Measuring the Quality of Physician Practice by Using Clinical Vignettes: A Prospective Validation Study

View larger version (40K): [in a new window]   Figure 1. Planned study design showing sites and physician sample by level of training and clinical case for the 3 quality measurement methods. 1 = simple case; 2 = complex case; COPD = chronic obstructive pulmonary disease; DM = diabetes mellitus; MCO = managed care organization; MD = medical doctor (physician); SP = standardized patient; VAMC = Veterans Affairs medical center.

View larger version (20K): [in a new window]   Figure 2. Direct comparison of scores, overall and by disease, using 3 measurement methods: standardized patients, vignettes, and chart abstraction. Scores are expressed as percentage correct; error bars represent upper bound of 95% CIs. Per analysis of variance model, P < 0.001 (overall); P > 0.2 for method by disease interaction. COPD = chronic obstructive pulmonary disease.

View larger version (20K): [in a new window]   Figure 3. Comparison of vignette scores to standardized patient and chart scores, stratified by case complexity and training level. Scores are expressed as percentage correct; error bars represent upper bound of 95% CIs. Per analysis of variance models, P < 0.001 (overall); P > 0.2 for method by case complexity interaction; P > 0.2 for method by physician training level interaction.

View larger version (13K): [in a new window]   Figure 4. Comparison of variations among and within the 4 sites by measurement method. Boxes represent interquartile range, and stems describe 5th to 95th percentile range.

View larger version (9K): [in a new window]   Figure 5. Distribution of unnecessary items ordered by participants while caring for cases depicted by vignettes compared with all tests and referrals entered directly into the medical record after standardized patient visits.