Background Condom use problems including breakage, slippage, or partial use (delayed application or early removal) jeopardise condom effectiveness. Our research aims were to: 1) assess the prevalence of condom use problems among urban US STD clinic users; 2) identify participant and partner characteristics and Behaviours associated with condom use problems; and 3) examine how these characteristics are associated with the odds of experiencing condom breakage/slippage or partial use during vaginal sex.

Methods Patients (n=1609) attending STD clinics in 3 cities (Denver, Long Beach, and San Francisco) between June 2004 and May 2005 were enrolled in a study evaluating the behavioural effects of a video-based waiting room intervention modelling couples overcoming barriers to safer sexual behaviours. Two surveys were conducted (baseline and 3-months) measuring behaviours during the previous 3 months. Bivariate analysis using χ² and multivariable analysis using logistic regression were conducted.

Results At baseline, 767 men and women (median age=26 years) reported using a condom at least once during vaginal sex with their most recent partner (64.4% main and 35.5% non-main) in the preceding 3 months. A majority did not use condoms consistently (62.3%). Among 100% condom users, 152 (52.6%) reported no errors, while 137 (47.4%) experienced errors (56 breakage, 49 partial use, and 32 both errors). Among all users, the per-condom use rates of breakage/slippage, but not partial use, varied significantly by partner type (5.96% main and 9.35% non-main). Multivariable analysis revealed the following characteristics associated with increased odds for condom breakage/slippage: African American race (OR=2.0; CI 1.3 to 3.1), Latino ethnicity (OR=2.0; CI 1.3 to 3.1), drunk/high during sex (OR=1.5; CI 1.1 to 2.1), STI among recent sex partner (OR=1.7; CI 1.2 to 2.3); and main partner status (OR=1.8; CI 1.3 to 2.6); and for partial use: female gender (OR=1.4; CI 1.0 to 1.9), drunk/ high during sex (OR=1.5; CI 1.1 to 2.0), and main partner status (OR=1.4; CI 1.0 to 2.0).

Conclusions In this population of condom users at high risk for STI/HIV, inconsistent condom use and consistent condom use with errors were reported frequently. These results suggest that clinicians should not assume that patients use condoms correctly, and that patients may benefit from condom use counselling tailored to individual and partnership characteristics and behaviours.