers acknowledged gaps in knowledge on HIV transmission risks among MSM, and both health care workers and MSM cited the need for additional structural supports for providing services to MSM, including HIV/sexually transmitted infection risk counseling specifically addressing the needs of MSM.³⁵ Even in the context of novel HIV prevention approaches, the regular use of condoms and CCL remains a core component of HIV prevention for MSM. To better understand the barriers and facilitators to CCL use, this study explores prevalence and correlates of CCL, specifically WBL use, among MSM in Blantyre, Malawi.

Materials and Methods

Study design, population, and recruitment

The methods for this study, along with respondent-driven sampling (RDS) diagnostics, have been previously described in more detail elsewhere.³² The primary aim of the original study was to better understand the role condoms play in prevention of HIV acquisition among MSM in Blantyre, Malawi. In brief, this cross-sectional study enrolled 338 MSM between April 2011 and March 2012 in Blantyre, Malawi. To be considered eligible, participants had to be ≥18 years of age, assigned male gender at birth, and fluent and able to provide verbal consent in either Chichewa or English. In addition, participants must have reported having anal sex with another man in the past 12 months and have no prior participation in this study.

Recruitment to the study was conducted through RDS. Ten individuals, or “seeds,” were purposively chosen, provided three recruitment coupons, and then asked to recruit other MSM into the study. Recruited MSM were also given three coupons each and asked to recruit MSM from their own network. A double-incentive structure was used, in which participants received reimbursement for travel costs as well as for recruitment for each additional participant. This process continued until a minimum sample size of 345 participants was reached. The sample size and power calculations for this study were originally based on the difference in HIV prevalence between consistent condom users and nonconsistent condom users. This analysis was conducted post hoc.

This study was conducted in collaboration with the Johns Hopkins Bloomberg School of Public Health (JHSPH), the Centre for the Development of People (CEDEP), and the Malawi College of Medicine, University of Malawi. Study activities were approved by the University of Malawi College of Medicine Research and Ethics Committee (COMREC) and for secondary analysis by the JHSPH Institutional Review Board.

Procedure

Trained members of CEDEP and the University of Malawi College of Medicine administered sociobehavioral questionnaires to eligible participants. Participants were administered a study questionnaire, which included indicators focused on sociodemographic characteristics, mental health and depressive symptoms, substance use, sexual relationships, access to health care and disclosure of sexual orientation to health care workers, family members, and peers. After completing their questionnaires, participants received HIV and syphilis testing and counseling from trained College of Medicine nurses. Study design and demographics are further described elsewhere.³²

Measures

Measures of sexual practices included practices among different male and female partner types (such as main and casual), partner characteristics, condom use, and number of sexual partners. Quantitative measures of HIV prevention knowledge were also assessed and included knowledge on correct lubricant use. Specifically, respondents were asked what they believed was the safest lubricant for use during anal sex. Responses were categorized as “petroleum jelly or Vaseline,” “body creams/fatty creams,” “water-based lubricant,” “saliva,” “no lubricant use,” “no response,” or “don't know.” As silicon-based lubricants were not widely available during the time of this study, survey questions regarding lubricant use were focused on WBL rather than CCL broadly. Finally, the survey included quantitative measures on human rights violations, both experienced and perceived.

The outcome of interest for this study was based on participants' response to the question “Which type of lubricant do you generally use for anal sex with men?” Response options were “petroleum jelly or Vaseline,” “body creams/fatty creams,” “water-based lubricant made for sex (such as K-Y Jelly or Durex),” “saliva,” “no lubricant use,” or “other.”

While income in the past month was reported as a continuous variable, for this analysis, income was examined by tertile, with individuals being grouped into the 33rd percentile and below, between the 33rd and 66th percentiles, and above the 66th percentile.

Additional correlates included consistent condom use in the past 6 months, familial knowledge that respondent has sex with men, access to lubricant, and receipt of information on how to prevent HIV infection from sex between men. Condom use was measured with the question “In general, how often have you used a condom in the past 6 months?” A four-point Likert-type scale was used to gauge use, but for the purposes of this study, “always” was coded as 1 and all other responses “0.” Familial awareness that respondent has sex with men was measured with the question “Does anyone in your family know that you have sex with other men?” for which respondent provided a yes/no response. For access to various forms of lubricants, participants were asked “What kind of access to lubricants do you have when you need them?” Responses included “no access,” “difficult or little access,” “some access,” “very easy access,” “no response,” and “don't know.” For analysis, “some access” and “very easy access” were coded as 1 and all other responses as 0. Receipt of information on how to prevent HIV from sex between men was measured with the question “In the last 12 months, have you received information on how to prevent HIV infection from sex between men?”

