brachytherapy versus pelvic external beam radiotherapy for patients with endometrial cancer of high-intermediate risk (PORTEC-2): An open-label, non-inferiority, randomised trial, Lancet 375 (2010) 816–823. 43 Chapter 3 ASSESSING NEEDS AND DEMAND FOR RADIOTHERAPY M. Barton, M. Williams 3.1. INTRODUCTION Cancer services, such as screening, surgery, chemotherapy or radiotherapy, should be delivered in the type and amount that meet local demand. Estimating demand requires knowledge of the types and numbers of cancers and the indications for services. For example, the demand for breast screening can be calculated by determining the number of women aged 50 to 70 years old. It is more complicated to determine the demand for services, such as radiotherapy or chemotherapy, that have a large number of indications relevant to small proportions of the cancer population. Different populations will have different incidence rates of cancer, and the proportions of the common types of cancer may vary. Cancer registries provide information on the types and frequency of cancer in a population. They may also record data about stage at presentation, which has a critical influence on the outcomes. In addition, factors relating to specific groups of patients, such as performance status and co-morbidities, may alter treatment recommendations. Unfortunately all these details are often poorly recorded by cancer registries. Nevertheless, planning of sufficient services to meet the needs of the treatment population is vital in providing optimal care. This chapter describes an evidence based approach to estimating the demand for radiotherapy, and its application to different treatment modalities and different populations. The work was done mainly for Australia [3.1], but has been used in Europe [3.2, 3.3] and North America [3.4]. Cancer services include all cancer control interventions, such as screening, early detection, diagnosis, treatment, palliation and rehabilitation. The estimation of the demand for radiotherapy will be described in detail, and examples given of how this approach has been adapted to other modalities and other populations. 44 3.2. INDICATIONS FOR RADIOTHERAPY Radiotherapy has a role to play in the treatment of nearly all forms of cancer [3.5]. It is the treatment of choice when it offers: — The best chance of cure; — The best improvement in local cancer control; — The best chance of preventing organ/function loss; — The best palliation; — The fewest side effects. Radiotherapy may be used alone or in combination with surgery and/or chemotherapy. It may be used for cure or to relieve symptoms (palliation). The use of radiotherapy has been examined in the treatment of every cancer that makes up 1% or more of all cancers notified to central cancer registries [3.1, 3.6–3.14]. Data on indications can be combined with the epidemiological data on the frequency of each indication for radiotherapy to produce an overall estimate of the proportion of cancer patients for whom radiotherapy is indicated as the treatment of choice at least once during their illness (Table 3.1). The study was subject to sensitivity analysis and extensive peer review by surgeons, radiation oncologists, medical oncologists and epidemiologists. Overall, 52% of new cancer patients would benefit from radiotherapy at least once [3.1]. Radiotherapy is indicated to cure or improve survival in nearly 85% of the cases where radiotherapy is indicated [3.15], but may be less effective when the stage at presentation is more advanced. Other authors have also developed epidemiologically based estimates [3.4, 3.16, 3.17]. A criterion based benchmarking method has been used to estimate need on the basis of utilization in regions where radiotherapy was assumed to be optimal [3.18]. This method uses the highest utilization rates as a benchmark, but does not relate this to the clinical evidence. In developed countries, radiotherapy demand is dominated by prostate, breast and lung cancer, and the three methods for estimating demand have been compared with uptake in British Columbia, Canada. The criterion based method fitted actual practice better. This is something of a self-fulfilling prophecy and is