Indian men's use of commercial sex workers: prevalence, condom use, and related gender attitudes

Abstract

Background/Objectives

Commercial sex represents a critical context for HIV transmission within India and elsewhere. Despite research and programmatic attention to commercial sex workers (CSWs), less is known concerning the male CSW clients considered a bridge population for HIV transmission to the general population and thought to drive demand for the sex trafficking of women and girls. The current study assesses the prevalence of past-year CSW contact, condom non-use therein, and associations with demographic characteristics and gendered attitudes among a national sample of Indian men.

Methods

The nationally representative Indian National Family Health Survey-3 (NFHS-3) was conducted across all Indian states in 2005-2006; the current sample was limited to 46,961 sexually active men. Analyses calculated the prevalence of past-year CSW contact and inconsistent condom use; adjusted logistic regression models were used to evaluate associations of demographic characteristics, sexual entitlement and justification of wife abuse with past-year CSW contact and inconsistent condom use.

Results

Approximately 1 in 100 ( 0.9%) Indian men reported past-year CSW contact; over half of such men reported inconsistent condom use with CSWs. CSW contact was most common among men ages 15-24 (3.6%) and never-married men (9.9%). Men's CSW contact related to higher levels of sexual entitlement (AOR=1.64; 95% CI 1.24, 2.17) and justification of violence against wives (AOR=1.41; 95% CI 1.03, 1.93).

Conclusions

Men's past-year CSW contact was concentrated among young and unmarried Indian men; condom non-use with CSWs was common. Traditional gender ideologies appeared to support men's CSW contact, bolstering consideration of this behavior as a gendered form of HIV risk. Findings provide direction for interventions to reduce men's CSW contact in the Indian context by describing high-risk sub-populations and indicating that gender ideologies should be addressed.

Background

The global HIV epidemic affects over 33 million people and, in multiple national contexts, is directly attributable to commercial sex involving male clients.^1-3 India, home to the third largest HIV epidemic with an estimated 2.4 million infected,⁴ is considered uniquely propelled by the sex industry with an estimated HIV infection prevalence ranging from 5% to over 50% among female commercial sex workers (CSWs).^5-7 Despite extensive research and programmatic attention to CSWs in India,^8-10 far less is known about male CSW clients. Evidence of inconsistent condom use and elevated STI/HIV prevalence among male clients^11-22 indicates the need for etiological research concerning this population, particularly as male CSW clients are thought to function as a bridge population by which HIV infection may spread to both commercial and non-commercial sex partners (e.g., wives),^{1, 11, 16, 23} and, thus, the general population.²⁴

In contrast with other national contexts,^{19, 25-28} little research has assessed Indian men's involvement with CSWs, and none at a population-level. That which is known suggests that CSW involvement may be common among younger men,^29-31 often constituting their first sexual experience,^{30, 32} as well as unmarried men.¹⁵ However, samples studied have typically been drawn from higher-risk groups (e.g., STI clinic patients, truck drivers),^{15, 29, 30, 32, 33} leaving the extent to which this behavior persists in the general population remains unclear. To date, no population-based studies have explored either the prevalence or relevance of demographics or etiologic factors (i.e., related exposures, behaviors, attitudes or beliefs) to Indian men's CSW contact, a primary HIV risk behavior within India.

Critical to such investigation, and increasingly considered central to HIV prevention research and programmatic efforts related to heterosexual transmission within India and elsewhere, are disparate gender dynamics and traditional gender ideologies.⁴ Men's sexual entitlement (i.e., attitudes of entitlement to sexual access and control over sex with female partners) and culturally-sanctioned norms of masculinity that foster power imbalances within sexual relationships are consistently found relevant to men's sexual risk behavior, e.g., multiple sexual partners and condom non-use.^34-39 Men's use of CSWs has been similarly posited as an expression of masculinity,^39-42 and men's perceptions of masculinity and entitlement to sexual access appear to relate to their CSW contact,^{16, 43, 44} and to condom non-use among male CSW clients.^{16, 45} Sexual entitlement also appears to promote male client preference for very young CSWs,⁴⁶ with such male clients increasingly considered to drive market demand for the trafficking of minor girls into sexual exploitation.^{47, 48} Thus, gender ideologies are hypothesized to confer risk for men's CSW contact and higher-risk sex within this context.

