|
|
| |||||||||||||||||||||||||||||||||||||||
Background Allogeneic bone marrow transplantation, a sophisticated and expensive procedure, is the only curative therapy for chronic myeloid leukemia (CML). We examined the availability and appropriateness of allogeneic bone marrow transplantation for CML in 10 economically advanced countries with diverse health care systems. For each country we obtained data on the likelihood of transplantation to treat CML in patients under the age of 55 years, the length of time from diagnosis to transplantation, and the stage of disease at the time of transplantation.
Methods Data were collected on 9873 allogeneic bone marrow transplantations performed at 208 centers in 10 countries from 1989 through 1991. Data were acquired from transplantation registries and by means of a mailed survey of all centers and teams that did not contribute data to registries. Data on the incidence of disease were drawn from national and regional cancer registries.
Results Among the 10 countries there was a twofold difference between the lowest and highest rates of transplantation to treat CML (0.26 to 0.54 per 100,000 population per year); Swedish patients were the most likely to receive a transplant, and German patients the least likely. The median length of time from diagnosis to transplantation ranged from 6.8 to 15.4 months. In all countries, most transplantations were performed in the chronic phase of the disease, but as many as a third of patients received transplants in the less favorable accelerated or blast phase. The values for the United States fell near the middle of those for the 10 countries on all measures.
Conclusions Our findings challenge the assumption that the United States is unique in providing broad access to high-technology treatments. On no measure of the availability or appropriateness of transplantation for CML did it surpass the other nine countries studied.
Bone marrow transplantation for CML, a prototype of high-technology care, is a complex and costly treatment. It is also the only curative therapy for CML, and treatment in the early stage of the disease is advantageous1,2,3,4,5. After the transplantation of bone marrow from an HLA-matched sibling, the likelihood of survival without leukemia at five years is about 50 percent when the transplantation is performed in the chronic phase of the disease, 30 percent when it is performed in the accelerated phase, and 15 percent when it is performed in the blast phase1,2,3,4,5.
The outcome is best for patients who receive a marrow transplant in the chronic phase of disease within one year of diagnosis2,5. As compared with recipients of marrow from HLA-matched siblings, patients who receive marrow from unrelated donors or HLA-mismatched related donors fare less well6,7,8. Unlike a bone marrow allograft, conventional chemotherapy cannot cure CML; patients treated with hydroxyurea, busulfan, or interferon survive in the chronic phase of disease for a median of three to six years9,10,11,12.
Because CML usually progresses slowly, patients who are eligible for a marrow transplant can be kept waiting if the resources for transplantation are relatively scarce, though the risk that the disease will reach a less curable stage increases. Therefore, by assessing rates of transplantation, the stage of CML at transplantation, and the waiting times in different countries, we were able to evaluate the influence of different medical care systems on the availability and clinical appropriateness of marrow transplantation for CML.
Methods
Incidence
The U.S. Bureau of the Census and the British Information Services provided national population data. Data on the incidence of leukemia were obtained from cancer registries in all the countries we studied except Germany. For the United States, we used data compiled by the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute, which cover approximately 10 percent of the population. For other countries we used data from the International Agency for Research on Cancer. Cancer registries in Sweden and the Netherlands provided data on incidence in the entire populations of those countries. In France, population-based registries in six departments in widely separated sections of the country covered about 7 percent of the population. In Australia, six registries covered 84 percent of the population. For both France and Australia, we estimated the national incidence of CML by extrapolating data from the registries, weighted by the size of the populations they covered.
For Germany, we limited our analyses to the Lander (states) that constituted the German Federal Republic before 1990. To estimate the incidence of CML in Germany (where the sole registry represented only 1.7 percent of the population), we used the average incidence of CML in the nine other countries. We have confidence in this estimate because there is substantially less variation among countries in the incidence of leukemia than is the case for other cancers,13 because the number of cases of CML we assumed approximates estimates by German oncologists, and because the ranking of Germany in our analysis was unaffected whether we used transplantation rates based on populations (for which we had precise data) or rates based on the incidence of disease (assuming an average incidence of leukemia in Germany).
The numbers of cases of CML in candidates for transplantation (those 54 years of age or younger) were estimated from the age-standardized rates for males and females reported by each registry. We assumed that if one country had an overall age-standardized rate that was 50 percent higher than another's, the incidence of CML in that country would be 50 percent higher in each five-year age category. The SEER Program provided age-specific rates of incidence of CML for each five-year age group from 0 to 54 years. Using age-standardized rates for males and females in the United States, we derived a proportional age- and sex-specific rate for each country. This rate was used to estimate the number of patients with newly diagnosed CML who could have been candidates for transplantation in each country. The rate for each age group was multiplied by the actual male or female population of that age in the country; the sum of the results for all the age and sex categories through the age of 54 years produced the estimate we used. The validity of these estimates is indicated by their close correspondence to the actual numbers of cases of CML in the Netherlands and Sweden, for which exact counts of cases were available.
