Introduction: Little is known about event-level condom use errors during vaginal sex.
Aims: The aim of this article is to examine factors associated with complete condom use compared to forms of incomplete condom use, including delayed condom application and early condom removal, over a 30-day period.
Methods: Data were collected in 2008 from 1,875 men during a prospective daily diary study of sexual behavior and condom use. Incomplete condom use was assessed during vaginal intercourse considering situational, relational, intrapersonal, and behavioral predictors. Analyses were generalized estimating equation multinomial logistic regression.
Main outcome measures: Condom use during vaginal sex: complete use, delayed application, early removal, and both delayed application/early removal.
Results: Of 29,714 diary reports, 30% (8,876) included vaginal intercourse. Analyses were conducted (in 2009) on 83.2% of these intercourse events (6,325), during which condoms were used and for which participants provided complete data. Incomplete condom use included penetration before applying condoms during 12% of events (749), removing condoms before intercourse ended during 3.1% (198), and both types of incomplete use during 2% (129) of events. Erection quality (less than typical erect penile length and circumference, less rigidity, and difficulty maintaining erection) was associated with greater odds of incomplete use, as was a participant's history of other recent incomplete use. Perceptions of condom comfort and recent previous complete use were protective factors against incomplete use.
Conclusions: Men's incomplete condom use is associated with recent condom use patterns, as well as with their physiological characteristics and perceptions of condoms during sexual intercourse. Event-level analyses such as these help describe how incomplete condom use occurs among individuals incorporating condoms into their sexual activities.
© 2010 International Society for Sexual Medicine.