Context: Utilization review (UR) seeks to improve quality and cost-efficiency of health care. However, how well the process works in practice has not been assessed.
Objective: To describe the outcomes of a sample of physician reviews in terms of the explicit criteria that the UR was designed to implement.
Design: Retrospective analysis of transcripts of precertification reviews.
Participants and setting: California physicians employed by a UR firm conducted 96 interviews from April 1990 to July 1991 with attending physicians who had proposed to insert tympanostomy tubes on a patient younger than 16 years and whose proposals had been found to be inappropriate on an initial screen.
Main outcome measures: The appropriateness rating assigned to each case by the physician-reviewer and by the investigators using explicit criteria. Logistic regression identified factors associated with the reviewers' recommendations to perform surgery and with recommendations at variance from the criteria.
Results: The reviewers recommended 78% of cases for surgery, of which only 29% were supported by the criteria or had extenuating circumstances. The criteria concurred with all 30 of the reviewers' recommendations against surgery. Two factors, female sex (odds ratio [OR], 8.2; 95% confidence interval [CI], 1.2-53.8) and previous tympanostomy tube insertion (OR, 30.9; 95% CI, 2.4-394.8) were associated with reviewer recommendations in favor of surgery that were at variance from the criteria, despite the lack of evidence for either as a mitigating circumstance.
Conclusion: Physician reviewers were more lenient than the explicit criteria that the reviews were designed to implement. In no cases did the reviewers depart from the criteria's recommendations in favor of surgery.