|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Factors Affecting the Use of Palliative Radiotherapy in OntarioFrom the Radiation Oncology Research Unit, Department of Oncology, Queen’s University, Kingston Regional Cancer Center, Kingston, Ontario, Canada. Address reprint requests to William J. Mackillop, MD, Radiation Oncology Research Unit, Kingston General Hospital, Apps Level 4, Kingston, Ontario, K7L 2V7 Canada; email william.mackillop@ krcc.on.ca.
PURPOSE: To describe the use of palliative radiotherapy (PRT) and to identify factors associated with the use of PRT. MATERIALS AND METHODS: The Ontario Cancer Registry was used to identify 193,253 adult patients who died of cancer between 1986 and 1995. Radiotherapy records from all Ontario cancer centers and the data on socioeconomic status (SES) from the Canadian Census were linked to the Ontario Cancer Registry data. The proportion of cases who received at least one course of PRT at any time within 2 years of death (PRT2Y) was used as a primary measure of the use rate of PRT. RESULTS: Overall, 26.4% of cases underwent at least one course of PRT. PRT2Y remained relatively constant over the study period. PRT2Y was disease-specific and ranged from 4% for pancreatic cancer to 41% for prostate cancer. Age was negatively associated with PRT2Y (adjusted odds ratio [OR], 4.5 for the youngest group), and SES was positively associated with PRT2Y (adjusted OR, 1.2 for patients from wealthy communities). Patients who were initially diagnosed in a hospital affiliated with a cancer center (adjusted OR, 1.4) or who lived in a county in which a cancer center is located (adjusted OR, 1.2), or who resided in certain regions (adjusted OR, 1.20 for Hamilton and 1.17 for Kingston), were more likely to be treated with PRT. CONCLUSION: The use of PRT varied across the dispersed population in Ontario and was influenced by factors unrelated to the patient’s needs. An effort should be made to reduce barriers to access for disadvantaged groups.
WITH MORE PEOPLE dying from chronic or progressive diseases, palliative care, which addresses the physical, psychologic, social, and spiritual needs of patients with advanced disease and their families, has received great interest over the last decade.1 In developed countries, an increasing share of health care resources has been devoted to palliative care programs that emphasize maximizing the quality of life remaining to patients.2,3 Palliative radiotherapy (PRT) plays an important role in any comprehensive palliative care program. PRT is known to be effective in alleviating a variety of cancer-related symptoms, including pain resulting from bone metastases,4-6 neurologic deficits resulting from brain metastases and spinal cord compression,7 and the symptoms of locally advanced lung cancer, such as cough, dyspnea, and hemoptysis.8 It has been estimated that approximately one half of all radiotherapy is prescribed with palliative intent. However, there is little information available about the use of PRT in routine practice at a population level.9 In Ontario, all radiotherapy is provided by a network of provincial cancer centers, the aim of which is to provide equitable access to high-quality radiotherapy for all residents of the province.10 Ontarians make no direct payments for medical services or for radiation treatment, and thus, there are no direct financial barriers to access to radiotherapy. Despite the many strengths of this system,10 there is evidence that access to radiotherapy in Ontario has not always been optimal. The rate of radiotherapy use in the initial treatment of cancer has been shown to be generally low by international standards, and very low indeed in areas remote from the regional cancer centers.11 Furthermore, increasing demand for radiotherapy, coupled with resource limitations, led to the development of long waiting lists for radiotherapy at many Ontario cancer centers in the late 1980s, a problem that has since persisted despite large capital investments in the system in the 1990s.12 Waiting times for radiotherapy in many parts of the province have often been longer than is considered acceptable by the majority of radiation oncologists, and it has been suggested that waiting lists may have become an implicit form of rationing.13 There is some evidence that the use of PRT at Ontario cancer centers may have diminished because of this.14 Given the importance of PRT in the symptomatic care of patients with advanced cancer and these general concerns about the adequacy of access to radiotherapy in Ontario, this study was undertaken to describe the use of PRT in Ontario and to identify aspects of the health care system that influence access to PRT.
