Spatial patterns of prostate-specific antigen testing in asymptomatic men across Australia: a population-based cohort study, 2017–2018

• Published in: Public health (London) 2023-04, Vol.217, p.173-180
• Main Authors: Kohar, A., Cramb, S.M., Pickles, K., Smith, D.P., Baade, P.D.
• Format: Article
• Language: English
• Subjects: Aged; Australia; Bayes Theorem; Bayesian; Cohort Studies; Early Detection of Cancer; Geographical; Humans; Male; National Health Programs; Prostate-Specific Antigen; Prostatic Neoplasms - diagnosis; Prostatic Neoplasms - epidemiology; Retrospective Studies; Screening; Small area; Victoria
• ISSN: 0033-3506; 1476-5616
• DOI: 10.1016/j.puhe.2023.01.039

In Australia, while prostate-specific antigen (PSA) testing rates vary by broad area-based categories of remoteness and socio-economic status, little is known about the extent of variation within them. This study aims to describe the small-area variation in PSA testing across Australia. This was a retrospective population-based cohort study. We received data for PSA testing from the Australian Medicare Benefits Schedule. The cohort included men (n = 925,079) aged 50–79 years who had at least one PSA test during 2017–2018. A probability-based concordance was applied across multiple iterations (n = 50) to map each postcode to small areas (Statistical Areas 2; n = 2,129). For each iteration, a Bayesian spatial Leroux model was used to generate smoothed indirectly standardized incidence ratios across each small area, with estimates combined using model averaging. About a quarter (26%) of the male population aged 50–79 years had a PSA test during 2017–2018. Testing rates among small areas varied 20-fold. Rates were higher (exceedance probability>0.8) compared with the Australian average in the majority of small areas in southern Victoria and South Australia, south-west Queensland, and some coastal regions of Western Australia but lower (exceedance probability