Abstract: FR-PO897

Effects of Predictive Modeling Assisted Care Interventions on Hospitalization Rates in Hemodialysis Patients

Session Information

Category: Dialysis

  • 607 Dialysis: Epidemiology, Outcomes, Clinical Trials - Non-Cardiovascular

Authors

  • Sweet, David F., Fresenius Medical Care North America, Waltham, Massachusetts, United States
  • Chaudhuri, Sheetal, Fresenius Medical Care North America, Waltham, Massachusetts, United States
  • Han, Hao, Fresenius Medical Care North America, Waltham, Massachusetts, United States
  • Chamberas, Anthony, Fresenius Medical Care North America, Waltham, Massachusetts, United States
  • Maddux, Dugan, Fresenius Medical Care North America, Waltham, Massachusetts, United States
  • Reviriego-Mendoza, Marta, Fresenius Medical Care North America, Waltham, Massachusetts, United States
  • Larkin, John W., Fresenius Medical Care North America, Waltham, Massachusetts, United States
  • Usvyat, Len A., Fresenius Medical Care North America, Waltham, Massachusetts, United States
  • Maddux, Franklin W., Fresenius Medical Care North America, Waltham, Massachusetts, United States
Background

Hemodialysis (HD) patients continue to have high hospitalization rates regardless of advances in the treatment of end stage renal disease. We have developed and deployed a predictive model including more than 200 variables to identify in-center Hemodialysis (HD) patients at an increased risk for a hospitalization. In the pilot Dialysis Hospitalization Reduction Program (DHRP), we perform early identification of patients at risk for a critical event which may aid clinicians to intervene in a timely manner. We thus investigated the impact of DHRP on hospitalization rates in HD patients.

Methods

A total of 72 Fresenius Kidney Care clinics were included in the pilot; 141 clinics in neighboring regions to DHRP clinics were used as controls. We implemented the pilot in South Alabama/Florida Panhandle Region since Jan of 2015. For patients identified at high risk of hospitalization, Social Workers assessed and placed them on an intensive program employing community resources and integrated care. The dietitians utilized a high risk assessment looking at weight, nutrition, and access to food and supplements. The resident nurses assessed current care for anemia, adequacy, access, prior hospitalizations, and blood pressures. We compared mean quarterly hospital admission and readmission rates per patient year between groups.

Results

We analyzed the results 24 months after the start of the program and we found that DHRP patients exhibited: 1) 23% reduction in average yearly hospital admission rate compared to controls, and 2) 3.6% lower average yearly hospital readmission rate compared to controls. When fitting a linear regression model, both the quarterly hospital admission rate and quarterly readmission rate for the control group indicated an increasing trend with time, while the DHRP group indicated a small decreasing trend. Quarterly differences between groups for both admission rate and readmission rate increased over time (p< 0.003).

Conclusion

These findings suggest that care coordination based on predictive risk assessment in the DHRP is associated with lower hospitalization rates in HD patients, compared to controls. Further studies are needed to confirm these results.

Funding

  • Commercial Support