Data analyses

The choice of covariates for these analyses was informed by findings from previous literature and the Modified Social Ecological Model.³⁶ Individualized weights, computed with the use of Stata's RDS data-smoothing algorithm, were applied to each variable.³⁷ Seed participants were included in the analyses. The RDS-II weights were then used in bivariate analyses, although both unadjusted and RDS-adjusted estimates are reported. RDS-adjusted 95% confidence intervals (CIs), as seen in Table 1, were determined through bootstrapping using 1,000 iterations. RDS weighting specific to the outcome of interest was then used in bivariate logistic analysis. Covariates achieving statistical significance of p ≤ .125 or other variables considered important because of a priori inferred relationship were then included in an RDS-weighted multivariable logistic analysis. The final multivariable regression model included age, income, self-reported condom use in the last 6 months, and “outness” among family members. Since the variable based on participants' responses to the question “Which is the safest lubricant to use during anal sex?” was highly correlated with the outcome in bivariate analysis, and it was ultimately excluded from multivariable analysis due to collinearity. All analyses were conducted using Stata 13.1 (College Station, TX).³⁸

Table 1.

General Demographic Characteristics of Men Who Have Sex with Men in Blantyre, Malawi, 2011–2012

Characteristic	Non-RDS-adjusted proportion (n)	RDS-adjusted proportion, %	RDS-adjusted 95% confidence interval
Age, years
18–25	56.8% (192/338)	58.5	50.2–66.8
≥26	43.2% (146/338)	41.5	33.2–49.8
Gender identity
Male	77.8% (256/338)	80.2	74.8–85.6
Female	19.2% (65/338)	17.0	11.8–22.1
Transgender	2.9% (10/338)	2.8	0.74–5.5
Sexual orientation
Gay or homosexual	62.3% (210/337)	60.8	53.6–68.1
Bisexual	37.1% (125/337)	36.3	29.1–43.4
Heterosexual or straight	0.6% (2/237)	2.9	0.0–6.7
Nationality at birth
Malawi	98.8% (325/329)	N/A	N/A
Zimbabwe	0.6% (2/329)	N/A	N/A
Other African	0.6% (2/329)	N/A	N/A
Highest level of education (completed)
Some secondary high school or lower	43.2% (146/338)	46.2	39.3–53.3
Completed secondary	38.5% (130/338)	40.8	34.0–48.0
Postsecondary	18.3% (62/338)	13.0	9.4–17.7
Income in the last month
≤6,000 MKW	34.0% (113/332)	32.1	32.1–46.3
6,001–15,000 MKW	34.3% (114/332)	31.2	31.2–45.0
>15,000 MKW	31.6% (105/332)	18.6	18.6–29.1
Employment status
Unemployed	46.8% (158/338)	51.7	44.2–59.3
Employed (self and by other)	40.2% (136/338)	36.8	29.6–44.1
Student	13.0% (44/338)	11.5	0.74–15.5
Marital status
Married, cohabiting	16.0% (54/338)	16.2	9.7–22.7
Single/divorced/separated	84.0% (284/338)	83.7	77.4–90.1
Type of lubricant primarily used for anal sex with men
Petroleum jelly or Vaseline	45.3% (149/329)	49.9	42.7–57.1
Body creams/fatty creams	3.7% (12/329)	2.7	1.1–6.2
Water-based lubricant	32.2% (106/329)	25.4	20.3–31.4
Saliva	1.8% (6/329)	4.4	1.6–11.7
No lubricant use	12.8% (42/329)	13.6	9.6–19.0
Other	4.3% (14/329)	2.0	2.0–7.5
Identified water-based lubricant as safest lubricant for anal sex	42.5% (130/306)	34.3	27.9–41.2
Reported access to lubricants (any type)
No/difficult access	58.8% (191/325)	60.6	53.6–67.2
Some/very easy access	41.23% (134/325)	39.4	39.4–46.4
Tested HIV positive as part of the study	15.4% (52/329)	12.5	7.2–17.8

MKW, Malawi Kwacha; RDS, respondent-driven sampling.

