Send to

Choose Destination
Big Data. 2015 Dec;3(4):238-48. doi: 10.1089/big.2015.0038.

Understanding Emergency Department 72-Hour Revisits Among Medicaid Patients Using Electronic Healthcare Records.

Author information

1 Rensselaer Institute for Data Exploration and Application, Rensselaer Polytechnic Institute , Troy, New York.
2 Department of Mathematical Sciences, Rensselaer Polytechnic Institute , Troy, New York.
3 Department of Computer Science, Rensselaer Polytechnic Institute , Troy, New York.


Electronic Healthcare Records (EHRs) have the potential to improve healthcare quality and to decrease costs by providing quality metrics, discovering actionable insights, and supporting decision-making to improve future outcomes. Within the United States Medicaid Program, rates of recidivism among emergency department (ED) patients serve as metrics of hospital performance that help ensure efficient and effective treatment within the ED. We analyze ED Medicaid patient data from 1,149,738 EHRs provided by a hospital over a 2-year period to understand the characteristics of the ED return visits within a 72-hour time frame. Frequent flyer patients with multiple revisits account for 47% of Medicaid patient revisits over this period. ED encounters by frequent flyer patients with prior 72-hour revisits in the last 6 months are thrice more likely to result in a readmit than those of infrequent patients. Statistical L1-logistic regression and random forest analyses reveal distinct patterns of ED usage and patient diagnoses between frequent and infrequent patient encounters, suggesting distinct opportunities for interventions to improve efficacy of care and streamline ED workflow. This work forms a foundation for future development of predictive models, which could flag patients at high risk of revisiting.


big data analytics; business intelligence; data mining; machine learning


Supplemental Content

Full text links

Icon for Atypon
Loading ...
Support Center