Over the course of the Spring 2015 semester, students will be producing a piece of software for Dr. Matt Ferrari that will provide a more meaningful medium for displaying Measles projection data for every country in the world. Dr. Ferrari composes this data each year and presents it to the World Health Organization so that they can make more effective vaccine deployment strategies. At present, there is not a better way to display his data than as a collection of 193 very opaque spreadsheets. Our goal is to address this need by building a platform with which one can easily visualize the data and compare multiple countries on demand, without the needless obfuscation of 193 distinct text documents.
Dr. Ferrari's summary of the project is as follows:
Surveillance and diagnostic capacity for quantifying changes in infectious diseases has advanced faster than the analytical capacity to process and evaluate these data to inform public health decision-making. Theory predicts that improved vaccination should reduce the incidence of disease, but that this reduction 1) should proceed in a non-linear fashion through time, 2) should be accompanied by changes in the age distribution of infection, and 3) should be sensitive to local demographic rates. Thus, quantitative evaluation of the impact of vaccination programs requires summarizing surveillance data relative to dynamic changes expected due to demographic and health system change.
I propose to develop an interactive platform for visualizing disease incidence data in tandem with model projections that utilize locally-specific demographic and vaccination program data. This project will leverage software that I have developed for both retrospective and prospective evaluation of measles and rubella vaccination strategies, to allow visualization of surveillance data relative to model predictions.
The proposed visualization platform will illustrate projected dynamics under alternate proposed vaccination scenarios both on a map, and over time – allowing users to visualize individual countries, compare countries, select particular windows of time, project both the variation in several disease outcomes over time (e.g. incidence, mortality, age of cases), and overlay observed disease surveillance over model predictions to provide a visual representation of model fit. The test-case will be visualization of proposed measles and rubella vaccination policy in sub-Saharan Africa through 2030 using the currently proposed Global Alliance for Vaccines and Immunizations (GAVI) funding schedule for vaccine campaigns. The proposed framework will be designed to facilitate active discussion with health policy stake-holders (e.g. World Health Organization, GAVI) by visualizing the impact of proposed health policy over time and on a regional map. While this implementation is proposed as a test-case, the framework for visualization can be subsequently applied to other regions and proposed vaccinations.