Economic Burden of Illness in Canada, 1993

Direct Costs

Drug Expenditures

Methods

The 1993 total national drug expenditure is derived from estimates provided by Health Canada's National Health Expenditures in Canada, 1975-1994 (NHEC) ¹ and Health Canada's Bureau of Drug Policy and Coordination. The NHEC provides cost estimates of non-prescription drugs and personal health supplies. The Bureau of Drug Policy and Coordination provides estimates primarily for 1993 Canadian prescription drug costs to the consumer. These estimates are based on three IMS audits ^2-4 and Statistics Canada's Hospital Morbidity.^5,6 (Refer to Appendix 2 for definitions.)

The Canadian Compuscript (CS)² and Canadian Disease and Therapeutic Index (CDTI)³ audits are used to calculate the cost of drugs purchased directly by the consumer. The CS audit estimates the dollar value of prescriptions dispensed by Canadian pharmacies to the consumer by therapeutic class; the CDTI provides estimates of the frequency with which office-based physicians recommended drugs by diagnostic category, age and sex. These data are used to distribute the dollar value of prescriptions filled for each therapeutic class to a diagnostic category, age group and sex.

The Canadian Pharmaceutical Market: Drug Store and Hospital Purchases (CDH)⁴ and CDTI ³ audits and Hospital Morbidity ^5,6 are used to calculate the cost of drugs used in hospitals. The CDH estimates the dollar value of drug purchases for hospitals by therapeutic class. These costs have been increased by 33% to approximate retail values, assuming hospital dispensing costs (pharmacists, storage, etc.) to be equivalent to retail dispensing fees and mark-ups.^{7 *} The CDTI is used to assign dollar values of drugs in each therapeutic class to diagnostic category. Age and sex distributions are based on Statistics Canada's data on days in hospital for each diagnostic category.

^* This figure is based on the experience of pharmacists and others in the health care setting. Although we were unable to obtain data to support this assumption, the figure appears to be agenerally accepted estimate.⁷

Assumptions

• The 1993 national drug expenditure ¹ is assumed to include all drugs sold in retail drug stores as estimated by the Bureau of Drug Policy and Coordination.
• The distribution of drugs within a therapeutic class for each diagnostic category is assumed to be similar for office-based physicians and hospitals. Although the specific drugs used may differ, therapeutic drug classes are fairly broad.
• The distribution of days in hospital by diagnostic category, age group and sex is assumed to reflect the distribution of hospital drugs.

Limitations

• Part of the national drug expenditure could not be allocated by diagnostic category. This represents most non-prescription drugs and all personal health supplies.
• The extent to which NHEC overlaps with the three IMS audits is unknown. The latter represents mainly prescription drugs and some non-prescription drugs sold in retail drug stores.
• IMS audits are based on samples. Drug costs for some categories or subcategories may be underestimated.
• CDTI data reflect drugs prescribed, whether or not the prescription is actually filled.
• CS data include non-prescription drugs only if prescribed by a physician and dispensed by a pharmacist in a drug store.
• CDH data do not include the activities of specialty health clinics; thus some diagnostic categories may be underestimated (e.g. cancer, pregnancy).

Results

The 1993 national expenditure for drugs is estimated at $9.9 billion. An estimated $7.2 billion of this expenditure can be allocated to diagnostic category. This figure represents mainly prescription drugs dispensed in retail drug stores ($6.12 billion) or used in hospitals ($1.04 billion). An additional $2.7 billion, which primarily represents non-prescription drugs and personal health supplies, could not be categorized.

Cardiovascular diseases represented the largest share ($1.6 billion) of the $7.2 billion classified by diagnostic category, followed by respiratory diseases ($962 million) and mental disorders ($602 million).

Coronary heart disease ($239 million) accounted for 15.3% of the pharmaceutical cost of cardiovascular diseases. Drugs used for the treatment of chronic bronchitis, emphysema and asthma ($401 million) made up 41.7% of the amount spent on respiratory diseases. Diabetes ($161 million) was responsible for 32.2% of the cost of drugs for the treatment of endocrine and related diseases. Female cancers ($60 million) accounted for one quarter (25.7%) of drug costs to treat cancer.

