Rates of complication in first-trimester manual vacuum aspiration abortion done by doctors and mid-level providers in South Africa and Vietnam: a randomised controlled equivalence trial.

BACKGROUND: We assessed whether the safety of first-trimester manual vacuum aspiration abortion done by health-care providers who are not doctors (mid-level providers) is equivalent to that of procedures done by doctors in South Africa and Vietnam, where mid-level providers are government trained and accredited to do first-trimester abortions. METHODS: We did a randomised, two-sided controlled equivalence trial to compare rates of complication in abortions done by the two groups of providers. An a-priori margin of equivalence of 4.5% with 80% power and 95% CI (alpha=0.05) was used. 1160 women participated in South Africa and 1734 in Vietnam. Women presenting for an induced abortion at up to 12 weeks' gestation were randomly assigned to a doctor or a mid-level provider for manual vacuum aspiration and followed-up 10-14 days later. The primary outcome was complication of abortion. Complications were recorded during the abortion procedure, before discharge from the clinic, and at follow-up. Per-protocol and intention-to-treat analyses were done. This trial is registered at with the identifier . FINDINGS: In both countries, rates of complication satisfied the predetermined statistical criteria for equivalence: rates per 100 patients in South Africa were 1.4 (eight of 576) for mid-level providers and 0 for doctors (difference 1.4, 95% CI 0.4 to 2.7); in Vietnam, rates were 1.2 (ten of 824) for mid-level providers and 1.2 (ten of 812) for doctors (difference 0.0, 95% CI -1.2 to 1.1). There was one immediate complication related to analgesics. Delayed complications were caused by retained products and infection. INTERPRETATION: With appropriate government training, mid-level health-care providers can provide first trimester manual vacuum aspiration abortions as safely as doctors can.

Morbidity and mortality patterns in HIV-1 seropositive/ seronegative women in Kampala and Harare during pregnancy and in the subsequent two years.

OBJECTIVE: To compare birth outcomes, hospital admissions and mortality amongst HIV-1 seropositive and HIV-1 seronegative pregnant women in Kampala, Uganda and Harare, Zimbabwe. DESIGN: In Kampala and Harare about 400 HIV-1 seropositive and 400 HIV-1 seronegative pregnant women were recruited at initial visit for antenatal care into a prospective study and followed for two years after delivery. The women were classified as HIV-1 seropositive at recruitment if initial and second ELISA tests were positive and confirmed by Western Blot assay. Data on demographic, reproductive, contraceptive and medical histories were obtained using a comprehensive questionnaire at entry, 32 and 36 weeks gestation, at delivery and at six, 12, and 24 months post delivery. In addition, a physical examination and various blood tests were performed at each antenatal and post natal visit. RESULTS: During the two years after delivery, HIV-1 seropositive women had higher hospital admission and death rates than HIV-1 seronegative women. HIV-1 seropositive mothers had a two-fold increase in risk of being admitted to hospital (Kampala: RR = 2.09; 95% CI = 0.95 to 4.59; Harare: RR = 1.98; 95% CI = 1.13 to 3.45). In the six weeks after delivery eight deaths occurred, six of which were among HIV-1 seropositive women and in the period from six weeks to two years after delivery, 53 deaths occurred, 51 of which were among HIV-1 seropositive women (Kampala: RR = 17.7; 95% CI = 4.3 to 73.2; Harare: RR = 10.0; 95% CI = 2.3 to 43.1). However, there was no difference in hospital admission rates between HIV-1 seropositive and seronegative women during pregnancy itself and there was only one death during that period (in a HIV-1 seronegative woman). There was no difference in the frequency of complications of delivery between HIV-1 seropositive and HIV-1 seronegative women and the outcome of births were also similar. CONCLUSIONS: A significant number of HIV-1 positive pregnant women presented at both Harare and Kampala although there was no difference in the number of hospital admissions or mortality between HIV-1 seropositive and HIV-1 seronegative women during pregnancy. Although there were no differences in complications during pregnancy or outcome at delivery, in the two years after delivery, HIV-1 seropositive women in both centres were at increased risk of being admitted to hospital and of dying.

Safety and efficacy of levonorgestrel implant, intrauterine device, and sterilization.

OBJECTIVE: To evaluate safety and efficacy of levonorgestrel-releasing contraceptive implants (Norplant; Leiras Oy, Turku, Finland) in developing countries. METHODS: We used controlled cohort methodology. Women attending family planning clinics in eight developing countries selecting Norplant were enrolled, together with women of similar age choosing intrauterine devices (IUDs) or surgical sterilization. Participants were interviewed and examined at semi-annual visits and followed-up for 5 years regardless of change of contraceptive methods. Incidence rate ratios of health events were estimated for initial and current method use. RESULTS: Altogether, 7977 women initiated Norplant, 6625 IUD, and 1419 sterilization. The overall follow-up rate was 94.6% and 78,323 woman-years of observation were accumulated. Pregnancy rates for Norplant, copper IUDs, and sterilization each averaged less than 1 per 100 woman-years. With two exceptions, no significant excess risk of serious morbidity was detected for Norplant users compared with controls. The incidence of gallbladder disease was higher in women who initiated Norplant use than in controls (rate ratio 1.52, 95% confidence interval [CI] 1.02, 2.27), as was the incidence of hypertension and borderline hypertension in current implant users (rate ratio 1.81; CI 1.12, 2.92). Other new findings were increased risks of respiratory diseases and decreased risks of inflammatory disease of the genital tract in Norplant users compared with IUD users and sterilized women. CONCLUSION: The study confirms the safety with respect to serious disease and the high contraceptive efficacy of Norplant, copper IUDs, and sterilization.