Missing data

Initial analysis for missing data showed that nine participants did not respond to the question about lubricant use, the outcome of interest. Since these participants represented <3% of the data on lubricant use, no imputation was completed and participants with incomplete data on lubricant use were excluded from multivariable analysis. In the final multivariable model, 300 participants had data available for complete case analysis.

Results

Table 1 displays both the unweighted and RDS-weighted estimates for characteristics of the 338 MSM participating in this study. The mean age of participants was 25 years (range 18–49). Seventy-eight percent (263/338) of participants were identified as male, while 22% (75/338) reported identifying as either female or transgender. Further sociodemographic and population characteristics are described elsewhere.³²

In unweighted estimates, 45.3% (149/338) of participants reported using petroleum jelly or Vaseline during anal sex with men, whereas 32.2% (106/338) reported WBL. In addition, 3.6% (12/338) reported use of body creams/fatty creams, 1.8% (6/338) saliva, 12.8% (42/338) no lubricant use, and 4.3% (n = 14) other. When using RDS weights, the estimated proportion of men using WBL was 25.4% (95% CI: 20.3–31.4). Unadjusted estimates of the percentage of MSM self-reporting WBL use by each correlate are also shown in Table 2.

Table 2.

Unadjusted Percentage of Men Who Have Sex with Men in Blantyre, Malawi with Selected Characteristics Comparing Those Who Self-Reported Primarily Using Water-Based Lubricant for Anal Sex with Men with Those Who Did Not, 2011–2012

Variable	Those who primarily used WBL for anal sex with men (self-reported), %, (n), N = 106	Those who did not primarily use WBL for anal sex with men (self-reported), % (n), N = 223
Age, years
18–25	58.5 (62)	50.2 (112)
≥25	41.5 (44)	49.8 (111)
Highest level of education completed
Some secondary high school or lower	30.2 (32)	49.3 (110)
Completed secondary	41.5 (44)	37.2 (83)
Postsecondary	28.3 (30)	13.5 (30)
Income in last month
≤6,000 MKW	21.4 (22)	40.3 (89)
6,001–15,000 MKW	29.1 (30)	35.3 (78)
>15,000 MKW	49.5 (51)	24.4 (54)
Knowledge that WBL is the safest lubricant for anal sex
Incorrect knowledge	2.9 (3)	86.5 (173)
Correct knowledge	97.1 (101)	13.5 (27)
Access to lubricants
No/difficult access	67.6 (71)	53.3 (113)
Some/very easy access	32.4 (34)	46.7 (99)
HIV prevention for sex between men in the last 12 months
Did not receive information	67.3 (70)	81.9 (181)
Received information	32.7 (34/74)	18.1 (40)
Knowledge that anal sex puts them at greatest risk for HIV
Incorrect knowledge	71.9 (69)	91.4 (181)
Correct knowledge	28.1 (27)	8.6 (17)
Family is aware participant has sex with men	68.0 (70)	46.8 (102)
Participant afraid to seek health care because has sex with men	28.3 (30)	16.1 (36)
Tested HIV positive as a part of the study	20.8 (22)	13.0 (29)

WBL, water-based lubricant.

Correlates of water-based lubricant use

In bivariate logistic regression with RDS-weighting, the following variables were found to be positively associated with WBL use at the p < .05 level: (1) having an income >15,000 Malawi Kwacha (MKW) per week compared to an income of equal to or <6,000 MKW per week [odds ratio (OR): 3.6; 95% CI: 1.7–7.7]; (2) always using condoms in the last 6 months with any partners compared to not always using condoms in the last 6 months with all partners (OR: 3.0; 95% CI: 1.6–5.6); and 3) family is aware that the participant has sex with other men as opposed to family being unaware that the participant has sex with other men (OR: 2.6; 95% CI: 1.4–4.8). Only older age was found to be inversely associated with WBL use (OR: 0.9; 95% CI: 0.9–1.0), with the likelihood of WBL use decreasing by 6% per every year increase in age (Table 3). Gender identity was not found to be associated with WBL use.

Table 3.

