Advancing Sexual and Reproductive Health and Rights
International Perspectives on Sexual and Reproductive Health
Volume 37, Number 3, September 2011

Poverty, Multiple Partners Linked to HIV Infection Among Indian Women

In India, women who are poor or have had multiple sexual partners, and those whose husbands are nonmonogamous or sexually coercive, have an increased likelihood of testing positive for HIV, according to a national household-based survey.1 Among both women overall and those who were married, the odds of having HIV were elevated if their household income was below the national poverty line (odds ratios, 1.6–1.8) or if they had had more than one lifetime sexual partner (5.2–6.0). Other risk factors included being aged 26–35, and having a husband who had other wives or partners, had been tested for HIV or had ever forced them to perform unwanted sexual acts.

The data came from the third Indian National Family Health Survey, which was conducted in 2005–2006 in every state except Nagaland. Men aged 15–54 and women aged 15–49 were eligible for the survey. Because HIV transmission to women often occurs within marriage in India (only 6% of HIV-positive women in the survey had never married or cohabited), the investigators assessed risk factors for infection not only among all sexually experienced women, but also among a subsample of women whose husband had participated in the survey. Respondents were interviewed about their socioeconomic, demographic and behavioral characteristics and had blood taken for HIV testing. The final sample consisted of 37,781 sexually active women, including 22,684 married women whose husband also completed the survey. The researchers used logistic regression to identify associations between women's and husbands' characteristics and women's HIV status.

The vast majority of women were married (93%). Roughly one-fourth were aged 15–25, two-fifths were aged 26–35 and one-third were aged 36–49. About half lived in an urban area (47%), and one in four belonged to a household that qualified for government assistance (24%). A substantial minority (25%) had never heard of AIDS. Very small proportions reported having had a genital sore in the past 12 months (2%) or more than one lifetime sex partner (2%). Among married women, 2% reported that their husband was not monogamous and 7% had a partner who had taken an HIV test in the past. Seven in 10 husbands had at least a secondary education.

A logistic regression model that included more than two dozen variables—all of those associated with HIV incidence in univariate models, as well as factors identified in previous studies—revealed that the odds of infection were higher among married than unmarried women (odds ratio, 2.5), and were especially elevated among those who were separated, divorced or widowed (9.5–22.7).

Among both all women and those who were married, women aged 26–35 were more likely to test positive than older women (odds ratios, 2.5–3.7). The likelihood of infection was also elevated among women living below the poverty line (1.6–1.8), those who had had a genital sore in the past 12 months (3.0–3.2) and those who had had more than one lifetime sexual partner (5.2–6.0).

In analyses restricted to married women whose husband had completed the survey, the odds of HIV infection were elevated among those whose husband had other wives or partners (3.4), had been tested for HIV (2.8) or had ever forced the respondent to perform any unwanted sexual acts (2.6). Women whose husband had a secondary or higher education were less likely than those whose husband had no schooling to be HIV-positive (0.4).

The investigators caution that although the study was based on a large national sample, the generalizability of the findings is limited because of the exclusion of households in Nagaland, which has a very high HIV rate, and because the study excluded women with missing data (who differed from those in the analysis in several respects). Other limitations, they add, are the cross-sectional design of the survey and the possible impact of social desirability bias. Nonetheless, the authors suggest that the associations between husbands' behavior and women's infections warrant particular attention; they advocate increased HIV education efforts that promote "condom use both within and outside marriage" and "emphasize respect for women and wives."—S. Ramashwar


1. Ghosh P et al., Factors associated with HIV infection among Indian women, International Journal of STD & AIDS, 2011, 22(3):140–145.