An organization-wide analysis of laboratory test cost variables and costs as displayed in the electronic health record: first steps
Karen Fang, Noah Hoffman, Niklas Krumm, Patrick Mathias- General Medicine
Abstract
Displaying the cost of laboratory tests and medications in the electronic health record (EHR) at the time of order may influence ordering practices. Our organization’s EHR was configured to display a semi-quantitative value adjacent to laboratory orders: “$” tests imply relatively little cost and “$$$$” tests imply resource intensive tests. Cost display is associated with charges from hospital or clinic fee schedules by default but can also be configured to reflect direct laboratory costs. To understand our system’s configuration and potentially improve the information being provided to our clinicians upon ordering, all the test ordering data was compiled and analyzed for all individual “orderable” tests within the year of 2021 (Jan 1st 2021 – Dec 31st 2021). Variables analyzed included frequency of test ordering, average test turnaround time, and each of the various possible measures of “cost”: direct costs based on labor and reagents/supplies (when available), reimbursement from the Clinical Laboratory Fee Schedule, and pricing from the fee schedule of the largest hospital in our organization (chargemaster). Results showed an extensive rightward skew in each of the “cost” variables, with most routine tests, like the complete blood count (CBC) or complete metabolic panel (CMP), falling in the lower price range and a number of outlier tests with much higher costs (newer forms of molecular testing). There was minimal linearity in correlations between the various price variables. Correlation between chargemaster price by quintile and actual dollar signs displayed in the EHR showed that the two often disagree. No definitive discernable logic was thus identified for the number of dollar signs shown in the EHR based on the possible “cost” variables analyzed here. Between the lab’s best estimate for direct costs and chargemaster pricing, a rough correlation with slope greater than 1 was identified, suggesting that chargemaster pricing is systematically higher than the cost of running each test. Between the cost of running each test and reimbursement via the Clinical Laboratory Fee Schedule, a rough correlation with slope less than 1 was identified, suggesting that lab costs are systematically higher than reimbursement. The current lack of strong correlations between these various “cost” variables is helpful as a first step in examining how to improve our organization’s EHR system, assessing our laboratory’s cost effectiveness, and discussing potential future considerations for hospital chargemaster pricing as well as the US system of reimbursement. Further work is needed to clarify the relationship between cost displays in EHR and physician ordering practices.