Pelvic infection: a comparison of the Dalkon shield and three other intrauterine devices.

A detailed analysis was undertaken of reports of possible pelvic infection in relation to the use of four commonly fitted intrauterine contraceptive devices during 1971 to 1978 in the United Kingdom. The four devices were the Dalkon shield, Lippes loops 3C and 2D, and the Gravigard (copper 7), and data used were those collected systematically through the UK intrauterine device research network. Prospective reports that the Dalkon shield was uniquely related to high levels of infection when compared with other intrauterine devices were not substantiated in this prospective study among 13 349 users. Though some factors such as social class and previous experience of abortion appeared to influence the rate of infection, the type of intrauterine device being worn did not appear to be a significant factor. Various methods of analysis were used including life table, regression, and discriminant analysis, using information relating to the type of intrauterine device worn, the characteristics of the user, the fitting centre, and the pattern of diagnosis and treatment of reported or suspected pelvic infection. The results of this study suggest that fears that the Dalkon shield may be associated with a higher incidence of pelvic infection than other intrauterine devices may have been unjustified.