Steroid hormone contraception and bone mineral density: a cross-sectional study in an international population. The WHO Study of Hormonal Contraception and Bone Health.

OBJECTIVE: To evaluate relationships between bone mineral density and use of steroid hormonal contraceptives. METHODS: This was a multicenter cross-sectional study in seven centers in three regions of the developing world from April 1994 to June 1997. Women 30-34 years old attending family planning clinics, with at least 24 months of lifetime use of combined oral contraceptives (OC), depot-medroxyprogesterone acetate (DMPA), or levonorgestrel implants, or no or only short-term (less than 6 months) use of steroid hormonal contraceptives, had bone mineral density (BMD) measured at the distal radius and the midshaft of the ulna using single-photon x-ray absorptiometry. RESULTS: In the study, 2474 women were examined. For OC use, adjusted mean BMD was significantly higher in short-term, current users compared with women who never used hormonal contraceptives. For DMPA and levonorgestrel implants, adjusted mean BMD was statistically significantly lower in short-term current users compared with those who never used hormonal contraceptives. For all three hormonal methods, there were no significant differences in BMD between past users of hormonal contraceptives and never users, even among those who had used the methods for 4 or more years. The magnitude of changes in BMD was small and less than one standard deviation (SD) from the mean of those who never used steroid contraceptives. CONCLUSION: This study suggests that hormonal contraceptive use by young adult women is associated with small changes in BMD that occur early after initiation of use and are reversible.

Risk of cardiovascular diseases associated with oral progestagen preparations with therapeutic indications.

Data from the WHO Collaborative Study suggest that use of progestagen preparations with therapeutic indications are associated with a significantly increased risk of venous thromboembolism but not stroke or myocardial infarction.

Effect on stroke of different progestagens in low oestrogen dose oral contraceptives. WHO Collaborative Study of Cardiovascular Disease and Steroid Hormone Contraception.

The multicentre WHO collaborative case-control study had limited power to differentiate stroke risks by progestagen component in low dose combined oral contraceptives. Given this caveat, odds ratios for ischaemic, haemorrhagic, and all strokes combined associated with use of oral contraceptives containing desogestrel or gestodene ("third generation") were found to be similar to those containing levonorgestrel in this study.

Cardiovascular disease and combined oral contraceptives: reviewing the evidence and balancing the risks.

Cardiovascular risks have been a concern since combined oral contraceptives (OCs) were first introduced. In the past four years new, mostly reassuring information on the safety of modern, low oestrogen dose OCs has become available. However, in 1995 the new information showed higher venous thromboembolism (VTE) risk for OCs containing desogestrel and gestodene compared with levonorgestrel- or norethindrone-containing OCs. The controversial responses by national authorities, their scientific and public health merits were hotly debated and many considered the differences in risk small and resulted from bias and/or confounding. We discuss these arguments and conclude they lack empirical support or cannot account for the 2-fold increased risk. The risk of ischaemic stroke and myocardial infarction (MI) associated with low oestrogen dose OCs are very small in women without cardiovascular risk factors, while increased risk of haemorrhagic stroke is confined to women >35 years of age. Applying the most recent risks to models of OC-attributable events and deaths, OC-attributable mortality in women <35 years is estimated to be <3 per million users annually, rising to about 10 per million users annually among smokers. In the context of external cause mortality (about 90 per million women of reproductive age annually in the UK) such risks appear small. Over the age of 35 years, OC-attributable mortality is a more important concern, particularly among smokers. In the absence of any appreciable OC-attributable mortality in young healthy women, the additional VTE risk for third compared with second generation OCs should be considered when women choose which OC to use.

Norplant consensus statement and background review.

This review has highlighted the attributes of a very important new method of contraception. The signatories to this document agree that, with the provision of appropriate information and instruction for the user, Norplant is a good contraceptive choice to be made available worldwide in family planning programs that have the resources for appropriate training and counseling. The signatories to this document are acting in their own personal capacity and not as representatives of any particular organization.

