BACKGROUND: High-quality community-level data on the distribution of morbidity and mortality, and their associated behavioral, environmental, economic and social determinants, are needed to effectively plan and evaluate public health programs. Cost data associated with these health outcomes and public health programs are needed for economic evaluation including for assessing the relative cost effectiveness of different interventions/programs. The objective of the California Community Burden of Disease and Cost Platform (CCBDCP) project is to provide a platform/tool, using open source software, to compile and process such data into a web-based interactive environment.
METHODS: The CCBDCP is a collaborative effort between multiple Centers within the California Department of Public Health (CDPH) and several California Local Health Jurisdictions (LHJs). The platform is built primarily with R and the “R Shiny” package, using an “agile” approach of distributed work and iterative enhancements to functionality. Phase I includes California mortality data, American Community Survey socioeconomic data, and high-level cost data associated with a range of conditions. Mortality ICD-10 codes are aggregated using Global Burden of Disease and US Centers for Health Statistics schema; California “communities” are based on aggregation of census tracts; cost data are from multiple sources, including Medicare and the AHRQ Healthcare Cost and Utilization Project. In Phase I, mortality-related measures are aggregated by community and LHJ, and include cause-specific death rates and cause-specific average and total years of life lost.
RESULTS: Interactive CCBDCP web-based applications with death, socioeconomic, and cost data are live and in use by CDPH and LHJ partners and include static and interactive univariate and bivariate maps, ranking charts within and between LHJs and communities, and scatterplots with assessment of correlations. Preliminary observations have proven useful to partners (e.g. high violence-associated years of life lost in specific communities) and are informing ongoing work on the Platform. A prototypic version of Platform can be seen at https://phdataviz.shinyapps.io/PLACE/.
CONCLUSIONS: Through collaboration, an initial phase of a platform was developed that appears to have great potential value for public health partners. The use of open source tools allows the platform to be easily and freely shared, and can allow for collaborative enhancement. All computer code and documentation can be shared with any interested parties. This tool is a work in progress and additional measures (e.g. Disability Adjusted Life Years, life expectancy, and “Public Health Geocoding Disparities Project” measures), algorithms, and functionality will be added in Phase II, likely by mid-2017.