Achieving Rapid Improvement in Hospital Readmissions: A Project Boost Pilot As a Driver of System Wide Change

1North Memorial Medical Center, Robbinsdale, MN
2Johns Hopkins, Baltimore, MD

Meeting: Hospital Medicine 2012, April 1-4, San Diego, Calif.

Abstract number: 97699


A mentored Project BOOST pilot was undertaken at North Memorial Medical Center, a level I trauma center serving north Minneapolis and the northwestern metropolitan area on the A6 cardiac telemetry unit. BOOST target tool triggers of medications, patient support status, and prior hospitalization within 3 months were identified for weekday interdisciplinary rounds by the electronic medical record. Problem diagnoses of CHF, pneumonia and COPD were identified during interdisciplinary rounds. The primary intervention specific to the pilot unit was bedside interdisciplinary rounding around care transition plans. Additional interventions triggered by the target tool were teach back on diagnosis and medication, arranged follow–up appointments within 7 days, home health nursing visits within 2 days and follow up phone calls at 24 hours by home health liaisons. The implementation phase was January through May 2011 with the operational phase from June through October of 2011. The baseline number of hospital readmissions was 2584 per year.


To reduce hospital readmissions.


Readmissions on the pilot unit in the implementation phase were 173 compared to131 in the operational phase (24% reduction), whereas readmissions on the comparison unit went from 163 to 199 (22% increase). The pilot unit also achieved at least a 10% point differential improvement in willingness to recommend scores and a trend toward decreased length of stay while the comparison unit had a steady to increased length of stay. Medication reconciliation improved to 95% as compared to 82% hospital wide. Hospital wide interventions during this time consisted of electronic medical record triggers for home health care referral and an improved medication reconciliation process. Hospital wide readmissions as measured by PPR (Potentially Preventable Readmissions) showed a reduction in the observed to expected readmissions ratio from 0.99 to 0.92 by the end of the pilot. The pilot drove further analysis of hospital wide readmissions. 56% of readmissions had a psychiatric or substance abuse co morbidity. While 34% had a diagnosis of CHF or pneumonia, sepsis was the most frequent admission diagnosis. Anticoagulant use was a significant predictor of readmisssion (P < 0.05); over more than 62% of readmitted patients were prescribed an anticoagulant compared to about 48% of non–readmitted patients. An audit of 15 sepsis readmissions revealed the presence of advanced illness. Eight of 15 patients died with their readmission event.


System changes driven by the BOOST pilot results include propagation of bedside interdisciplinary rounds, reorganization of system wide coordination and behavioral health services, hospitalist collaboration in transition communication and advanced illness screening.

To cite this abstract:

Larson C, Howell E, Degelau J, Vinson K, Johnson K, Rivers S, Abrahamson V. Achieving Rapid Improvement in Hospital Readmissions: A Project Boost Pilot As a Driver of System Wide Change. Abstract published at Hospital Medicine 2012, April 1-4, San Diego, Calif. Abstract 97699. Journal of Hospital Medicine. 2012; 7 (suppl 2). Accessed March 28, 2020.

« Back to Hospital Medicine 2012, April 1-4, San Diego, Calif.