Sources of Data The Ontario Cancer Registry is a population-based registry that covers the province’s entire population of approximately 11 million people. The Ontario Cancer Registry routinely collects information about the demographic and clinical characteristics for all incident cases. Information about vital status of all cases is routinely updated by linkage to the death records of the Ontario Registrar General. The completeness of cancer registration, as measured by capture-recapture methodology, is greater than 95% for all sites combined.15 Details on the registration procedures and further background information on the registry were given elsewhere.16 Radiotherapy records in Ontario were provided by eight regional cancer centers and the Princess Margaret Hospital in Toronto. Since the early 1980s, all treatment centers in Ontario have kept a computerized summary of every course of radiotherapy in standard format. The record includes the following variables: intent of treatment, number of fractions, dates of first and last radiation treatments, total dose, and body region irradiated. A review of 1,122 paper charts over the period from 1982 to 1991 indicates a high level of completeness (95%) and accuracy (99%) with respect to the variables essential to this study.12 The Canadian Census provided an electronic file that contained descriptors of community socioeconomic status (SES), such as median household income for 1991, the midpoint of the study.17 These descriptors were available at the level of the census enumeration area and at the level of the census subdivision. To measure the effect of community SES on access to care, these SES data were linked to each case in the Cancer Registry, based on the person’s residence at the time of diagnosis, as described previously.18
Study Population
Definition of Study Variables PRT2Y. The proportion of cases that had at least one course of PRT in the last 2 years of life (PRT2Y) was used to describe access to PRT. In establishing the 2-year cutoff point, we considered how far back in time before death we had to track the patients to identify most of those who had at least one course of PRT. We wanted to be able to identify the majority of patients who ever had PRT without following back any further than necessary, because this would limit the usefulness of this indicator in practice. Figure 1 shows the results of an analysis in which we systematically altered the cutoff point used, based on the patients who died of cancer between 1990 and 1991, the midpoint of the study. The observed PRT rate changed little with further follow-back beyond 2 years. This does not mean that few patients received PRT before this, only that if they did, they also received additional PRT later on.
SES. Median household income was used to describe the SES of the community in which the patient resided at the time of diagnosis. For the purpose of this study, three SES groups were defined as follows: low (household income < Can $20,000), medium (Can $20,000 to $50,000), and high (> Can $50,000). Actual distributions of income within these groups have been published elsewhere.18 Catchment regions of the cancer centers. The catchment areas of each cancer center were defined by assigning each census subdivision to the cancer center most frequently used by its residents, as described previously.11
Statistical Analysis To allow comparisons against variables examined, rates were age-standardized by the direct method using the study population of all cases at the time of death. Rates of use of PRT were mapped at the county level using the ArcInfo Geographic Information System (Environmental Systems Research Institute, Inc, Redlands, CA). Multivariate analysis using the logistic regression model was also performed to evaluate the role of each factor after controlling for the effects of the remaining variables. The regression analysis was performed using the backward stepwise procedure.
Univariate Analysis A total of 193,253 cancer patients who died of cancer were included in the study. Of these, 51,019 (26.4%) received at least one course of PRT to any anatomic region for any clinical indication within 2 years of death. The majority of them (61.7%) received only one course of PRT in their lifetime, 22.9% received two courses, and 15.4% received more than two courses. Skeletal metastasis was the most common indication for PRT, accounting for 44% of all courses. The proportions of courses to other body regions were as follows: chest, 20%; brain, 14%; abdomen and pelvis, 6%; superficial lymph nodes, 5%; and skin and superficial soft tissue, 2%. Figure 2 shows the frequency distribution of fractions per course by selected body regions. As demonstrated, treatment with palliative intent was delivered, on average, using six fractions per course to all body regions combined, 6.6 fractions to the brain, head, or skull, 6.5 fractions to the chest in patients with lung cancer, and 3.8 fractions to bone.
The PRT2Y rates in both males and females decreased with age. The proportion of male patients who ever received PRT was 34.6% in those younger than 40 years and only 13.7% in those older than 80 years. The corresponding figures for females were 42.5% and 9.8%, respectively. Because of the large differences among age groups, the following results have been age-standardized. Table 1 gives the age-standardized rates of PRT2Y and their 95% confidence intervals (CIs) for subgroups of cases defined by the main study variables. PRT2Y decreased from 27.0% (95% CI, 26.4% to 27.6%) in 1986 to 25.0% (95% CI, 24.4% to 25.6%) in 1990 and increased slightly thereafter to 26.9% (95% CI, 26.3% to 27.5%) in 1995.
