Skip to main content
NCBI home page
Journal of Oncology Practice logoLink to Journal of Oncology Practice
. 2012 Jul 24;8(6):358–362. doi: 10.1200/JOP.2012.000617

Technology Resource Planning in Radiation Oncology: Application of a Needs-Based Analytic Framework to Radiosurgery Planning in Ontario

Jeffrey N Greenspoon 1,✉, Daria O'Reilly 1, James R Wright 1, Anthony Whitton 1, Jonathan Sussman 1, Stephen Birch 1
PMCID: PMC3500481  PMID: 23598845

The authors demonstrate through the example of single brain metastases in Ontario that it is feasible to perform explicit needs-based resource planning in radiation oncology.

Abstract

Purpose:

With the emergence of radiosurgery as a new radiotherapeutic technique, health care decision makers are required to allocate capital radiotherapy resources to meet both current and future radiosurgery requirements. The goal of this article is to demonstrate the feasibility of applying an explicit, needs-based model to resource planning in radiation oncology.

Methods:

Using an analytic model that relates radiosurgery need to population size, epidemiology, level of service planned, and productivity, the current radiosurgical need for single brain metastases in Ontario was estimated. The model was populated using Ontario-specific data where possible and supplemented with information from the published literature. Multiway sensitivity analyses were performed to calculate the minimum and maximum technology requirements.

Results:

The calculated number of full-time radiosurgical units required to treat patients with single brain metastases in Ontario was 5.9. Sensitivity analyses performed varying both level of service planned and productivity yielded a range of requirements from 2.5 to 12.2 full-time radiosurgery units.

Conclusion:

We have shown through the example of single brain metastases in Ontario that it is feasible to perform explicit, needs-based resource planning in radiation oncology. As the availability of new specialized technology increases, health care decision makers may use this approach to ensure the needs of their population are met while maximizing productivity and minimizing opportunity cost.

Introduction

Radiosurgery is a rapidly emerging technology in radiation oncology in Ontario. It offers a noninvasive alternative to surgery in properly selected patients.1–3 Although the majority of patients treated with radiosurgery in Ontario are those with brain metastases from a primary tumor, the clinical indications for radiosurgery are expanding. Encouraging clinical trial results have been reported for the primary treatment of early-stage lung cancer, prostate cancer, liver metastases, bone metastases, and a number of benign tumors.4 There are open clinical trials evaluating new indications for radiosurgery in Ontario as well.5–7 Although the treatment of brain metastases is of palliative intent, aggressive local management with radiosurgery has led to both improved survival and quality of life.2,3,8

Radiosurgery can be delivered in two broad ways. Individual cancer centers can purchase equipment to modify a traditional linear accelerator to provide radiosurgery on a conventional machine. This flexible unit can then provide conventional radiotherapy when it is not performing radiosurgery.9 Alternatively, a cancer center can allocate a permanent part of its radiation treatment capacity to performing radiosurgery through the purchase of a dedicated radiosurgery unit. Although this unit will be unable to perform conventional radiation treatments, it has the ability to perform complex radiosurgery treatments.9 A dedicated radiosurgery unit can likely provide radiosurgery in a shorter time than a flexible unit through streamlined treatment planning and machine quality assurance.7,9

In Ontario, radiation treatment delivery resources (megavoltage linear accelerators) are assigned to individual cancer centers based on population size and radiation treatment utilization by Cancer Care Ontario (CCO). In 2010 in Ontario, the average number of radiation courses delivered by a megavoltage machine was 550.10 Radiation therapy wait times are also closely monitored by CCO. In nonurgent cases, it is expected that patients will begin radiotherapy within 2 weeks of the decision to proceed with treatment. To assess the implications of a dedicated radiosurgery unit in Ontario, we adapted the needs-based analytic human resource planning model by Birch et al11 to radiosurgery technology. This allowed for explicit planning of both human and technologic resources needed for radiosurgery in Ontario. We also discuss the opportunity costs of decisions regarding radiosurgery technology in a publically funded health care system with limited resources.

