Watching the doctor-watchers. How well do peer review organization methods detect hospital care quality problems?

摘要: Objective. —To determine how well one state's peer review organization (PRO) judged the quality of hospital care compared with an independent, credible judgment care. Design. —Retrospective study comparing a PRO's review, including initial screening, physician and final judgments, independent "study judgment" based on blinded, structured, implicit reviews records. Setting. —One medical surgical Medicare hospitalizations during 1985 through 1987 audited randomly by PRO. Sample. —Stratified random sampling records: 62 records that passed PRO screening process were not referred for review; 50 failed screen then confirmed physicians to be "quality problems." Intervention. —None. Main Outcome Measure. —A below standard or above mean overall ratings five internists in diagnosis related groups (DRGs) surgeons DRGs. Each step was evaluated many passing failing (positive negative predictive value) classified (sensitivity specificity). Results. —An estimated 18% reviewed according judgment, 6.3% problems (difference, 12%; 95% confidence interval, 1 23). The detect refer two three standard. In addition, only judgment. Therefore, agreed little more than expected chance, especially about poor Although correctly above, it detected 11% study. Conclusions. —Most all, this would improved additional preliminary screens identify 67% but its screening. also must accurate order cost-effective, as slightly better at identifying More reproducible is needed might accomplished reviewer selection training, structured method, reviewers per record. ( JAMA . 1992;267:2349-2354)