The relationship of depot-medroxyprogesterone acetate (D-MPA) use to risk of cervical carcinoma in situ was investigated using data from a large multi-national, hospital-based case-control study. To avoid possible detection bias from Pap smear screening, final analyses were restricted to a subset of cases with symptoms at the time of their diagnosis of cervical carcinoma in situ. Relative to nonusers, the risk was elevated in women who had ever used DMPA and increased with duration of use. Decreasing trends in relative risk with times since first and last uses were observed in long-term users. Results from another portion of this same study did not show a relationship of invasive cervical cancer to DMPA use. These findings suggested that if DMPA increases the risk of cervical carcinoma in situ then either this is a reversible effect, or the cervical lesions induced by DMPA tend not to progress to invasive disease.