To inform the development of much-needed programs to reduce men's CSW contact, the current study describes among a national, population-based sample of Indian men 1) the prevalence of past-year CSW contact and inconsistent condom use in this context, 2) demographic correlates of these behaviors, and 3) associations of CSW contact and related inconsistent condom use with major aspects of men's traditional gender ideologies (i.e., sexual entitlement, endorsement of violence against wives).

Methods

Sample

Data resulting from the India National Family Health Survey 3 (NFHS-3), a national survey conducted in all 29 states of India from November 2005 to August 2006 by the International Institute for Population Sciences (IIPS) and Macro International, was utilized for the present study. The NFHS (referred to as the “Demographic and Health Survey” or “DHS” in other national contexts) is regularly conducted in many developing countries to obtain population-based estimates of major health concerns and risk behaviors. A nationally representative household-based sample was created via a stratified, multistage cluster sampling strategy. Within each state, 2-stage (rural areas) and 3-stage (urban areas) procedures selected a total of 3,850 primary sampling units (PSUs) comprised of 1 or more villages in rural areas, and census enumeration blocks within wards in urban areas; PSU selection probability was proportional to population size. Household enumeration conducted within each PSU formed the sampling frame for systematic selection of households. Participants were recruited in their homes by trained research assistants who asked them if they would be willing to participate in a national study on health. Given the sensitive nature of the data collected, interviews were conducted privately. Prior to data collection, participants were read a standard informed consent document, had an opportunity to have questions answered, and provided written consent for the survey. Further details of data collection and management procedures are available elsewhere.⁴⁹ These procedures identified 85,373 men ages 15-54 eligible for participation, of which 74,369 completed the survey for a response rate of 87% (74,369/85,373).⁴⁹ The analytic sample was limited to men who reported having had sex in the 12 months prior to the survey. Of the 74,369 participants, 63% (n=46, 961) reported having had sex in the prior 12 months, thus meeting inclusion criteria for the current analytic sample.

Data Collection and Measures

Questionnaires were administered verbally via a trained interviewer to minimize potential literacy barriers in either English or the principal language of each Indian state based on the preference of household members. Data were entered into CSPro software and cleaned by IIPS in order to create the final public-use dataset used in the current investigation.

Demographics including age, education, martial status and religion were assessed via single items. A relative index of household wealth was calculated based on interviewer-observed assets including ownership of consumer items and dwelling characteristics; individuals were ranked based on their household score and divided into quintiles with 1 = poorest and 5 = wealthiest 20% of households.⁵⁰

The Justification of Wife Abuse module was based on a modification of the Battering is Justified subscale of the Attitudes and Believes about Wife Beating Scale;⁵¹ respondents were asked if husbands were justified in hitting or beating wives across seven contexts, e.g., “if she neglects the house or the children”. Participants were given the options of “no”, “don't know or depends” and “yes”; Cronbach's alpha for this measure was 0.83. Responses were summed such that a higher score indicated greater endorsement of justification of wife abuse; this scale was subsequently categorized into tertiles for analysis. Men's sexual entitlement was assessed via standard DHS items; participants’ endorsement of four different husband reactions to wives’ refusal of sex, e.g., “use force and have sex with her even if she doesn't want to. Response options were “no”, “don't know or depends” and “yes” (Chronbach's alpha =0.61). Responses were summed such that a higher score indicated greater sexual entitlement. As the distribution of the sexual entitlement scale was skewed, responses were considered bivariately for analysis with the uppermost 25% representing high endorsement of sexual entitlement with the remaining 75% serving as the referent group.

