Objective: To determine whether physicians’ posttest probability estimates are influenced by receiving test characteristics and impact their subsequent clinical decisions.
Design: Questionnaire-based randomized controlled trial.
Setting: Mailed survey with a vignette describing an infant whose pretest likelihood of pertussis was 30% and direct fluorescent-antibody (DFA) test was negative for pertussis.
Subjects: Nationally representative sample of U.S. pediatricians (N = 1502).
Interventions: Random receipt of either: 1) no additional information (controls); 2) the DFA’s sensitivity & specificity (TC group); or 3) the test’s sensitivity & specificity with their definitions (TCD group).
Main Outcome Measures: Estimated posttest probability (PTP) of pertussis; PTP of 0.50; “nearly correct” PTP (correct PTP ± 5%).
Analyses: Chi-square and t-tests.
Results: With 635 participants, the survey’s estimated response rate was 49.7%. Despite the negative DFA result, 67% of participants estimated a PTP higher than the pretest probability of 30% (mean PTP = 0.41; SD ±0.26). The mean PTP of the TCD group was significantly higher than controls (0.45 vs. 0.38, p<0.001), while those of TC and controls did not differ significantly (0.41 vs. 0.38, p = 0.16). Decision support had 2 distinct effects on participants; compared to controls, significantly more TC and TCD participants estimated the PTP to be 0.50 (38% vs. 17%, p<0.0005; 41% vs. 17%, p<0.0005), and also estimated a nearly correct PTP more often (20% vs. 13%, p=0.06; 19% vs. 13%, p=0.08, respectively).
Conclusions: Receiving information about test characteristics increased errors in post-test probability estimation for many pediatricians, while it reduced errors for others.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.