Volume 23
Number 3
2002

[Table of Contents]

The Medical Care Cost of Childhood and Adolescent Cancer in Manitoba, 1990–1995

Abstract

The main purpose of this study is to estimate the medical care costs of childhood and adolescent cancer in Manitoba, and to determine the elements that influence these costs. Retrospective chart reviews were done to obtain all the information. A total of 118 childhood (age 0–14 years) and 41 adolescent (age 15–19 years) cancer patients were included. For childhood cancer, in-patient hospitalizations accounted for 59% of the total cost, followed by bone marrow transplant (BMT) (9%), medications (8%), laboratory investigations (7%) and physician fees (7%). For adolescent cancer, in-patient hospitalization accounted for 37% of the total cost, followed by BMT (25%), physicians' fees (11%), medications (9%) and laboratory investigations (7%). Overall, the average cost for the first, second and third year following diagnosis was $50,902 (median 35,708), $13,939 (4,127) and $6,769 (2,565) respectively for childhood cancer patients, and $57,354 (24,192), $16,888 (3,267) and $3,436 (3,267) respectively for adolescent cancer patients. Further work involving long-term data linkage of medical charts with hospital and clinic financial billing codes is needed to provide more accurate estimates of the costs of childhood and adolescent cancer care.

Key Words: adolescent cancer; childhood cancer; cost

Introduction

Cancer is the most common fatal disease of childhood and adolescence; only accidents kill more Canadian children and adolescents aged 1–19 years.¹ Recent improvements in the prognosis for children with cancer, due to advances in treatment,² has focused attention on the burden of therapy on the children and their families,³ and the long-term effects of the disease and its management.^4–5 Costs to the health care system have been less well studied.

Although reports on childhood cancer often include adolescents, teenagers have a mix of cancers which differ from those of both adults and children. The embryonal cancers found in young children (e.g., neuroblastoma, retinoblastoma, and hepatoblastoma) are almost unheard of among those aged 15–19 years, while the epithelial carcinomas of adults (e.g., lung, breast, colon) are equally rare. Patterns of care also differ, as adolescent cancer patients may be treated either in pediatric oncology centres using pediatric protocols or adult oncology centres using either pediatric or adult protocols. Adolescent patients are much less likely than children with cancer to be enrolled in clinical trials.⁶

To date, there have been no published estimates of the direct medical care costs of childhood and adolescent cancer in Canada. Information on the cost of treating childhood and adolescent cancer patients may be useful to health care planners to direct and plan priorities of childhood cancer control programs and to make the most effective use of limited medical care budgets.

Only a handful of studies have been conducted to estimate the medical care cost of childhood and adolescent cancer. In 1983, Lansky et al.⁷ reported medical costs for childhood cancer patients in the US that ranged from $100 (histiocytosis X) to $1800 (lymphoma) per month. In-patient hospital charges were the most expensive component (mean cost $400 per month), followed by pediatricians' fees ($100 per month). The estimation of medical costs was based on a randomly chosen four-week period in the years 1978 to 1980 and the sample size (N=64) was small.

In 1985, also in the US, Bloom et al.⁸ estimated that the mean cost of care for children with cancer was $29,708 per patient-year, based on a followup time of six months. The authors concluded that direct medical expenditures were about two thirds of the total costs.

Birenbaum and Clarke-Steffen⁹ conducted a study (N=19 families) to describe the health care costs for childhood cancer patients in the terminal phase of illness in the US in 1987. Direct health care costs, defined as expenses billed for the provision of health care services for the dying child from the time of admission to the study until death, accounted for $8,456, or 76% of the total costs. The median costs of outpatient and in-patient services were $949 (range: $55–$2,138) and $6,877 (range: $0–$79,027) respectively.

Unlike the US, which relies upon a mix of privately and publicly funded health care, Canada has a publicly funded health care system which provides universal access to hospitalization and physician care. However, the nature of childhood cancer care in Canada and the US is similar; in both countries, treating institutions are almost always members of the two major childhood cancer cooperative study groups (Pediatric Oncology Group and Children's Cancer Group), that have merged recently to form the Children's Oncology Group.

