Prostate cancer treatment in private and public health services

More evidence is required to explain the differences in cancer treatment reported by data linkage studies

Investigations linking cancer registry and administrative data sets provide evidence about outcomes that can highlight differences in treatment. What should follow is further quantitative and qualitative research, not speculation, to identify evidence‐based reasons for the differences, and to suggest remedies.

In this issue of the Journal, te Marvelde and colleagues report that, by linking Victorian Cancer Registry data to the Victorian Admitted Episodes and Radiotherapy Minimum Datasets, they identified differences between patients diagnosed with prostate cancer in public and private health services in the treatment they received during the 12 months after diagnosis. Private patients were more likely to have radical prostatectomies and less likely to be treated with curative radiotherapy.1 How should we interpret this result? The authors point out that factors other than disease severity and comorbidity seem to influence treatment choices and doctors and patients will have differing viewpoints. Without further information it is not possible to tell whether one group is being overtreated or the fault is underservicing of the other, or whether the difference is not attributable to either of these causes.

In general, one reason for differences in treatment choices is the lack of high level evidence on which to base the choices. For example, there has been a long running debate on the merits of measuring prostate‐specific antigen (PSA) levels. Australia has had a high rate of overdiagnosis and overtreatment of low risk prostate cancer.2

It was hoped that two randomised studies of PSA screening would provide definitive evidence about the benefit of PSA testing, but they did not lead to standardisation of practice. The European Randomised Study of Screening for Prostate Cancer (ERSPC) found that PSA testing was associated with a survival advantage;3 the United States Prostate, Lung, Colorectal and Ovarian (PLCO) trial did not, although subsequent analyses have suggested that differences in trial execution and compliance may account for the differing findings of the two studies.4,5 However, as any survival benefit accrues over a period of years whereas treatment toxicity is more immediate, clinical judgement is required.

The evidence‐based PSA testing guidelines of the National Health and Medical Research Council provide some consensus guidance. At least the paucity of evidence has spurred research to find better biomarkers of disease.6

If surgery or radiotherapy for early prostate cancer are chosen instead of active surveillance or watchful waiting, it is imperative that patients understand the risks and benefits of each modality, with equivalent outcomes, to be able to make informed choices.7 However, patients in rural areas may not have the same access to radiation therapy or favour spending prolonged periods away from home, and this will influence their choices.

It is also worth exploring whether public hospital surgeons are more likely than private practitioners to work in multidisciplinary teams or to participate in multidisciplinary meetings including radiation oncologists, and consequently more likely to refer patients for radiotherapy. Multidisciplinary meetings are less common in prostate cancer care than for many other cancer types, but there is evidence that clinicians benefit from the treatment discussions and that the meetings positively affect patients’ perceptions of treatment recommendations.8,9

Similarly, it has been reported in the US that converting a urology practice to an integrated prostate cancer centre offering intensity‐modulated radiation therapy (IMRT) led to increased use of IMRT and fewer prostatectomies in the centre and in surrounding practices.10 Whether these changes were related to patient preferences, clinical factors, physicians’ financial incentives, or other factors could not be determined.

The socio‐demographic features of patients also contribute to differences between the public and private sectors. In one study, regional differences between US states indicated the lack of clear evidence for choosing radiotherapy or surgery; being older, single, or black (for men under 60) were each associated with greater use of curative radiotherapy.11 In Brazil, a multivariate analysis found that higher age, being black, and higher PSA levels were predictors of non‐surgical treatment; compared with private patients, public patients were older and less likely to have prostatectomies, but the authors could not separate the relative effects of biological and socio‐economic factors on treatment differences.12

In the study by te Marvelde and colleagues,1 private patients were generally younger, more likely to live in cities and in higher socio‐economic areas, had fewer comorbid conditions, and were more often diagnosed with lower grade prostate cancer than public patients. However, after adjusting for age and comorbidity, the prostatectomy rate was still higher in the private sector, confirming other data.13 They also found that treatment differences between private and public practices for patients with International Society of Urological Pathology grade 1 disease had declined during 2011–2017, which is a positive finding. Further research is needed to identify factors associated with differences in treatment patterns, rather than assuming that practitioners have a bias toward their own specialty.13

As with PSA testing, regular evidence updates derived from analyses of data in population registries and clinical registries such as the Australia and New Zealand Prostate Cancer Outcome Registry, guideline updates, and studies based on linkage of registries and administrative data sets should inform the choice of prostatectomy or radiotherapy, in both the public and private sectors.