Article Text

Download PDFPDF

Forecasting annual incidence and mortality rate for prostate cancer in Australia until 2022 using autoregressive integrated moving average (ARIMA) models
  1. Arul Earnest1,
  2. Sue M Evans1,
  3. Fanny Sampurno1,
  4. Jeremy Millar2
  1. 1 Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
  2. 2 Alfred Hospital, Melbourne, Victoria, Australia
  1. Correspondence to Professor Arul Earnest; arul.earnest{at}


Objectives Prostate cancer is the second most common cause of cancer-related death in males after lung cancer, imposing a significant burden on the healthcare system in Australia. We propose the use of autoregressive integrated moving average (ARIMA) models in conjunction with population forecasts to provide for robust annual projections of prostate cancer.

Design Data on the incidence and mortality from prostate cancer was obtained from the Australian Institute of Health and Welfare. We formulated several ARIMA models with different autocorrelation terms and chose one which provided for an accurate fit of the data based on the mean absolute percentage error (MAPE). We also assessed the model for external validity. A similar process was used to model age-standardised incidence and mortality rate for prostate cancer in Australia during the same time period.

Results The annual number of prostate cancer cases diagnosed in Australia increased from 3606 in 1982 to 20 065 in 2012. There were two peaks observed around 1994 and 2009. Among the various models evaluated, we found that the model with an autoregressive term of 1 (coefficient=0.45, p=0.028) as well as differencing the series provided the best fit, with a MAPE of 5.2%. External validation showed a good MAPE of 5.8% as well. We project prostate cancer incident cases in 2022 to rise to 25 283 cases (95% CI: 23 233 to 27 333).

Conclusion Our study has accurately characterised the trend of prostate cancer incidence and mortality in Australia, and this information will prove useful for resource planning and manpower allocation.

  • adult oncology
  • health informatics
  • epidemiology

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

View Full Text

Statistics from


  • Contributors AE conceived the study, collated the data, analysed and wrote the initial draft as well as the final manuscript. SME provided critical input in the design of the study and writing the manuscript. JM provided critical input in the design of the study and writing the manuscript. FS provided critical input in the design of the study and writing the manuscript.

  • Funding statement The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval Ethics approval or this study was provided by Monash University Ethics Committee (Project ID: 12654).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available on reasonable request. Extra data are available by emailing

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.