Bivariate and Multivariable Logistic Regression Analysis of Selected Characteristics Associated with Water-Based Lubricant Use Among Men Who Have Sex with Men in Blantyre, Malawi, 2011–2012, with Respondent-Driven Sampling-Weighting (N = 300)

Variable	OR (95% CI)	aOR (95% CI)	p
Age (per 1 year increase)	0.94 (0.88–1.00)	0.89 (0.83–0.95)	<.001
Highest education
Did not complete secondary or high school	Ref.	—	–
Completed secondary or high school	1.18 (0.61–2.30)	—	—
Some tertiary education or higher	2.09 (0.91–4.82)	—	—
Income in the last month
≤6,000 MKW	Ref.	Ref.	—
6,001–15,000 MKW	1.67 (0.76–3.67)	2.04 (0.89–4.70)	.092
>15,000 MKW	3.56 (1.65–7.69)	5.9 (2.48–14.19)	<.001
Reported always using condoms in the last 6 months	3.00 (1.62–5.56)	1.27 (1.06–1.52)	.010
Received information on HIV prevention for sex between men	1.57 (0.75–3.30)	—	—
Family is aware participant has sex with men	2.57 (1.39–4.75)	2.52 (1.29–4.92)	.007
Participant afraid to seek health care because has sex with men	1.48 (0.73–3.03)	—	—
Tested HIV positive as a part of the study	1.46 (0.60–3.55)	—	—

Bold figures significant at the p < 0.01 level in the multivariable regresion.

95% CI, 95% confidence interval; aOR, adjusted odds ratio; OR, odds ratio.

Notably, in bivariate regression, testing positive for HIV as a part of the study, fear of seeking health care, receipt of information on HIV prevention with men, and education level were not found to be associated with self-reported WBL use.

In multivariable logistic regression with RDS-weighting, all of the previous variables retained statistical significance (Table 3). The odds of WBL use for income >15,000 MK compared to 6,000 MKW or less per week [adjusted OR (aOR) 5.9; 95% CI: 2.5–14.2] increased when controlling for other factors. Family being aware that participant has sex with men (aOR: 2.5; 95% CI: 1.3–4.9) and reporting always using condoms in the last 6 months (aOR: 1.3; 95% CI: 1.1–1.5) were still positively associated with WBL use. The negative association between WBL and increasing age (per 1 year increase in age; aOR: 0.9; 95% CI: 0.8–0.9) became slightly more significant when holding other factors constant.

Discussion

There has been an increasing recognition of the specific HIV prevention and treatment needs of MSM across sub-Saharan Africa. To reduce HIV incidence, mathematical modeling has shown that countries will need to adopt multiple HIV prevention interventions, including increased HIV testing frequency, preexposure prophylaxis (PrEP), and condom and CCL distribution and uptake.^6,39

–42 And while there is increasing study of other determinants of the HIV prevention and treatment cascade, there remain limited data available characterizing correlates of WBL use in most settings. The results presented here demonstrate that higher income, increased disclosure of sexual orientation to family, younger age, and increased use of condoms were associated with increased use of WBL.

In these data, less than one-third of the MSM respondents (25% with RDS-weighting) reported using WBL during anal sex with men. Encouragingly, the proportion of MSM reporting using WBL appears to have significantly increased from a study conducted among MSM in Malawi in 2008³³ before the introduction of CCL as a PEPFAR-approved commodity. While not possible to attribute a causal effect, the increase does suggest that increasing availability of funding for CCL is associated with increased availability of both CCL for MSM and uptake of this HIV prevention tool. However, the observed use of WBL is significantly lower than a study among MSM in Nigeria and also than a study of MSM in Tanzania.^24,28 Thus, while the data support an increasing trend in usage between 2008 and 2012, there remain significant unmet needs for WBL overall among MSM included in these studies and very little data on this subject since.

This study found that only about one in ten of participants reported no lubricant use, which regardless of use of CCL, speaks to the interest and need for lubricants during anal sex. In addition, knowledge about the relative advantages of the use of WBL was highly correlated with the actual reported use of these lubricants during sex. Taken together, these results support the notion that if men are made aware of the benefits of proper lubricant use, they will be more likely to use CCL. Across many settings, there is still a dearth in basic HIV prevention education available to MSM.²¹ Yet, there is also supporting evidence that recipient education in the benefits of HIV prevention approaches will make individuals be more likely to use them.⁴³ For example, a study conducted among MSM in Cameroon found that ever accessing a community-based HIV service was positively correlated with ever HIV testing.⁴⁴ Due to the criminalized nature of same-sex practices in Malawi, previous studies have found that health care providers are themselves afraid to provide prevention information specific to MSM due to courtesy or secondary stigma associated with being seen as enabling illegal activity.³⁵ Furthermore, several providers in the same study were unaware of the specific risks of HIV among MSM and encouraged the development of further studies of MSM in Malawi to inform health care decision-making. The results presented here in this context reinforce the need for improved provider education about the epidemiology of and clinical care for HIV MSM and other key populations, even in the context of more generalized HIV epidemics.