Not surprisingly, the PRT2Y rates varied from one disease group to another (Table 1). The rates were high (> 30%) for breast, lung, and genitourinary malignancies, intermediate (20% to 30%) for head and neck malignancies, and low for three other disease groups (< 20%). When age-standardized rates of PRT for each separate disease (defined by the International Classification of Diseases, Ninth Revision) were explored, a broad spectrum of rates were observed within each disease group. The values of PRT2Y for the genitourinary cancers were as follows: prostate, 40.7%; kidney, 26.4%; and bladder, 20.2%. The values of PRT2Y for the gastrointestinal malignancies were as follows: rectum, 24.5%; colon, 9.7%; stomach, 8.5%; and pancreas, 4.0%. The values of PRT2Y for the hemopoietic diseases were as follows: myeloma, 28.3%; Hodgkin’s Disease, 10.7%; Non-Hodgkin’s lymphoma, 20.5%; and leukemia, 5.1%. PRT2Y rates differed by median household income, but the difference was not statistically significant on univariate analysis. PRT2Y varied across regions of Ontario, from 23.2% in northern Ontario to 29.6% in Kingston, 27.9% in Ottawa, and 28.3% in Hamilton. This significant interregional variation is illustrated in Fig 3, which displays PRT2Y for each county and region. Generally, eastern and southwestern Ontario demonstrated higher use rates in comparison with the rest of province, and the rates were usually higher in counties where a cancer center was located.
Table 1 also indicates that patients diagnosed in a hospital affiliated with a cancer center or residing in a county in which a cancer center is located had a higher rate of use of RT compared with others. Further exploration of intercounty variation is presented in Fig 4, in which PRT2Y is plotted against the number of patients who died of cancer in each county. The horizontal line is the average rate for the whole province, and the shaded area indicates the 95% CI for the expected rate of each county, assuming the absence of nonrandom variation. In general, rates in those counties where cancer centers were located (marked by solid dots) were significantly higher than the provincial average, except for Toronto, which had a lower rate than the provincial average.
Multivariate Analysis All study variables were included in a multiple logistic regression model, which is presented in Table 2. The variations in PRT2Y for all the variables examined are shown as adjusted odds ratios (OR) and their 95% CIs after allowing for the effects of other variables. All factors demonstrated significant independent effects in association with variation of PRT2Y. Patients who died in 1990 tended to be less likely to have been treated with PRT in comparison with those who died at other times. Age was negatively associated with PRT2Y. Patients younger than 40 years were approximately 4.5 times more likely to receive PRT than those patients older than 80 years.
Once other variables were controlled for, SES was positively associated with PRT2Y in the multivariate analysis. Patients from wealthy communities were 1.17 times more likely to receive PRT than those in the low-income group. Patients who were initially diagnosed in a hospital affiliated with a cancer center (OR, 1.4), who lived in a county in which a cancer center was located (OR, 1.2), or who resided in certain catchment regions, such as Hamilton (OR, 1.20) and Kingston (OR, 1.17), were more likely to be treated with PRT.