Past-year CSW contact was assessed via two items regarding recent sexual contacts; men reporting that any one of their three most recent sexual contacts was a commercial sex worker or indicating via a second item that they had “paid for sex in the past 12 months” were classified as having had sex with a CSW in the past 12 months. Any report of condom non-use either within an CSW reported as one of their 3 most recent sexual contacts or within a follow-up item concerning condom use with CSWs in the past 12 months indicated inconsistent condom use with a CSW in the past 12 months.

Data collection procedures were approved by the ORC MACRO Institutional Review Board; analyses of these data were deemed exempt from human subjects concerns by the Harvard School of Public Health IRB based on the anonymous nature of the database.

Analysis

The prevalence estimates of past year CSW contact and condom non-use therein were calculated for the overall sample and by demographics. Differences based on demographics were assessed via Wald chi-square analyses; the significance level for all the analyses was p<0.05. To determine the demographic characteristics most relevant to CSW contact and condom non-use, multivariate regression models were constructed to simultaneously consider all demographic characteristics assessed; adjusted odds ratios (AORs) and 95% confidence intervals (CIs) were estimated for each demographic factor. Prevalence estimates were calculated for each scale item, and each level of endorsement across both gendered attitude scales. Differences in prevalences of CSW contact and condom non-use with CSWs based on level of endorsement of each attitude scale were calculated, and logistic regression models estimated the associations of higher levels of endorsement with each outcome, using men with low levels of endorsement of these attitudes in the referent category. Models were adjusted for all assessed demographics (age, education, urban residence, martial status, religion and household wealth). To preserve statistical power, missing values concerning education (n=17) and religion (n=9) were imputed to the largest categories ( no/primary education and Hindu, respectively). Risk estimates generated were evaluated for statistical significance based on 95% confidence intervals not including 1.0. Statistical analyses were performed with use of STATA Version 9⁵² so as to appropriately account for the complex sampling design of the NFHS-3. All analyses were weighted to account for selection probability and non-response using the sample weight for the entire standardized to the current analytic sample size.

Results

Past-Year CSW Contact, Inconsistent Condom Use, and Relations with Demographic Characteristics (Table 1)

Table 1.

Demographics and associations with past year CSW contact, and inconsistent condom use with CSWs among sexually active Indian men (n=46,961)

