EVALUATION OF CLINICAL LABORATORY-PHARMACY LINKAGE DECISION SUPPORT IN THE USE OF POTASSIUM SUPPLEMENTS
Author(s)
Yu S, Galanter W, Lin FJ, Lambert BUniversity of Illinois at Chicago, Chicago, IL, USA
Presentation Documents
OBJECTIVES: Clinical decision support (CDS) has been utilized to link laboratory and pharmacy data to optimize medication therapy. This study aimed to evaluate the effect of synchronous and asynchronous CDS in inpatients receiving potassium supplements. METHODS: The synchronous and real-time asynchronous lab-pharmacy CDS was implemented in our 450-bed academic teaching hospital in June 2003. Non-hemolyzed serum potassium ([K+]) values and medication orders for potassium supplements, 12 months prior to and after CDS implementation, were analyzed. A Cox proportional hazards model was constructed to assess the effect of CDS in improving clinicians’ response time at the presence of hyperkalemia. Potassium ≥5.0 mEq/L and >5.4 mEq/L were used to define high normal and elevated hyperkalemia. Response time was measured from the time of the first instance of hyperkalemia to cancelation of the medication order. Response time was censored at the time of a consequent normal potassium or patient discharge. RESULTS: In the pre-CDS period, 12.5% (1439/11512) of the potassium supplement orders were followed by at least one abnormal serum potassium value; whereas in the post-CDS period, 10.96% (1206/11005) of the orders were alerted by CDS. While controlling for patient characteristics, posting time of the hyperkalemic results, and severity of hyperkalemia (5.0 mEq/L≤[K+]≤5.4 mEq/L vs. [K+]>5.4 mEq/L), potassium supplement orders were cancelled more rapidly after CDS implementation (hazard ratio[HR]=1.14, p=0.007). Moreover, clinicians responded sooner if the result with hyperkalemia was posted during 6am-noon vs. midnight-6am (HR=1.36, p<0.001), while the response time was longer if the result was posted during noon-6pm (HR=0.73, p=0.004) or 6pm-Midnight (HR=0.57, p<0.001). Patient age, sex, race, and the severity of hyperkalemia had no significant effect on clinicians’ response time. CONCLUSIONS: The synchronous and asynchronous real-time lab-pharmacy linkage decision support helped clinicians to manage potassium supplements in a more timely manner in patients with high normal or elevated potassium.
Conference/Value in Health Info
2011-05, ISPOR 2011, Baltimore, MD, USA
Value in Health, Vol. 14, No. 3 (May 2011)
Code
DM4
Topic
Health Service Delivery & Process of Care
Topic Subcategory
Health Care Research, Hospital and Clinical Practices, Prescribing Behavior, Quality of Care Measurement
Disease
Multiple Diseases