Back

HRSA Population Health Dashboard

Nguyen, C.

2023-04-24 health informatics
10.1101/2023.04.17.23288701 medRxiv
Show abstract

Project PurposeTo develop a quality dashboard tool for the Health Resources and Services Administration (HRSA) community health centers to examine health center characteristics associated with performance excellence and predict future healthcare cost. BackgroundHRSA is the primary federal agency for improving health care to people who are geographically isolated, economically or medically vulnerable. HRSA-funded health centers are required to upload to Uniform Data System (UDS). HRSA health data is rich; however, their website provides limited meaningful visualization and data-driven analysis. MethodThe project reviewed data from Uniform Data System between 2017 and 2019. The data was cleaned using Excel. Visualization was displayed on Tableau, and the cost prediction modeling was completed using Python Machine Learning libraries. All the findings are available for public view on Weebly-hosted website. OutcomesThe Tableau dashboard showcases the effectiveness of the preventive intervention and chronic disease management initiatives over the years. This tool focuses on comparison between outcomes in health care programs and cost benefits in all states. The dashboard also uses the most fit Machine Learning model and the most important variables to predict the cost per patient of a health center. DiscussionThe lack of similar performance comparison dashboards by HRSA suggests a need for HRSA Population Health Dashboard to understand the performances of HRSA-funded health centers or regions through times. It shows the ineffectiveness of the current healthcare programs in reducing disease growth and healthcare cost over the course of 3 years. Next Steps/OpportunitiesThe final dashboard will be available for public use and will continue to aggregate more real-time data using Application Programming Interface (API) to improve its accuracy of prediction model. The HRSA Population Health Dashboard will be beneficial to management at both HRSA federal- and health center-level to make valuable strategies.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.