The main purposes of this population-based study are to estimate the medical care costs of childhood and adolescent cancer in Canada, and to determine which elements influence these costs from the perspective of health care providers. The study was undertaken in Winnipeg, the capital city of the province of Manitoba, as all children with cancer in the province are referred to the Children's Hospital in Winnipeg. All of the costs included in this study refer to 1992 Canadian dollars.

Materials and methods

Medical care components

Information on all newly diagnosed cases of cancer (excluding skin carcinomas) in children and adolescents aged 0–19 years between January 1, 1990 and December 31, 1992 was obtained retrospectively from the population-based Manitoba Cancer Registry. These cancer cases were followed for three years from diagnosis, up to December 31, 1995. The diagnoses of childhood (age 0–14 years) and adolescent (age 15– 19 years) cancer were grouped using the International Classification of Childhood Cancer¹⁰ which is based on the 2^nd edition of the International Classification of Disease for Oncology¹¹ morphology and topography codes.

Approximately 85% of the population of Manitoba lives within 100 kilometres of Winnipeg. As a result, childhood cancer care in Manitoba is centralized; all childhood cancers in the province are treated at the Children's Hospital in Winnipeg. Adolescent cancer patients are treated either in the Children's Hospital or in an adult setting at the Health Sciences Centre, also in Winnipeg. At the time of this study, the use of outreach centres in Manitoba was minimal.

Medical care utilization information was collected in five settings: the Children's Hospital in Winnipeg, the Health Sciences Centre in Winnipeg, the CancerCare Manitoba clinic, and the emergency room (ER) of the Children's Hospital and the Health Science Centre. All medical care information was abstracted retrospectively from the patients' medical charts using a data abstraction form developed for this purpose by a research assistant at the Manitoba cancer registry.

Medical care information was collected for the following components: length of hospital stay; medications administered (antibiotics, chemotherapy and other supportive pharmacological treatment); laboratory tests performed; physician visits by type of visit (clinic, hospital, emergency room, consultation or administration of chemotherapy); diagnostic tests (radiology and other diagnostic tests); diagnostic and operative surgeries; blood products administered and radiation treatments; and the type of visit: in-patient, outpatient or emergency room. Bone marrow transplantation was costed separately.

Information on any visits to an outreach centre or followup visits to medical professionals in private practice was not collected in the study.

Unit costs

We based our study on costs in 1992, which was roughly the midpoint of the study period. A cost was assigned to each component abstracted from the patient's chart, along with the date when the cost was incurred. Total costs were obtained by summing the individual costs. To avoid double counting, each cost component was estimated independently.

Bone marrow transplant

The average cost of a bone marrow transplant (BMT) was estimated by the department of pediatric BMT in the Children's Hospital at $108,990. This cost included comprehensive care for the time of transplant and for a 100 day period thereafter. To avoid double counting, no medical care utilization data were collected on the questionnaire during the BMT period. As a result, medical care costs are included within the comprehensive care costs component of total BMT costs.

Laboratory investigations

The costs per 100 workload units, which included salaries, supplies and overhead for each type of laboratory test, were obtained from Westman Laboratory Services (operated by Manitoba Health),¹² while the workload units for each specific test were obtained from the Canadian Institute for Health Information.¹³ Workload units are defined as the number of minutes of direct labour time required to perform a specific test or procedure. The cost of a specific test was calculated by multiplying the cost per workload unit by the number of workload units utilized. The hospital pays for laboratory tests based on workloads, so unit costs were not available and had to be estimated.

Medications and blood products

The costs of chemotherapeutic agents, antibiotics, and other supportive pharmacological treatment were obtained from the Health Sciences Centre Drug Formulary, 1991–1992.¹⁴ The price of blood products was provided by the Blood Services Centre¹⁵ in 1994 and 1995 dollars. These were converted to 1992 dollars using the pharmaceutical component of the Industrial Product Price Index.¹⁶

Physician fees

There were five categories of physician payments in this study: consulting physician fees, hospital in-patient fees, ER physician fees, physician fees for chemotherapy, and sessional physician fees.

Consulting physician fees were obtained from the billable fee-for-service charges (e.g., surgeons, radiologists).¹⁷ Hospital in-patient physician fees per patient day were calculated based on the concomitant care fees, (i.e., hospital physician fees varied by days of hospitalization).¹⁷ ER physician fees for each visit to ER were based on the fee per ER visit.¹⁷ Physician fees for the administration of chemotherapy were obtained from the Manitoba Health Services Insurance Plan.¹⁷ The CancerCare Manitoba clinic has full-time pediatric oncologists on staff to deliver outpatient care; this care was billed as sessional physician fees, which were based on the frequency of clinic visits during the study period.

