Online Program

Return to main conference page

All Times ET

Thursday, June 3
Practice and Applications
Data-Driven Healthcare
Thu, Jun 3, 1:10 PM - 2:45 PM
TBD
 

Did Increasing Continuity of Care Protect Patients with Chronic Disease from Emergency and Hospitalization Readmission? A Cohort Spatial-Temporal Study in Mississippi (309680)

*Phi Le, University of Mississippi Medical Center 
Joshua Mann, University of Mississippi Medical Center 

Keywords: Continuity of care, emergency visit, hospitalization, Mississippi

Introduction: Some studies have shown associations between high Bice-Boxerman continuity of care (COC) index and lower odds of emergency department (ED) visits or inpatient hospitalizations for patients with chronic diseases. We would like to know whether COC reduces the risks of readmissions to emergency visits and hospitalizations.

Method: We did a spatial longitudinal cohort study for patients with chronic diseases living in central Mississippi. We collected medical records of patients who visited the ED and the inpatient department each year for 75 zip code regions from 2015 to 2018. The outcomes were the yearly dependent means of emergency visits and inpatient hospitalizations of patients for each zip code area. And the corresponding dependent variables were the means of COC, BMI, race, gender, and uninsured rates. We applied spatial-temporal models to find associations between outcomes and dependent variables.

Results: We did not have evidence that continuity of care affected the population with chronic diseases from readmissions to emergency visits [95% credible from -1.194 to 3.659]. BMI, insured status, and time also had no effects on the risk of revisiting the emergency department. However, we found that regions with a 1% increase in the white population would reduce the readmissions to ED by 252 for every 1,000 visits a year [95% credible interval from -0.386 to -0.119]. Males had fewer ED visits than females, a 1% increase in the male population will decrease ED visits by 361 for every 1000 visits a year for each region [95% credible interval from -0.666 to -0.055]. And one year older in population would increase by 11 emergency visits for that group for every 1,000 visits a year [95% credible interval from 0.004 to 0.018]. Regarding hospitalizations, we did not see evidence that COC, BMI, gender, time, uninsured rate affected hospitalization readmissions.