Cancer auditing - how accurate are your data?

AIM

Human involvement in the collection and entering of information into a database leads to a degree of error. The aim of this study was to assess the concordance between two individuals blinded from each other who independently collected information on the same set of patients and entered it into a colorectal neoplasia database.

METHOD

A colorectal research nurse and a surgeon independently maintained an electronic database on all new patients admitted with colorectal neoplasia under the surgeon over a 5-year period. Twenty-three key endpoints were selected from the database in order to determine the agreement between the two observers. The κ statistic (for nominal and ordinal data) and the concordance correlation coefficient (for interval data) were used to determine the level of agreement between the two data sets.

RESULTS

Both observers recorded 432 new referrals during this period. There was only complete concordance between the two databases with respect to the number of new patients and returns to theatre within 30 days. Nonetheless, there was almost perfect concordance between the two data sets for a majority of the endpoints. The most important areas of variance were in the length of stay (κ=0.78), the American Society of Anesthesiology grade (κ=0.41), emergency surgery (κ=0.36), nodal staging (κ=0.54) and time to recurrence (κ=0.77).

CONCLUSION

This study highlights a number of important areas of data inaccuracy in a prospective colorectal database. The inaccuracies were due to observer bias, issues of data interpretation, or just difficulty in collecting the information accurately.

Colorectal cancer surgery 2000-2008: evaluation of a prospective database

BACKGROUND

  Colorectal cancer is a common cause of cancer death in Australia and is primarily managed operatively. Surgical databases are valuable in monitoring performance in cancer treatment and detecting problems and trends.

METHODS

  Diagnostic and treatment variables and short-term outcomes were gathered prospectively for patients undergoing resection for colorectal cancer over a 9-year period. Survival data were obtained by linkage to state and interstate death indices.

RESULTS

  Eight hundred and five patients underwent resection for colorectal cancer during the study period. Overall 5-year survival was 61%. Five-year cancer-specific survival was 73%. Five-year cancer-specific survival for Australian Clinico-Pathological Staging (ACPS) stages A, B, C and D was 96, 80, 61 and 19%, respectively (P < 0.0001). Emergency presentations showed diminished survival (59% versus 75%, P < 0.0001) after controlling for age and stage (hazard ratio (HR) 1.78, P= 0.005), as did transfusion recipients (63% versus 74%, P= 0.0014; HR 1.78, P= 0.004). Anastomotic leak did not affect survival in multivariable analysis. Non-cancer causes accounted for 26% deaths, primarily comprising cardiovascular deaths in the elderly.

DISCUSSION

  High case ascertainment, data completeness and accuracy can be obtained with prospective, independently gathered data linked electronically to national death records. Survival for colorectal cancer in South Australia continues to improve. Close follow-up for disease recurrence is warranted for transfusion recipients, emergencies and advanced disease. Locally managed databases with linkage to state registries and other institutions are powerful methods to improve data quality and surgical care at a national level.