In studies across sub-Saharan Africa and in more stigmatizing settings, it is generally easier to engage younger MSM compared with older men, reflected in the composition of study samples, suggesting that younger men are better served by existing programs for MSM^9,45

–49 As hypothesized, this study has found that younger MSM in Blantyre are more likely to use WBL. Given that HIV is a lifelong infection with cumulative HIV acquisition risks, older men have been found to have increased burden in almost every setting, including Malawi. Thus, given the risks associated with nonuse of WBL with condoms during anal sex, these findings suggest a significant risk of onward HIV transmission for those who are not virally suppressed. Moreover, there was limited disclosure of sexual orientation to health care workers with only about one in five participants reported ever telling a health care worker that they had sex with other men. With limited disclosure, there cannot be open dialogue between providers and clients about appropriate risk reduction methods, including CCL use. Moreover, limited disclosure limits the ability of the provider to provide appropriate referrals to community and peer-driven services where CCL may be available and further risk reduction strategies can be discussed. Given the limited disclosure of sexual practices and use of WBL, health care providers do not know of men's sexual orientation and therefore no measures of health care-related stigma were associated with the use of WBL.

Although this study shows high overall interest in the use of lubricants among MSM, proper use of WBL was limited. Therefore, ensuring availability, accessibility, and proper education of CCL use is vital to encouraging further uptake. In accordance with previous studies, higher income when compared to the lowest income was also associated with WBL use independent of knowledge about the utility of WBL. In lower income settings with significant HIV prevalence and incidence, these results would suggest that scaled up CCL provision, including providing this commodity free of charge or at subsidized cost, may be needed to address the needs of those most marginalized. Encouragingly, consistent condom use was also positively associated with WBL use, although because of the cross-sectional nature of the data, it is not possible to determine the direction of the association. Since there are still conflicting findings regarding the safety of CCL use during CAI,²³ prevention messaging should still encourage the use of CCL with a condom.

Limitations

Since this was a cross-sectional study, causality and direction of association between the variables and the use of WBL use cannot be inferred. In addition, the sample size and power calculations were originally based on the difference in prevalence of HIV between consistent condom users and inconsistent users. Thus, these represent post hoc analyses. In addition, the validity of the findings in this study also inherently depends on correct responses provided by the participants, which are subject to social desirability, recall bias, and perceived disclosure issues due to the legal context at the time of the study. To minimize these, we used highly trained and MSM-friendly interviewers, including members of the community and allies. Moreover, the low levels of reported WBL use compared to other studies of MSM and the limited consistent increase since 2008 in Malawi suggest consistency of responses. These indicators have been used across sub-Saharan Africa for MSM and have also been informed with significant formative research across the continent suggesting validity of the measures.^26,50,51 The question used for the outcome of interest did not ask the participants whether they used lubricant with a condom during anal sex. Thus, it is possible that participants reported use of WBL without a condom. Frequency of WBL use during anal intercourse, along with sexual positioning (insertive versus receptive sex), was also not assessed. Also, to create categorical variables, exploratory analysis showed that the categorization for income that best showed its relationship with WBL use was at approximately the 33rd and 66th percentiles, although this does not necessarily align with country standards for socioeconomic levels. Finally, there remains limited consensus around interpretation of the results using RDS-weighting.⁵² Specifically, there is concern that even with RDS-weighting, nonresponse or network size bias may exist or may produce CIs that are too narrow.⁵³ There were also n = 38 participants with incomplete data excluded from the final multivariable analysis. Although this represents a fairly large proportion of missing data (11%), the proportion of data missing for each individual variable was low. The variable for “access to lubricant” was missing the most values (n = 21) with ∼6% missing. Importantly, these data were collected in 2012 and the social contexts affecting gay men and other MSM in Malawi are dynamic. However, in March of 2019, arguments made reminding viewers of laws criminalizing same-sex practices were similar to public discussions in 2012 suggesting the relevance of the data being presented here to inform programs addressing the HIV prevention needs of MSM in Malawi in 2019.⁵⁴