In Ontario, the overall rate of PRT in the 2 years preceding death from cancer was 26.4%, but it varied widely across the province. Although several demographic factors and the primary sites of disease emerged as factors associated with the variation, after adjusting for these variables, the effects of catchment region, hospital of diagnosis, and county of residence persisted and SES was positively associated with the use of PRT. Before examining the implications of our results, the strengths and limitations of this study should be considered. The present study has the advantage of being population-based. We had information on all cases of incident cancers, including information about radiotherapy. Given the characteristics of our data sources, near-completeness of case ascertainment and a high level of accuracy as documented in the methodology of this study, the selection biases that can occur in a hospital-based series were avoided. PRT is mainly used for the management of advanced cancer, and a large proportion of incident cases that present at an early stage may not require PRT during their lifetime. Hence, we confined our study to those patients who died of cancer instead of overall incident cases, which are more commonly used when studying the initial management of cancer cases. In this way, interpretation with regard to the temporal trends and comparisons in the use of PRT was more straightforward. This approach is not sensitive to differences in case mix at diagnosis because the cases studied were all within 2 years of death from cancer. PRT2Y changed over time. The 2% lower PRT rate in 1990 (compared with that of 1986 or 1995) corresponded to approximately 380 patients who had not been treated by PRT in that year, assuming a constant practice of PRT over the study period. This was, probably, the direct effect of the long waiting lists for radiotherapy that prevailed at many cancer centers at that time.12 During the study period, there were no advances in the palliative care that reduced the need for PRT. In fact, the important role of PRT in palliating symptoms such as cough, hemoptysis, dyspnea caused by advanced lung cancer, and pain resulting from skeletal metastasis was well-documented by randomized trials.19,20 Apparently, the evidence from these studies did not trigger an increase in the use of PRT in Ontario. The use of PRT varied markedly across age groups in Ontario. We believe that the strong negative association observed between age and the use of PRT cannot be fully explained by the decreasing functional status with aging. There is evidence that the decline in the use of PRT for advanced breast cancer, prostate cancer, and myeloma is more rapid than the decline in functional status with age.21 Moreover, a number of clinical studies have demonstrated that radiotherapy is tolerable for older patients and that advanced age alone does not necessarily make a patient ineligible for PRT.22,23 It is also unreasonable to assume that the difference in the use of PRT by age would be mainly attributable to the variation of treatment preferences among differing age groups, although limited studies in this field are available for reference. Given the absence of other plausible explanations for the age effect, we postulate that older patients may be treated differently from young patients for reasons unrelated to the patients’ needs and wishes, and this is borne out by the evidence that the referral rate for PRT by primary care givers in the oldest patients was only one half of that in the young patients in this population.21 The variation in the use of PRT by socioeconomic groups is consistent with our previous study on adjuvant radiotherapy for breast cancer.24 SES is known to be positively related to other demographic factors, such as level of education, which is commonly characterized as a determinant of health care access in literature. It is probable that patients from wealthy communities with a relatively high level of education tend to be more informed and demanding. We postulate that patients with a high household income and a high level of education are better equipped to negotiate with their physicians in obtaining a referral to a radiation oncologist and, thus, are better at gaining access to treatment. Our hypothesis is consistent with the observation that poorer, less educated groups receive more acute hospital care and have more contacts with general practitioners, whereas wealthier, more educated groups have a higher frequency of specialized consultations in Canada and Switzerland, which both have health insurance schemes designed to provide equitable access.25,26 The relationship between the use of PRT and the factors of county of residence, catchment region, and hospital of diagnosis reinforced and extended our findings from an earlier study based on the initial management of radiotherapy.11 Patients who lived in a county without a cancer center were 24% less likely to be treated with PRT, which is consistent with the findings from the United States and Australia.27,28 Generally, living in a county without a cancer center implies a long distance and inconvenient access to PRT. We are not surprised that such patients were less likely to be treated with PRT, because the trade-off between the discomfort of travel and the benefit of treatment could be less favorable, especially for weak and sick patients with advanced cancer. Furthermore, patients who resided in a remote area may have had fewer channels to access medical information and knowledge. It is expected that patients who are unaware of the options and merits of PRT are less likely to be treated with PRT. Although it is reasonable to assume that the nature of a centralized radiotherapy system may reduce the use of PRT among patients who reside far away from treatment facilities, patients living near a cancer center do not necessarily have a higher rate of use of PRT unless there is a good system of outreach and adequate availability of radiotherapy facilities in their catchment regions. This was suggested by the lower rate of PRT in Toronto, an urban community in which two cancer centers are located. In exploring the underlying reasons for the interregional variation in the use of PRT in Ontario, we do not rule out the potential effects of factors, such as doctors’ management preferences on the access to PRT, which may have contributed to a large component of the regional variation in the management of advanced non–small-cell lung cancer.29 However, in the context of PRT, the interregional variation may occur if implicit rationing is prompted by long waiting lists for radiotherapy, as we have observed in some regions in Ontario since the 1980s.12 We recognized that immediate intervention is essential for PRT, and it is difficult, both psychologically and physically, for symptomatic patients with advanced cancer to tolerate a long delay in treatment. Thus, when there are resource shortages for radiotherapy, referring physicians may resort to alternative, but less optimal, approaches, such as narcotics used for palliating painful skeletal metastasis. On the other hand, because an outreach radiotherapy program has not been applied in some regions, it is reasonable to assume that variations in referrals for PRT by primary care givers may have led, at least partly, to the unequal use of PRT across the province. Our study also shows that patients who were initially diagnosed in a hospital with a cancer center were approximately 35% more likely to be treated with PRT. Generally, in a hospital with a cancer center, a multidisciplinary team continuously delivers the multimodality health care services for cancer patients and an integration of the opinions of surgical, medical, and radiation oncologists in formulating the best cancer treatment plan is the norm. The cooperative approach in such hospitals may increase each doctor’s awareness of his or her colleagues’ potential contribution to disease management, which ensures a better functional relationship between radiation oncologists and other medical professionals. Hence, it is expected that patients diagnosed in such hospitals may have more optimal access to PRT.
