Could changes in clinical practice reduce waiting lists for radiotherapy?

OBJECTIVES

To describe variation in the practice of radiation oncology among the nine cancer centres in Ontario, and to explore the impact of variations in the number of treatments prescribed per case on the overall demand for radiotherapy in the province.

METHODS

Prospectively collected, electronic records of all courses of radiotherapy given at the nine radiotherapy centres in Ontario between 1 January 1996 and 31 December 1997 were analysed to describe inter-centre variations in the number of radiation treatments (fractions) prescribed per case. The effect of the observed inter-centre variations in practice on the total provincial workload was modelled.

RESULTS

During the two-year study, 672,292 fractions were administered as 44,096 courses of treatment. On average there were 63.6 high-energy treatment machines operating in the province over this period. The mean number of fractions per course for the province as a whole was 15.3, and ranged from 10.9 at centre D to 16.0 at centre F. The inter-centre variation in the number of fractions per course was higher when radiotherapy was used palliatively than curatively. The range of variation in the number of treatments per curative course was disease-specific; it was highest for breast cancer, and lowest for the gynaecological malignancies. If each clinical problem had been treated everywhere in the province as it was at the centre which treated it with the fewest treatments, 77,274 fewer radiation treatments each year would have been required to treat the same number of cases. This is equivalent to the capacity of 14.6 treatment machines, and exceeds the estimated shortfall in the current supply of radiotherapy in the province. Province-wide adoption of more parsimonious approaches to the use of palliative radiotherapy, or to the use of curative (adjuvant) radiotherapy in breast cancer, would produce dramatic decreases in the overall demand for radiotherapy machine time. In these particular situations, the literature indicates that shorter courses of treatment are just as effective, and as well tolerated, as longer courses.

CONCLUSIONS

Variation in the way that oncologists prescribe radiotherapy is an important determinant of overall radiotherapy machine workload. Modest changes in prescribing by radiation oncologists, even within the range of current practice in Ontario, and within the range of fractionation schemes that have been shown to produce optimal results, have the potential to reduce waiting lists for radiotherapy.