To examine whether the risk of pelvic inflammatory disease associated with intrauterine device (IUD) use varies with a woman's sexual behavior, we analyzed data from the Women's Health Study, a hospital-based, case-control study carried out in the United States from 1976-1978. The cases were 657 women hospitalized with pelvic inflammatory disease; controls were 2566 women hospitalized with nongynecologic conditions. After controlling for confounding factors, we found no consistent differences in the risk of pelvic inflammatory disease associated with IUD use among women in different categories of gonorrhea history, frequency of intercourse, or number of recent sexual partners. However, among women with only one sexual partner, married and cohabiting women had little appreciable increased pelvic inflammatory disease risk associated with IUD use compared with those using no contraception, whereas previously and never-married women using IUDs had relative risk estimates of 1.8 and 2.6, respectively. These results suggest that women at low risk of acquiring sexually transmitted infections have little increase in the risk of pelvic inflammatory disease from use of an IUD.