Overall, only one quarter of patients who died of cancer in Ontario ever received PRT. To our knowledge, there are no national or international standards available for the use of PRT. The lack of reliable data on the prevalence of symptoms for which PRT is indicated precludes the possibility of projecting an appropriate target PRT rate in the study population at this time. However, based on the study results, we were able to identify a subgroup of patients who had more optimal access to PRT. They were younger (age From a public health policy perspective, how can we achieve an acceptable level of radiotherapy services to a dispersed population? It seems unrealistic to expect every medium-sized rural community to be provided with a cancer center at this time. However, our results do suggest that concentrating on strengthening the outreach system, educating physicians about the potential benefits of PRT, and developing strategies to mitigate resource shortages may be useful in improving access to PRT in Ontario.
1. Davis RM, Wagner EH, Groves T: Managing chronic disease. BMJ 318: 1090-1091, 1999 (editorial)
2.
Lubitz JD, Riley GF: Trends in medicare payments in the last year of life. N Engl J Med 328: 1092-1096, 1993 3. Seale C, Cartwright A: The year before death. Aldershot, United Kingdom, Avebury, 1994 4. Ciezki J, Macklis RM: The palliative role of radiotherapy in the management of the cancer patient. Semin Oncol 22: 82-90, 1995 (suppl 2)[Medline] 5. Bone Pain Trial Working Party. 8 Gy single fraction radiotherapy for the treatment of metastatic skeletal pain: Randomized comparison with a multifraction schedule over 12 months of patient follow-up. Radiother Oncol 52:111-121, 1999 6. Chahbazian C: Radiation therapy for the palliation of pain in advanced cancer. Am J Clin Oncol 16: 444-445, 1993[Medline] 7. Paszat L, Shenouda G, Blood P, et al: The role of palliative radiotherapy for brain metastases. Can J Oncol 6: 48-53, 1996 (suppl) 8. Brundage MD, Bezjak A, Dixon P, et al: The role of palliative thoracic radiotherapy in non-small cell lung cancer. Can J Oncol 6: 25-32, 1996 (suppl) 9. Frodin JE, Jonsson E, Moller T, et al: Radiotherapy in Sweden: A study of present use in relation to the literature and an estimate of future trends. Acta Oncol 35: 967-979, 1986 10. Hayter C: Historical origins of current problems in cancer control. CMAJ 158: 1735-1740, 1998[Abstract]
11.