	Sample (n=46,961)	CSW contact in past 12 months (n=46,961)		Inconsistent Condom Use with CSWs (n= 470 men who had sex with CSWs)
	%* (95% CI)	%** (95% CI)	AOR*** (95% CI)	%** (95% CI)	AOR*** (95% CI)
Total		0.9 (0.8, 1.1)		52.3 (45.7, 58.8)
Age
15-24	12.2 (11.8, 12.7)	3.6 (3.0, 4.4)	2.18 (1.15, 4.12)§	41.3 (32.3, 50.9)	-ref-
25-34	33.8 (33.2, 34.5)	0.8 (0.6, 1.1)	1.80 (1.10, 2.96)§	56.3 (44.8, 67.2)	1.86 (0.90, 3.87)
35-44	33.2 (32.6, 33.9)	0.4 (0.3, 0.6)	1.21 (0.71, 2.04)	69.6 (55.0, 81.1)	2.07 (0.78, 5.52)
45 and above	20.7 (20.2, 21.2)	0.3 (0.2, 0.5)	-ref-	72.5 (50.0, 87.5)	3.24 (0.79, 13.36)
p value		<0.001		0.003
Residence
Rural	66.2 (64.0, 68.2)	0.8 (0.7, 1.0)	-ref-	58.9 (49.6, 67.6)	-ref-
Urban	33.9 (31.8, 36.0)	1.2 (1.0, 1.4)	1.35 (0.99, 1.84)	43.5 (34.9, 52.6)	1.08 (0.58, 1.99)
p value		0.006		0.020
Education
Higher	11.7 (11.1, 12.4)	0.5 (0.3, 0.8)	-ref-	19.7 (8.3, 40.1)	-ref-
Secondary	45.8 (44.9, 46.6)	1.1 (0.9, 1.3)	2.01 (1.20, 3.38)§§	45.8 (38.0, 53.8)	3.87 (1.29, 11.60)§
Primary/none	42.5 (41.5, 43.6)	0.9 (0.7, 1.1)	3.27 (1.84, 5.80)§§§	65.4 (54.1, 75.2)	5.28 (1.54, 18.06)§§
p value		<0.001		0.001
Wealth Index
Poorest	17.7 (16.7, 18.7)	0.7 (0.5, 1.1)	-ref-	81.7 (56.0, 94.0)	5.40 (1.24, 23.47)§
Poorer	19.2 (18.4, 20.0)	0.9 (0.7, 1.2)	1.30 (0.75, 2.24)	52.5 (37.5, 67.0)	1.29 (0.45, 3.69)
Middle	20.4 (19.7, 21.2)	0.8 (0.6, 1.1)	1.17 (0.68, 2.02)	60.4 (46.0, 73.1)	2.44 (0.97, 6.16)
Richer	21.0 (20.2, 21.8)	1.2 (0.9, 1.5)	1.64 (0.94, 2.86)	47.2 (35.7, 59.0)	1.66 (0.75, 3.69)
Richest	21.8 (20.6, 23.0)	1.0 (0.8, 1.3)	1.60 (0.87, 2.93)	34.6 (23.3, 47.9)	-ref-
p value		0.204		0.004
Marital status
Married currently	94.5 (94.1, 94.8)	0.4 (0.4, 0.5)	-ref-	66.5 (56.3, 75.3)	1.61 (0.74, 3.47)
Never married	4.9 (4.7, 5.2)	9.9 (8.4, 11.8)	18.60 (11.35, 30.50)§§§	38.7 (31.0, 47.1)	-ref-
Married formerly	0.6 (0.5, 0.7)	3.6 (1.7, 7.6)	8.16 (3.58, 18.59)§§§	88.3 (47.1, 98.5)	5.30 (0.71, 39.63)
p value		<0.001		0.001
Religion
Hindu	82.5 (81.3, 83.7)	0.8 (0.7, 1.0)	-ref-	57.3 (49.3, 65.2)	-ref-
Muslim	12.0 (10.9, 13.1)	1.2 (0.9, 1.7)	1.31 (0.92, 1.87)	37.6 (25.4, 51.5)	0.50 (0.23, 1.10)
Other	5.5 (5.0, 6.0)	1.6 (1.1, 2.3)	1.49 (0.96, 2.32)	37.3 (21.7, 56.2)	0.64 (0.25, 1.69)
p value		0.016		0.021

^*column percent, i.e., values total 100%

^**row percent, i.e., interpreted as the % among the row descriptor

^***adjusted for all variables in column

^§p<0.05

^§§p<0.01

^§§§p<0.001

Across the total sample, 0.9% of participants reported past-year CSW contact, with just over half (52.3%) of such men reporting at least one instance of condom non-use with CSWs during this period. Chi square analyses revealed several demographic differences in past-year CSW contact and inconsistent condom use. In subsequent adjusted logistic regression analyses, several independent factors related to past-year CSW contact were identified, specifically being ages 15-24 years (AOR 2.18, 95% CI 1.16, 4.10), being ages 25-34 years, (AOR 1.82, 95% CI 1.11 2.99), attaining secondary education (AOR 2.08, 95% CI 1.23, 3.50) or primary education (AOR 3.46, 95% CI 1.96, 6.12), and being never-married (AOR 18.9, 95% CI 11.5, 31.0) or formerly-married (AOR 8.14, 95% CI 3.58, 18.51). Among men reporting past-year CSW contact, factors independently related to inconsistent condom use with CSWs included attaining secondary education (AOR 3.95, 95% CI 1.28, 12.13), primary education (AOR 5.08, 95% CI 1.46, 17.62), and the lowest level of wealth (AOR 4.31, 95% CI 1.13, 16.48).

