Primary content

Does travel distance affect radiotherapy utilisation in NSW and the ACT?

Authors

G Gabriel (Lead)

M Barton

G Delaney

Aims

To calculate actual RTU and identify factors affecting RTU with special emphasis on the effect of distance.

Method

Data collected from all radiotherapy departments (RTD) were linked to NSW/ACT Central Cancer Registries (CCR). Geographic Information System (GIS) software was used to calculate road distance between patient residence and the closest RTD. Patients were excluded from the study if their nearest RTD was outside NSW.

Results

Road distance from patient residence to the nearest RTD ranged from 0.1-772km.

Overall RTU rate was 26%. This represents 54% of the estimated optimal RTU. RTU rates decreased with increasing distance from patient residence to the nearest RTD (p <0.0001).

Multivariate logistic regression shows that male gender, younger age and shorter travel distance were significantly associated with receiving radiotherapy. Patients were 10% less likely to receive radiotherapy for each extra 100km they had to travel to the nearest RTD (p<0.001).  Patient diagnosed with regional or distant stage were more likely to receive radiotherapy than patients diagnosed with unknown stage (p<0.001). Survival analysis shows increased survival for patients receiving radiotherapy.

Implications

This study is unique because individual radiotherapy data were collected from all public and private RTDs and were matched to NSW-ACT CCR records. By using GIS software, it has been possible to calculate actual road distance from patient residential address to the nearest RTD and to identify the magnitude of cross-border flows.

Our study showed that actual RTU rate was 26%. This represents only half of the estimated optimal RTU rate. Based on the optimal and actual rates, it is estimated that 22% of the study cohort or 21,560 patients have missed out on the potential benefits of evidence-based radiotherapy during the study period.

While our actual RTU rate is low compared to the optimal utilisation rate, it is comparable with other Australian and the North American data.