Home >  Incite >  Issue 7 > 

Evidence-based benchmarks for cancer services

Research updates

Issue 7, June 2011

Professor Michael Barton discusses the different structures for cancer services in NSW and how we can make sure everyone has the same access to treatment.

There is still considerable variation in access to cancer services, despite increased investment and the availability of evidence-based guidelines from reputable international cancer organisations. For example, the proportion of people with new cases of cancer who received radiotherapy in NSW in 1999 ranged from 27 per cent on the Mid-North Coast to 49 per cent in Northern Sydney.[ 1] Without a benchmark it is difficult to tell if there is under- or over-servicing of a region. Over the past decade, we have developed a series of evidence-based benchmarks to provide a basis for the interpretation of activity data and to assist the planning of cancer services.

Our approach uses a model that combines indications for a treatment with information on the proportions of cancer patients with these indications. We defined a treatment indication as a condition for which the treatment was the treatment of choice; when it offered the best survival, local tumour control, functional result, quality of life or side effect profile.

The proportion of cancer cases with attributes for each treatment decision was determined from epidemiological databases and clinical reports. Indications were mapped out in a 'tree' structure using standard decision evaluation software. The tree was broken into treatment decision points. By summing the proportions for each branch with an indication for a treatment, we can estimate the optimal treatment rate.

Using these methods, we have estimated the proportion of people who should receive radiotherapy or chemotherapy if the guidelines were followed, which we defined as the optimal utilisation.

We examined more than 2,000 published reports and constructed utilisation decision trees for each cancer site. The estimated optimal chemotherapy and radiotherapy utilisation rates for all cancers were 51.0 per cent (95% Confidence Intervals 50.6-51%)[ 2] and 52.3 per cent (95% Confidence Intervals 51.7-53.1%) respectively.[ 3] Indications for treatment were based on level I or II evidence for 83 per cent of chemotherapy indications and 56 per cent of radiotherapy indications. The optimal chemotherapy utilisation rates by tumour sites ranged from a low of 13 per cent for thyroid cancer to a high of 94 per cent for myeloma; and for radiotherapy, from 0 per cent for liver cancer to 92 per cent for cerebral cancers. Actual chemotherapy utilisation varied from 29 per cent in the United Kingdom to 42 per cent in Canada. Actual radiotherapy utilisation varied from 24 per cent in the USA to 43 per cent in Sweden.

There are a number of limitations to these models:

• The epidemiological data are specific to individual populations and some information on rare indications is difficult to obtain.
• The model of radiotherapy only covered cancers that are notified to a cancer registry.
• Non-melanomatous skin cancer and benign conditions were excluded because the incidence of these conditions is unreported and evidence-based treatment guidelines do not exist for most non-malignant conditions.

Optimal utilisation models provide a basis for planning cancer services. Our benchmark for radiotherapy has been adopted nationally, in Europe[ 4] and by the International Atomic Energy Agency, to determine the need for radiotherapy services based on the number and types of cancer in a region.

The utilisation models can be adapted for different populations with different distributions of cancer,[ 5] and to estimate the benefits of cancer treatments.[ 6]

With improved survival of cancer patients, more patients are now having more than one treatment episode.

When there are differences between optimal utilisation estimates and actual practice, the models can also be used to identify the specific indications where that discrepancy occurs. Comparison of patterns-of-care data in lung cancer[ 7] showed that there was under-utilisation of chest radiotherapy in Stage 1 lung cancer.

The question remains whether optimal is really achievable. Levels of radiotherapy utilisation approach the optimal figure in some well-resourced health regions.[ 1] The barriers to optimal utilisation rates are not just resources such as facilities, equipment and staff; some well-resourced health areas have low utilisation rates for some tumours, suggesting that referrer preference may reduce appropriate utilisation.[ 8]

While we acknowledge that models are never perfect, we've based our models of optimal service utilisation on the best available evidence and they are transparent and extensively peer-reviewed. This approach can readily incorporate new knowledge, but is not excessively sensitive to changes that only affect a few indications. It has been successful in persuading policy makers and funders to support the expansion of cancer services.

With improved survival of cancer patients, more patients are now having more than one treatment episode. We need further research to characterise retreatment activity and to develop models that predict retreatment utilisation.

References

1. Barton M. Radiotherapy utilization in New South Wales from 1996 to 1998. Australas Radiol 2000;44:308-14.
2. Ng, W, Jacob, S, James, M, Delaney, G, and Barton, MB. Chemotherapy in cancer care: estimating the optimal chemotherapy utilisation rate from a review of evidence-based clinical guidelines.  1-8-2008. Sydney,  CCORE.
3. Delaney G, Jacob S, Featherstone C, Barton M. The role of radiotherapy in cancer treatment: estimating optimal utilization from a review of evidence-based clinical guidelines. Cancer. 2005;104:1129-37.
4. Slotman BJ, Cottier B, Bentzen SM, Heeren G, Lievens Y, van den BW. Overview of national guidelines for infrastructure and staffing of radiotherapy. ESTRO-QUARTS: Work package 1. Radiother Oncol. 2005;75:349-6.
5. Barton MB, Frommer M, Shafiq J. Role of radiotherapy in cancer control in low-income and middle-income countries. Lancet Oncol. 2006;7:584-95.
6. Shafiq J, Delaney G, Barton M. An evidence-based estimation of local control and survival benefit of radiotherapy for breast cancer. Radiotherapy & Oncology 2007;84:11-7.
7. Vinod SK, Barton M. Actual versus optimal utilisation of radiotherapy in lung cancer - Where is the shortfall? Asia Pacific Journal of Clinical Oncology. 2007;3:1-7.
8. Vinod SK, Hui AC, Esmaili N, Hensley MJ, Barton MB. Comparison of patterns of care in lung cancer in three area health services in New South Wales, Australia. Intern.Med.J 2004;34:677-83.