Methods

Analytic Approach to Technologic Requirement Planning

Using the framework for workforce planning by Birch et al,11 we performed an explicit, needs-based assessment for radiosurgery in Ontario. The needs-based approach to workforce planning can be described as follows:

graphic file with name jop00612-2921-m01.jpg

This equation relates current health care system need Nt to the size of the population Pt multiplied by three factors that relate population to need. H represents the level of health (or illness) in the population Pt. (H/P)t is an epidemiologic term to describe the current average health status of individuals in the population. Explicitly defining an epidemiologic term allows for needs adjustments resulting from a change in health status when planning between current populations. This term also allows for epidemiologic changes when planning for the future population Pt+1. For example, a new surgical technique may significantly decrease local tumor recurrences and therefore decrease future illness (Ht+1). Thus, despite a future population growth, the average need per person of our future population would decrease (H/P)t+1. Qt represents the current quantity of health services delivered. (Q/H)t represents the quantity of service provided for a specific health state. This term allows for the explicit consideration of changes in the indications for care. For example, until recently, patients with small lung tumors were not considered appropriate for radiosurgery. However, with the publication of clinical efficacy trials, it is now known that selected patients with small lung tumors benefit from radiosurgery.4 Therefore, whereas 10 years ago the planned amount of radiosurgery administered for small lung tumors was (Q/H)t−10 = 0, there are now selected patients with small lung tumors for whom radiosurgery is the treatment of choice (Q/H)t > 0. The third term in the equation relates to productivity of a treatment modality. For the purposes of this article, it is the inverse of the productivity of each radiosurgery method being examined (N/Q)t. Thus, as the productivity of a treatment modality increases, the number of machines needed to fulfill the population need decreases, assuming all other aspects of health care need are equal. Although there are currently no randomized comparisons of two radiosurgery techniques in the Ontario environment, subgroup analyses of a randomized trial have demonstrated equivalence in terms of efficacy,2 and there is an ongoing open randomized trial in Ontario. This trial will compare the process of delivering radiosurgery for both a dedicated radiosurgery unit and a flexible unit. This will inform the community of the relative productivity of each unit, which will enable more accurate needs assessments and future economic evaluations to be performed.7

Analytic Approach to Radiosurgery Planning for Brain Metastases

Despite the rapid expansion of radiosurgical indications, in Ontario, a majority of patients who currently receive radiosurgery do so for brain metastases.4,10 At the Juravinski Cancer Centre in 2011, 67% of patients treated with radiosurgery had brain metastases.10 The clearest evidence-based indication for radiosurgery is in the management of a single brain metastasis,2,3 and we therefore focus the remainder of this discussion on resource planning for single brain metastases. However, the principles discussed could be applied to other common and growing indications for radiosurgery (eg, multiple brain metastases, lung cancer, prostate cancer, liver metastases, and bone metastases) to create a comprehensive radiosurgery needs assessment. We explicitly describe each term in the analytic model to describe the current need for radiosurgery for brain metastases. We also hypothesize how each term may change over the 10-year life of capital radiotherapy technology in Ontario.

The most straightforward term is Pt. The Ontario population in 2009 was approximately 13.0 million.12 Using an estimated 1% population growth per year, the Ontario population at the end of our 10-year technology cycle would be 14.4 million.

The epidemiologic term for need (H/P)t can be estimated using the current published literature. We do not believe that estimating the current need for radiosurgery based on use is appropriate. Current use may underestimate the current epidemiologic need for two major reasons. First, radiosurgery is a relatively new technology, and some primary oncologists may not be aware of its indications. Second, radiosurgical resources in Ontario are located exclusively in large urban cancer centers, which may limit the access of some members of the population to this resource.13 For these reasons, we rely on the published literature to assess the current epidemiologic need.

Brain metastases are reported to occur in up to 50% of patients with the diagnosis of cancer.14,15 Additionally, up to 40% of patients with brain metastasis will present with a single lesion.15,16 It is estimated that there are approximately 60,000 new cases of cancer per year in Ontario, resulting in up to 12,000 patients per year presenting with a single brain metastasis.10 Approximately 50% of these patients will require an operative procedure for emergent symptom resolution or tissue diagnosis, leaving 6,000 patients in Ontario every year for the elective management of a single brain metastasis from any primary tumor site.10,16,17 With the current size of the population at 13.0 million, there are 6,000 potential candidates for radiosurgery for a single brain metastasis, or 0.05% of the population. If one had access to high-quality local data regarding the incidence of urgent and elective presentations of single brain metastases, one could substitute local data to calculate this percentage.

The incidence of brain metastases is currently rising.18 It has been hypothesized that a decrease in cancer-specific mortality, improvements in systemic therapy, and the relative protection of the brain from systemic therapy are responsible for the rising incidence of brain metastases.3,4,18 We therefore hypothesize that the future epidemiologic need for radiosurgery for single brain metastases will increase over time [(H/P)t+10 > (H/P)t].

