The influence of waiting time on treatment choices among prostate cancer patients

Objectives. (1) To examine the spatial and temporal patterns in the length of time patients waited for treatment for prostate cancer from 1996 to 2001 in Ontario. (2) To evaluate the influence of recent treatment waiting times, at the time a patient was diagnosed with prostate cancer, on the type of treatment selected.;Results. 24% of the study population selected radiotherapy as the curative treatment option and 48% selected surgery as the curative treatment option. Median waiting times were 148 and 91 days respectively. Differences in median (actual) waiting time for radiotherapy or surgery treatment of greater than 2 weeks were observed according to CCO treatment region and calendar year.;The prevailing waiting times for radiotherapy and for surgery were not related to the choice of any treatment (radiotherapy or surgery) versus no treatment, nor related to the choice of radiotherapy versus surgery. Among the covariates investigated, decreasing age, higher SES of residence county, more recent year, and specific CCO regions were associated with the selection of any treatment versus no treatment. Increasing age, further distance to a comprehensive cancer centre, and specific CCO regions were associated with selection of radiotherapy versus surgery. Patients with a higher socio-economic residence area and diagnosis at a prostatectomy centre were more likely to have surgery than radiotherapy.;Conclusion The prevailing waiting time at diagnosis did not influence treatment choices among patients with localized prostate cancer.;Methods. A retrospective cohort study was conducted on 17,626 prostate cancer patients diagnosed between 1996 and 2001 in Ontario. To address the first objective, the individual waiting time was defined as the interval from the date of diagnosis to the start of the initial treatment. The spatial and temporal patterns of interest were investigated. To address the second objective, the exposure of interest was defined as the prevailing waiting time for radiotherapy and surgery in the 3 months prior to diagnosis. The individual prevailing waiting time assignment was dependent on a patient's county of residence and the calendar month and year at diagnosis. Separate logistic regression models were established to examine the effect of prevailing waiting time on treatment selected---curative treatment (radiotherapy/surgery) versus neither treatment, and radiotherapy versus surgery, after controlling other covariates.