Diagnostic tests, surgeries, radiation therapy

The unit costs of these procedures were obtained from the Manitoba Health Services Insurance Plan.¹⁷

In-patient hospital care

Hospital in-patient charges were based on the Children's Hospital per diem rate, which included the cost of room and board, nursing, intensive care, radiology tests, laboratory investigations, drugs, use of the operating rooms and any additional treatment. From these charges we were able to separate out the costs of radiology tests, laboratory investigations, drugs and use of operating rooms. Because we were unable to determine the specific in-patient hospital care costs of room and board, nursing, intensive care and additional treatment associated with treating children with cancer, and because these costs for children with cancer may have differed from the costs of other hospitalized children, we conducted sensitivity analyses based on adding or subtracting 30% of the reported costs.

Results

A total of 118 childhood cancer patients (57 boys and 61 girls) and 41 adolescent cancer patients (23 boys and 18 girls) were included in the study with a mean age at diagnosis of 6.7 and 17.0 years respectively. Table 1 presents the distribution of cancers in childhood and adolescence by diagnosis. The four most common childhood cancers were central nervous system (CNS) tumours (N=33), leukemias (N=30), lymphomas (N=14) and sympathetic nervous system (SNS) tumours (N=13), which accounted for more than 70% of all childhood cancer cases. In comparison, lymphomas (N=12), CNS tumours (N=7), soft-tissue sarcomas (N=7) and leukemias (N=6) were the four most common cancers among adolescents, accounting for 78% of all cases. At the end of three years of follow-up, 91 childhood cancer patients and 31 adolescent cancer patients were alive, corresponding to a three-year survival rate of 77% and 76% respectively.

FIGURE 1
Distribution of total medical care costs of childhood and adolescent cancer, Manitoba, 1990–1995

The average number of hospitalization days differed dramatically by diagnosis for both childhood and adolescent cancers (Table 2). Among the more common childhood cancers (those listed in Table 2), the average number of days was highest for SNS tumours, and lowest for lymphomas. The average number of days was highest for leukemias, and lowest for CNS tumours for adolescent cancers. Hospitalizations were concentrated in the first year after diagnosis (data not shown).

Overall, the average costs for the first, second and third years following diagnosis were $50,902, $13,939 and $6,769 respectively for childhood cancers (Table 3). In comparison, the average costs for the first, second and third years following diagnosis were $57,354, $16,888 and $3,437 respectively for the adolescent cancers (Table 4). However, these costs varied by diagnosis and by vital status at the end of the study period. Among the four most common childhood cancer sites, SNS tumours had the highest average cost ($86,715) for the first year after diagnosis followed by leukemias ($59,595), CNS tumours ($42,859) and lymphomas ($35,834). The average costs of treating leukemias, lymphomas and CNS tumours dropped by 67%, 39% and 71% respectively between the first and second years, while for all other cancer sites (except kidney, for which costs declined 55%), there was more than a 90% decline between these time periods.

Among the four most common adolescent cancer sites, leukemias had the highest average cost ($165,800) for the first year following diagnosis, reflecting the use of BMT, followed by soft-tissue sarcomas ($73,355), lymphomas ($36,092) and CNS tumours ($16,365). The average cost of treating lymphomas, CNS tumours and leukemias dropped by 30%, 95% and 89% respectively between the first and second years, while for all other cancer sites except bone tumour, for which costs declined 53%, there was more than an 85% decline between these time periods.

For both childhood and adolescent cancer patients, the average medical costs in the first year after diagnosis for patients who survived were lower than for those who did not (Figure 2). This was particularly true of childhood leukemia cases; those who died had medical costs ($89,707) which were almost double those of the cases who survived ($52,066). For adolescent soft-tissue sarcomas, those who died had medical costs ($103,804) that were triple those of the cases who survived ($32,757). For childhood cancer cases, no real differences were observed for lymphomas ($36,586 versus $35,708) and SNS tumours ($88,316 versus $86,182), while cases with CNS tumours who survived actually cost more than those who died ($44,913 versus $39,435). The average cost of treating childhood bone and soft-tissue sarcoma patients who died was more than double that of those who survived, as were the treatment costs of adolescent CNS tumours, bone tumours and leukemia patients. Higher costs for cancer patients who did not survive reflect more hospitalization days than those who survived.