Well-patient care (defined here by ICD-8 as special conditions and examinations without sickness) had a significant impact on drug costs ($280 million). This category includes contraceptives, infant formulae and vitamins, biologicals and prophylactic antimalarials.

[Figures]

---

Physician Care Expenditures

Methods

The 1993 total national expenditure for physician services is derived from Health Canada's National Health Expenditures in Canada, 1975-1994. ¹ This estimate includes professional health services provided by both physicians and psychologists. (Appendix 2 provides more detailed definitions.) The largest component of this expenditure is professional fees paid primarily through provincial medical care insurance plans.

Seven provinces (Manitoba, Saskatchewan, Alberta, BC, Nova Scotia, PEI and Newfoundland) have provided total expenditures primarily for physician fee-for-service expenditures by diagnostic category and sex. The distribution of costs is similar for the seven provinces with the exception of well-patient care and ill-defined conditions (see figure in Appendix 3). Provincial officials suggest that disparities in these two categories can be attributed to overclassification of many diagnoses to one of these two categories that capture poorly defined diagnoses.²

The Manitoba data have been validated;^3,4 Manitoba Health's distribution of 1993/94 physician expenditures ⁵ is thus used to reflect national medical care costs by diagnostic category, age and sex.^4,5 This distribution is applied to the 1993 national physician care expenditure to estimate the physician care expenditure by diagnostic category, age and sex for Canada in 1993. A non-ICD category, "Without Diagnosis," was added to correspond to Manitoba's categories.

Assumptions

• Most provinces could only provide 1991/92 data at the time of our comparison of provincial medical care expenditures. We assumed the distribution within each province to be fairly constant between 1991/92 and 1993/94.
• We assumed the distribution of medical care expenditures for Ontario, Quebec, New Brunswick and the territories to be similar to that of the seven other provinces.
• Manitoba's distribution of medical service costs by diagnosis, age group and sex is assumed to reflect that of the national physician expenditure, based on comparison with seven other provinces.

Limitations

• Provincial data files used here are designed to administer payment of services and not to collect patient information. Although most provincial medical care plans collect data about patient diagnosis, not all physicians are required to record diagnostic coding. Limitations of coding include changes to initial diagnoses upon further investigation, use of a limited set of diagnostic codes, overclassification to categories that capture poorly defined conditions (i.e. ill-defined, without diagnosis, well-patient care) and lack of validation.
• Remuneration for services of psychologists represents a very small portion of the national expenditure for physician care. Any private expenditures for services of psychologists would be included in the category of Other Professionals in the cost component Additional Direct Health Expenditures. Expenditures for services of psychologists in hospitals and other institutions are included in Hospital Care Expenditures and Expenditures for Care in Other Institutions.
• Physicians paid by salary, session or contract are counted elsewhere (i.e. Hospital Care Expenditures and Expenditures for Care in Other Institutions).
• Medical expenditures by diagnostic category for Ontario, Quebec, New Brunswick and the territories, representing 67.5% of the national physician expenditure,¹ were unavailable; the distribution of disease costs for these areas will have an important impact on the national distribution.

Results

The 1993 cost of medical services provided by physicians was $10.4 billion. Well-patient care (defined by ICD-9 as "factors influencing health status and contact with health services") was the leading medical care cost ($1.1 billion), accounting for 10.6% of the expenditure for physician services. Well-patient care includes general medical examinations and special investigations. Other major contributors to the physician care expenditure were respiratory diseases and diseases of the nervous system and sense organs, each costing approximately $1 billion.

Diabetes accounted for 47.3% of the national physician expenditure for endocrine and related diseases. Coronary heart disease and stroke ($338 million) made up almost 40% of the total medical care expenditure for cardiovascular diseases. The majority of this was for coronary heart disease ($263 million).Chronic bronchitis, emphysema and asthma ($289 million) were responsible for 28.3% of the medical care costs for treating respiratory diseases.

