Big Data

My day began with a 1:1 meeting with Trevor Mundel who heads Global Health at the Foundation. The conversation turned to scenarios and forecasting and I learned about Chris Murray at the Institute of Health Metrics and Evaluation, a visionary who has led the establishment of a global model of disease burden.  I'm looking forward to reading Chris's recently published biography, 'Epic Measures'. You can visit the IHME website to play with amazing interactive visualizations of disease patterns and causes of death of men and women of all ages around the world.

The statistics on causes of disease and death are gathered from various local sources in each country and often require trickery such as Bayesian statistics to fill in the gaps. For example in many countries HIV deaths are often not reported as such, so the 'official' data must be interpreted accordingly based on assumptions about HIV prevalence from antenatal screening and other health programs.  IHME is constantly adding new data and refining the model to go from historical reporting to forecasting future patterns of disease and layering in additional factors including socio-economic, environmental and others.  The biggest contributions to reducing infant mortality include increased personal income, lower birth rates, better nutrition and education of mothers as well as innovations in treatment and healthcare systems. Shockingly, the factors that are driving increases in death rates among adults in the developing world include the spread of HIV / AIDS, tobacco use, the obesity epidemic and alcohol abuse.