Conclusion

The last decade has increased our understanding of the complex epidemiology of HIV across sub-Saharan Africa with specific risks of HIV acquisition and transmission in specific populations as in every other part of the world. Given the anatomy of the anus and rectum and the biology of HIV, in the absence of PrEP, CAI with serodiscordant and viremic partners is the most effective mode of sexual transmission of HIV.³ In addition, no natural lubricants are available during anal sex necessitating the use of lubricants for both safer and more pleasurable sex. In the context of condom programming, CCL represents a crucial intervention component both specifically for MSM as well as for other key populations. The data presented here can inform programming aiming to decrease CAI by specifically supporting older men who are currently underserved and also consider distribution free to those who are unable to pay. In the context of novel and emerging HIV prevention approaches for MSM, including PrEP, decreasing condomless sex remains an important goal reinforcing the relevance of achieving sufficient coverage of CCL.

Footnotes

Acknowledgments

Research for this study was funded through the U.S. Agency for International Development under USAID | Project SEARCH, Task Order No. 2, Contract No. GHH-I-00-07-00032-00, beginning September 30, 2008, and is supported by PEPFAR. The Research to Prevention (R2P) Project is led by the Johns Hopkins Center for Global Health and is managed by the JHSPH Center for Communication Programs (CCP). The authors also acknowledge United Nations Development Programme (UNDP), United Nations Population Fund (UNFPA), UNAIDS, and Voluntary Services Overseas (VSO) for their additional support. The content and ideas expressed within the article do not engage the responsibility of UNAIDS, USAID, VSO, or UNDP.

Author Disclosure Statement

No competing financial interests exist.

References

UNAIDS: The gap report. Contract No. JC2656. Joint United Nations Programme on HIV/AIDS (UNAIDS), Geneva, Switzerland, 2014.

Baral

, Grosso

, Holland

, Papworth

: The epidemiology of HIV among men who have sex with men in countries with generalized HIV epidemics. Curr Opin HIV AIDS, 2014; 9:156–167.

Patel

, Borkowf

, Brooks

, Lasry

, Lansky

, Mermin

: Estimating per-act HIV transmission risk: A systematic review. AIDS, 2014; 28:1509–1519.

Baggaley

, White

, Boily

: HIV transmission risk through anal intercourse: Systematic review, meta-analysis and implications for HIV prevention. Int J Epidemiol, 2010; 39:1048–1063.

Beyrer

, Baral

, van Griensven

, et al.: Global epidemiology of HIV infection in men who have sex with men. Lancet, 2012; 380:367–377.

Sullivan

, Carballo-Dieguez

, Coates

, et al.: Successes and challenges of HIV prevention in men who have sex with men. Lancet, 2012; 380:388–399.

Brown

, Grosso

, Adams

, et al.: Characterizing the individual, social, and structural determinants of condom use among men who have sex with men in Swaziland. AIDS Res Hum Retroviruses, 2016; 32:539–546.

Twahirwa Rwema

, Lyons

, Ketende

, et al.: Characterizing the Influence of structural determinants of HIV Risk on consistent condom use among female sex workers in Senegal. J Acquir Immune Defic Syndr, 2019; 81:63–71.

Stahlman

, Grosso

, Ketende

, et al.: Depression and social stigma among MSM in Lesotho: Implications for HIV and sexually transmitted infection prevention. AIDS Behav, 2015; 19:1460–1469.

10.

Wendi

, Stahlman

, Grosso

, et al.: Depressive symptoms and substance use as mediators of stigma affecting men who have sex with men in Lesotho: A structural equation modeling approach. Ann Epidemiol, 2016; 26:551–556.

11.

Department of Reproductive Health and Research, World Health Organization: Use and procurement of additional lubricants for male and female condoms: WHO/UNFPA/FHI360: WHO Advisory Note. WHO, Geneva, Switzerland. 2012.

12.

Rosen

, Rosen

: Study of condom integrity after brief exposure to over-the-counter vaginal preparations. South Med J, 1999; 92:305–307.

13.

de Wit

, Sandfort

, de Vroome

, van Griensven

, Kok

: The effectiveness of condom use among homosexual men. AIDS, 1993; 7:751–752.

14.

