Abstract
Previous reports suggest that age, household income and travel distance from residence to treatment facility are barriers which influence access to radiotherapy. The purpose of this study was to analyse the utilisation of palliative radiotherapy in a health care system where the main goal is equal access despite place of residence. The Authors collected prospective data on the use of palliative radiotherapy in adult cancer patients over a 12-month period in 2007/2008. All patients were treated in northern Norway and had unlimited, rapid access to treatment. Efforts were made to account for potential overuse. The patients were divided into three groups according to travel distance. The majority of irradiated patients had bone metastases, followed by non-bony thoracic targets and brain metastases. No statistically significant differences in the utilisation rates were detected, the latter ranging from 129 annual treatment courses per 100,000 inhabitants in the region furthest away from the hospital to 142 courses in that closest to it. The median age of the irradiated patients in the three groups was similar. In conclusion, these data suggest that the utilisation of palliative radiotherapy in a health care system without financial barriers to treatment no longer varies with distance between residence and hospital.
Despite the fact that palliative radiotherapy has long been an established and important part of multimodality cancer treatment, several barriers have been identified which influence access to treatment. Among these barriers, age, socioeconomic status and distance from treatment facility play important roles (1-6). In Ontario, Canada, patients with bone metastases whose place of residence was furthest from a radiotherapy center were most likely to receive a single fraction as compared to multiple fractions (7). Exact numbers from the general population on the variation with travel distance in administration of palliative radiotherapy are lacking. To the Authors' best knowledge, the present comprehensive analysis is the first to address this issue.
Patients and Methods
Prospective data were collected in a predominantly rural area of Norway that is served exclusively by one oncology care provider, thereby assuring complete information on the use of radiotherapy in the population. To begin with, the Norwegian health care system and local situation are briefly described. The state is responsible for hospital services through state ownership of regional health authorities. Within these, psychiatric and somatic hospitals are organised as health trusts. One of these is Nordland Hospital, which provides oncology services to the complete population of the county of Nordland. The municipalities have responsibility for primary health care. No individual health insurance is needed to access all these public services as all Norwegians are covered by the national public insurance system. Free hotel accommodation is provided to patients who are not able to travel back and forth due to distance. Travel expenditures are, for practical reasons, reimbursed as well. The Radiation Oncology Unit at Nordland Hospital, Bodø, Norway prospectively recorded all referrals for palliative radiotherapy among patients ≥18 years of age. The study was run in the 12-month period between September 01, 2007 and August 31, 2008. Depending on indication, pain level etc. the waiting time for palliative radiotherapy during this period was 0-14 days. There was unlimited access to treatment, i.e. no priority was given to patients with less advanced tumours, younger patients, those in better general condition etc. This excellent service was possible as the Department was opened in early 2007 and its capacity was very good. Nordland Hospital established its radiation treatment facility in close collaboration with the University Hospital of North Norway in order to improve the treatment of cancer patients in northern Norway (mimimised waiting times, shortened distance of travel to radiotherapy facilities). Both institutions follow the same guidelines and procedures regarding radiation and medical cancer treatment. National guidelines form the backbone of all institutional guidelines. All medical records and information on date of death were available in the hospitals' data systems.
For the purpose of this analysis, patients from three different target regions were analyzed. Region one consisted of the city of Bodø with a margin of 70 km (close to treatment facility). Region two was on average 150 km away from the facilities (intermediate, car or train transportation possible). Region three consisted of the islands of Lofoten and Vesterålen, where travel involves air or sea transportation (Figure 1). Patients from the southern- and northernmost areas of the county of Nordland were excluded as we know from clinical practice that varying proportions of these patients travel to the neighbouring hospitals located in Tromsø and Trondheim, respectively. Due to Norwegian law regulations, we had no access to reliable data from Trondheim. The data on palliative radiotherapy during the 12-month period at the University Hospital in Tromsø were collected retrospectively after the initial analysis of the Bodø data suggested significantly lower utilisation of radiotherapy in region three, i.e. the region that is slightly closer to Tromsø than region 2. Thus, the Authors ascertained complete information on all 3 regions for a second analysis. The radiotherapy utilisation rates in the 3 regions were compared with a likelihood ratio test (chi-square test with 2 degrees of freedom). Wilcoxon and Kruskal-Wallis tests were used to compare the baseline characteristics between different groups. A p-value <0.05 was considered statistically significant.
Results
By July 01, 2008, the official number of inhabitants in each of the 3 target regions was 53,639-57,858 (information released by the municipalities, accessible at www.ssb.no/folkendrkv/2008k2/kvart18.html). The numbers of treatment courses and target volumes (combined data from Bodø and Tromsø) are shown in Table I, which also contains information on treatment policies. As can be seen in the Table, no statistically significant differences in patient numbers and radiotherapy courses were found between the regions. Thirty-eight percent of the treatment courses in patients from region 3 were given in Tromsø, while 62% were given in Bodø. Regarding patients from the two other regions, more than 80% of the radiotherapy courses were administered in Bodø. Although difficult, we tried to estimate possible overuse of palliative radiotherapy. The American Society of Clinical Oncology (ASCO) states that cancer patients should not receive chemotherapy during the last 2 weeks of their life. Adopting this policy for radiotherapy might be a way to calculate the number of unnecessary treatments or overuse, which is also displayed in Table I. Importantly, some patients in good general condition died from unforeseeable events such as pulmonary embolism and ileus. In other situations, such as superior vena cava obstruction, emergency radiotherapy is the only way to prolong the life of a proportion of the afflicted patients, although it is clear that non-responders will have very short survival. It is thus not possible to completely avoid ‘overtreatment’.
