Improving Laboratory Monitoring Compliance When Utilizing Clozapine

Background: Clozapine is an effective medication that requires close monitoring for potentially serious side effects. The complexity of laboratory monitoring results in missed or late testing that raises the risk of adverse events. In addition, clozapine remains underutilized, possibly due to its unique side-effect profile and complex monitoring requirements. Methods: An order set was created within the Epic electronic health record (EHR) to automate lab orders for clozapine monitoring. Compliance with an established clozapine laboratory monitoring protocol was assessed for all patients initiated on clozapine during pre-intervention (6/1/2020-6/30/2021) and post-intervention (1/1/2022-6/30/2022) periods through manual chart review. Baseline demographics were compared with Mann-Whitney-U and chi-square tests as appropriate. Compliance for individual labs (primary outcomes) and rate of clozapine initiation (secondary outcome) were compared using Fisher exact test. A Benjamini-Yekutielli test was used to correct for multiple testing. All calculations were performed using R (version 4.1.1). Results: Statistically significant differences between pre and post-intervention groups were found in median age, median days admitted, gender, and race. For all labs, there was improved compliance (reductions in missed and late orders). Two labs had improved compliance with statistical significance: troponin (OR 25.00, 95% CI [3.67-1076.31], adj. p<0.001***) and CRP (OR Inf, 95% CI [5.73-Inf], adj p<0.001***). Following the creation of the order set, the odds of initiating clozapine in patients appeared higher but did not reach statistical significance (OR 0.57, 95% CI [0.26-1.28], p=0.15). Discussion: Laboratory monitoring compliance improved following the creation of a clozapine order set, most significantly for CRP and troponin tests. The most notable difference in baseline characteristics between pre- and post-intervention groups was race, the latter being more racially diverse (74% versus 65% white, respectively). Past research suggests that if anything, clozapine may be less likely to be initiated in more racially diverse populations; yet our data suggested a possible increased rate of initiation. Taken together, our findings support the use of automated order sets within an EHR to improve clozapine laboratory monitoring compliance.