Variation and predictors of vaginal douching behavior

Vaginal douching is a widespread practice among American women. Little research has been done examining variation in the practice or identifying risk factors.

We collected data on douching, as well as hypothesized predictors of vaginal douching, as part of a cohort study on preterm birth. African-American women residing in Baltimore City, Maryland, were enrolled if they received prenatal care or delivered at The Johns Hopkins Medical Institution. Interview data were collected on 872 women between March 2001 and July 2004, with a response rate of 68%. Logistic regression analysis was selected to identify factors associated with douching in the 6 months prior to pregnancy.

Almost two thirds of women reported ever douching and more than two thirds of those women reported douching in the 6 months prior to pregnancy. Variation was seen in the practice of douching with regard to frequency as well as technique. After adjusting for several confounders, prenatal enrollment (odds ratio [OR], 1.80; 95% confidence interval [CI], 1.29, 2.53), more unmet needs for time for "non-essentials" (OR, 1.83; 95% CI, 1.27, 2.63), smoking in the year prior to the birth (OR, 1.78; 95% CI, 1.22, 2.60), and age >19 years (OR, 2.60; 95% CI, 1.36, 4.97) were significant predictors of douching in the 6 months prior to pregnancy.

We identified considerable heterogeneity in the practice of vaginal douching in a cohort of low income African-American women.

Future studies should incorporate measures of the predictors of douching and detailed exposure information to determine the independent contribution of vaginal douching to health outcomes.