As we have described, approximately 6,000 patients in Ontario are potentially eligible for radiosurgery for single brain metastases [potential (Q/H)t]. There are other treatment options available to patients with a single brain metastasis (eg, neurosurgery, conventional radiation, supportive care). There are many patient and physician factors involved in treatment modality decisions. It is beyond the scope of this article to estimate how this decision making may change over time. We therefore did not assume a change in patient or physician preference when estimating (Q/H)t+10. Alternatively, radiosurgery is a relatively new modality in Ontario. Within 1 year of the introduction of radiosurgery to the Juravinski Cancer Centre at McMaster University, we have observed an increase in the referral rate for radiosurgery for a single brain metastasis.19 This is thought to be the result of the education of community oncologists, the creation of multidisciplinary team care for metastatic disease, and a population-based focus on quality of life. Although there is likely a ceiling effect once most oncologists and patients are educated and informed of the treatment options, it seems reasonable to assume that the proportionate amount of radiosurgery performed for patients with a single brain metastasis will increase over the next 10 years [(Q/H)t+10 > (Q/H)t].

The final term to discuss is related to technical productivity. In radiosurgery, the technical productivity (Q/N)t is measured as the number of courses of radiotherapy delivered per year. In Ontario, CCO assumes a 10-hour workday, 5 days per week. Technical productivity is a simple measure of a complex process. Each patient requires immobilization device fitting, radiotherapy planning, and quality assurance before beginning therapy. The average number of courses of radiotherapy delivered per unit yearly in Ontario is 550: (Q/N)t = 550. When planning for radiosurgical needs, as the technical productivity improves, the number of physical units needed to deliver the necessary care decreases. Therefore, the term in the resource planning model is the inverse of technical productivity (N/Q)t. Although radiosurgery can be time consuming and complex, it is reasonable to assume that the technical productivity of radiosurgery will improve analogous to the technical productivity of conventional radiotherapy. In Ontario, there has been a 9% improvement in technical productivity from 2008/2009 to 2009/2010.10 It seems unreasonable to plan for a continued 9% yearly productivity improvement. Decision makers would be in a better position to plan for future productivity; however, for modeling purposes, we assumed a 1% productivity improvement per year over the 10-year life cycle of our radiosurgical equipment. This means that in 10 years, we could expect the average technical productivity to be 607 yearly courses/unit (Q/N)t+10.

Results

Using the information described, we were able to populate the analytic model introduced by Birch et al11 to calculate the current radiosurgical need in Ontario for patients with a single brain metastasis. Pt = 13 million, (H/P)t = 0.05% = 0.0005, (Q/H)t = 50% (range of values tested, 25% to 75%) = 0.5(0.25 to 0.75), (N/Q)t = 1/550 (range tested, 1/400 to 1/650) = 0.0018 (0.0015 to 0.0025). Using these, values the number of full-time radiosurgical units required in Ontario to meet the needs of patients with single brain metastases is Nt = 5.9. Performing an extreme value sensitivity analysis with these ranges of values, we see that the range of radiosurgical machine requirements for single brain metastases is from Nt(low) = 2.5 to Nt(high) = 12.2. We allowed one decimal point in our calculation of full-time radiosurgery machine requirements, because as we have described, some linear accelerators have the flexibility to perform both conventional radiotherapy as well as radiosurgery. This large range is driven by the range of values tested for (Q/H)t. To narrow this range, one could seek local patient and physician preferences to accurately estimate the local (Q/H)t. Using this model, we can incorporate expected future changes into each term in the needs assessment to allow decision makers to plan for the future technology and human resource needs in radiosurgery in Ontario. For example, if the population of patients with a single brain metastasis increases, and radiosurgery machine productivity increases by a similar proportion, there may not be a need to increase future radiosurgery capacity.

Discussion

There are 14 cancer centers in Ontario providing radiotherapy. Each cancer center serves a different population, and the capacity of each treatment center ranges from two to 17 linear accelerators. Each cancer center has a relatively fixed number of linear accelerators and thus a fixed capacity to perform radiotherapy. When a cancer center replaces an existing flexible linear accelerator with a dedicated radiosurgery unit, it is making a tradeoff. This tradeoff is known as the opportunity cost.20 The opportunity cost of increasing radiosurgery capacity is that the cancer center will have a decreased capacity to provide conventional radiation over the next 10 years. Each cancer center could perform an explicit assessment of its current and future needs and examine different methods of meeting the radiosurgical needs of its community. For small catchment areas, it may minimize the opportunity cost of providing radiosurgery by providing a limited flexible capacity for radiosurgery with one flexible linear accelerator, and if the need for radiosurgery exceeds the local capacity, patients could travel to centers with larger radiosurgery capacity.

