ORIGINAL RESEARCH

The Economic Burden of Schizophrenia
in Canada

Ron Goeree, MA1, Bernie J O’Brien, PhD2, Paula Goering, RN, PhD3, Gordon Blackhouse, MBA4,
Karen Agro, PharmD5, Anne Rhodes, MSc6, Jan Watson, RPN7


Objective: To estimate the financial burden of schizophrenia in Canada in 1996.

Method: Using a prevalence-based approach, all direct health care costs, administrative costs of income assistance plans, and costs of incarceration attributable to schizophrenia were determined. Also included was the value of lost productivity associated with premature mortality and morbidity. In addition to using published papers and documents, direct contact was made with representatives from various provincial and federal programs for estimates of the direct health care and non-health care costs.

Results: The estimated number of persons with schizophrenia in Canada in 1996 was 221 000, with equal distribution between males and females. The direct health care and non-health care cost was estimated to be $1.12 billion in 1996. In addition, another $1.23 billion in lost productivity associated with morbidity and premature mortality was attributable to schizophrenia.

Conclusions: The total financial burden of schizophrenia in Canada was estimated to be $2.35 billion in 1996. The largest category of cost was morbidity (52%), followed by acute care and psychiatric hospital admissions (14% and 10% respectively). Given the magnitude of these cost estimates, there are large potential cost savings with more effective management and control of this debilitating disease.

(Can J Psychiatry 1999;44:464–472)

Key Words: schizophrenia, cost, productivity cost, friction-cost method, Canada

Schizophrenia is too often a debilitating disease, which can have severe mental, physical, and emotional consequences. Because of its early age of onset, it is frequently characterized as a young-adult disease. However, this characterization is misleading. For many, the disease is chronic in nature, having a permanent impact on family and work relations, social interaction, and quality of life in general. For some chronic sufferers, the disease is unrelenting, while for others, the chronic course is represented by periods of remission and acute exacerbation of symptoms. It is important to document the economic impact of this disease on society so that appropriate decisions about health care priorities can be made.

Estimates of schizophrenia prevalence vary by geography and time. The Epidemiologic Catchment Area (ECA) study reported an average 6-month prevalence rate of 9 per 1000 population across 5 cities in the United States (US) (1). A rate of 9.2 per 1000 for males and 9.0 per 1000 for females was also estimated in a recent World Health Organization (WHO) review of the global burden of disease (2). In another extensive review of international studies by the WHO, the age-corrected prevalence rate of schizophrenia, whether measured at a point in time or over a 1-year period, was found to be 5.8 per 1000 population (range 0.9 to 17.4 per 1000) (3). A similar median rate of 5.9 per 1000 population was found by Dohrenwend in a review of US and European studies (4,5). The results from these review articles and increasingly more studies on the prevalence of schizophrenia (1,6–19) suggest that the 6-month to 1-year prevalence is typically between 3 and 12 per 1000 population (8).

Perhaps the most disturbing consequence of schizophrenia is a higher than normal rate of premature mortality. In 1992, schizophrenia was listed as the underlying cause in 1% of all deaths in the US and as a contributing cause in another 13% (20). Although some evidence suggests that schizophrenia is associated with higher rates of accidents and other causes of death, most research has shown suicide as the principal reason for premature mortality. The lifetime risk of suicide in the general population is approximately 0.5% to 1.0% (21,22). According to a Health Canada report on suicide, the relative risk of suicide in persons with schizophrenia is 15 to 25 times that in the general population (22–31).

Although the disease affects only a small percentage of people, the financial burden on patients, family members, and the health care system are substantial and can be overwhelming. Estimates of the financial burden of schizophrenia range from 0.5% to as high as 3.5% of gross national product (32–35). This study estimates, from a societal perspective, the financial burden of schizophrenia in Canada in 1996 using a prevalence-based approach. All direct health care costs, administrative costs of income assistance plans, and incarceration costs attributable to schizophrenia were included in the analysis. Also estimated was the value of lost productivity associated with premature mortality and morbidity for patients with schizophrenia using the friction-cost (FC) method.

