From the 2018 HVPAA National Conference
Christopher Grondin (Michigan Medicine), James Henderson (Michigan Medicine), Lauren Heidemann (Michigan Medicine), Kathryn Levy (Michigan Medicine), Matthew Johnson (Michigan Medicine), Jenna Keedy (Michigan Medicine), Robert Chang (Michigan Medicine), Kayla McAleenan (Michigan Medicine), Eve Kerr (Michigan Medicine), Lee Schroeder (Michigan Medicine), Scott Owens (Michigan Medicine), Brian Tolle (Michigan Medicine), Linda Bashaw (Michigan Medicine), Jeffrey Lott (Michigan Medicine), Christopher Petrilli (Michigan Medicine)
Background
The Hospital Medicine Choosing Wisely Campaign recommends against ordering daily labs in the face of clinical and laboratory stability. Daily labs can lead to unnecessary blood draws, direct financial costs, and downstream testing due to spurious results or iatrogenic anemia. We aimed to reduce inappropriate daily complete blood count (CBC).
Objectives
To reduce unnecessary ordering of routine morning phlebotomy with a novel best practice advisory (BPA) targeting CBC for patients with laboratory stability.
Methods
CBC orders were tabulated from May 1st, 2017 to September 30th, 2017 on Michigan Medicine’s five general medicine resident services. A CBC was defined as low-value if it met all three of the following criteria: 1) It was the third CBC ordered within a 72-hour period, 2) The previous two CBCs were both acceptable (i.e., Hemoglobin > 10, white blood cells and platelets in normal ranges), and 3) Hemoglobin was stable (i.e., not decreased by more than 1.5 g/dL when comparing two recent CBCs). A BPA was developed to alert clinicians ordering a low-value CBC (FIGURE 1). On July 12th, 2017 the BPA was activated on three of the five resident general medicine services. The BPA would only fire when a CBC order was placed; i.e., it would not fire for standing orders (CBCs that recur daily without having to reorder).
Results
The BPA was not visible to providers when a standing CBC was ordered (87% of CBC orders). There was a reduction in 12/180 (7%) of low-value CBCs. The BPA fired for 41 of 180 low-value CBCs resulting in 29 acknowledgements/overrides (29/41; 71%) and 12 order removals (12/41; 29%). 77% of low-value CBCs occurred without an opportunity for BPA firing.
Conclusion
A BPA that fires when a low-value CBC is ordered had no significant impact on ordering rates at our institution. This is largely due to the high prevalence of standing CBC orders, which provided no opportunity for the pilot BPA to fire. When the BPA was seen, providers did remove 29% of low-value CBC orders, which is very promising.
Implications for the Patient
Our results reveal the magnitude of standing daily labs and expose challenges in targeting these labs for electronic health record (EHR) interventions. When developing solutions in the EHR, rapid Plan/Do/Check/Act-cycles are necessary to pick-up critical, limiting and unforeseen barriers such as alerts not appearing in setting of standing lab orders.
