Background Research indicates perception of risk is necessary for adoption of health protective behaviour. Many studies report optimistic bias in risk perceptions, ie, actual risk exceeds perceived risk. This study assessed the relationship between sexual risk behaviours and perceived risk for acquisition of an STI in a cohort of sexually experienced young women.

Methods Female Marine Corps recruits volunteered to participate in a cognitive-behavioural intervention to decrease STIs and unintended pregnancies that took place during Recruit Training (RT). Data are from the follow-up questionnaire (median=9.8 months post RT) for women who reported vaginal/anal sex since RT (N=1192). Perceived risk was assessed by, I think my chance of getting an STI during my next 6 months in the Corps is about." (0%, 1%, 5%, then 10%–100% in 10% jumps). Sexual risk behaviours since completing RT included # sex partners (1/2/3/4/5+), # casual partners (0/1/2/3+), % of time used condoms during sex (100/90–99/60–80/50/10–40/0), frequency of sex after drinking alcohol/using drugs (never/almost never/sometimes/usually or always), and whether any partner had an STI or was having sex with others (no/possible/yes). Bivariate relationships between sexual risk behaviour and perceived risk were evaluated using negative binomial regression.

Results Perceived risk was low (mean=6.05). All risk behaviours were significantly associated with perceived risk (p<0.001). In general, perceived risk increased as behavioural risk increased except for condom use. Pairwise comparisons show the 0% group mean (2.41) is higher than for the 100% group (1.47, p<0.05), but lower than the 90–99% group (3.96, p<0.05), and much lower than the 60-80%, 50%, and 10-40% groups (9.03, 8.88, and 11.05 respectively; all p<0.001). Restricting the analysis to participants with multiple partners or to unmarried women yielded the same results. Associations between sexual risk behaviour and two other measures of perceived risk (own risk relative to other female Marines, risk of behaviour for getting an STI; not shown) parallel the results described here.

Conclusions There is a major disconnect between women's perception of risk and their condom use, the most effective means for preventing common STIs. Since women must negotiate with and rely on male sexual partner(s) for condom use, future research should further examine social and contextual factors that influence young women's perceptions of risk in relation to condom use.