Figure 1 represents a production possibility frontier for a hypothetic cancer center that has the capacity to treat one patient per hour per unit. A production possibility frontier represents the maximum number of patients treated daily using the available equipment.20 The center is equipped with three units, and it is assumed that radiotherapy and radiosurgery are equally substitutable at all levels of production (hence the straight-line negative slope for the production possibility frontier). Hypothetically, if this cancer center replaces a conventional radiotherapy unit with a dedicated radiosurgery unit, the maximum capacity for radiotherapy decreases from 30 patients daily to 20 patients daily, and the maximum capacity for radiosurgery increases from no patients daily to 10 patients daily. This tradeoff in case mix represents the opportunity cost of purchasing a dedicated radiosurgery unit. If this cancer center had previously used its radiotherapy resources to the maximum capacity, it would treat 30 patients daily (represented by the triangle on the production possibility frontier). If the cancer center, once it substitutes one conventional unit with a dedicated radiosurgery unit, is able to treat 20 patients with conventional radiation daily and 10 patients with radiosurgery daily, this would represent a new point on the production possibility frontier (circle). At this point, the cancer center has changed its production mix; however, it continues to produce with maximum efficiency, because both the triangle and circle are located on the frontier. If, however, the cancer center replaces a conventional radiotherapy unit with a dedicated radiosurgery unit and is only treating five patients daily with radiosurgery, the new point of production on the production possibility frontier is represented by the star. At this point, the cancer center is producing less efficiently than either the triangle or the circle, because the star is located within the frontier. The area bounded by the arrows and the frontier line (lightning bolt) represents points of improved efficiency over the star. The only point bounded by this area achievable with the dedicated radiosurgery machine is the circle. All other points (including those on the frontier) would only be achievable if the cancer center were to elect to purchase a flexible machine capable of producing both radiosurgery and conventional radiotherapy. This hypothetic example highlights the importance of performing a technology needs assessment before resource allocation decisions. A technology needs assessment has the ability to help minimize the opportunity cost of providing new treatment options in both publically funded and fee-based health care systems.

Figure 1.

Figure 1.

Open in a new tab

Production possibility frontier for radiotherapy and radiosurgery. Circle represents the maximum production when one radiotherapy unit is substituted for a dedicated radiosurgery unit. Star represents a point of inefficient production. Area bounded by the two arrows (containing lightning bolt) represents the area of improved efficiency over the star.

In conclusion, in any health care system, it is necessary to consider the opportunity cost of introducing a new health care technology. Radiation oncology provides a setting in which, because of the relatively fixed capacity and known technology utilization, one can use the analytic approach introduced by Birch et al11 to perform a needs assessment. A needs assessment should ideally be performed before a formal economic evaluation of treatment options. We have shown that using a needs assessment as part of resource planning, we can minimize the opportunity cost of providing radiosurgery. This technology needs assessment can be used during workforce planning to ensure that there are appropriate human resources available to appropriately use the new technology. Using this analytic model for technology needs assessment, decision makers can make evidence-informed plans regarding the current and future technology expansion in radiation oncology.

Author's Disclosures of Potential Conflicts of Interest

The author(s) indicated no potential conflicts of interest.