[Figures]

---

Hospital Care Expenditures

Methods

The 1993 total national hospital expenditure is derived from Health Canada's National Health Expenditures in Canada, 1975-1994.¹ The cost of hospital drugs ($1.04 billion) calculated in our cost component Drug Expenditures has been deducted from this total.

Statistics Canada routinely collects hospital statistics from all public, private and federal hospitals operating in Canada. More than 90% of hospital beds are represented in Hospital Statistics: Preliminary Annual Report, 1993-94;² however, only public hospitals are required to report financial information. The total 1993/94 hospital expenditure by hospital type is estimated by multiplying the reporting public hospital's cost per bed for each type of hospital by the total number of operating beds within the hospital type. Drugs dispensed in hospitals have been deducted from the total.

Hospital expenditures by diagnostic category, age group and sex are estimated separately for three broad hospital categories: acute, long-term and psychiatric hospitals. (Refer to Appendix 2 for definitions.) The distribution of Statistics Canada's 1993/94 hospital expenditures to these three categories was determined and applied to Health Canada's 1993 hospital expenditure.

Acute care hospital expenditure is distributed using 1993 total resource intensity weights (RIWs) by diagnostic category and sex from the Canadian Institute for Health Information (CIHI).³ Total RIWs represent both typical and atypical acute inpatient cases, taking into account both the total cost of acute care (fixed and variable) and the total of services used for acute care. Acute care hospital expenditure is distributed by age group using days-in-hospital data.

Long-term care hospital expenditure is distributed using the 1993/94 one-year pattern of days in general and allied special hospitals by diagnostic category, age group and sex for stays of 100 or more days for patients who were discharged during 1993/94. ⁴

The total psychiatric hospital expenditure, categorized as Mental Disorders, is distributed by age group and sex using the 1993/94 one-year pattern of days spent in psychiatric hospitals for patients who separated from hospital during 1993/94. ^5,6

Assumptions

Acute Care Hospitals

• We assume the acute care hospitals identified above to be the same as those used in calculating total RIWs; the majority of operating costs for "non-teaching hospitals with long-term care units" and "teaching" hospitals are assumed to be for treatment of acute disease. Long-term care, outpatient and emergency ward costs (included in the national expenditure) are assumed to have diagnostic category, age and sex distributions similar to acute care costs. The cost of hospital drugs (included in total RIWs) is assumed to have little impact on the overall distribution of acute care hospital costs; drugs represented 4% of the total hospital expenditure before excluding them.

Long-term Care Hospitals

• There is no clear definition of "long-term" care; however, 100 days reflects the average length of stay in extended care hospitals and the criteria for copayment in some provinces.⁷ The 1993/94 one-year pattern of days in hospital for stays of 100+ days for patients who separated during the 1993/94 year is assumed to reflect the distribution of patients hospitalized in "long-term" hospitals. The average cost per day in long-term care hospitals is assumed to be similar for all diagnostic categories. The pattern of hospitalization in public hospitals (on which hospital morbidity is based) is assumed to reflect that of private and federal hospitals.

Psychiatric Hospitals

• The 1993/94 one-year pattern of days spent in psychiatric hospitals for patients discharged during 1993/94 is assumed to reflect the age and sex distribution of patients hospitalized in psychiatric hospitals.

Limitations

• Statistics Canada ² collects financial information only from reporting public hospitals. Private and federal hospitals, which represent 4% of hospital beds, do not report financial information.

Acute Care Hospitals

• CIHI ³ provides about 80% coverage of Canada and less in Quebec, Manitoba and Nova Scotia. It is not possible to organize hospitals neatly into categories (i.e. non-teaching hospitals with long-term care units provide both acute and long-term care). The hospitals used in calculating total RIWs may vary from those included in our acute care hospital category. Total RIWs include the cost of drugs dispensed in hospitals.
• More than 20% of beds in general and other short-stay hospitals (i.e. acute care hospitals) are long-term care beds. Total RIWs pertain only to the acute care provided within these hospitals.

Long-term Care Hospitals

• Statistics Canada ⁴ does not provide hospital morbidity by hospital type (i.e. extended care, rehabilitation). Data also exclude the Yukon and the Northwest Territories and all federal and proprietary hospitals.

