Feasibility of Clinically Prioritized Colorectal Cancer Quality Indicators With a New South Wales Population-Based Linked Dataset.

Aim: To determine the feasibility of using population-based linked data to measure an Australian multidisciplinary set of 26 colorectal cancer (CRC) quality indicators.

Methods: Data were obtained on adult patients diagnosed with CRC (ICD-10-AM codes C18-C20) between July 1, 2005 and December 31, 2019 from the New South Wales (NSW) Cancer Registry. The NSW Cancer Registry data were linked to the Clinical Cancer Registry, Admitted Patient Data Collection, and death records. The feasibility assessment included (1) mapping required variables to available data, (2) review of publicly available reports to identify routine reporting of the indicators, (3) assessment of data completeness and coverage using proportional analyses, and (4) pilot test calculation of feasible indicators where data exist.

Results: Data mapping found that 14 indicators were potentially feasible. Linked data were available for 38,430 patients to test eight surgical indicators and 8489 patients to test six neoadjuvant therapy indicators. The data required to measure these indicators had significant limitations in data coverage, completeness, and quality, rendering the calculations unreliable and some implausible. The data completeness for staging ranged from 74% to 85%, and almost one half of diagnosis dates were illogical. Overall, six of the 26 indicators were feasible and reliable to measure. These addressed unplanned reoperation/readmission, colonoscopy, surgical mortality, and survival.

Conclusion: This study identified six clinically relevant quality indicators feasible to measure using NSW population-based data. However, these indicators were surgical processes and outcomes. There are insufficient data to produce adequate and clinically meaningful quality measurements for a multidisciplinary CRC team, particularly in diagnostic workup, neoadjuvant therapy, and supportive care.