Crosby

, Diclemente

, Yarber

, Snow

, Troutman

: An event-specific analysis of condom breakage among African American men at risk of HIV acquisition. Sex Transm Dis, 2008; 35:174–177.

15.

Voeller

, Coulson

, Bernstein

, Nakamura

: Mineral oil lubricants cause rapid deterioration of latex condoms. Contraception, 1989; 39:95–102.

16.

Office of the U.S. Global AIDS Coordinator (OGAC): Technical guidance on combination HIV prevention for men who have sex with men. OGAC, Washington, DC, 2011.

17.

USAID: Technical issue brief, condom use: How it relates to HIV and STI prevention. USAID, 2013. Available at www.usaid.gov/sites/default/files/documents/1864/CondomSTIIssueBrief.pdf

18.

USAID | DELIVER PROJECT Task Order 5: USAID contraceptive and condom catalog 2011. USAID, Arlington, VA, 2011.

19.

Geibel

: Condoms and condiments: Compatibility and safety of personal lubricants and their use in Africa. J Int AIDS Soc, 2013; 16:18531.

20.

Sieff

: Tanzania suspends U.S.-funded AIDS programs in a new crackdown on gays. The Washington Post, 2016. Available at www.washingtonpost.com/world/africa/tanzania-suspends-us-funded-aids-programs-in-a-new-crackdown-on-gays/2016/11/23/ec6ced6e-ab5c-11e6-8f19-21a1c65d2043_story.html?noredirect=on&utm_term=.2485de9fe972 accessed July 24, 2019 .

21.

Wilson

, Santos

G-M

, Hebert

, Ayala

: Access to HIV prevention services and attitudes about emerging strategies: A global survey of men who have sex with men (MSM) and their health care providers. MSMGF, Oakland, CA, 2011.

22.

Arreola

, Hebert

, Makofane

, Beck

, Ayala

: Access to HIV prevention and treatment for men who have sex with men: Findings from the 2012 Global Men's Health and Rights Study (GMHR). MSMGF, Oakland, CA, 2012.

23.

Rebe

, De Swardt

, Berman

, Struthers

, McIntyre

: Sexual lubricants in South Africa may potentially disrupt mucosal surfaces and increase HIV transmission risk among men who have sex with men. S Afr Med J, 2014; 104:49–51.

24.

Ayoola

, Sekoni

, Odeyemi

: Transactional sex, condom and lubricant use among men who have sex with men in Lagos State, Nigeria. Afr J Reprod Health, 2013; 17:90–98.

25.

Stromdahl

, Onigbanjo Williams

, Eziefule

, et al.: Associations of consistent condom use among men who have sex with men in Abuja, Nigeria. AIDS Res Hum Retroviruses, 2012; 28:1756–1762.

26.

Baral

, Burrell

, Scheibe

, Brown

, Beyrer

, Bekker

: HIV risk and associations of HIV infection among men who have sex with men in peri-urban Cape Town, South Africa. BMC Public Health, 2011; 11:766.

27.

Jones

, Weiss

, Chitalu

, et al.: Acceptability and use of sexual barrier products and lubricants among HIV-seropositive Zambian men. AIDS Patient Care STDS, 2008; 22:1015–1020.

28.

Romijnders

, Nyoni

, Ross

, et al.: Lubricant use and condom use during anal sex in men who have sex with men in Tanzania. Int J STD AIDS, 2016; 27:1289–1302.

29.

Fay

, Baral

, Trapence

, et al.: Stigma, health care access, and HIV knowledge among men who have sex with men in Malawi, Namibia, and Botswana. AIDS Behav, 2011; 15:1088–1097.

30.

Bureau of Democracy Human Rights and Labor: Country reports on human rights practices for 2011. Malawi. United States Department of State, Washington D.C., 2011.

31.

Sun

: Malawi suspends anti-LGBT law. 2014. Available at www.hrc.org/blog/malawi-suspends-anti-lgbt-law accessed July 24, 2019 .

32.

Wirtz

, Jumbe

, Trapence

, et al.: HIV among men who have sex with men in Malawi: Elucidating HIV prevalence and correlates of infection to inform HIV prevention. J Int AIDS Soc, 2013; 16(Suppl 3):18742.

33.

Baral

, Trapence

, Motimedi

, et al.: HIV prevalence, risks for HIV infection, and human rights among men who have sex with men (MSM) in Malawi, Namibia, and Botswana. PLoS One, 2009; 4:e4997.

