Background As STIs continue to increase in the United States, one possible explanation is that declines in condom use have contributed to these increases. To date, condom use trends from nationally representative data on individuals with opposite-sex partners have not been examined.

Methods We used data from the National Survey of Family Growth (2002, 2006–10, 2011–15) to examine trends in condom use at last sex among unmarried, non-cohabiting women and men aged 15–44 with past-year opposite-sex partners, by race/ethnicity (Hispanic, non-Hispanic white, non-Hispanic black), age (15–24, 25–29, 30–44), any past-year non-monogamy (two or more partners or perceived partner non-monogamy), and past-year STI testing. Year of survey was included as a categorical variable. Chi-square tests and adjusted prevalence ratios were used to test differences in the prevalence of condom use in 2002, 2006–2010 and 2011–2015.

Results Overall, condom use prevalence remained stable among women (2002, 35.5%; 2006–10, 39.1%, 2011–15, 37.4%) and among men (2002, 49.4%; 2006–10, 53.3%, 2011–15, 53.3%), with no differences in temporal trends by race/ethnicity or age in adjusted models. Any reported non-monogamy was also not associated with changes in condom use over time for any group. There was a significant decline in condom use among women aged 30–44 who reported STI testing, from 2006–10 to 2011–15 (36.3%, 95%CI: 29.4–43.1; 25.0%, 95%CI: 20.8–29.2, respectively), and among non-Hispanic black men who reported STI testing, from 2002 to 2006–10 (77.9%, 95%CI: 68.9–86.8; 61.8%, 95%CI: 54.6–69.0, respectively).

Conclusion Overall and for most subgroups, condom use has remained stable over time, suggesting it is not contributing to increases in STI. Models adjusting for demographics suggest these results are not due to demographic shifts. However, certain sub-groups may be using STI testing as a protective strategy when not using condoms. Examination of other explanatory factors is needed.

Disclosure No significant relationships.