Endorsement of Violence Against Wives and Sexual Entitlement (Table 2)

Table 2.

Attitudes of Sexual Entitlement and Justification of Violence Against Wives among sexually active Indian men (n=46,961)

	% (95% CI)	% (95% CI)	% (95% CI)
Sexual Entitlement
If wife refuses sex, husband has right to:	Yes	Don't know or depends	No
Get angry	20.5 (19.6, 21.5)	0.7 (0.5, 0.8)	78.9 (77.9, 79.8)
Refuse financial support	5.9 (5.5, 6.4)	0.9 (0.7, 1.0)	93.2 (92.7, 93.7)
Use force for unwanted sex	6.0 (5.5, 6.6)	1.1 (1.0, 1.3)	92.8 (92.3, 93.3)
Have sex with another woman	4.2 (3.8, 4.7)	1.4 (1.2, 1.6)	94.4 (93.9, 94.9)
Violence against Wives is Justified if	Yes	Don't know or depends	No
Wife goes out without telling him	23.6 (22.7, 24.5)	0.3 (0.2, 0.3)	76.2 (75.2, 77.1)
She neglects children	28.2 (27.2, 29.2)	0.2 (0.2, 0.3)	71.6 (70.6, 72.6)
She argues with him	25.5 (24.5, 26.5)	0.5 (0.4, 0.6)	74.1 (73.1, 75.0)
She refuses sex	7.7 (7.2, 8.2)	0.6 (0.5, 0.7)	91.8 (91.2, 92.3)
She burns food	12.2 (11.5, 13.0)	0.3 (0.2, 0.3)	87.6 (86.8, 88.3)
She is unfaithful	23.5 (22.5, 24.5)	1.4 (1.2, 1.7)	75.1 (74.1, 76.1)
She is disrespectful towards in-laws	35.8 (34.7, 36.9)	0.6 (0.5, 0.8)	63.6 (62.5, 64.8)

Participant endorsement of attitudes of sexual entitlement varied across the survey items; 20.5% reported that they felt that husbands had the right to get angry if a wife refuses sex, and 4.2% felt that husbands had the right to have sex with another woman when faced with sexual refusal from wives. Participants expressed endorsement of violence against wives in response to wives showing disrespect towards in-laws (35.8%) neglecting children (28.2%), arguing with husband (25.5%) and being unfaithful (23.5%).

Associations between traditional gender ideologies and CSW contact and inconsistent condom use with CSWs (Table 3)

Table 3.

Associations of Attitudes of Sexual Entitlement and Justification of Violence Against Wives with CSW contact and inconsistent condom non-use with CSWs among sexually active Indian men (n=46,961)

	Sample (n=46,961)	CSW Contact in Past 12 Months (n=46,961)			Condom non-use with CSWs past 12 months (n=470)
	%* (95% CI)	%** (95% CI)	OR (95% CI) p value	AOR*** (95% CI) p value	%** (95% CI)	OR (95% CI) p value	AOR*** (95% CI) p value
Sexual Entitlement
Low	73.9 (72.9, 75.0)	0.7 (0.6, 0.9)	-ref-	-ref-	50.7 (42.6, 58.8)	-ref-	-ref-
High	26.1 (25.0, 27.1)	1.5 (1.2, 1.9)	2.12 (1.63, 2.77) <0.001	1.63 (1.23, 2.16) <0.001	54.4 (43.8, 64.7)	1.16 (0.68, 1.98) 0.587	0.77 (0.43, 1.37) 0.376
Justification for Wife Abuse
Low	49.3 (48.2, 50.5)	0.7 (0.6, 0.9)	-ref-	-ref-	46.5 (37.1, 56.1)	-ref-	-ref-
Moderate	22.0 (21.3, 22.8)	0.9 (0.7, 1.2)	1.28 (0.89, 1.85) 0.182	1.06 (0.73, 1.52) 0.768	38.7 (25.6, 53.7)	0.70 (0.34, 1.44) 0.335	0.64 (0.30, 1.38) 0.255
High	28.6 (27.6, 29.7)	1.3 (1.1, 1.6)	1.91 (1.41, 2.59) <0.001	1.41 (1.03, 1.93) 0.034	63.9 (53.4, 73.2)	1.97 (1.11, 3.51) 0.021	1.46 (0.80, 2.66) 0.211