Bone Marrow Transplantation
Data on patients were obtained from three sources: the International Bone Marrow Transplant Registry (IBMTR; see the Appendix), the Societe Francaise de Greffe de Moelle, and a mailed survey of bone marrow transplantation centers that did not routinely contribute data to the IBMTR.
Our initial source of information was the IBMTR, based at the Medical College of Wisconsin, which contains data on more than 20,000 allogeneic bone marrow transplantations (for any indication) performed at approximately 200 transplantation centers worldwide and which represents more than 50 percent of the bone marrow-transplantation teams worldwide14. The IBMTR provided data on 2559 transplantations performed at 71 centers during the period we studied.
The Societe Francaise de Greffe de Moelle collects data on bone marrow transplantations performed at French institutions; 35 centers contributed data on allogeneic transplantations from 1989 through 1991. Since only 14 of these centers participated in the IBMTR during this time, we obtained all our data on transplantations in France (1708 cases) from the society.
The IBMTR's statistical center also surveyed all allogeneic bone marrow-transplantation teams in countries other than France for which the IBMTR data base could not provide comprehensive information. With information from the IBMTR's directory of bone marrow-transplantation teams and interviews with authorities on bone marrow transplantation and health care in each country, we identified teams at more than 100 institutions for which data were needed, including teams that participated in the IBMTR but whose data were incomplete. Eighty-five teams (84 percent of those surveyed) responded with information on a total of 4981 cases. If a team's records did not permit the ready retrieval of information on each patient, the survey requested summary information.
For 17 of the 18 centers that did not respond in time to the mailed questionnaire or that provided incomplete information, we obtained data from a worldwide survey conducted by the IBMTR in 1991 and an annual study of European transplantation centers15,16,17; data on 1 center were obtained from the Dutch Health Council. The data thus obtained did not include information on the stage of disease or the length of time from diagnosis to transplantation, and the 1991 IBMTR survey covered only the period from 1988 through 1990. We estimated the number of transplantations performed in 1991 at each center on the basis of data from the annual European Bone Marrow Transplant Group study17 or (for non-European centers) by assuming that the number of transplantations performed at a center in 1991 equaled its average for the preceding two years.
In these ways we obtained data from 208 transplantation centers on a total of 9873 allogeneic bone marrow transplantations performed for a variety of indications. Data on the stage of disease and the length of time from diagnosis to transplantation were complete for 8169 cases (83 percent). Among the transplantations, 2995 (30 percent) were performed in patients with CML.
Calculation of Rates of Transplantation
To determine the likelihood that patients with CML would receive transplants, we calculated the proportion of patients with new cases of CML who would have undergone transplantation by the end of the second full calendar year after the calendar year in which the diagnosis was made. For example, for patients whose disease was diagnosed in 1989, we calculated the proportion that would have undergone transplantation by the end of 1991. (According to the IBMTR data, about 90 percent of transplantations for CML are performed within three years of diagnosis.) To obtain the rate of transplantation, we first stratified cases according to the year in which they were diagnosed. Next, we counted the number of patients with transplants in whom CML was diagnosed in a given year who were less than 55 years old and who underwent transplantation in the same year, in the next year, or in the second year after diagnosis. Using data from the International Agency for Research on Cancer, we estimated the incidence of CML in each country for persons under the age of 55. For each year in which we had data on both diagnosis and treatment (1987 through 1991), we divided the number of transplant recipients whose diagnosis was made in a given year (regardless of the year when the transplantation was performed) by our estimate of the total number of newly diagnosed cases of CML in that country each year. Finally, we averaged the ratios calculated for each year to estimate the overall proportion of patients with CML who had undergone transplantation by the end of the second calendar year after diagnosis.
We performed no tests of significance because our data covered entire populations, not samples, and represented the actual number of transplantations in each country.
