Clinical Epidemiology (Nov 2018)

A validated algorithm to identify recurrence of bladder cancer: a register-based study in Denmark

  • Rasmussen LA,
  • Jensen H,
  • Virgilsen LF,
  • Jensen JB,
  • Vedsted P

Journal volume & issue
Vol. Volume 10
pp. 1755 – 1763

Abstract

Read online

Linda Aagaard Rasmussen,1 Henry Jensen,1 Line Flytkjær Virgilsen,1 Jørgen Bjerggaard Jensen,2,3 Peter Vedsted1 1Research Centre for Cancer Diagnosis in Primary Care (CaP), Research Unit for General Practice, Department of Public Health, Aarhus University, 8000 Aarhus C, Denmark; 2Department of Clinical Medicine, Aarhus University, 8200 Aarhus N, Denmark; 3Department of Urology, Aarhus University Hospital, 8200 Aarhus N, Denmark Purpose: Recurrence of cancer is not routinely registered in the national registers in Denmark. The aim of this study was to develop and validate a register-based algorithm to identify patients diagnosed with recurrence of invasive bladder cancer (BC). Materials and methods: We performed a cohort study based on data from Danish national health registers. Diagnosis codes and procedural codes in the Danish National Patient Register and Systematized Nomenclature of Medicine codes in the Danish National Pathology Register were used as indicators of cancer recurrence. Status and date of recurrence as registered in the Danish Bladder Cancer Database (DaBlaCa-data) were used as the gold standard of BC recurrence to ascertain the accuracy of the algorithm. Results: The algorithm reached a sensitivity of 85% (95% CI: 78–91), a specificity of 90% (95% CI: 79–96), and a positive predictive value of 95% (95% CI: 89–98). The algorithm demonstrated superior performance in patients undergoing cystectomy compared to patients undergoing radiotherapy as primary BC treatment. The concordance correlation coefficient for the agreement between the recurrence dates generated by the algorithm and the gold standard was 0.96 (95% CI: 0.95–0.98), and the estimated date was set within 90 days of the gold standard date for 90% of patients. Conclusion: The proposed algorithm to identify patients diagnosed with BC recurrence from Danish national registries showed excellent performance in terms of ascertaining occurrence and the timing of BC recurrence. Keywords: urinary bladder neoplasms, recurrence, algorithms, validation studies, registers, Denmark

Keywords