34.

Malawi National Government: Global AIDS response progress report: Malawi progress report for 2013. Malawi National AIDS Commission, Lilongwe, Malawi, 2014.

35.

Wirtz

, Kamba

, Jumbe

, et al.: A qualitative assessment of health seeking practices among and provision practices for men who have sex with men in Malawi. BMC Int Health Hum Rights, 2014; 14:20.

36.

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.

37.

Schonlau

, Liebau

: Respondent driven sampling. Stata J, 2012; 12:72–93.

38.

StataCorp: Stata 13.1. StataCorp, College Station, TX, 2014.

39.

Mukandavire

, Walker

, Schwartz

, et al.: Estimating the contribution of key populations towards the spread of HIV in Dakar, Senegal. J Int AIDS Soc, 2018; 21(Suppl 5):e25126.

40.

Shubber

, Mishra

, Vesga

, Boily

: The HIV modes of transmission model: A systematic review of its findings and adherence to guidelines. J Int AIDS Soc, 2014; 17:18928.

41.

MacFadden

, Tan

, Mishra

: Optimizing HIV pre-exposure prophylaxis implementation among men who have sex with men in a large urban centre: A dynamic modelling study. J Int AIDS Soc, 2016; 19:20791.

42.

Boily

, Pickles

, Alary

, et al.: What really is a concentrated HIV epidemic and what does it mean for West and Central Africa? Insights from mathematical modeling. J Acquir Immune Defic Syndr, 2015; 68(Suppl 2):S74–S82.

43.

Yasin

, Delegchoimbol

, Jamiyanjamts

, Sovd

, Mason

, Baral

: A cross-sectional evaluation of correlates of HIV testing practices among men who have sex with men (MSM) in Mongolia. AIDS Behav, 2013; 17:1378–1385.

44.

Park

, Papworth

, Billong

, et al.: Correlates of prior HIV testing among men who have sex with men in Cameroon: A cross-sectional analysis. BMC Public Health, 2014; 14:1220.

45.

Siegler

, Voux

, Phaswana-Mafuya

, et al.: Elements of condom-use decision making among South African men who have sex with men. J Int Assoc Provid AIDS Care, 2014; 13:414–423.

46.

Holland

, Papworth

, Billong

, et al.: Access to HIV services at non-governmental and community-based organizations among men who have sex with men (MSM) in Cameroon: An integrated biological and behavioral surveillance analysis. PLoS One, 2015; 10:e0122881.

47.

Baral

, Ketende

, Mnisi

, et al.: A cross-sectional assessment of the burden of HIV and associated individual- and structural-level characteristics among men who have sex with men in Swaziland. J Int AIDS Soc, 2013; 16(Suppl 3):18768.

48.

Hakim

, Aho

, Semde

, et al.: The epidemiology of HIV and prevention needs of men who have sex with men in Abidjan, Cote d'Ivoire. PLoS One, 2015; 10:e0125218.

49.

Rao

, Stahlman

, Hargreaves

, et al.: Sampling key populations for HIV surveillance: results from eight cross-sectional studies using respondent-driven sampling and venue-based snowball sampling. JMIR Public Health Surveill, 2017; 3:e72.

50.

Baral

, Adams

, Lebona

, et al.: A cross-sectional assessment of population demographics, HIV risks and human rights contexts among men who have sex with men in Lesotho. J Int AIDS Soc, 2011; 14:36.

51.

Mason

, Ketende

, Peitzmeier

, et al.: A cross-sectional analysis of population demographics, HIV knowledge and risk behaviors, and prevalence and associations of HIV among men who have sex with men in the Gambia. AIDS Res Hum Retroviruses, 2013; 29:1547–1552.

52.

McCreesh

, Frost

, Seeley

, et al.: Evaluation of respondent-driven sampling. Epidemiology, 2012; 23:138–147.

53.

Magnani

, Sabin

, Saidel

, Heckathorn

: Review of sampling hard-to-reach and hidden populations for HIV surveillance. AIDS, 2005; 19(Suppl 2):S67–S72.

54.

Nyondo

: CEDEP describes running mates stance on same-sex marriages “razzle-dazzle”. Nyasa Times, 2019. Available at www.nyasatimes.com/cedep-describes-running-mates-stance-on-same-sex-marriages-razzle-dazzle/ accessed July 24, 2019 .