Introduction Little research has explored how love affects condom use among women engaging in transactional sex. Most interventions for these women focus on non-romantic partners who may not confer the most sexual risk for HIV/STI.

Methods 26 women were enrolled into 4-week study during which event-level information about mood and sexual intercourse was collected twice daily via cell phone. We used descriptive statistics and binary logistic regression with generalised estimating equation correction to test associations between person and event-level factors and condom use during transactional sex events.

Results Participants reported 88 paid/traded vaginal intercourse events; 50.2% were condom protected. After controlling for partner type, feeling in love on a given day was significantly associated with higher odds of condom use, while time of day and day of were significantly associated with lower odds of condom use. There was a significant interaction between being in love and using condoms (p = 0.03). When participants reported being in love, they used condoms during 57.1% of events with romantic partners and 61.1% of events with non-romantic partners. When women did not report being in love, they reported condom use in 64.3% of events with non-romantic partners and 54.0% of events with romantic partners.

Conclusion There was a strong association between being in love on a given day and condom use during transactional events. This association was moderated by partner type such that when women reported being in love they were more likely to use condoms with non-romantic partners and less likely to use condoms with romantic partners. Love may mark increased STI/HIV risk in romantic relationships, especially if condoms are seen as barriers to intimacy that distinguishes romantic from non-romantic partners. Unpacking women’s partner-specific perceptions of risk may be help women make the link between affective states and preventive behaviours.

Disclosure of interest statement The study was funded by the Indiana University Purdue University Indianapolis Centre for Urban Health and NIDA grant T32 DA 023356.