Mackillop WJ, Groome PA, Zhang-Solomons J, et al: Does a centralized radiotherapy system provide adequate access to care? J Clin Oncol 15: 1261-1271, 1997 12. Mackillop WJ, Fu H, Quirt CF, et al: Waiting for radiotherapy in Ontario. Int J Radiat Oncol Biol Phys 30: 221-228, 1994[Medline] 13. Mackillop WJ, Zhou Y, Quirt CF: A comparison of delays in the treatment of cancer with radiation in Canada and the United States. Int J Radiat Oncol Biol Phys 32: 531-539, 1995[Medline] 14. Mackillop WJ, Zhou S, Groome P, et al: Changes in the use of radiotherapy in Ontario 1984-1995. Int J Radiat Oncol Biol Phys 44: 355-362, 1999[Medline] 15. Robles SC, Marrett LD, Clarke EA, et al: An application of capture-recapture methods to the estimation of completeness of cancer registration. J Clin Epidemiol 41: 495-501, 1988[Medline] 16. Clarke EA, Marrett LD, Krieger N: Cancer registration in Ontario: A computer approach, in Jenson O, Parkin DM, MacLennan R et al (eds): Cancer Registration Principles and Methods. Lyon, France, International Agency for Research on Cancer, 1991, pp 246-257 17. Statistics Canada: Census Handbook. Ottawa, Ontario, Canada, Industry Canada, publication 92-352-XPE, 1997, 1996 18. Mackillop WJ, Zhang-Salomons J, Groome PA, et al: Socioeconomic status and cancer survival in Ontario. J Clin Oncol 15:1680-1689, 1997 19. A Medical Research Council. Randomized trial of palliative radiotherapy with 2 factions or a single fraction in patients with inoperable non-small cell lung cancer and poor performance status. Br J Cancer 65:934-941, 1992 20. Hoskin PJ, Price P, Easton D, et al: A prospective randomised trial of 4 Gy or 8 Gy single doses in the treatment of metastatic bone pain. Radiother Oncol 23: 74-78, 1992[Medline] 21. Tyldesley S, Zhang-Salomons J, Groome PA, et al: The association between age and the utilization of radiotherapy in Ontario. Int J Radiat Oncol Biol Phys 47: 469-480, 2000[Medline] 22. Souchon R: Palliative radiation oncologic therapy: Is patient’s age a determining factor of feasibility? A 1-year analysis (1997) at a radiotherapy clinic of an academic teaching hospital. Strahlenther Onkol 175: 218-224, 1999[Medline]
23.
Mitsuhashi N, Hayakawa K, Yamakawa M, et al: Cancer in patients aged 90 years or older: Radiation therapy. Radiology 211: 829-833, 1999 24. Paszat LF, Mackillop WJ, Groome PA, et al: Radiotherapy for breast cancer in Ontario: Rate variation associated with region, age and income. Clin Invest Med 21: 125-134, 1998[Medline] 25. Roos NP, Mustard CA: Variation in health and health care use by socioeconomic status in Winnipeg, Canada: Does the system work well? Yes and no. Milbank Q 75: 89-111, 1997[Medline] 26. Gutzwiller F, La Vecchia C, Levi F, et al: Education, disease prevalence and health service utilization in the Swiss National Health Survey "SOMIPOPS". Prev Med 18: 452-459, 1989[Medline] 27. Guidry JJ, Aday LA, Zhang D, et al: Transportation as a barrier to cancer treatment. Cancer Pract 5: 361-366, 1997[Medline] 28. Denham JW: How do we bring an acceptable level of radiotherapy services to a dispersed population? Australas Radiol 39: 171-173, 1995[Medline] 29. Mackillop WJ, Dixon P, Zhou Y, et al: Variations in the management and outcome of non-small cell lung cancer in Ontario. Radiother Oncol 32: 106-115, 1994[Medline] Submitted January 13, 2000; accepted July 27, 2000.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
, counties without cancer centers.
69), residents of medium and high income communities, initially diagnosed in a hospital with a cancer center, and residing close to a cancer center in one of the following regions, where overall access to PRT was better (Kingston, northwestern Ontario, and London). The PRT rate for this subgroup, when disease-standardized using the study population of all patients who died of cancer, was 57.8% (95% CI, 55.3% to 60.3%). There is no a priori reason to expect any overtreatment in Ontario; the system of payment of radiation oncologists avoids supplier-driven demands. Hence, this subgroup may serve as a surrogate for the ideal population with an approximation to optimal access to PRT in a current context. The overall rate of 26.4% in the use of PRT in Ontario, which was less than one half of the ideal rate, strongly suggests that PRT was underused in Ontario. 
