Statistical methods for mortality estimation in data-sparse settings
Conference
Category: International Association for Official Statistics (IAOS)
Abstract
Mortality is the most direct indicator of health at the population level. Globally, two-thirds of deaths are unreported and we know little about the timing and cause of these deaths. This lack of vital statistics critically limits the ability to monitor population health and evaluate public health interventions, especially in low- and middle-income countries (LMICs) where they are most needed. In most settings without fully functioning civil registration and vital statistics (CRVS) systems, data from surveys and censuses are usually the only source of information to estimate basic demographic indicators. Such data are usually messy, sparse, and do not provide enough granularity to directly derive reliable and disaggregated estimates. In this session, we propose four talks from speakers in both academia and government, on various approaches to combine information from multiple sources to combat the issue of data sparsity.