Psychiatric Hospitals

• Separation data are counts of patients showing deaths and discharges during the reporting year and may not represent the total number of patients being treated in hospital during that year. Our age and sex cost distributions may therefore vary from the actual distribution in psychiatric hospitals during the year.

Results

The total national hospital expenditure for 1993 was an estimated $26.1 billion. Acute care, long-term care and psychiatric hospitals represented approximately $22.6 billion, $1.8 billion and $1.7 billion of this expenditure, respectively.

Cardiovascular diseases represented almost one fifth of the total hospital expenditure, costing $4.9 billion. Mental disorders, cancer and injuries cost $3.6 billion, $2.5 billion and $2.3 billion, respectively.

Coronary heart disease and stroke together ($2.83 billion) made up 58.2% of the total hospital expenditure for cardiovascular diseases. Diabetes accounted for 52.1% of the total hospital expenditure for endocrine and related diseases. Chronic bronchitis, emphysema and asthma represented 35.6% of respiratory diseases.

[Figures]

---

Expenditures for Care in Other Institutions

Methods

The 1993 total national expenditure for care in other institutions is derived from Health Canada's National Health Expenditures in Canada, 1975-1994.¹ Health Canada defines "other institutions" as residential care facilities (RCFs) that are approved, funded or licensed by provincial/territorial departments of health and/or social services. (Appendix 2 provides more detailed definitions.)

Residents of these facilities generally are chronically ill or disabled and reside in the facility more or less permanently. Usually RCFs do not provide the level of medical care and supporting diagnostic and therapeutic services provided by hospitals, although there is some overlap in services. Drugs prescribed in these institutions are included in this expenditure.

Statistics Canada's 1993/94 inventory of all RCFs in Canada, reported annually by provinces and territories, provides an expenditure breakdown by predominant type of care and principal characteristic of the predominant group of residents within the reporting facility.² Facilities providing Type II care or higher correspond to facilities included in Health Canada's definition and are used to distribute the 1993 Health Canada figures by principal characteristic of residents.

Assumptions

• The 1993/94 distribution of total expenditures in reporting RCFs that provide predominantly Type II care and higher ² is assumed to represent the 1993 distribution for other institutions.¹
• The distribution of RCFs by predominant type of care and principal characteristic of residents ² is assumed to represent that of Quebec.

Limitations

• The 1993 expenditure for care in other institutions ¹ could not be categorized by diagnostic category.
• The Statistics Canada 1993/94 expenditure for reporting RCFs ² includes only 70% of operating facilities and 75% of all operating beds, and it excludes Quebec. Thus, it underestimates the total national RCF expenditure.
• Quebec did not report the predominant type of care in a facility.

Results

The 1993 national expenditure for care in other institutions was $7.0 billion. The majority of this expenditure ($6.1 billion) was for homes for the aged. Another $831 million (12.1%) was spent for the care and treatment of those with conditions classified as mental disorders: the developmentally delayed, the psychiatrically disabled, clients with alcohol and drug addictions, and emotionally disturbed children.

[Figures]

---

Health Science Research Expenditures

Methods

The 1993/94 health science research expenditure is derived from the Reference List of Health Science Research in Canada, 1993-94, ¹ the Traffic Injury Research Foundation (TIRF) 1993 Annual Report ² and the Association of Canadian Medical Colleges (ACMC) 1993/94 Canadian Medical Education Statistics.³

The Reference List provides the cost of health science research grants and awards made in Canada by federal, provincial and voluntary agencies for the fiscal year 1993/94 (see Appendix 4). Unfortunately, the grants and awards are not classified by diagnostic category. We have apportioned these costs to the relevant area of research, assigning projects to an appropriate diagnostic category based on project title. Diagnostic subcategories were created for diagnoses with a large number of grants/awards, and additional categories were created for grants/awards that could not be classified by diagnosis (see Appendix 2).

The 1993/94 cost of motor vehicle traffic accident research from the TIRF Annual Report ² is added to the category of injuries.

