Prospective associations between perceived barriers to condom: Use and "perfect use"

Background Identifying malleable predictors of condom use in a clinic-based population may benefit efforts to prevent sexually transmitted infections (STIs). Purpose To prospectively test associations between three measures (relational-trust factors, fit and feel, and dislike of condom use) and perfect condom use in patients attending clinics diagnosing STIs. Methods A convenience sample was recruited from five clinics in three U.S. cities. Data were collected from December 2007 through April 2011. Daily electronic diaries were completed for up to 180 days. Occasions of penile-vaginal intercourse (PVI) involving condom use without any of four errors/problems were classified as "perfect use." Three subscales (relational-trust factors, fit and feel, and dislike of condom use) were developed from baseline data. Generalized estimating equations were used to account for non-independence of PVI events. Results Among 17,156 reported occasions of PVI, condoms were either not used or used with errors/problems in 8,857 (51.6%) instances. The remaining 8,829 occasions (48.4%) involved perfect use. Relational-trust (p=0.054) and fit and feel (p=0.13) issues were not significantly associated with perfect use. Dislike of condom use (p=0.005) was significantly associated with perfect use (estimated OR=0.93, 95% CI=0.89, 0.98). Significant interactions with race, age, or gender were not observed. Conclusions Clinic attendees may be more likely to use condoms perfectly if three perceptions are reduced in magnitude: I won't use condoms, condoms spoil the mood, and I get turned off when my partner suggests we use condoms. Findings support a paradigm shift in the way clinics promote condom use to patients.