Extrarelational Sex Among Mexican Men and Their Partners' Risk of HIV and Other Sexually Transmitted Diseases

Abstract

Objectives. This study explored the risk of HIV and other sexually transmitted diseases (STDs) among married and cohabiting women in Mexico City, Mexico, derived from their partners' sexual behaviors.

Methods. Results were derived from the first population-based household survey in Mexico that investigated male sexual behavior. Analyses were restricted to sexually active married or cohabiting men (n = 3990).

Results. Fifteen percent of the men reported extrarelational sex during the past year, 9% reported condom use during last intercourse, and 80% perceived no HIV risk. Most secondary partners were coworkers, mistresses, or friends.

Conclusions. Targeted HIV and STD prevention efforts appear necessary because a substantial number of women may be at risk.

Predominant global AIDS prevention strategies encourage monogamous sexual relationships. However, this strategy has been inadequate for many women. Women's main risk for HIV and other sexually transmitted diseases (STDs) is often their male partners' sexual behavior, and negotiation of safer sex is frequently controlled by men.1–5

The current investigation explored the risk of HIV and other STDs among married and cohabiting women in Mexico by estimating the prevalence of men's sexual behavior outside of their primary relationships. Associations between extrarelational sex and men's sociodemographic characteristics were determined. Other indicators of HIV and STD risk, including condom use rates, the identity of secondary partners, and whether the partners knew of each other, are reported.

METHODS

A multistage, stratified probability household survey was conducted in the Mexico City, Mexico, metropolitan area from June 1992 to March 1993 by National Council for Prevention and Control of AIDS (CONASIDA), at the Mexican Ministry of Health, and the Population Council. A detailed description of the sampling technique is given elsewhere.6

Face-to-face interviews were carried out with a 25-minute structured survey. Of 13 713 eligible men, 8600 men were contacted in person, and only 532 (6%) refused to participate, for an overall response rate of 59% (n = 8068). Analyses indicated no evidence for bias because of nonresponse or underreporting.6 The analysis was restricted to men aged 17 to 60 years who had been married or cohabiting for at least 1 year and who were sexually active with their wife or cohabiting partner during the year before the survey (n = 4099). Thirty-seven men had been eliminated from the analysis because they reported sex with a secondary partner only. Men who did not supply an answer for the outcome variable extrarelational sex (n = 7) or other variables used in the multiple regression analysis (n = 102) were removed from the analysis.

The final sample size of 3990 constitutes 97% of the originally specified sample. Data were weighted to reflect the 1990 census and the probability of selection. The complex survey design (i.e., the clustered sample) was taken into account via SUDAAN.7 Extrarelational sex was defined as intercourse with more than 1 female partner during the year before the interview.

RESULTS

Sociodemographic Characteristics

Respondents ranged in age from 17 to 60 years (mean = 37 years). Eighty-eight percent were married, 31% had a primary-school education or less, and 42% had at least some university education. Almost half (41%) of the respondents were blue-collar workers (e.g., street vendors), and the other half (52%) were white-collar workers (e.g., managers).

Extrarelational Sex

Fifteen percent of the married and cohabiting Mexican men reported extrarelational sex. The number of extrarelational partners ranged from 1 to 30. Only 3 men reported any same-sex sexual behavior during the past year (prevalence = 0.1%). Of the men who reported extrarelational sex, 28% stated that their last sexual partner had been other than their wife or cohabiting partner. The most commonly selected category was “friend” (54%), followed by “mistress” (17%) and “coworker” (14%). Only 5% reported that their last sexual partner had been a “prostitute,” although the number of sexual encounters with prostitutes during 1 year ranged from 1 to 30.

A logistic regression model predicting extrarelational sex was estimated, controlling for education, occupation, age, and marital status (Table 1 ▶). A higher education level and a cohabiting relationship were associated with extrarelational sex.

TABLE 1—

Multivariate Logistic Regression Results Predicting Extrarelational Sex During the Past Year Among Mexican Men (n = 3990)^a: Mexico City, 1992–1993

	OR	95% CI
Age, y
17–29	0.82	0.62, 1.08
30–44	1.05	0.82, 1.34
45–60	1.00
Education
≤Primary	1.00
Secondary	1.56*	1.21, 2.01
≥University	1.51*	1.17, 1.95
Occupation
Blue collar	1.00
White collar	0.98	0.79, 1.21
Other	0.74	0.50, 1.11
Marital status
Married	1.00
Cohabiting	1.81*	1.40, 2.33

Note. OR = odds ratio; CI = confidence interval.

