Unwanted and mistimed pregnancies are still common everywhere, even though modern family planning has helped families control their fertility. In low income countries access to family planning is more limited and more than 100 million women in low income countries are classified as having unmet need because they say they don't want to get pregnant but do not have access to modern family planning methods. The US government, through USAID spends $649 million to support family planning programs in low income countries. Enthusiasm for spending on family planning services for the poor is diminished for some policymakers by ambiguity over whether family planning is just a life style choice or indeed has important affects on the health of children and families. It turns out that the evidence on the health impact of family planning remains vulnerable to serious bias. There have been no randomized trial based studies to show that family planning could improve population health by helping women prevent unwanted pregnancies. All prior studies of the topic are subject to bias because of either retrospective recall of whether a birth was wanted or the problem where women who have more unwanted births are also prone to have more unmeasured problems in their lives that also impair the health of their families. This proposal will answer whether unwanted children and their siblings and mothers have worse health outcomes and school attainment. The quasi-experimental allocation of family planning and health services that was started in Matlab, Bangladesh in 1985 allows tremendous progress in providing a better answer than ever before. The study is able to make 3 innovations: 1) Wantedness is measured prenatally;2) The bias due to confounding of wantedness with unobservables is broken using the Matlab quasi-experimental assignment, geospatial coordinates, and health worker assignment as identifying instruments;3) Long term follow up measures of children and families out to 18 years after an unwanted birth is available through the demographic surveillance system. This project will make significant gains in our ability to say whether investments in family planning are cost effective compared to other strategies for achieving Millennium Development Goals for maternal and child health.
This project will use data on the health and school attainment of children and mothers in Matlab, Bangladesh from 1990 to 2008 in order to find out if children who were unwanted have worse health and schooling. The study is set in an area that experienced the random assignment of different communities to receive door to door delivery of family planning services back in 1985 This random assignment will help the study produce definitive results on whether population health can be improved by measures that help poor families control their experience of unwanted fertility.
|Bishai, David; Razzaque, Abdur; Christiansen, Susan et al. (2015) Selection bias in the link between child wantedness and child survival: theory and data from Matlab, Bangladesh. Demography 52:61-82|