This workshop will bring together leading international researchers in the area of spatial statistics and spatio-temporal modelling as applied to mortality in low and middle income countries (LMIC’s). Topics to be considered include statistical inference for noisy and irregular spatio-temporal data, models allowing for informative sampling designs, and uncertainty in environmental exposures.
As a substantial proportion of deaths in LMIC’s occur outside of the health care system, administrative data gives incomplete information regarding mortality in these countries. Obtaining an accurate picture of the causes of death requires the use of health surveys, such as the Demographic and Health Surveys (DHS) and the Indian Million Deaths Study. When measuring exposure to environmental risk factors in LMIC’s, resorting to data collected by satellites (which is plentiful but inaccurate) is often the only option. Specialized Statistical methods for research on global mortality, methods distinct from those developed for rich-world administrative health data, are therefore required.
The workshop will seek to identify open methodological issues which will, when resolved, contribute to closing the gap between our understanding of mortality in LMIC’s and in the developed world. Health researchers at the forefront of global mortality research will convey the current state-of-the-art and describe problems which current methods are unable to address. Statistical researchers active in this area will show how recent advances in spatial modelling and Bayesian computation have been able to accommodate the particularities of mortality data in the global context.
A number of this meeting’s participants will also be attending the Causal adjustment in the presence of spatial dependence workshop in Montreal on 11-13 June.