Quality assessment in surgery: riding a lame horse.

BACKGROUND: Quality assessment in surgery is paramount for patients and health care providers. In our center, quality assessment is based on the recording of preoperative risk factors of each patient and a well-established grading system to track complications. Our prospective quality database is administrated by residents. However, the validity of such data collection is unknown.
METHODS: To evaluate the validity of the recorded data, a specially trained study nurse audited our prospective quality database over a 6-month period. In the first 3 months, the audit was done in an undisclosed manner. Then, the audit was disclosed to the residents who were again subjected to a teaching course. Thereafter, the audit was continued in a disclosed manner for another 3 months, and data were compared between the 2 periods. Furthermore, we inquired about the strategies to assess surgical quality in 108 European medical centers.
RESULTS: Surprisingly, residents failed to report most complications; 80% (164/206) and 79% (275/347; P = 0.27) of the negative postoperative events were not recorded during the first and the second period, respectively. When captured, however, grading of complications was correct in 97% of the cases. Moreover, comorbidities were incorrectly assessed in 20% of the patients in the first period and in 14% thereafter (P = 0.07). The survey disclosed that residents and junior staff are responsible of recording surgical outcome in 80% of the participating European centers.
CONCLUSIONS: Recording of outcome by surgical residents is unreliable,despite active and focused training. Hence, surgery should be evaluated by dedicated personnel.