^*column percent, i.e., values total 100% within column

^**row percent, i.e., interpreted as the % among the row descriptor

^***adjusted for age, education, wealth, urban residence, marital status and religion

Sexually active Indian men demonstrating higher levels of sexual entitlement (26.1% of the sample) were more likely to report past-year CSW contact as compared to those with lower levels of entitlement (1.5% vs.0.72%; OR 2.12 ; 95% CI 1.63, 2.77; AOR 1.6; 95% CI 1.23, 2.16). Similarly, men reporting the highest level of endorsement of violence against wives (28.6% of the sample) demonstrated a greater risk of past-year CSW contact as compared to those reporting lowest levels of endorsement of wife abuse (1.3% vs. 0.7%; AOR 1.41; 95% CI 1.03, 1.93). No such differences were detected among men with moderate levels of endorsement of wife abuse.

Among those men reporting past-year CSW contact, inconsistent condom use was most common among men with the highest endorsement of violence against wives (63.9% vs. 46.5%; OR 1.97, 95% CI 1.11, 3.51), however this relationship attenuated into non-significance in adjusted analyses. No relationships were observed between condom non-use with CSWs and sexual entitlement.

Discussion

Findings from this first population-based investigation of CSW contact among Indian men indicate that approximately 1% purchased sex from prostituted women and girls in the past year. Notably, less than half of men reporting past-year CSW contact indicated consistent condom use in these encounters, suggesting considerable risk for STI/HIV transmission or acquisition within this context. Current evidence that Indian men's attitudes supportive of sexual entitlement and abuse of wives relate to their CSW contact adds to the growing body of work indicating the role of socially-constructed attitudes concerning male entitlement to sexual access to and violence against female partners in promoting HIV risk.^34-39 Given the role of commercial sex in facilitating the spread of HIV in the Indian context and elsewhere,² findings provide needed direction for HIV prevention efforts to address men's CSW contact by informing both the targeting of such efforts as well as their scope.

Consistent with prior work,²⁹ Indian men's involvement with CSWs varied considerably across age categories and marital status with 3.6% of men ages 15-24 reporting past-year commercial sex contact, almost one in 10 (9.9%) never married men reporting this behavior, and 3.6% of formerly married men engaging in this form of HIV risk. The overall population prevalence of past-year CSW contact was low, but was concentrated among distinct demographic groups, suggesting the utility of targeted intervention efforts. Current identification of higher prevalence groups should assist in guiding policies and interventions seeking to reduce this behavior among the general public, as efforts to reduce men's CSW contact or promote condom use therein could have a potentially large impact on the heterosexual spread of HIV.²

Evidence that two thirds of currently married men reporting past-year CSW contact do not consistently use condoms in these encounters bolsters concerns that male clients of CSWs represent a critical bridge population in the HIV epidemic,^{1, 11, 16, 23, 24} and again offers direction in targeting interventions to high-risk populations. In contrast, evidence that the youngest group of men is least likely to report unprotected sex with CSW is promising, and suggests their relatively greater awareness of the need to protect against STI/HIV. Men at the lowest levels of household wealth and education could similarly benefit from condom promotion efforts specific to CSWs.

