Accuracy of administrative coding data in colorectal cancer resections and short-term outcomes

BACKGROUND: Administrative data are routinely captured for each hospital admission and may serve as an alternative source for populating databases. This study aims to determine the accuracy of administrative data to provide tumour characteristics and short-term post-operative outcomes, after a colorectal cancer (CRC) resection, compared with clinical data. METHODS: A retrospective study of all CRC resections at a single hospital from 1 January 2008 to 31 December 2013 was conducted. Local administrative data were coded as per ICD-10-AM (International Classification of Diseases, Tenth Revision, Australian Modification) and Australian Classification of Health Interventions. Clinical data for all patients were extracted from the medical charts and compared with administrative data. Code combinations and algorithms were used to improve the accuracy of administrative data. RESULTS: A total of 436 patients were identified. The accuracy of algorithms combining tumour location and type of operation for right colon, left colon and rectum were 93, 89 and 88%, respectively. The accuracy of histological type was 89%, lymph node status 92% and metastasis status 88%. The accuracy of return to theatre and in-hospital mortality was 100%. CONCLUSION: Administrative data can provide reliable information on tumour details and short-term post-operative outcomes. The potential for administrative data to validate data captured in registries and be used independently for audit and research should be further explored