PIP: The Norplant contraceptive implant has been registered in 60 countries and used by over 6 million women worldwide. Clinical studies have repeatedly confirmed this method's long-term efficacy and safety when used appropriately. Preliminary findings of the Post-Marketing Surveillance study of Norplant, a 5-year prospective study conducted in 32 family planning clinics in 8 developing countries in 1987-97, indicate an intrauterine pregnancy rate of 0.23/100 woman-years, an ectopic pregnancy rate of 0.03/100 woman-years, and a 67.3% continuation rate at 5 years. No significant excess of malignant neoplastic or cardiovascular disease has been observed. The major side effect is an irregular pattern of uterine bleeding, associated with about 25% of the discontinuations after 5 years of use. The quality of the family planning service is a major determinant of successful Norplant use and the degree of user satisfaction. Informed choice, the quality of follow-up care, easy access to removal services, and provider skills and attitudes are also critical. The signatories to this Consensus Statement (acting as individuals rather than representatives of their organizations) agree that, with the provision of appropriate information to users, Norplant is a good contraceptive choice that should be made available globally in all family planning programs with resources for appropriate training and counseling.

Combined oral contraceptives, smoking, and cardiovascular risk.

STUDY OBJECTIVE: To assess age specific incidence and mortality of stroke, acute myocardial infarction (AMI), and idiopathic venous thromboembolism (VTE) associated with use of modern low dose combined oral contraceptives (OCs) and the interaction with smoking. DESIGN: Hospital-based case-control study. SETTING: Hospitals in Oxford region in the United Kingdom, which covered a defined population, during the period 1989-1993. METHODS: Relative risk estimates from the WHO Collaborative Study and observed incidence rates from the Oxford region were used to estimate age specific incidence of each disease among women without cardiovascular risk factors and model total cardiovascular incidence and mortality. RESULTS: Among women who did not use OCs, smoke nor had any other cardiovascular risk factors, total incidence of stroke and AMI were less than 2 events per 100,000 woman years in those aged 20-24 years and rose exponentially with age to 8 events per 100,000 among women aged 40-44 years. Incidence of idiopathic VTE among women who did not use OCs rose linearly with age (from 3.3 per 100,000 at ages 20-24 years to 5.8 per 100,000 at ages 40-44 years). The increased risk of idiopathic VTE associated with OC use among non-smokers constituted over 90% of all cardiovascular events for women aged 20-24 years and more than 60% in those aged 40-44 years. Fatal cardiovascular events were dominated by haemorrhagic stroke and AMI, and among OC users who smoked these two diseases accounted for 80% of cardiovascular deaths among women aged 20-24 years, rising to 97% among those aged 40-44 years. Cardiovascular mortality associated with smoking was greater than that associated with OC use at all ages. Attributable risk associated with OC use was 1 death per 370,000 users annually among women aged 20-24 years, 1 per 170,000 at ages 30-34 years, and 1 per 37,000 at ages 40-44 years. Among smokers, the cardiovascular mortality attributable to OC use was estimated to be about 1 per 100,000 users annually among women aged less than 35 years, and about 1 per 10,000 users annually among those above the age of 35 years. CONCLUSION: The incidence of fatal cardiovascular events among women aged less than 35 years is low. The VTE risk associated with OC use is the largest contributor to OC induced adverse effects. The potentially avoidable excess VTE risk associated with the newer progestogens desogestrel and gestodene would account for a substantial proportion of total cardiovascular morbidity in this age group. For women over age 35 years the absolute risks associated with OC use and smoking are greater because of the steeply rising incidence of arterial diseases. The combination of smoking and OC use among such women is associated with particularly increased risks. Any potential reduction in AMI or stroke risk with use of third generation OCs would be a more important consideration among older compared with younger women, particularly if they smoke. However, the mortality associated with smoking is far greater than that associated with OC use (of any type) at all ages.

Twenty years of epidemiology in fertility regulation.

This paper summarizes findings from epidemiological research in fertility regulation which have helped in formulating policies nationally and internationally, without pretending to be all-encompassing; rather, it should help in demonstrating the usefulness and importance of epidemiological research and in discussing outstanding issues of public health relevance.

PIP: Epidemiological research serves to continue surveillance of fertility regulation agents once phase 3 clinical trials have been completed and the contraceptives have been marketed. Epidemiologic research conducted during the past 20 years has had a major impact on family planning program policies and clarified many concerns about contraceptive side effects that emerged in the early 1970s. Central has been the reassessment of the risk of malignant neoplasms associated with use of hormonal contraception. Data from large-scale studies such as the Cancer and Steroid Hormone Study in the US and the World Health Organization Collaborative Study involving 9 developing and 2 developed countries suggest that combined oral contraceptives (OCs) decrease the risk of ovarian and endometrial cancer, especially in long-term users. Although there is emerging evidence that OC use exceeding 5 years is associated with a modest increase in cervical cancer risk, the causality of the association is questionable given the probable influence of confounding factors such as sexually transmitted diseases. Moreover, epidemiologic studies launched in the 1980s confirmed that the previously noted association between OCs and cardiovascular diseases has been reduced as a result of lower doses of ethinyl estradiol and revised prescribing practices. Other foci of epidemiologic investigations have included possible side effects associated with natural family planning, the impact of modern IUDs on pelvic inflammatory disease and ectopic pregnancy risks, and the association between vasectomy and testicular and prostate cancer. Given the observational nature of epidemiologic research and the potential for bias, findings from several studies addressing the same research question with different methodological approaches are generally assessed.