FIGURE 2
The average cost of the first year after diagnosis by vital status and cancer site for childhood and adolescent cancer

Discussion

To reduce the possibility of selection bias, which may affect hospital-based studies, this study used population-based cancer registry data to identify all children and adolescents aged 0–19 years diagnosed with malignant neoplasms in Manitoba. Three years of incident cases provided enough data to allow costs to be categorized by cancer type, while three years of followup provided a good overview of the various components of care.

In our study, the most expensive component was in-patient care, which accounted for 59% and 37% of the total costs for childhood and adolescent cancer respectively. The other expensive components were: BMT, medications, laboratory investigations and physicians' fees for both childhood and adolescent cancers. These relative proportions are similar to those noted by Bloom et al.⁸

We estimated the costs for five childhood cancer patients for whom information was unavailable in the medical charts, based on treatment procedures received for patients with similar diagnoses. It is likely that this resulted in a minor underestimation of costs, as only major treatment components were costed (e.g., surgical procedures).

Similarly, 30% of laboratory costs had to be estimated because there was insufficient information on the medical charts to determine which specific tests were undertaken. Expenditures for these tests were imputed from the mean cost of tests for the same diagnostic category (e.g., renal, microbiological) for both childhood and adolescent cancer patients. This may also mean that costs were underestimated.

Likewise, while hospital care was assigned from a global hospital figure, cases of cancer in childhood and adolescence have a higher-than-average relative intensity weighting (a measure of the complexity of the care required and so the nursing hours involved and so the cost), and this may have also contributed to an underestimation of the generalizable average costs. For this reason, we conducted sensitivity analyses by adding and subtracting 30% of the reported hospital care. The usefulness of such sensitivity analyses may be appreciated from a earlier Canadian report of an economic evaluation of allogeneic BMT.¹⁸

We did not include the costs of outreach centres (rural areas) and private health professionals. Those costs should be minimal as almost all childhood and adolescent cancer cases were treated in the Children's Hospital or the Health Sciences Centre in Winnipeg.

The average medical cost of patients who survived to the end of three years was less than that of patients who did not survive, especially for childhood leukemia and adolescent soft-tissue sarcoma patients. Much of the difference occurred because patients who died had more days of hospitalization than patients who survived. Successful treatment not only saves lives, but reduces the cost of treating childhood and adolescent cancer patients over the initial three years.

Among the four most common types of childhood cancers, costs were highest in the first year for SNS tumours (almost all of which were neuroblastomas), followed by leukemias, CNS tumours and lymphomas. The high cost of treating neuroblastomas reflects the greater number of days of hospitalization for these patients. This finding differs from those of Lansky et al.⁴ and Bloom et al.;⁸ the leading average costs in these studies being for lymphomas and malignant bone tumours respectively. However, the present study represented the costs in the initial three years after diagnosis while the Lansky study focused on children with cancer treated as outpatients. In the study by Bloom et al., hospital visits were used to identify participants rather than using a registry of all children with malignant diseases. This could have reduced the probability of including low-cost patients who return for visits less frequently than once every six months.

In contrast, among the four most common types of adolescent cancers, costs were highest in the first year after diagnosis for leukemias, followed by soft-tissue sarcomas, lymphomas and CNS tumours. The high cost of treating leukemia reflects the greater number of days of hospitalization and relatively high use of BMT for these patients.

The present study only estimated the medical care cost of childhood and adolescent cancers for a three-year period after diagnosis. We were not able to estimate long-term increased health care costs in long-term survivors, or to estimate the additional costs which some survivors of childhood and adolescent cancer may require for special education or home care services, or the non-medical costs borne by the families of these children and adolescents.^3,19,20

Long term data linkage of medical charts with hospital and clinic financial billing codes would provide more accurate and valuable information for estimating the costs of childhood cancer care.

Acknowledgements

The authors gratefully acknowledge Dr. David Feeny for his comments on the manuscript. The assistance of Ms. Cathy Jonatanson and Ms. Jeri Kostyra in data collection is acknowledged with appreciation.