Distribution of patient characteristics. Statistically significant differences were found in the number of patients irradiated, number of treatment courses, number of target volumes irradiated and anatomical distribution of target volumes.
As shown in Table I, patients with skeletal targets contributed the majority of palliative indications, followed by those with non-bony thoracic targets and brain metastases. Emerging indications such as stereotactic radiotherapy of lung or liver metastases were not represented in the present study. In the Authors' institutions, hypofractionated regimens such as 8-Gy single fractions for uncomplicated, painful bone metastases, 2 fractions of 8.5 Gy for poor prognosis non-small cell lung cancer, or 10 fractions of 3 Gy for brain metastases were preferred. The patients continued on systemic treatment as appropriate (chemotherapy, hormonal treatment, targeted drugs, bisphosponates), or received best supportive care. Overall, this analysis arrived at a number of 142 palliative radiotherapy courses per 100,000 inhabitants per year in the region close to the hospital (95% confidence interval 126-160) (not included: radiosurgery for brain metastases and stereotactic radiotherapy for extracranial metastases). In the intermediate and distant regions, 116 (95% confidence interval 100-134) and 129 courses (95% confidence interval 113-147) were registered, respectively. The difference in the administration of palliative radiotherapy courses between the 3 regions is not statistically significant.
Nordland county in Norway (green, capital: Bodø), neighbouring counties in yellow (oncology services provided by the hospitals in the capitals Tromsø and Trondheim, respectively). The black arrow corresponds to a distance of approximately 300 km by car. The areas around the cities of Narvik and Mosjoen were excluded from the analysis as the population of these regions might prefer travel to the hospitals in Tromsø or Trondheim.
Discussion
The results of this analysis suggest that lung and prostate cancer patients represent the largest groups referred to palliative radiotherapy, followed by breast and bladder cancer. When using the definition of overuse that is derived from ASCO's criteria in quality assurance of chemotherapy, the present analysis arrives at low figures, which did not impact on the validity of the results. Based on the most recent population-based cancer registry data (2000-2004), on average 534 annual new cancer cases per 100,000 inhabitants were registered in northern Norway (www.cancerregistry.no, Cancer in Norway 2005). We found no significant differences in the utilisation of palliative radiotherapy in regions with different distances to the treatment facility, although the highest utilisation rates were seen in the region closest to the hospital. All patients had free access to treatment, housing and travel (no need for insurance or payments). In addition, hospital staff arrange for room and flight reservation and other possible burdens. No significant age difference between the patient groups from the three regions was found, i.e. the patients who travelled furthest were not younger than the ones treated close to their homes. No attempt was made to look at small differences in travel distance, e.g. less than 20 km versus more than 20 km, as such stratification of the patients would result in very small groups and considerable problems with statistical power.
Of course, the results have to be compared to benchmark data. Ideally, benchmark data would be derived from a nationwide population-based cancer registry with complete recording of treatment details, where all cancer care providers follow the same evidence-based guidelines, all patients have free access to treatment (irrespective of age, socioeconomic parameters, distance to treatment facility etc.), and no financial incentives might bias the choice of treatment. The historical data that match these requirements best were derived from a Swedish survey in the year 2001 (8) and a Norwegian survey in the year 2004 (9). The Swedish survey covered a population of 8.89 million inhabitants. Skeletal metastases were the target in 62% of palliative treatments. Approximately 13% of treatments were given for brain metastases. No information is available on potential overuse, number of patients <18 years of age, and on the exact number of target volumes, but one has to assume that some palliative radiotherapy courses for bone or lymph node metastases were given to more than one target volume, as in the present study. Overall, 87 annual palliative treatments per 100,000 inhabitants were given (95% confidence interval 85-89), including the use of radiosurgery for brain metastases. Although up from previous Swedish surveys, this figure still is indicative of underuse. In addition, the reluctance to offer reirradiation has decreased in recent years (10, 11). The most recent Norwegian data from 2004 allow for calculation of the figures for the 3 northern counties (approximately 460,000 inhabitants), which at that time were served by the University Hospital of North Norway in Tromsø. As mentioned previously, the facilities in Bodø, which now serve a part of this population (in particular the 3 regions evaluated here), were opened in 2007. In 2004, 114 annual palliative treatments per 100,000 inhabitants were administered in northern Norway, not corrected for overuse and including all age groups, but without radiosurgery for brain metastases (95% confidence interval 107-123, significantly higher than the Swedish rate). Data on numbers of target volumes, exact anatomical sites and travel distances are not available.
These considerations lead us to conclude that medical doctors are well aware of the potential of palliative radiotherapy, patients are willing to travel to quite distant facilities to receive therapy and that no relevant variations (underuse) in the utilisation of palliative radiotherapy exist when the health care providers take care of the cost of travelling and other possible burdens. Different results, i.e. considerable variations with distance, were previously identified in northern England for radiation treatment of breast, colon, rectum, lung, ovary and prostate cancer (5), in the United States of America for radiotherapy of breast cancer (2, 3), and in northwestern Italy for radiotherapy in general (4).
Conclusion
Unlike previous studies with different scenarios, the present data do not confirm that the utilisation of palliative radiotherapy in adult cancer patients varies with distance between residence and treatment facility, at least in a health care system without financial barriers to treatment. This poses challenges to health care systems where barriers to access for disadvantaged groups were identified.
- Received February 24, 2009.
- Revision received April 8, 2009.
- Accepted April 13, 2009.
- Copyright© 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
