Sexual intercourse remains the leading route of HIV transmission among women in the USA, accounting for more than 80% of new diagnoses.³ Many women become infected through their main partner due to low rates of consistent condom use in primary relationships.^5,6 Because male condom use is not directly under a woman's control, gender inequalities may be particularly important in shaping this sexual risk behavior.

The theory of gender and power provides a framework that may help explain condom use within romantic relationships. It asserts that power dynamics between men and women can be explained by three major structures: sexual division of power (i.e., male partner dominance within relationships), sexual division of labor (i.e., economic inequality), and structure of cathexis (i.e., gender norms). These structures are maintained through societal and institutional social mechanisms and are manifest in relationship control, women's economic potential, and expectations of women's role in society.^7,8

The sexual division of power is particularly relevant for understanding risk among low-resource women in the context of interpersonal relationships. Several studies have identified this power imbalance and have described its impact on sexual risk behavior. For example, in a study of urban Black men ages 18–24, the overwhelming majority believed that condom use was the man's decision. Only a small percentage conceded that their female partner may play a “small role” in choosing to use a condom.⁹ Another study of Black teenage women reported that 26% felt little control over condom use. Sixty-six percent felt that a male sex partner would be hurt, insulted, angry, or suspicious if questioned about his HIV risk factors.¹⁰ Additionally, studies report that demanding condom use would show a lack of trust in the relationship.^10–12 Furthermore, in one study, Black teenage women sometimes feared rejection or violence when considering asking a partner to wear a condom.¹⁰ Therefore, urging women to demand condom use and encouraging monogamous relationships may not be an effective prevention strategy for women that do not perceive control in sexual decision-making.

Economic inequalities derived from the sexual division of labor may further exacerbate the power imbalance in romantic relationships. Women who are economically dependent on their partners may yield less power in romantic relationships and therefore less control over sexual decision-making. Empirical studies have linked economic inequality and poverty with increased HIV risk.^1,13,14 For example, populations with lower incomes have been found to use condoms less and have higher rates of HIV/AIDS than populations with higher incomes.^1,14 In a recent qualitative study with African American women, lack of economic opportunities and economic dependency were cited as compelling reasons for remaining in risky sexual relationships.¹³ Reduced economic autonomy may lead to the inability to maintain monogamous relationships or assert their sexuality. Additional literature suggests that one of the components of relationship power may be an imbalance of valued resources (e.g., economic and emotional) between partners.^7,15,16

Twenty-eight percent of women reported being economically dependent on a male partner. In bivariate analysis, those dependent on a partner were more likely than those not dependent on a male partner to be Hispanic/Latina and White/Other, to have abstained from alcohol in the past 6 months, to be in a longer relationship, to be in a very committed relationship, and to have low sexual assertiveness (p<0.05; Table 1).

In unadjusted analysis, young women dependent on a male partner were 1.6 times more likely to report not using a condom the last time they had sex with that partner than women not dependent on their partner (95% CI=1.11–2.30; Table 2, Model I). Specifically, of those who were economically dependent on their partner, 67.0% did not use a condom compared to 55.8% who were not economically dependent on their partner (p=0.010; data not shown).

Forward stepwise regression including potential confounders resulted in a final multivariate model that retained economic dependence and age in years. After adjusting for age, results remained virtually unchanged (AOR=1.61; 95% CI=1.11–2.32; p=0.01; Table 2, Model II).

Over a century of research has been conducted on social determinants of health, highlighting the importance of social context in shaping health and health behaviors.^27–29 Specifically, the decision to use a condom, while often understood as an individual choice, may be strongly impacted by many external factors. As a result, attempts to improve condom use by intervening solely at the individual or interpersonal levels may have limited effectiveness. In fact, as this study highlights, improving condom use may require a multilevel approach that accounts for social factors, such as economic dependence on a male partner.¹¹

This study is an important contribution to the literature as it addresses a novel research question for a widespread public health problem in a highly underserved and vulnerable population—young, low-income women in early motherhood. However, it is important to note a few limitations. First, this study relied on self-report, introducing the possibility of recall bias. Second, the data analyzed were cross-sectional; therefore, the temporality of associations cannot be assumed. Third, as the sample consisted of young mothers 1-year postpartum, findings may not generalize to the general population of young women. Young parents may be more vulnerable to influences of economic dependence because of the desire to provide for their child. Similarly, the final sample may not be representative of those eligible for the parent study or of those who participated in baseline data collection. However, as previously described, no significant differences were seen between the final sample and baseline-only participants in terms of baseline sociodemographic characteristics and baseline economic dependence and condom use. Additionally, because this was a secondary data analysis, only a single variable was available to measure the construct of economic dependence. This may have led to non-differential misclassification, thus potentially underestimating the relationship. However, this study is able to show the utility of a brief measure of economic dependence on sexual risk of young women, having relevance for clinicians, public health practitioners, and researchers because they can use this as a quick assessment of economic dependence and potentially uneven economic relationships between men and women. Similarly, while used in other studies,²⁰ the variables used to measure sexual assertiveness may not accurately reflect the construct.