Since 2010 the rate of increase in life expectancy in the United States (US) has stagnated and even declined, reversing for the US the trend toward increased lifeexpectancy that is still continuing in most nations. The goal of this project is to develop an interactive tool, MortalityMinder, to explore trends in mortality, and identify their associated community level social determinants (SD). A key aspect of the tool is its ability to identify and explore subpopulations (i.e. groups of counties) exhibiting different mortality trends since 2000. For example, recent studies have found that “Deaths of Despair” due to suicide and substance abuse have increased dramatically amongst white males between the ages of 25-64 particularly in rural America.(1) Other studies found increases in maternal mortality in the US.(2) The identification of subpopulations experiencing different trendsin mortality and the characterization of their distinct demographics and SD can be used to guide the design of interven-tions and policies to decrease death and morbidity in the US,to deliver better healthcare more efficiently, and to improvethe daily lives of Americans. However, identifying more spe-cific and discriminating subpopulations is a major analytic challenge.
Analyzing the SDs of subpopulations with troubling mortality trends can lead to an understanding of unmet healthcareneeds and community factors at the root of the problem. Providing state-of-the-art machine learning and data visualization capabilities in an open source environment, Mortality-Minder features a multi-window interactive dashboard whichintegrates diverse public health and socioeconomic data re-sources to help decision makers solve these problems.