Dr. Leontine Alkema, assistant professor of biostatistics in the UMass Amherst School of Public Health and Health Sciences, is lead statistician on a new analysis of global and regional levels and trends in abortion for the 25 years from 1990-2014 conducted with colleagues at the Guttmacher Institute, the World Health Organization (WHO) and other institutions. The study, which appeared in a recent issue of The Lancet, benefited from improved statistical modeling according to Dr. Alkema. Findings offer policymakers and stakeholders involved in directing resources toward sexual and reproductive health services a more accurate assessment of the prevalence and distribution of abortion worldwide.
[Photo: Dr. Leontine Alkema]
Overall, lead author Dr. Gilda Sedgh of the Guttmacher Institute, New York City, and colleagues found, “The abortion rate declined significantly in the developed world, but not in the developing world, between 1990 and 2014.” They say findings highlight the need for increased investment in family planning services and safe abortion and post-abortion care in developing countries. Dr. Alkema and colleagues also note that the estimation of abortion incidence, especially in the developing world, is challenging because of data scarcity.
She says, “Abortion data are scarce and it’s difficult to get good quality information, especially where abortions are illegal or if there is a stigma attached to it. Given this, we had to find new ways to include all available information we have on abortion incidence. For example, in the current study, we were able to include data on subgroups of married and unmarried women, as well as data providing a minimum or a maximum boundary on abortion incidence only.”
In addition to making use of an extended database, the authors also developed a statistical model to construct estimates with missing or incomplete information.
Dr. Alkema explains, “The statistical model allowed us to produce estimates with uncertainty bounds that communicate the possible range of outcomes for all populations, including those with no or limited data. While having more data is still a much-needed goal to work towards, the good thing about using a model is that it allows for the exchanging of information across countries with and without data in a systematic and documented manner.”
She adds, “Hence, we were able to check and communicate the sensitivity of results to model assumptions and to carry out validation exercises, for example to check that the model was able to accurately predict abortion incidence for countries with data, if those data were temporarily left out during the model fitting.”
Another advance of the new statistical approach is that the researchers were able to track incidence among different subgroups, such as married and single women. “The model helps to identify who is having abortions,” Dr. Alkema notes, “which is helpful to policymakers and stakeholders in reproductive health.”
Findings make it clear that abortions occur everywhere, and that women across all groups, even contraceptive users in developed countries, are among those who have unintended pregnancies and abortions, she adds.
Overall, the authors state, “The findings underscore that investments are needed to meet women and couples’ contraceptive needs, especially in the developing world, where abortion rates are high and many abortions are unsafe. Reliable estimates of abortion incidence in the developing world are scarce and additional research in this area is needed to improve our ability to monitor trends in this region.”
This work is a product of a collaborative working group on estimating abortion incidence and safety co-led by the Guttmacher Institute and the Department of Reproductive Health and Research at WHO.