Reducing Preventable Readmissions for Patients with Diabetes on the Parkland Hospitalist Units

BACKGROUND: High rates of readmission are detrimental to both the patient and the hospital, and they are associated with decreased patient satisfaction, diminished quality of life, and substandard overall care. Diabetes remains one of the greatest risk factors for 30-day all-cause readmissions. LOCAL PROBLEM: Under the Affordable Care Act (ACA), the Centers for Medicare and Medicaid Services (CMS) established the Hospital Readmissions Reduction Program (HRRP), which penalizes hospitals for high readmission rates related to heart failure, COPD, acute myocardial infarction, pneumonia, and stroke. Because diabetes was not a disease scrutinized under the HRRP at the start of the project period, Parkland was not specifically focused on reducing readmissions for patients with diabetes. METHODS: This quality improvement project utilizes the DMAIC framework. The proper context and measures were defined, and baseline process and outcome measures were obtained. A quality gap analysis was completed, and an FMEA was used to identify the gaps that needed to be addressed. Outcome and process measures were analyzed using chi-squared analysis and segmented control charts, and the balancing measures were analyzed using a continuous control chart. INTERVENTIONS: The first intervention was a rearrangement of the EMR nursing flowsheet drop-down menu used to document inpatient diabetes education to highlight diabetes survival skills first. The second intervention was the creation of the Diabetes Hospital Education and Resource Officer (HERO) Program which provided self-selected nurse champions across each hospital unit to be leaders in diabetes patient-care. RESULTS: Nine months after both interventions, there was a significant decrease in the 30-day all-cause readmission rate from the Parkland hospitalist unit by 5%. The documentation rates for insulin administration and hypoglycemia or hyperglycemia education increased significantly five months after the first intervention, and nine months after both interventions when compared to the baseline. Correlation analysis showed that with education, there was a decrease in readmission rates, but the changes were not significant. All three balancing measures remained in control during the project period. CONCLUSIONS: Changes to the EMR can create an immediate impact while continuous improvements need to be sustained by a leadership program with human factors.