Methods

Prevalence Estimates of Schizophrenia

Little data exist on the prevalence of schizophrenia in Canada. Based on the US ECA study and a study in Edmonton, Bland estimated the 6-month (current) prevalence of schizophrenia in Canada to be between 3 and 12 per 1000 population (8). To estimate the number of cases of schizophrenia prevailing in Canada in 1996, we combined the 6-month age- and gender-specific prevalence rates from the US ECA study (1,36) with Canadian population data by age and gender (37).

Premature Mortality

Data on the annual number of deaths due to schizophrenia are reported by Statistics Canada (38). However, this may be an underestimate of mortality because a proportion of deaths coded as “suicide” will be attributable to schizophrenia. A US study found that 10% of suicides were attributable to schizophrenia (39,40). In Canada, a Quebec study found that 6.7% of suicides were attributable to schizophrenia and another 2.7% were due to schizophreniform disorder, for a total of 9.3% (41). In Edmonton, Dyck found that 7.8% of suicide attempts were made by persons diagnosed with schizophrenia (27). In the present study we assumed that 10% of Canadian suicides are attributable to schizophrenia. In a sensitivity analysis, we explored the impact of alternative attribution rates of 5% and 15%.

Health Care Costs

Health care costs attributable to schizophrenia were estimated from several primary and secondary data sources. Health care use and unit costs were collected from national databases or from each individual province for the following resources: acute-care hospitalizations, psychiatric hospitalizations, homes for mentally disabled individuals, seniors’ homes, community mental health services, psychiatric clinics, prescription medicine, and physician visits. Most provinces could provide a breakdown of resource use by mental health diagnosis for all health care services. For provinces that could not provide a breakdown of health care use for patients diagnosed with schizophrenia, the percentage of total resource use for patients with schizophrenia from other provinces was used for allocating expenses.

The number of hospital separations and patient days in acute care hospitals was obtained through secondary analysis of the Canadian Institute for Health Information (CIHI) (42) and Medecho (43) hospital-discharge databases, using schizophrenia as the most responsible diagnosis. The average daily cost of care for schizophrenia in public acute-care hospitals was obtained from a hospital participating in the Ontario Case Costing Project (OCCP) (44).

The number of hospital separations and patient days in provincial psychiatric hospitals for patients with a primary diagnosis of schizophrenia was abstracted from a Statistics Canada publication on mental health hospitalizations (45). To cost these psychiatric hospital admissions, we used published estimates of the average daily cost per patient from 38 psychiatric hospitals in Ontario, Quebec, Alberta, British Columbia, and the Maritimes (46).

Respective departments in each province were contacted by mail and through follow-up telephone interviews for information on use of residential treatment and seniors’ homes by patients with schizophrenia. We apportioned the annual cost of running these facilities to schizophrenia through either the percentage of patient days or the percentage of residents diagnosed with schizophrenia. The method for apportioning costs to schizophrenia varied in each province depending on data availability.

A similar approach was used to collect information on and apportion costs of provincial mental health services and psychiatric freestanding or hospital-based clinics. Each province was contacted for information on the total annual cost of community mental health services and psychiatric clinics. To apportion these annual costs to schizophrenia, information on the number of clients, the number of client visits, and the number of clinic visits for patients diagnosed with schizophrenia was used.

Information on prescription drug use is available at the national level through national physician and pharmacy surveys. We obtained information on drug use for the treatment of schizophrenia from an audit of physician prescribing in Canada (47). The number of scripts written by drug class was combined with costs per script to estimate total prescription drug cost for schizophrenia. The estimated cost per script, by drug class, was based on the cost of the most commonly prescribed drug in the class (47).

In Canada, either physicians and other health professionals are employed directly by outpatient programs, or they bill the provincial governments directly for services. The costs of professional services for physicians employed by hospitals and programs are already included in the operating budgets of the hospital or outpatient program. For physician-claim data to provincial governments, we contacted each provincial insurance plan and requested billing data by age and gender for all patients, for mental health patients, and for patients with schizophrenia.