Consistent with prior research in India and other settings,^{34, 35, 38} results indicate high levels of endorsement of violence and sexual entitlement, with such attitudes focused on control over women's autonomy (e.g., arguing with husband, disrespect towards in-laws). Traditional gender ideologies have previously been found relevant to men's perpetration of physical violence and sexual coercion,^{34-36, 45} thus patterns observed may help explain the high prevalence of violence victimization consistently documented among both women in India⁵³ as well as CSWs.^{7, 54, 55}

New to this body of work, the current evidence that such ideologies relate to past-year CSW contact provides the first empirical support in the Indian context for concern that social norms of male control and sexual privilege foster men's use of women in prostitution.⁴² These findings add to a growing international body of evidence^{16, 44} indicating that men's use of CSWs is gendered, and that intervention efforts targeting this high risk behavior should consider the continuum of men's entitlement to power and control over women's sexuality.⁴⁴

Findings are particularly timely as HIV prevention strategies^{34, 56} begin to incorporate modification of norms that legitimize men's control and sexual risk-taking. Within India, efforts targeted to young men with the goals of promoting gender equity and address masculinity⁵⁷ to reduce HIV risk have demonstrated preliminary effectiveness in modifying gender attitudes.³⁴ Currently these programs do not explicitly target men's CSW contact, but current evidence that major elements of masculinity (attitudes of sexual entitlement and justification of wife abuse) relate to men's CSW contact, coupled with identification of youth as a high-risk sub-population for such behavior, indicates that HIV prevention programs focused on promoting gender equity among youth may also be effective in reducing CSW contact. Beyond direct HIV risk considerations, efforts to reduce men's CSW contact have been recommended^{23, 48, 58, 59} based on the reported role of men's demand for commercial sex in promoting coercive, forced, and fraudulent entry of women and girls into prostitution (i.e., sex trafficking).^{48, 58, 59} Thus, current findings regarding sub-populations at greatest risk for CSW contact, as well as evidence of elevated risk based on traditional gender ideologies, may also inform programmatic efforts to decrease men's CSW contact as a means to reduce such sexual exploitation.

Findings should be considered in light of several limitations. All measures were self-reported; in particular, accurate assessment of CSW contact may hindered by factors such as the sensitive nature of discussing sexual activity, including that with CSWs.⁶⁰ Moreover, as transactional sex may also include exchanges of sex for resources such as food, safety, or other material goods,⁴² the current assessment is limited in referring only to paid sex. These considerations likely render the currently calculated prevalence of CSW contact an underestimate; future research will benefit from a more inclusive definition as well as assessment of additional qualities (e.g., chronicity, age at first CSW contact, context of CSW contact) which may serve to normalize such experiences and thus improve accuracy of reporting. The household-based sample may have also contributed to underestimation of CSW contact; both truckers and men who migrate for labor have demonstrated high levels of CSW contact^{33, 61} yet may be underrepresented within the current sample. Neither occupation nor migration-related characteristics were available, however these factors are recommended for further study to best tailor intervention efforts. Use of a past-year assessment may have limited the ability to detect effects of gendered norms on this behavior, as men who may have had CSW contact prior to the past year may report similar levels of endorsement of gender role attitudes. Cross-sectional analyses do not allow determination of the relative chronology of CSW involvement in relation to endorsement of gender ideologies; it is possible that men's gender ideologies were influenced as a result of their CSW contact.

Currently, the vast HIV prevention infrastructure addressing commercial sex as a context for transmission almost exclusively targets female CSWs,⁴² with few efforts targeting male clients and no such interventions identified within the peer-reviewed literature. Thus, the current study begins to fill a critical gap by describing this population and identifying both high-risk groups as well as modifiable risk factors for such behavior, i.e., socially-constructed gendered attitudes concerning male entitlement to sexual access to, and violence against, women. Given the major role of heterosexual transactional sex in facilitating the spread of STI/HIV in India and elsewhere,² as supported by present findings of extensive condom non-use with CSWs, programmatic and research efforts to address men's CSW contact as a critical form of HIV risk behavior should be prioritized.