Quantifying the Clinical Significance of Drug—Drug Interactions: Scaling Pharmacists' Perceptions of a Common Interaction Classification Scheme
OBJECTIVE: To develop a precise, interval level scale of the clinical significance of drug—drug interactions that reflects the professional judgments of practicing pharmacists. PARTICIPANTS: A convenience sample of 63 practicing pharmacists representing hospital (clinical and staff) and retail (chain and independent) practice settings. METHOD: Pharmacists judged the similarity among 15 interaction categories that have been commonly used to classify drug—drug interactions. A multidimensional scaling technique produced a spatial representation (i.e., a psychological map) of the structure inherent in those similarity judgments. Pharmacists' ratings of clinical significance were projected onto that same spatial representation using a multiple regression procedure, and the resulting information was used to develop a scale of clinical significance. RESULTS: The clinical significance scale developed from pharmacists' judgments was substantially different from a comparison scale published in a popular reference. The new scale was more precise than the comparison scale, and it also approximated an interval level of measurement. The judgments used to produce the new clinical significance scale were not reliably influenced by pertinent demographic characteristics of the sample. CONCLUSIONS: Inconsistencies between published clinical significance scales and the professional judgments of practitioners could affect patient care to the degree that a summary measure of clinical significance affects a practitioner's response to a potential drug—drug interaction. The clinical significance scale developed in this study has good measurement characteristics and reflects the professional judgments of practicing pharmacists. Use of the new scale is recommended on these grounds, although further assessment of its generality is warranted.