Physician attitudes, self-estimated performance and actual compliance withlocally peer-defined quality evaluation criteria

Citation
Pj. Saturno et al., Physician attitudes, self-estimated performance and actual compliance withlocally peer-defined quality evaluation criteria, INT J QUAL, 11(6), 1999, pp. 487-496
Citations number
32
Categorie Soggetti
Public Health & Health Care Science
Journal title
INTERNATIONAL JOURNAL FOR QUALITY IN HEALTH CARE
ISSN journal
13534505 → ACNP
Volume
11
Issue
6
Year of publication
1999
Pages
487 - 496
Database
ISI
SICI code
1353-4505(199912)11:6<487:PASPAA>2.0.ZU;2-1
Abstract
Background. Physicians' agreement with quality evaluation criteria, and est imates of their own and their colleagues' compliance with these criteria we re compared with actual compliance. Methods. Physicians practicing in 10 health centers in Spain defined 13 qua lity evaluation criteria for two patient conditions (upper respiratory infe ctions and high serum cholesterol). Compliance with criteria was measured b y an external team, using random samples of medical records stratified by c ondition in each health center (n=1000). Concurrently, physicians were surv eyed regarding agreement with the criteria, and were asked to estimate thei r own and their health center's rate of compliance with these criteria. Results. Agreement ratings varied from 5.9 to 9.1 on a 10-point scale. Actu al compliance rates ranged from 1.8 to 91.7% of records. Agreement correlat ed significantly with self-reported compliance but not with actual complian ce. Estimates of one's own and one's health center compliance were positive and significantly correlated for all criteria, but were significantly high er Tur oneself than for one's health center for sis of 13 criteria. Conclusions. Wide variation in physicians' agreement on quality criteria an d in actual performance reveal a lack of clear guidelines. Agreement on cri teria did not always translate into compliance with criteria. Physicians te nded to rate their own performance as better than the average of their peer s, suggesting that aggregate data may not influence physicians to change. S elf-estimate of one's own or one's colleagues performance is not a good pro xy for actual performance so that peer ratings are of dubious value for per formance appraisal.