P044: Factors influencing laboratory test ordering by physicians and nurses in the emergency department

Abstract

Introduction: Understanding factors that influence laboratory test ordering in emergency departments (EDs) can help to improve current laboratory test ordering practices. The aim of this study is to compare patterns and influences in laboratory test ordering between emergency physicians and nurses at two ED sites, Halifax Infirmary (HI) and Dartmouth General (DG). Methods: A mixed-methods approach involving administrative data and telephone interviews was employed. Data from 211,279 patients at HI and DG EDs were analyzed. Chi-square analysis and binary logistic regression were used to determine significant factors influencing whether a test was ordered, as well as significant factors predicting likelihood of a nurse or a physician ordering a test. All significant associations had a p-value of <0.0001. Interviews were conducted (n=25) with doctors and nurses in order to explore areas of potential influence in a clinician’s decision-making process, and discuss what makes decision making difficult or inconsistent in the ED. These interviews were analyzed according to the Theoretical Domains Framework. The interviews were coded by two individuals using a consensus methodology in order to ensure accuracy of coding. Results: Overall, laboratory tests were more likely to be ordered at DG than at HI (OR=1.52, 95% CI: [1.48, 1.55]). Laboratory tests were more likely to be ordered by nurses at DG than at HI (OR=1.58, 95% CI: [1.54, 1.62]). Laboratory tests were more likely to be ordered if the ED was not busy, if the patient was over 65, had a high acuity, had a long stay in the ED, required consults, or was admitted to hospital. Doctors were more likely to order a laboratory test in patients over 65, requiring consults or hospital admission, whereas nurses were more likely to order laboratory tests in patients with high acuity or long stays in the ED. Data from the interviews suggested differing influences on decision making between nurses and doctors, especially in the areas of social influence and knowledge. Conclusion: Currently, there is limited research that investigates behaviour of both emergency physicians and nurses. By determining barriers that are most amenable to behaviour change in emergency physicians and nurses, findings from this work may be used to update practice guidelines, ensuring more consistency and efficiency in laboratory test ordering in the ED.