References

1. Huchcroft S, Clarke A, Mao Y, et al. This battle which I must fight: Cancer in Canada's children and teenagers. Ottawa: Supply and Services Canada; 1996. 110p.

2. Craft AW, Pearson ADJ. Three decades of chemotherapy for childhood cancer: from “cure at any cost” to “cure at least cost”. Cancer Surv 1989;8:605–629.

3. Barr RD, Furlong W, Horsman J, et al. The monetary costs of childhood cancer to the families of patients. Int J Oncol 1996;8: 933–940.

4. Hawkins MM, Stevens MC. The long-term survivors. Br Med Bull 1996 Oct;52(4): 898–923.

5. Blatt J, Copeland DR, Bleyer WA. Late effects of childhood cancer and its treatment. In: Pizzo PA, Poplack DG, eds. Principles and Practice of Pediatric Oncology. 3^rd ed. Philadelphia, Lippincott-Raven Publishers, 1997. 1303–1329p.

6. Bleyer WA, Tejeda H, Murphy SB, et al. National cancer clinical trials: children have equal access; adolescents do not. J Adolescent Health 1997;21:366–373.

7. Lansky SB, Black JL, Cairns NU. Childhood cancer medical costs. Cancer 1983;52: 762–766.

8. Bloom BS, Knorr RS, Evans AE. The epidemiology of disease expenses – the costs of caring for children with cancer. JAMA 1985; 253:2393–2397.

9. Birenbaum LK, Clarke-Steffen L. Terminal care costs in childhood cancer. Pediatr Nurs 1992;18:285–288.

10. Kramárová E, Stiller CA, Ferlay J, et al. Editors. International Classification of Childhood Cancer. IARC technical report No. 29, Lyon; 1996. 15–18p.

11. Percy C, Van Holten V, Muir C. International Classification of Diseases for Oncology, second edition. World Health Organization, Geneva, 1990. 144p.

12. Westman Regional Laboratory Services Incorporated. Annual Report 1992–1993. 13p.

13. MIS Guidelines. Ottawa: Canadian Institute for Health Information, 1999. Chapter 4 – Workload Measurement Systems.

14. Hall K, Nicolle L, Sochasky CM. Winnipeg Health Sciences Centre – Formulary, 1991– 1992, 6^th Edition.

15. Canadian Blood Services, Unit cost model, Ottawa: Canadian Blood Services; 1998.

16. Statistics Canada: Industrial Products Price Index, December 1995. Ottawa: Minister of Industry, Science and Technology; 1996. p 64.

17. Manitoba Health Services Insurance Plan, Physician's Manual, April 01, 1992. The Minister of Health, Manitoba, 1992.

18. Barr RD, Furlong W, Henwood J, et al. Economic evaluation of allogeneic bone marrow transplantation: A rudimentary model to generate estimates for the timely formulation of clinical policy. J Clin Oncol 1996; 14:1413–20.

19. Canadian Coordinating Office for Health Technology Assessment, Guidelines for Economic Evaluation of Pharmaceuticals: Canada, 2^nd Edition. Ottawa: Canadian Coordinating Office for Health Technology Assessment, November, 1997. (This document can be downloaded from: http://www.ccohta.ca).

20. Drummond MF, O'Brien B, Stoddart G, and Torrance GW, Methods for the Economic Evaluation of Health Care Programmes. Second Edition. Oxford: Oxford University Press, 1997.

Author References

Wei Luo, Rachel Lane, Howard Morrison, Dena Schanzer, Cancer Division, Public Health Agency of Canada, Health Canada, Ottawa, Ontario

Dr. Kent Stobart , CancerCare Manitoba, Winnipeg, Manitoba

Dr. Ronald Barr, McMaster University, Hamilton, Ontario

Dr. Mark Greenberg, Hospital for Sick Children, Toronto, Ontario

Correspondence: Wei Luo, Cancer Division, Public Health Agency of Canada, Health Canada, Tunney's Pasture, PL 0601C1 Ottawa, Ontario, Canada K1A 0L2; Fax: (613) 941-1732; E-mail: Wei_Luo@hc-sc.gc.ca

[Previous] [Table of Contents] [Next]