| Abstract: | This controlled cohort study aimed to evaluate the safety and efficacy of Norplant contraceptive implants in developing countries. Women initiating Norplant implants were index subjects and women initiating intrauterine devices (IUDs) or surgical sterilization were controls. Consenting participants at 32 clinics in eight developing countries were admitted and followed-up every 6 months for 5 years. Major and less serious health events during follow-up were recorded. Incidence rate ratios of health events adjusted for clinic were estimated for initial and current method use. This paper reports non-reproductive health events. The study involved 7,977 women initiating use of Norplant, 6,625 of IUD, and 1,419 of sterilization. Five years follow-up was completed for 94.6% of the women. The study accumulated 78,323 woman-years of observation. The initial method chosen accounted for 84.4% or more of observed woman-years in users of Norplant, IUD, or sterilization. Twenty-two of the recorded 34 deaths were due to accidents, suicide or homicide. Few deaths or major health events were due to cancer or acute cardiovascular diseases and were not associated with the contraceptive method used. The incidence rates of major health events were low and with two exceptions, there was no significant excess risk of serious morbidity for Norplant users compared with controls; among Norplant initiators gallbladder disease occurred at an incidence rate of 1.5 per 1,000 woman-years and was weakly associated with use of Norplant (rate ratio 1.52 95% C.I. 1.02, 2.27]). For current Norplant users compared to controls, the rate ratio of a combined variable of hypertension and borderline hypertension was significantly elevated (1.81, 1.12, 2.92]). The occurrence of less serious health events was also low and several of them were significantly more often reported among Norplant users. Headache-migraine, weight gain, mood disturbances, pruritus, eczema, and acne had incidence rates among Norplant users of 11.5, 4.5, 2.8, 1.5, 1.4, and 0.9 per 1,000 woman-years, respectively, and were significantly higher than in controls. Respiratory health problems, nonspecific symptoms, and several ill-defined conditions were also significantly more often reported for Norplant users, but some of the excess incidence may be attributable to reporting and detection bias. The study confirms the safety with respect to serious disease of Norplant, IUDs, and sterilization. |