Administrative Costs of Disability Income-Assistance Plans

Although payments such as employment insurance (EI) are financial consequences of diseases, they are not considered costs from a societal perspective because resources are simply transferred from some members of the economy to others. However, since these transfers are not costless to make, the resource costs associated with administration and delivery of these payments are a legitimate cost of the disease. Some provincial programs like welfare and EI do not collect “reasons” or “diagnoses” for payment from recipients. Similarly, information by reason or diagnosis for Workers’ Compensation Board (WCB) or employer-sponsored sick leave or long-term disability is either not available or difficult to collect at a national level.

For these reasons we collected information regarding the number of recipients with schizophrenia receiving payments from federal and provincial disability income-assistance plans only. We contacted representatives from the Canada Pension Plan (CPP), the Quebec Pension Plan (QPP), and each provincial income-assistance plan for information on the number of recipients with schizophrenia and the administration costs of running these programs. To calculate average per payment administration costs, we requested total administrative costs of the plan for a year and the total number of assistance cheques issued during this same period.

Costs of Incarceration

The rate of crime among persons with severe mental illness, including schizophrenia, is slightly higher than in the general population (48,49). Arguably, the excess crime rate among persons with schizophrenia could be included in a financial assessment of burden. Costs of policing, investigations, legal defence and prosecution, and incarceration and the cost of crime to victims in terms of damages, lost property, and time off work due to physical and emotional injury could all be included. Unfortunately, there are limited data on the frequency of contact with police and the criminal justice system for persons with schizophrenia. As a result, any attempt to allocate costs of law enforcement and justice to schizophrenia would be arbitrary.

For this analysis, we concentrated on the area where some data are available, namely the cost of incarceration in federal and provincial jails and prisons. We reviewed 5 Canadian studies examining the proportion of jail and prison detainees with schizophrenia (50–54). These studies found the prevalence of schizophrenia among detainees to range from 0.3% to 6.3% for males and 0.4% to 13.0% for females. We assumed 5% of inmate days were due to schizophrenia. The number and daily cost of federal and provincial jails and prisons was obtained from a Statistics Canada publication (55).

Productivity Costs Due to Premature Mortality

The traditional approach to the valuation of lost productivity that is foregone due to premature mortality or reduced productivity at work due to morbidity is the so-called human-capital approach. In this framework, the cost of premature mortality due to a disease equals the discounted present value of the projected future stream of production for that person (that is, if they had not died prematurely). This approach implicitly assumes that labour markets are in equilibrium with no unemployment and that if a person leaves the labour force they will not be replaced. Clearly, this is an unrealistic assumption for most industrialized countries, where an excess supply of labour exists.

In an attempt to modify the human-capital approach to allow for worker replacement, Koopmanschap has developed the FC method (56,57). In its simplest form, the FC method adjusts to the human-capital approach by stating that, when a person terminates employment (for example, due to premature mortality from disease) the productivity lost from that job will continue only until a substitute worker fills that vacant position (that is, the friction period). To estimate the productivity cost associated with premature mortality due to schizophrenia, we combined estimates of mortality due to schizophrenia by age and gender (38) with estimates of schizophrenia labour-force participation (17,58,59) and average income data for those employed (60). We assumed a 3-month friction period but also explored the impact on the results of a 1-month friction period in a sensitivity analysis.

Productivity Costs Due to Morbidity

The major difficulty in estimating morbidity costs associated with a disease is in assessing reduced effectiveness at work. In theory, to the extent that a person’s effectiveness at work declines due to disease and other reasons, the wage rates they can command in the marketplace will also decline. Based on this logic, Rice and Miller used a regression model with personal income as the dependent variable to assess the impairment effect of schizophrenia on wages (39). They used these impairment rates as adjustments to wage rates to measure the morbidity of schizophrenia for different age and gender groups. To estimate the productivity cost associated with schizophrenia morbidity in Canada, we used the age–gender impairment rates from this US study (39) and applied them to Canadian patients with schizophrenia using age- and gender-specific average income data for those employed (60). These morbidity cost estimates were further adjusted to account for age- and gender-specific Canadian labour-force participation (17,17,58,59).

