Optimizing Clinical Decision Support: The Appropriate Alert at the Appropriate Time to the Appropriate Clinician Equals Improved Quality of Care (A3 = I)
Kris Niemi, Pharm.D., Siobhan Geary, R.N.,
Mark Larrabee, Pharm.D., Barbara Quinn, R.N.,
Joy Bailey, Kevin Brown, Pharm.D.
Sutter Medical Center, Sacramento, California
We describe the creation and use of a computer database that receives input from multiple independent computer systems to identify Joint Commission and Center for Medicare and Medicaid Services core measure patients in real-time and produce alerts to the appropriate health care provider when core measures are not met. The system is called Core Measure Manager (CMM). CMM accepts data from the pharmacy, laboratory, radiology, heart catheterization lab, and Abstract Data Type (ADT) systems as well as from automated dispensing machines via Health System Seven (HL7) compliant interfaces. We developed rules to stratify electronic data based on a point system assigned for data elements received from each system. When certain point thresholds are met, the patient is identified as a potential pneumonia (PNA) or heart failure (HF) patient (or both). Once the patient is identified as a possible core measure patient, CMM compares current medication orders with core measure recommended medication use. If the current orders match core measure recommendations, no alert prints. If the current orders do not match core measures, an alert prints in the appropriate patient care area. The alert identifies the potential deficit and provides the core measure recommendations within sufficient time to take action to meet time sensitive performance measures.
We assessed CMM’s ability to identify pneumonia and heart failure patients compared to final ICD-9 discharge diagnoses. CMM successfully identified 92% of pneumonia patients and 94% of HF patients. We compared six core measures for seven months before and after implementation of the pneumonia and heart failure alerts. Sixty seven percent of the core measure indicators increased in the percentage of months spent in the top decile after implementation of CMM. The average increase was 26%. The greatest improvement was seen in HF discharge instructions. No improvement was seen in the core measures of antibiotic administration within 4 hours of admission or administration of correct antibiotics within 24 hours of admission.
The initial goals of identifying appropriate patients and triggers, and alerting appropriate clinicians have been met. The ability of CMM to identify patients by diagnosis has been documented. Staff response to the support provided by CMM has been very positive. We demonstrated that the implementation of a computerized database that integrates data from several separate computer systems can be used to successfully identify core measure patients who have not yet received recommended aspects of care. We used this system to increase the frequency with which we achieved top decile core measure performance by 26%.
View a pdf of the poster from the Midyear Meeting (2.8 MB).