The ACMC ³ provides additional 1993/94 cost estimates for biomedical research at Canadian faculties of medicine; these are not categorized by diagnostic category. Funding sources include federal and provincial governments, national and provincial non-profit and charitable organizations, local community sources, universities and unaffiliated hospitals, and foreign and miscellaneous sources.

ACMC research expenditures which overlap ⁴ with research expenditures in the Reference List ¹ have been excluded to avoid double counting. Similarly, research by private industry (i.e. pharmaceutical companies) is excluded as research costs would be recovered from product sales.

Assumptions

• The Reference List gives 1993/94 figures for most agencies. A few agencies, because of different fiscal years or other reasons, were unable to provide 1993/94 figures. In these cases, we reported 1992/93 grants and assumed the type and value of research to be similar for the two years.

Limitations

• Although these sources provide a conservative estimate, totals reflect the vast majority of health research in Canada for 1993/94 (see exclusions, Appendix 2).
• The Reference List ¹ was compiled in July 1993 and does not include any subsequent changes or additions.
• We were unable to adjust our estimates to a 1993 expenditure because the 1993/94 Reference List reported a combination of 1992/93, 1993 and 1993/94 expenditures. The Medical Research Council of Canada (MRC) has not produced the 1994/95 Reference List [at the time of writing]; thus, 1993/94 figures for agencies that reported 1992/93 expenditures are not available.
• Totals for diagnostic subcategories may underestimate costs as they reflect only projects identifiable by project title. Diagnostic category totals are more comprehensive.

Results

The 1993/94 national expenditure for Canadian health science research was an estimated $752 million, $512 million of which could be classified by diagnostic category. An additional $239 million was spent on biomedical research that could not be categorized.

The largest share (18.1%) of the $512 million we could classify by diagnostic category was for basic research ($93 million). Cancer, nervous system and sense organ diseases, and cardiovascular diseases accounted for $73, $62 and $60 million, respectively.

A large proportion of the total research expenditure for endocrine and related diseases and for infectious diseases was identified as research on diabetes (35.2%) and HIV/AIDS (25.5%), respectively. Dentistry (31.2%) accounted for almost half the research cost for digestive diseases.

[Figures]

---

Additional Direct Health Expenditures

Methods

The 1993 total national expenditure for care by other professionals and for other direct health expenditures and capital is taken directly from National Health Expenditures in Canada, 1975-1994 (NHEC).¹ (Refer to Appendix 2 for definitions.)

Although expenditures are not available by subcategory of expenditure for 1993, they are available and consistent for previous years.¹ The 1992 percentage distribution of expenditure is used to distribute costs to each 1993 subcategory. Several of these cost components represent either system costs that should not be attributed to specific diagnostic groups (e.g. public health, prepayment adminstration) or areas that could be allocated at aggregate levels (dentists and denturists: digestive diseases; eyeglasses and hearing aids: nervous system and sense organ diseases).

NHEC additional expenditures also include the cost of health science research; we have excluded research costs here and included them in the previous section, Health Science Research Expenditures.

Assumptions

• The 1992 distribution of expenditures for each subcategory ¹ is assumed to reflect the 1993 distribution, based on the minimal variation in this distribution over the last two decades.

Results

Additional direct health costs totalled $17.6 billion in 1993. These additional expenditures are grouped into two major cost components: health care-related ($10.2 billion) and administrative ($7.4 billion).

The health care-related component includes professional health services and medical appliances. Professional services of dentists and denturists cost $4.7 billion, almost 80% of the cost of other professionals. This could be allocated at aggregate levels to digestive diseases. Eyeglasses and hearing aids ($1.4 billion) were allocated to nervous system and sense organ disorders. The remaining health care-related expenditure could not be assigned to any diagnostic category.

Within the administration component, the largest expense was public health ($3.6 billion), followed by capital expenditures for medical facilities ($2.3 billion). Other administrative expenditures were for prepayment administration of insurance coverage and miscellaneous health costs: training for health workers, voluntary health organizations and occupational health expenditures.

[Previous][Table of Contents]