Mahesh Karra, Assistant Professor of Global Development Policy at the Frederick S. Pardee School of Global Studies at Boston University, had two papers published in JMIR Research Protocols and the Journal of the Royal Statistical Society.
In the first paper, titled “The Effect of Improved Access to Family Planning on Postpartum Women: Protocol for a Randomized Controlled Trial,” Karra and co-author David Canning – Richard Saltonstall Professor of Population Sciences, and Professor of Economics and International Health at Harvard T.H. Chan School of Public Health – aims to describe a randomized controlled trial that is being conducted to identify the causal impact of an intervention to improve access to postpartum family planning (PPFP) services on contraceptive use, pregnancy, and birth spacing in urban Malawi. The results of this study aim to fill the current knowledge gaps in the effectiveness of family planning interventions on improving fertility and health outcomes.
In his second paper, titled “Adding measurement error to location data to protect subject confidentiality while allowing for consistent estimation of exposure effects,” Karra and co-authors David Canning and Ryoko Sato – Research Associate at the Harvard T.H. Chan School of Public Health – explore the use of a perturbation vector in public health data collection. The inclusion of this data point can allow researchers to more precisely study location-dependent information while also protecting subject confidentiality.
An excerpt:
The direct use of perturbed location data to construct explanatory exposure variables for regression models will generally make naive estimates of all parameters biased and inconsistent.We propose an approach where a perturbation vector, consisting of a random distance at a random angle, is added to a respondent’s reported geographic co-ordinates.We show that, as long as the distribution of the perturbation is public and there is an underlying prior population density map, external researchers can construct unbiased and consistent estimates of location-dependent exposure effects by using numerical integration techniques over all possible actual locations, although coefficient confidence intervals are wider than if the true location data were known
The JMIR Research Protocols paper can be read here. The Journal of the Royal Statistical Society paper can be read here.