In Bangladesh, the maternal mortality ratio dropped by 66% between 1990 and 2010, a remarkable decline in light of the country’s poor health infrastructure and socioeconomic conditions and its low rate of institutional deliveries. This decline is even more surprising in a country in which abortion is legal only to save the woman’s life, and illegal procedures—usually performed by untrained providers in unsafe conditions—are common. The only legal alternative is menstrual regulation (MR), in which manual or vacuum aspiration is used to bring on delayed menses within 10 weeks of a woman’s last period. The first two articles in this issue examine changes over time in the mortality risk associated with each type of termination, as well as in the type of method used. Both studies use surveillance data from Matlab, a subdistrict in which one area receives standard government health care and the other receives enhanced maternal and child health and family planning (MCH-FP) services.

In the first article, Mizanur Rahman and colleagues found that the odds of death from MR were four times those from live birth in 1989–1999, but were no longer elevated in 2000–2010 [see article]. The odds of death from complications of unsafe abortion were 12 times those from live birth in the earlier period and five times those from live birth in the later period. The reduction in mortality risk was greater in the area with enhanced MCH-FP services than in the area with standard government services. In the accompanying article, Julie DaVanzo and Rahman report that between 1989 and 2010, the proportion of pregnancy terminations carried out using MR rose from 39% to 79% in the area with enhanced services, and from 32% to 84% in the area with standard services. Among women who terminated their pregnancies, those aged 25–44 were more likely to use MR than those aged 20–24, and those with any education were more likely to use MR than those with no education.

Media outlets in developing countries have suggested that increased over-the-counter access to emergency contraceptive pills has led to repeat use and "misuse" of the product, raising concerns about safety. A study by Dawn Quin-Chee and colleagues, however, showed that in two African cities, repeat use was not common [see article]. Some 18% of women interviewed in Nairobi, Kenya, and 17% of those in Lagos, Nigeria, had ever used emergency contraceptive pills. On average, respondents who had used the pills had done so less than once per month, well below the limit of 1–4 times per month suggested by the World Health Organization.

Despite steadily declining fertility in Ghana, use of modern contraceptives is low, even in urban areas, with a larger share of use accounted for by traditional methods. Using in-depth contraceptive life history interviews to explore the reasons for this unusual pattern, Ivy Frances Osei and colleagues found that patterns of method use varied according to the type and stage of a couple’s relationship [see article]. Couples might use condoms early in their relationship and frequently preferred modern methods for spacing after their first birth; however, side effects, concerns about protecting fertility and fear that the "chemicals" in modern methods could cause health problems often led couples to choose traditional over modern methods, particularly before starting childbearing and after completing their families. The authors recommend that contraceptive programs target messages according to stage of relationship and work with people’s desire to use traditional methods to ensure that they can do so effectively.

Method mix—the percentage distribution of contraceptive users in a given country by method—is one measure of the range of methods available in that country. A skewed method mix (one in which 50% or more of users rely on a single method) could signal a lack of alternate methods or provider bias. In a study of 109 low- and middle-income countries by Jane Bertrand and colleagues, 30% had a skewed method mix, down from 35% in a 2006 analysis [see article]. Countries with a skewed mix were almost evenly divided by dominant method, with seven each skewed toward traditional methods, the pill, the IUD and the injectable. In addition, three were skewed toward female sterilization, and two toward the condom. The authors note that although a skewed method mix may be a flag for lack of choice or provider bias, it may simply reflect cultural preferences, and investigation is needed to identify the cause.

—The Editors