Benchmarking can forecast radiation therapy utilization

Oct 4, 2010

In countries such as Canada that provide healthcare services to their citizens, it's critically important to accurately estimate healthcare utilization. Healthcare institutions that answer directly to their citizens, and the government agencies that oversee them, need to optimize resources invested in facilities, equipment, and staff while ideally providing equal and timely access to all.

Forecasting resource allocation is a tightrope act, and it becomes even more difficult when wide variations exist in the proportion of cancer patients who require radiation therapy treatment. Such is the situation with prostate cancer patients, with published utilization ranges of 20% to 50%.

Estimates based on evidence-based guidelines methodology and estimated rates of diagnosis for a specific population tend to be utilized for planning and forecasting. Over the past few years, a team of Canadian researchers has been testing the accuracy of a criterion-based benchmarking model to predict radiation therapy utilization requirements for different types of cancer, and the group believes this model to be robust. Their most recent comparison of criterion-based and evidence-based models focused on prostate cancer patients, in an analysis recently published online in Clinical Oncology (August 24, 2010).

The team, headed by Marc Kerba, MD, assistant clinical professor of oncology at the University of Calgary in Alberta and a member of the department of oncology at Tom Baker Cancer Centre in Calgary, suggested that the use of a criterion-based benchmarking model reflects a best-practice perspective of a population's need for radiation therapy. The model may be used to determine if a given population is underutilizing or overutilizing the available treatment resources.

Furthermore, by comparing the calculated measure of need for treatment to the actual utilization rate of a given population, an objective analysis can be made about whether this reflects the norm or under- or overutilization.

In industry, benchmarking measures the performance and capabilities of products and services and the best practices of industry leaders. In healthcare, Kerba and colleagues explained, benchmark criteria should be based on utilization data where radiation therapy services are freely accessible to any patient who needs them, and decisions about its use should be based on appropriate evidence-based guidelines. Additionally, there should be no financial barriers to patients regarding the healthcare treatment they need, and physicians should not influence the type of care they recommend based on financial remuneration for either themselves or the healthcare facility that provides it.

Applying theory to practice

This study's primary objective was to determine the initial and lifetime treatment estimates of the percentage of prostate cancer patients who undergo radiation therapy in the province of Ontario and in the U.S. Initial treatment, which included both external-beam radiation therapy and brachytherapy, was defined as occurring within 12 months of diagnosis.

The researchers collected and analyzed surgical and radiation therapy data of nearly 130,000 prostate cancer patients treated between 1997 and 2001, including 35,379 whose cases were recorded in the Ontario Cancer Registry and 93,275 patients included in the Surveillance, Epidemiology, and End Results (SEER-9) database. (SEER is a registry including data from five states and four cities representing 10% of the U.S. population.)

The demographics of both the Canadian and U.S. patient groups were very similar. The median age was 69 to 70 years. In Ontario, 94.4% of the patients had a pathological diagnosis of adenocarcinoma; in the U.S., this represented 95.4%.

The researchers' utilization rates were based on evidence-based estimates of the appropriate rate of radiation therapy for prostate cancer published in 2003 and 2005, respectively, in the International Journal of Radiation Oncology, Biology, Physics (2003, Vol. 55:1, pp. 51-63) and Cancer (2005, Vol. 103:3, pp. 462-473).

Benchmarking-model rates were based on the average of the individual radiation therapy rates of the four Ontario counties whose cancer treatment centers had lower-than-average wait times for patient treatment. The total number of patients treated in these counties was divided by the total number of eligible invasive prostate cancer cases diagnosed to calculate the benchmark rate.

The evidence-based estimates and the criterion-based benchmark estimates for the total population were 32.3% and 37.2%, respectively, for initial therapy. The respective rates for lifetime need for radiation therapy were 61.2% and 59.1%.

These results show that criterion-based benchmarking analysis is comparable to evidence-based modeling, according to the authors.

"A significant advantage of the criterion-based benchmarking approach over evidence-based estimates of need is that its assumptions are implicit, capturing patient preferences influencing radiotherapy utilization over the course of many individual patient and physician decisions," they wrote.

By Cynthia E. Keen
AuntMinnie.com staff writer
October 5, 2010