The algebraic formulas for calculating productivity costs due to schizophrenia mortality and morbidity are presented in the Appendix.

Results

Prevalence of Schizophrenia in Canada

The estimated 6-month prevalence of schizophrenia in Canada for 1996 is presented in Table 1. The estimated number of persons with schizophrenia in Canada in 1996 is 221 000, with equal distribution between males and females. The age-adjusted 6-month prevalence is estimated to be 9.6 per 1000 for males (95% CI 5.3–13.8) and 9.2 per 1000 for females (95% CI 5.9–12.6). The 95% confidence interval for the 6-month prevalence of schizophrenia in Canada in 1996 is from 132 000 to 312 000.

Table 1. Estimated number of adults with schizophrenia (95% CI) in Canada in 1996, by age and gender

Age-group

Male

Female

Total

15–24 years

23 736 (11 675, 35 797)

20 291 (8 691, 31 891)

44 027 (20 366, 67 688)

25–44 years

57 376 (37 939, 76 812)

75 408 (56 249, 94 567)

132 784 (94 188, 171 380)

45–64 years

27 783 (9 631, 45 934)

11 962 (5 884, 18 041)

39 745 (15 515, 63 975)

65+ years

1 933 (1 933, 1 933)

2 642 (0, 6 670)

4 575 (1 933, 8 603)

Total

110 827 (61 178, 160 477)

110 304 (70 824, 151 169)

221 131 (132 001, 311 646)

Age-adjusted rate per 1000

9.6 (5.3, 13.8)

9.2 (5.9, 12.6)

9.4 (5.6, 13.2)

Sources: Statistics Canada (37) and Regier (1,36).

Direct Costs

The total health care and non-health care direct costs attributable to schizophrenia in Canada are presented in Table 2. There were 25 602 separations in public acute-care hospitals in Canada in 1996, accounting for 681 985 patient days. At an average per diem of $470.94 (44), these acute hospitalizations were estimated to cost $321 million. In addition to these hospitalizations, another 9511 separations represented 809 538 patient days in provincial psychiatric hospitals. At an average per diem of $285.70 (46), these provincial psychiatric hospitalizations were estimated to cost $231 million.

Table 2. Total direct health care and non-health care costs for persons with schizophrenia in Canada in 1996, by resource


Resource

Cost
$, millions

% of total cost (%)

Acute-care hospitals

321.18

28.62

Provincial psychiatric hospitals

231.29

20.61

Seniors’ homes

160.40

14.29

Homes for mentally disabled individuals

98.52

8.78

Community mental health services

73.81

6.58

Psychiatric and general hospital clinics

47.39

4.22

Prescription medications

48.13

4.29

Physician billings

36.90

3.29

Administration of income-assistance plans

34.68

3.09

Incarceration

69.95

6.23

Total

1122.25

100.00

The estimated costs attributable to schizophrenia for homes for mentally disabled individuals, seniors’ homes, community mental health services, and psychiatric and general hospital clinics in 1996 are also presented in Table 2. Of these facilities and programs, seniors’ homes had the highest cost ($160 million), followed by homes for mentally disabled persons ($99 million), community mental health services ($74 million), and psychiatric and general hospital clinics ($47 million).

The Intercontinental Medical Statistics (IMS) audit of physician prescribing in Canada (47) indicated 2.26 million prescriptions written in 1996 across Canada for persons diagnosed with schizophrenia. The most frequently prescribed medications were haloperidol (219 000 scripts), benztropine (203 000 scripts), and risperidone (171 000 scripts). These prescription medicines were estimated to cost $48 million.

Representatives of physician-billing plans in each province indicated that there was $20.69 million in billings for male persons with schizophrenia and another $16.20 million for females in 1996. Total billings attributable to schizophrenia of $36.90 million represented 0.5% of total physician billing in Canada in 1996. Nearly two-thirds of the billing estimate for schizophrenia were for Canadians aged 20–45 years.

