From the 2022 HVPA National Conference
Joseph Talledo MS (NYC Health + Hospitals), Da Wi Shin BE, Mona Krouss MD, Hyung Cho MD, Nessreen Mestari MPA, Daniel Alaiev BBA, Sigal Israilov MD
Background
Use of heparin may result in Heparin-induced Thrombocytopenia (HIT), which can cause life threatening thrombosis. Testing for HIT is commonly done through the Enzyme-linked Immunosorbent Assay (ELISA) which tests for the presence of heparin-PF4 antibodies, also called heparin-induced platelet antibodies (HIPA). If the test is positive for the presence of the antibodies, the Serotonin Release Assay (SRA) is ordered to confirm a diagnosis of HIT. Often both HIPA and SRA are ordered at the same time instead of being ordered sequentially (HIPA followed by SRA). HIT testing is also sometimes done unnecessarily on low-risk patients. This results in not only unnecessary healthcare costs but also negative patient care outcomes, including false positives and resulting unnecessary treatment.
Objective
Our objective was to decrease the amount of unnecessary HIT testing at NYC Health + Hospitals, the nation’s largest urban safety net system. This entails both decreasing the incidence of simultaneous ordering of HIPA and SRA, as well as decreasing the number of overall orders for HIPA in patients who have a low probability for HIT.
Methods
We offered Clinical Decision Support (CDS) to ordering providers. This took the form of Electronic Medical Record (EMR) interventions. Through a modified order screen, we encouraged providers to compute the “4T” score to calculate the likelihood that a patient has HIT, including relevant information for the users to tabulate the score. In addition, we added a Best Practice Advisory (BPA) that alerted the provider attempting to order HIPA and SRA at the same time. The BPA nudged the provider to remove the SRA order and instead order the HIPA alone before confirming a positive result by ordering a subsequent SRA.
Our primary outcome measure was the rate of SRA and HIPA orders per 1000 patient days. The pre-intervention period (6/17/2020 – 2/28/2021, 257 days) was compared to the post-intervention period (3/1/2021 – 4/8/2022, 404 days). Results were compared with an unpaired t-test assuming unequal variance.
Results
Comparing pre-intervention to post-intervention, SRA decreased 28.1% from 0.44 to 0.32 orders per 1,000 patient days (p < 0.001). HIPA increased by 0.7% from 0.52 to 0.53 orders per 1,000 patient days (p = 0.9).
Conclusions
The BPA successfully nudged providers to reduce unnecessary SRA ordering. For HIPA, no change in ordering patterns was seen, perhaps owing to the fact that the ordering provider would still have to calculate the 4T score and determine the appropriateness of ordering the test. This may also be due to heightened awareness of HIT and therefore a greater reluctance not to perform the initial HIPA test.
Clinical Implications
This intervention increased patient safety by decreasing inappropriate testing for HIT and therefore false positives. A false positive would have resulted in the patient being unnecessarily switched from heparin to other riskier alternatives.
