Abstract
Most college students are sexually active, engage in serially monogamous relationships, and use condoms inconsistently. Little is known about how condom use changes during college, and even less about variables predicting changes in use. Latent growth modeling (LGM) was used to examine changes in condom use during the first year of college among 279 women (mean age = 18.0; 74% White), who provided monthly reports on condom use frequency. At study entry, participants also reported on theoretically suggested risk and protective factors. Predictors of changes in use were examined after controlling for use of alternative contraception and partner type. LGM showed that women decreased their condom use during the first year of college. Levels of condom use were initially lower among women with strong alcohol–sexual risk expectancies, women with more previous sexual partners, women who did not smoke marijuana, and African American women. Decreases in condom use were greater among women with lower grade point averages, women from lower socioeconomic status families, and women who engaged in binge drinking. Reductions in condom use may place women at greater risk of unplanned pregnancy and sexually transmitted infections. Identification of factors associated with decreases in condom use will enable targeted educational and intervention efforts.
Acknowledgments
This research was supported by Grant R21-AA018257 awarded to Michael P. Carey from the National Institutes on Alcohol Abuse and Alcoholism.
Notes
Note. Means and standard deviations (in parentheses) are reported, based on those providing data at each time point (T).
a Frequency of condom use is reported on a scale from 1 (never) to 5 (always).
b Average of reports for romantic and casual partners if both types of partners were reported.
c Reports are based only on those engaging in intercourse during each month.
d Reports are based only on those reporting a romantic relationship each month.
Note. N = 279. B = standardized regression coefficients; CI = confidence interval. Only significant coefficients are reported. Analyses control for use of alternative contraception and having casual partners. χ2 (290, N = 279) = 497.16, p < .001 (comparative fit index = .84, Tucker–Lewis index = .83, root mean square error of approximation = .05, standardized root mean square residual = .04).
*p < .05. **p < .01. ***p < .001. †p < .10.
Note. N = 222. B = standardized regression coefficients; CI = confidence interval. Only significant coefficients are reported. Analyses control for use of alternative contraception. χ2(213, N = 222) = 347.54, p < .001 (comparative fit index = .88, Tucker–Lewis index = .88, root mean square error of approximation = .05, standardized root mean square residual = .05).
*p < .05. **p < .01. ***p < .001. †p < .10.