^aAll estimates are weighted to reflect the population and adjusted for the sampling design with SUDAAN.

*P < .001.

Condom Use With Primary and Secondary Partners

Nine percent of the men who reported extrarelational sex indicated that they used a condom during their last intercourse—22% if their last sexual partner was a secondary partner and 4% if she was the primary partner (P < .001). Condom use rates varied among categories of secondary partners (P < .05). Men reported more condom use with friends (21%) and mistresses (17%) than with coworkers (10%). Although few men reported engaging in anal sex during their last sexual activity (n = 17), none of these respondents reported using condoms.

Associations between condom use and key sociodemographic characteristics were assessed via multiple logistic regression analyses (Table 2 ▶). Condom use was uniformly low across all sociodemographic strata, but a higher education level was associated with condom use with a primary partner, and a white-collar occupation was associated with condom use with a secondary partner.

TABLE 2—

Multivariate Logistic Regression Results Predicting Condom Use During Last Intercourse With Primary and Secondary Partners Among Mexican Men Who Reported Extrarelational Sex (n = 579)^a: Mexico City, 1992–1993

	Condom Use With Primary Partner (n = 415)b		Condom Use With Secondary Partner (n = 164)b
	OR	95% CI	OR	95% CI
Age, y
17–29	1.59	0.25, 10.22	1.44	0.45, 4.62
30–44	2.95	0.57, 15.38	0.79	0.30, 2.06
45–60	1.00		1.00
Education
≤Primary	1.00		1.00
Secondary	3.55	0.37, 33.81	1.18	0.38, 3.65
≥University	9.98*	1.37, 72.53	0.64	0.20, 2.08
Occupationc
Blue collar	1.00		1.00
White collar	0.90	0.35, 2.32	3.86**	1.60, 9.34
Marital status
Married	1.00		1.00
Cohabiting	0.28	0.04, 2.07	2.53	0.78, 8.23

^aAll estimates are weighted to reflect the population and adjusted for the sampling design with SUDAAN.

^bRates of condom use with primary and secondary partner were 4% and 22%, respectively.

^c“Other” occupation category was not included in this analysis because of small cell size when restricting the sample to men who reported extrarelational sex.

*P < .05; **P < .01.

Additional HIV and STD Risk Indicators

Of the men who reported extrarelational sex, 89% indicated that their primary partner was unaware of their secondary partner, and one third reported that their secondary partner did not know about the primary. Eighty percent reported no personal HIV risk, and 70% reported never placing another at risk.

DISCUSSION

This investigation explored the degree to which married and cohabiting women in Mexico may be exposed to HIV and other STD risk via the sexual behavior of their male partners. Fifteen percent of the total sample of married and cohabiting men in the Mexico City metropolitan area reported extrarelational sex, so almost 250 000 married and cohabiting men are estimated to have sex with secondary partners over 1 year. If only 5% of these men are exposed to an STD, including HIV, and transmit it to their primary partners, more than 10 000 wives and cohabiting partners could be infected each year. Given a context where the great majority of primary sexual partners and a third of secondary partners are unaware of each other, a substantial number of women would unknowingly be at risk. Importantly, unlike previous reports that have suggested that men's bisexual practices constitute an important pathway for HIV infection among women,8, 9 the present sample showed little evidence of risky bisexual activity.

It would be possible to protect primary partners, with whom almost no condom use was reported in this study, if condoms were always used with secondary partners. However, findings indicated that condom use rates with secondary partners also were low: less than 25%. Condom use was infrequent among men from all sociodemographic strata, although higher education and white-collar occupations were statistically associated with more condom use. None of the men who reported anal intercourse, a behavior particularly associated with HIV risk,10 reported using a condom. The lack of perceived HIV risk found among men likely contributed to the minimal use of condoms.11, 12

Secondary partners were most often friends, coworkers, or mistresses, so men appeared to encounter their secondary partners in their normal social circle. Men with a higher education reported more extrarelational sex than did men with a primary-school education. HIV and other STD prevention strategies should take these findings into account via, for example, workplace interventions. Because some respondents reported sex with up to 30 sex workers during 1 year, programs with sex workers and their clients also could be highly effective.

The combination of substantial amounts of extrarelational sex, minimal condom use, and lack of perceived HIV risk indicates that HIV and other STD prevention efforts that take into account the social context of these risky behaviors are required.