Results
Table 1 shows the total number of transplantations during the study period in each country studied, the population, rate of transplantation (transplantations per 100,000 population), and distribution of transplantations according to diagnosis. Although there was a difference of more than twofold between the highest and lowest rates of transplantation for any disease (in France and Germany, respectively), most values are spread evenly along a continuum, with the United States near the middle. In the 10 countries combined, 77 percent of allogeneic marrow transplantations were performed in patients with CML, acute lymphoblastic leukemia, or acute myelogenous leukemia, with the remainder in those with lymphoma and other types of cancer or aplastic anemia and other nonmalignant diseases. In the United States, relatively few allogeneic bone marrow transplantations were performed to treat disorders other than leukemia, whereas in France and Sweden a relatively large proportion of transplantation resources were used for other diseases.
|
We assessed the availability of bone marrow transplantation by measuring two factors: the proportion of all patients with CML under 55 years of age who received transplants, and the length of time from diagnosis to transplantation among patients who received an HLA-matched transplant from a sibling.
Table 2 shows the incidence of CML, estimated number of new cases annually, and rate of transplantation for the various countries. The likelihood that a patient with CML would receive a transplant by the end of the second year after diagnosis ranged from approximately 1 in 2 (in Sweden) to about 1 in 4 (in Germany). About one third of patients with CML in the United States received transplants during the study period. The length of time to transplantation (Table 3) was calculated only for transplants of marrow from related donors, in order to eliminate variations due to the search for suitable unrelated donors. The median length of time from diagnosis to transplantation ranged from 6.8 months in New Zealand to 15.4 months in Germany. The United States had the fourth-shortest median time (9.0 months). In Australia, Denmark, and Germany, the median length of time to transplantation was more than one year. By contrast, three fourths of all patients in New Zealand, the Netherlands, and Sweden received transplants within one year. In the United States there was more variation than in other countries (with the exception of Australia) in the amount of time patients waited for a transplant. Whereas many patients in the United States received transplants rapidly, more than six years passed from the time of diagnosis before 95 percent of patients had received their transplants.
|
|
Stage of Disease at Transplantation
When implanted during the chronic phase of CML, an HLA-matched marrow allograft offers the best chance for cure1,2,4. The proportion of patients with CML who receive a marrow transplant while they are in the chronic phase of the disease is therefore an important indicator of the appropriateness of the use of bone marrow transplantation. Our analysis of this aspect of the use of transplantation was limited to patients with HLA-matched related donors.
Table 4 shows the proportion of transplantations of marrow from related donors that were performed in the chronic phase of CML. Excluding New Zealand, where the total number of transplantations was small, there was a 17 percent difference in values among the remaining nine countries. Roughly four fifths of the Dutch and Canadian patients underwent transplantation in the chronic phase, as compared with about two thirds of the patients in France and Australia and just half the patients in New Zealand. This variation was substantially smaller than the variation in the median length of time to transplantation and in the proportion of patients receiving transplants within one year. Moreover, in the countries where the smallest percentage of transplantations were performed within a year of diagnosis (Germany and Denmark), the proportions of patients who received transplants while their disease was in the chronic phase were comparable with those in the United States and several other countries where patients underwent transplantation more quickly -- for example, in Sweden (Table 3).
|
In this study we assessed disease-specific measures of the availability and appropriateness of an advanced form of medical technology, bone marrow transplantation for CML15,16,17,18,19. We focused on 10 economically advanced democracies whose health care systems differ in organization, incentive structures, and per capita expenditures. We examined the use and timing of transplantation, not how transplantation was performed, because there was no universally accepted technique. Nor did we examine outcomes, which can be influenced by numerous factors irrelevant to our inquiry. The validity of our conclusions depends in part on the accuracy of the data bases we used and on the number of patients for whom data were reported in each country.
The United States fell roughly in the middle of the 10 countries on two measures of the availability of bone marrow transplantation. As measured by the rate of transplantation, patients with CML in seven of the countries were more likely to receive transplants than were their counterparts in the United States. Many patients in the United States received their transplants quickly, but for some there were lengthy delays between diagnosis and transplantation. The relatively large percentage of U.S. patients who received transplants within six months of diagnosis implies a ready availability of and adequate capacity for the procedure. That 5 percent of U.S. patients received transplants more than six years after diagnosis could reflect many factors, including regional shortages of resources, limitations on the ability of some patients to pay for the procedure, differences in patients' willingness to undergo transplantation, variations in physicians' referral patterns, and differences in treatment philosophy. (In Germany, for example, both the reliance on restricted capacity as the primary method of cost containment and the anomaly of having university hospitals financed largely by ministries responsible for higher education rather than by ministries of health were important factors in the limited availability of transplantation services).
