Usefulness of a state-legislated, comparative database to evaluate qualityin colorectal surgery

PURPOSE: Colorectal surgery, a high-volume procedure, has been targeted for performance improvement to reduce length of stay. Specific postoperative q uality indicators and readmission rates should be analyzed concomitantly to assure that adverse events are not associated with earlier discharge. METH ODS: From July 1, 1990, to June 30, 1997, 1,218 consecutive patients who un derwent transabdominal colorectal surgery were analyzed for length of stay, mortality,, morbidity, and discharge disposition. Each patient was assigne d an Admission Severity Group rating 0 to 4 using a hospital-based state-le gislated software system (Atlas(TM)) to validate comparative performance in ternally and externally. Readmission data within 120 days of discharge were available for the last 678 consecutive patients from July 1, 1993, to June 30, 1997, using Lastword(TM) (computerized medical records). RESULTS: The annual frequencies of the 1,218 procedures were 173, 183, 175, 146, 167, 18 9, and 185, respectively, from July 1990 through June 1997. Severity distri bution was 32 for Admission Severity Group 0, 517 for Admission Severity Gr oup 1, 540 for Admission Severity Group 2, 128 for Admission Severity Group 3, and 1 for Admission Severity Group 4, with no annual difference (P = 0. 012). There was a significant reduction in total length of stay of 3.1 (12. 9-9.8) days during the seven years (P = 0.001). The overall operative morta lity rate was 1.4 percent, and the morbidity was 2.6 percent, with no annua l differences (P = 0.655 and P = 0.033, respectively). The disposition to h ome did not change (P = 0.21). Of the 678 patients followed up for readmiss ion, 100 (14.7 percent) were readmitted within 120 days, with no annual dif ference (P = 0.302). CONCLUSION: Mortality, morbidity, disposition, and rea dmission rates were not affected by a decreased length of stay after colore ctal surgery.