Elective abortion is legal in Mexico City but highly restricted elsewhere in Mexico—a situation that forces women from around the country to journey to the capital if they wish to safely terminate an unwanted pregnancy. To shed light on which women are able to successfully navigate this process, Leigh Senderowicz and colleagues examined medical records from more than 20,000 women who sought an abortion in 2013–2015 at the four primary-level clinics that offer the service in Mexico City. Although the vast majority of abortion seekers came from within the city or the surrounding metropolitan area, 7% came from bordering states and 5% came from the rest of Mexico. On average, abortion seekers from the rest of Mexico had 1.4 more years of education (a proxy for socioeconomic status) than did those from Mexico City. Moreover, in regression models that adjusted for average educational attainment in abortion seekers' home communities, the further that women lived from Mexico City, the more educated they were relative to their local peers. These results, according to the authors, suggest that poor and less-educated women from outside of Mexico City are less able than their wealthier and more-educated counterparts to manage and afford the trip to obtain a legal abortion, and thus may be left to seek an unsafe procedure nearer to where they live.

Also in This Issue

• Few studies have examined the relationship between intimate partner violence and women's autonomy, and even fewer have explored autonomy beyond the individual level. Eric Y. Tenkorang used data from 2,289 Ghanaian women in 40 communities to examine the associations between experiences of physical, sexual, emotional and economic violence and three types of autonomy—economic decision making, family planning decision making and sexual autonomy—at the individual and community levels. According to the study, all three types of autonomy were associated with intimate partner violence, but in different ways. At the individual level, family planning decision making was associated with reduced odds of having experienced any type of violence, while economic decision making was associated with elevated odds of having experienced emotional and economic violence. At the community level, women living in communities with higher levels of sexual autonomy had reduced odds of having experienced physical or economic violence. According to the author, the study's findings suggest a need for "strategies and empowerment programs that enhance Ghanaian women's access to and control of resources and power within the household," as well as for finding "ways to empower communities to help women control and make independent decisions in their household."
• Although researchers and program managers have increasingly recognized the importance of quality of care in women's contraceptive uptake, a reliable index that allows routine and comprehensive estimation of the quality of national family planning programs has been unavailable. To fill this gap, Anrudh Jain has proposed a new measure, the National Quality Composite Index (NQCI), that assesses three different dimensions of program quality—structure, process and outcome—using data from Family Planning Effort Surveys and Demographic Health Surveys. Examining program quality in 30 developing countries, Jain found that overall scores on the 100-point NQCI ranged from 50 in Pakistan to 72 in Cambodia. National scores for the three dimensions varied even more widely and were not correlated with each other, suggesting that they measure distinct dimensions of program quality. Overall NQCI scores were correlated with existing measures of national-level quality, but not with total fertility rate and modern contraceptive prevalence rate. Jain notes that the NQCI can be used both to identify differences between countries in national-level quality of family planning programs and to monitor changes over time within countries.
• Collecting reliable data on abortion can be difficult. In countries where the procedure is legally restricted and highly stigmatized, indirect methods may be necessary, because some women may be uncomfortable with revealing their abortion history and thus may not answer direct survey questions about it honestly. In their study, Marziyeh Ghofrani and colleagues used two such indirect methods—the randomized response technique and the unmatched count technique—to estimate the prevalence of induced abortion among a sample of married women in Tehran, Iran, and to compare the methods' results and usability. They found that the two methods produced similar prevalence estimates—both higher than previous estimates from direct questioning. However, women indicated that they considered the unmatched count technique to be easier to comprehend and more trustworthy in terms of confidentiality, which leads the authors to suggest that the method may be preferable in low-literacy, abortion-restricted settings such as Iran.

—The Editors