To assess risk factors for vulvar lichen sclerosus data were collected in a case-control study of 75 cases and 225 age-matched control subjects. Questions concerned personal characteristics and habits, gynecologic and obstetric data, general indicators of sexual habits, and selected dietary habits. The risk of vulvar lichen sclerosus was greater in parous women than nulliparous women, but there was little tendency for the risk to increase with number of births; the finding was not significant when only married women were considered. There was no material difference between cases and control subjects in relation to major indicators of sexual habits (age at first intercourse and number of sexual partners). A smaller proportion of cases was married, but no difference was observed in the distribution of cases and controls with reference to education, smoking habits, body mass index, and previous history of diabetes. There was no association between retinoids and risk of vulvar lichen sclerosus, but intake of carotenoids was inversely and strongly associated with vulvar lichen sclerosus. This apparent protection did not materially change after adjustment for socioeconomic status and other potential confounding factors.