Clinical pathologists should limit modifier terms used to denote probability of a diagnosis: a survey-based study.

Abstract

To examine the probability estimates for modifying terms used by clinical pathologists when interpreting cytologic samples and compare these to probability estimates assigned to these terms by clinicians, and to provide restricted, standardizing terms used in cytology reports. 49 clinical pathologists and 466 Veterinary Information Network members responded to 2 similar surveys. Online surveys were distributed to diplomates of the European College of Veterinary Clinical Pathologists and clinician members of the Veterinary Information Network, made available between March 17, 2022, through May 5, 2022. Respondents assigned a range of probabilities to each of 18 modifier terms used by clinical pathologists to denote probability associated with diagnoses; clinicians identified terms that would affect their treatment decisions in cases of canine lymphoma. Respondents then provided thoughts about restricting and standardizing modifying terms and assigning numeric estimates in reports. 49 clinical pathologists and 466 clinicians provided responses. For many terms, probability ranges agreed between the 2 groups. However, differences in estimated probability inferred by a term existed for at least 6 terms. Modifying terms could be restricted to 7 largely nonoverlapping terms that spanned the range of probabilities. Clinicians preferred having numeric estimates of probability, but clinical pathologists resisted providing such estimates in reports. Reducing and standardizing the number of modifying terms to reflect specific probability ranges would reduce disagreement between the clinical pathologist's intended probability range and the clinician's interpretation of a modifying term. This could result in fewer errors in interpretation and better patient care.