The United States also fell roughly in the middle of the range in terms of our measure of appropriateness -- the percentage of patients with CML who underwent transplantation during the chronic phase of the disease. In several countries the use of marrow transplantation appeared to be more appropriate, as indicated by higher percentages of transplantations performed in the chronic phase of CML. Physicians in these countries may refer eligible patients quickly and have appropriate services available, or they may simply not perform transplantations in patients with advanced disease. In the United States, a disproportionate number of patients with CML underwent transplantation in the accelerated phase of the disease or during a blast crisis. This finding has negative implications for the quality of care because the overall rate of transplantation for CML was not particularly high. The high percentage of transplantations performed in patients with poor prognoses means that proportionally fewer patients with good prognoses received transplants in the United States than in other countries.
We have presented data showing how a single treatment is used in the management of a relatively uncommon disease. We cannot assume that our findings apply to other expensive and complex procedures, or to a wider range of medical interventions. What are the larger implications of our findings? What contribution do they make to the debate on health care reform?
We believe that these findings are not unique to bone marrow transplantation for CML. It is likely that the features of the health care system in the United States that limit the use of transplantation for patients with CML also have deleterious effects on the treatment of other diseases. Two such systemic features come immediately to mind.
First, the United States is unique among the countries we studied in its use of the patient's ability to pay as a critical factor in deciding who receives therapy. This issue is not relevant in countries with universal health care coverage, where physicians can rely primarily on clinical criteria in allocating transplantation resources.
Second, although there are oncologists in private practice and fee-for-service methods of reimbursement in some other countries, the United States is unique in having a large number of specialists who practice outside hospitals in fee-for-service settings. The combination of a tradition of independent practice and a financing structure that provides little incentive to refer patients quickly to hospital-based care may explain, in part, why some patients receive transplants after the optimal point in the progression of disease. Conversely, in systems that have placed responsibility for all patient care expenses within a single entity (the case in both national single-payer systems and prepaid-insurance plans), the logical goal is to ensure that patients who will eventually need a transplant receive one as quickly as possible.
Assertions that in the United States the single-payer financing, global budgeting, and universal coverage that characterize the Canadian and European health care systems would lead inevitably to reductions in the availability of sophisticated and expensive medical interventions are not supported by our data. Although our findings do not prove that universal coverage or global budgets are essential, they demonstrate that health care systems based on these elements can perform as well as or better than our current system.
We are indebted to Dr. Larry Kessler for his advice on estimating transplantation rates, to Dr. Charles Schiffer for his thoughtful review of the study results, and to Sharon K. Nell, D'Etta Waldoch Koser, and Edward R. Knutson for their assistance in the collection of data.
Source Information
From the General Accounting Office, Washington, D.C. (G.S., M.G.C., E.A.P., R.C.W.); the International Bone Marrow Transplant Registry, Health Policy Institute, Medical College of Wisconsin, Milwaukee (M.M.H.); the Fred Hutchinson Cancer Research Center, Seattle (F.R.A.); and the Cancer Therapy Evaluation Program, National Cancer Institute, Bethesda, Md. (B.D.C.). The views expressed herein are those of the authors and do not necessarily reflect those of the General Accounting Office, which sponsored the larger study on which this work is based.
Address reprint requests to Mr. Silberman at the Program Evaluation and Methodology Division, General Accounting Office, Rm. 5737, Washington, DC 20548.
References
The International Bone Marrow Transplant Registry is supported by a grant (PO1-CA-40053) from the National Cancer Institute, the National Institute of Allergy and Infectious Diseases, and the National Heart, Lung, and Blood Institute and by grants from the following: the Alpha Therapeutic Corporation, the Armour Pharmaceutical Company, the Lynde and Harry Bradley Foundation, Bristol-Myers, Burroughs Wellcome, the Charles E. Culpeper Foundation, the Eleanor Naylor Dana Charitable Trust, the Eppley Foundation for Research, Hoechst-Roussel Pharmaceuticals, Immunex Corporation, the Kettering Family Foundation, the Robert J. and Helen C. Kleberg Foundation, the Eli Lilly and Company Foundation, Marion Merrill Dow, the Ambrose Monell Foundation, the Samuel Roberts Noble Foundation, Ortho Biotech Corporation, the John Oster Family Foundation, the Jane and Lloyd Pettit Foundation, the RGK Foundation, the Roerig Division of Pfizer Pharmaceuticals, the Sandoz Research Institute, the Stackner Family Foundation, the Starr Foundation, the Joan and Jack Stein Charities, the Swiss Cancer League, and Wyeth-Ayerst Research.
| |||||||||||||||||||||||||||||||||||||||
Related Letters:
Treatment of Acute Lymphoblastic Leukemia in a Second Remission
Pinkel D., Johnson F. L., Thomas E. D., Barrett A. J., Pollock B. H., Buchanan G. R.
Extract |
Full Text
N Engl J Med 1995;
332:823-824, Mar 23, 1995.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