Author Contributions

Conception and design: Jeffrey N. Greenspoon, Anthony Whitton, Stephen Birch

Administrative support: Jeffrey N. Greenspoon

Collection and assembly of data: All authors

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

References

  • 1.Muacevic A, Wowra B, Siefert A, et al. Microsurgery plus whole brain irradiation versus gamma knife surgery alone for treatment of single metastases to the brain: A randomized controlled multicentre phase III trial. J Neurooncol. 2008;87:299–307. doi: 10.1007/s11060-007-9510-4. [DOI] [PubMed] [Google Scholar]
  • 2.Andrews DW, Scott CB, Sperduto PW, et al. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: Phase III results of the RTOG 9508 randomized trial. Lancet. 2004;363:1665–1672. doi: 10.1016/S0140-6736(04)16250-8. [DOI] [PubMed] [Google Scholar]
  • 3.Linskey ME, Andrews DW, Asher AL, et al. The role of stereotactic radiosurgery in the management of patients with newly diagnosed brain metastases: A systematic review and evidence-based clinical practice guideline. J Neurooncol. 2010;96:45–68. doi: 10.1007/s11060-009-0073-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Roberge D, Ménard C, Bauman G, et al. Radiosurgery scope of practice in Canada: A report of the Canadian Association of Radiation Oncology (CARO) radiosurgery advisory committee. Radiother Oncol. 2010;95:122–128. doi: 10.1016/j.radonc.2010.01.003. [DOI] [PubMed] [Google Scholar]
  • 5.Ryu S. Image-guided radiosurgery or stereotactic body radiation therapy in treating patients with localized spine metastases. http://clinicaltrials.gov/ct/show/NCT00922974. [Google Scholar]
  • 6.Lukka HR. Radiation therapy in treating patients with prostate cancer. http://clinicaltrials.gov/ct2/show/NCT01434290. [Google Scholar]
  • 7.Whelan T. Robotic compared to fixed gantry radiosurgery for brain metastases (TRICK) http://clinicaltrials.gov/ct2/show/NCT01353573. [Google Scholar]
  • 8.Sperduto PW, Berkey B, Gaspar LE, et al. A new prognostic index and comparison to three other indices for patients with brain metastases: An analysis of 1,960 patients in the RTOG database. Int J Radiat Oncol Biol Phys. 2008;70:510–514. doi: 10.1016/j.ijrobp.2007.06.074. [DOI] [PubMed] [Google Scholar]
  • 9.Boudreau R, Clark M, Nkansah E. Ottawa, Ontario, Canada: Canadian Agency for Drugs and Technologies in Health; 2009. TomoTherapy, GammaKnife and CyberKnife therapies for patients with tumours of the lung, central nervous system, or intra-abdomen: A systematic review of clinical effectiveness and cost-effectiveness. [Google Scholar]
  • 10.Cancer Quality Council of Ontario. Radiation machine efficiency. http://www.csqi.on.ca/cms/one.aspx?objectId=128250&contextId=126935. [Google Scholar]
  • 11.Birch S, Kephart G, Tomblin-Murphy G, et al. Human resources planning and the production of health: A needs-based analytical framework. Can Public Policy. 2007;33(suppl):S1–S16. doi: 10.1097/PHH.0b013e3181b1ec0e. [DOI] [PubMed] [Google Scholar]
  • 12.Statistics Canada. 2009 census. http://www.statcan.gc.ca/tables-tableaux/sum-som/l01/cst01/demo02a-eng.htm. [Google Scholar]
  • 13.Birch S, Chambers S. To each according to their need: A community-based approach to allocating health care resources. CMAJ. 1993;149:607–612. [PMC free article] [PubMed] [Google Scholar]
  • 14.Pickren JW, Lopez G, Tsukada Y, et al. Brain metastases: An autopsy study. Cancer Treat Symp. 1983;2:295–313. [Google Scholar]
  • 15.Posner JB, Chernik NL. Intracranial metastases from systemic cancer. Adv Neurol. 1978;19:579–592. [PubMed] [Google Scholar]
  • 16.Mehta M, Noyes W, Craig B, et al. A cost-effectiveness and cost-utility analysis of radiosurgery vs. resection for single-brain metastases. Int J Radiat Oncol Biol Phys. 1997;39:445–454. doi: 10.1016/s0360-3016(97)00071-0. [DOI] [PubMed] [Google Scholar]
  • 17.Caroli M, Di Cristofori A, Lucarella F, et al. Surgical brain metastases: Management and outcome related to prognostic indexes: A critical review of a ten-year series. ISRN Surg. doi: 10.5402/2011/207103. [epub ahead of print on September 20, 2011] [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Chang JE, Robins HI, Mehta MP. Therapeutic advances in the treatment of brain metastases. Clin Adv Hematol Oncol. 2007;5:54–64. [PubMed] [Google Scholar]
  • 19.Schultz C, Edwards W, Gauld S, et al. Stereotactic radiosurgery and radiation therapy experience at the Juravinski Cancer Centre. Presented at the Canadian Radiation Oncology Annual Scientific Meeting; September 14-17, 2011; Winnipeg, Manitoba, Canada. [Google Scholar]
  • 20.Morris S, Devlin N, Parkin D. Economic Analysis in Health Care. ed 2. West Sussex: United Kingdom, Wiley; 2007. [Google Scholar]

Articles from Journal of Oncology Practice are provided here courtesy of American Society of Clinical Oncology

RESOURCES