Related Experiment Video
Updated: May 28, 2026
Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
Using linked routinely collected health data to describe prostate cancer treatment in New South Wales, Australia: a
David E Goldsbury1, David P Smith, Bruce K Armstrong
1Cancer Epidemiology Research Unit, Cancer Council NSW, PO Box 572, Kings Cross, NSW 1340, Australia. davidg@nswcc.org.au
Background:
Population-based patterns of care studies are important for monitoring cancer care but conducting them is expensive and resource-intensive. Linkage of routinely collected administrative health data may provide an efficient alternative. Our aim was to determine the accuracy of linked routinely collected administrative data for monitoring prostate cancer care in New South Wales (NSW), Australia.
Methods:
The NSW Prostate Cancer Care and Outcomes Study (PCOS), a population-based survey of patterns of care for men aged less than 70 years diagnosed with prostate cancer in NSW, was linked to the NSW Cancer Registry, electronic hospital discharge records and Medicare and Pharmaceutical claims data from Medicare Australia. The main outcome measures were treatment with radical prostatectomy, any radiotherapy, external beam radiotherapy, brachytherapy or androgen deprivation therapy, and cancer staging. PCOS data were considered to represent the true treatment status. The sensitivity and specificity of the administrative data were estimated and relevant patient characteristics were compared using chi-squared tests.
Results:
The validation data set comprised 1857 PCOS patients with treatment information linked to Cancer Registry records. Hospital and Medicare claims data combined described treatment more accurately than either one alone. The combined data accurately recorded radical prostatectomy (96% sensitivity) and brachytherapy (93% sensitivity), but not androgen deprivation therapy (76% sensitivity). External beam radiotherapy was rarely captured (5% sensitivity), but this was improved by including Medicare claims for radiation field setting or dosimetry (86% sensitivity). False positive rates were near 0%. Disease stage comparisons were limited by one-third of cases having unknown stage in the Cancer Registry. Administrative data recorded treatment more accurately for cases in urban areas.
Conclusions:
Cancer Registry and hospital inpatient data accurately captured radical prostatectomy and brachytherapy treatment, but not external beam radiotherapy or disease stage. Medicare claims data substantially improved the accuracy with which all major treatments were recorded. These administrative data combined are valid for population-based studies of some aspects of prostate cancer care.
