PERCEPTIONS OF PHYSICIANS USING DIAGNOSTIC SUPPORT SYSTEMS

Purpose: Previous research has shown that patients give lower ratings of diagnostic ability, professionalism, and satisfaction to physicians who use computer-based diagnostic support systems (DSS) (Arkes, Shaffer, & Medow, 2007). To extend this research, two studies were designed to generalize the previous findings to nurse practitioners (Study 1) and mitigate the effect by framing DSS use as “hospital policy” (Study 2). Methods: In both studies, participants read a 1-page scenario describing an interaction between a medical professional and her patient. Study 1 employed a 2 (nurse practitioner or physician) x 2 (no DSS or DSS) factorial design with 132 participants randomly assigned to one condition. In study 2, 190 participants were randomly assigned to 1 of 3 conditions (all physicians): no DSS, DSS, or DSS as hospital policy. Participants in both studies rated the medical professional on 5 dependent variables (thoroughness, wait, professionalism, diagnostic ability, satisfaction) and provided demographic information. Participants in study 1 also completed the Subjective Numeracy Scale and Decision Making Styles Inventory. Results: In study 1, medical professionals described as using a DSS were given significantly lower ratings of diagnostic ability than medical professionals making an unaided diagnosis. However, ratings of physicians did not differ significantly from ratings of nurse practitioners. There were gender and ethnic differences; females and non-whites provided significantly lower ratings on a number of dependent variables. The main effect of gender was qualified by significant gender x health care professional interactions; males gave physicians higher ratings of diagnostic ability and overall satisfaction whereas females gave higher ratings to the nurse practitioners. In addition, decision making style and numeracy were significant predictors of ratings of diagnostic ability. In study 2, there were no significant main effects or interactions with group or ethnicity; however, there was a significant main effect of gender. Replicating study 1, females provided lower ratings on several dependent variables. Conclusions: Previous research demonstrated that patients give lower ratings to physicians using DSS; this was extended to nurse practitioners. Furthermore, this research suggests that individual differences impact ratings of physicians. In both studies, females gave lower ratings to health care professionals than males. This research also provided some evidence that ethnicity, subjective numeracy, and decision-making style may also affect patients' perceptions of their physicians.