Differences in treatment choices for localised prostate cancer diagnosed in private and public health services

In reply: We thank Mark and colleagues and Woo and Murphy for their comments. We agree that the lack of information on prostate‐specific antigen (PSA) and T stage of the tumours is a weakness of our study, which was acknowledged in the article. The International Society of Urological Pathology (ISUP) grade of the tumour is a reliable predictor of prognosis by itself and unarguably one of the key determinants for making treatment recommendations. The fact remains that there were stark differences between the private and public health systems in the type of treatment men with similar grade tumours chose, even when corrected for a number of possible variables. This choice depends largely on how a patient’s prognosis is assessed by his treating specialist using any one of many available and differing prognostic tools.

We also agree that the Victorian Admitted Episodes Dataset (VAED)‐derived Charlson Comorbidity Index may understate comorbidities for some men due to the select number of comorbidities included in the index, which could partly explain some of the differences observed, especially in older men. Differences in treatment use in men aged under 70 years remain stark.

The Prostate Cancer Outcomes Registry–Victoria (PCOR‐Vic) only captured data for around 50% of Victorian men with newly diagnosed early stage prostate cancer and used a voluntary opt‐in process to enter patients into the database, whereas our study is population‐based. Hopefully, more treating doctors will use databases like PCOR‐Vic in the future to help inform practice.

We agree that the availability of robotic surgery and magnetic resonance imaging are two of many other factors that may have influenced treatment decision making by some men. Whether they would be enough to explain the large differences we observed will hopefully be explained better by future prospective outcomes research.

Contrary to the statement that the we overstated the rate of overdiagnosis of early stage prostate cancer in Australia, the rate estimated in the study by Glasziou and colleagues1 referred to by Woo and Murphy was 42% when comparing the two years, 1982 and 2012, whereas the study by Culp et al2 that we cited used data up to 2018.