In 1996, 12 759 Canadians with schizophrenia received disability income-assistance from federal plans (CPP/QPP). Another 35 399 persons with schizophrenia were on provincial plans in 1996. At an average administrative cost per payment of $60, administrative costs of federal and provincial disability income-assistance plan payments to persons with schizophrenia were estimated to be $35 million in 1996. Although not included in our Canadian schizophrenia cost estimates from a societal cost perspective, total federal and provincial disability payments to these 48 158 recipients with schizophrenia in 1996 was $378.4 million.

In Canada in 1995, 13 000 persons were in federal jails and prisons and another 19 000 in provincial jails and prisons (55). These 32 000 inmates resulted in 12 355 440 inmate days. Assuming that 5% of inmate days are due to schizophrenia (50–54) and using an average cost per inmate day of $121.49 in federal jails and $107.26 in provincial jails (55), the cost of incarceration for persons with schizophrenia was estimated to be $70 million in 1996.

In summary, the total direct health care and non-health care cost for persons with schizophrenia in Canada in 1996 was estimated at $1.12 billion. The largest single category of direct cost was acute care hospitals (29%), followed by provincial psychiatric hospitals (21%), seniors’ homes (14%), and homes for mentally disabled individuals (9%).

Productivity Costs

Employment ratios for the general population and for persons with schizophrenia are presented in Table 3. Also presented in Table 3 are general-population average monthly earnings for those employed (60) and estimates for employed persons with schizophrenia using productivity weights from the Rice and Miller study (39). These employment ratios and average earnings were used for calculating productivity costs due to mortality and morbidity.

Table 3. Employment and earnings data for the general population and persons with schizophrenia in Canada in 1996, by age and gender

 

Percentage in paid employment (%)

   

Average monthly earnings for those employed ($)

 


General populationa


Schizophrenia populationb

Schizophrenia productivity weightsc (%)


General populationd


Schizophrenia populatione

Age-group

Male

Female

Male

Female

Male

Female

Male

Female

Male

Female

15–19 years

0.338

0.356

0.280

0.340

92.86

97.05

434

389

403

377

20–24 years

0.640

0.624

0.469

0.546

92.86

97.05

2015

1676

1871

1627

25–34 years

0.822

0.710

0.756

0.670

92.51

93.87

2936

2366

2716

2221

35–44 years

0.848

0.723

0.785

0.701

78.00

92.00

3661

2674

2856

2460

45–54 years

0.824

0.667

0.775

0.688

55.00

89.00

3930

2756

2161

2453

55–64 years

0.552

0.335

0.529

0.344

22.60

84.70

3664

2462

828

2085

aSource: Statistics Canada “The Labour Force,” 1996 (59).
bSource: Bland (58).
cSource: Rice and Miller (39).
dSource: Statistics Canada “Earnings of Men and Women,” 1995 (60).
eCalculated as the general-population earnings multiplied by the schizophrenia productivity weights.

The number of deaths in 1996 listed as due to schizophrenia along with estimates of suicide deaths due to schizophrenia are presented in Table 4. Including suicide deaths, there were 323 deaths for males and 121 deaths for females attributable to schizophrenia in Canada in 1996. Using a 3-month friction period, these 444 deaths (342 preretirement) were estimated to cost $1.53 million. Because of higher employment rates and average earnings for males, male deaths due to schizophrenia (73% of total) represented 81% of total mortality productivity costs. If a 1-month friction period is used, the total cost estimate decreases to $0.51 million. The assumed proportion of suicides due to schizophrenia also had a large impact on total cost estimates, ranging from $0.74 to $2.23 million for estimates of 5% and 15% respectively.

Table 4. Premature mortality attributable to schizophrenia and the value of production lost in Canada in 1996, by age and gender

 

Deaths from schizophreniaa

Deaths from suicide attributed to schizophreniaa,b

Value of production lost due to premature mortalityc

Age-group

Male

Female

Male

Female

Male

Female

Total

10–14 years

0

0

4

1

—

—

—

15–19 years

0

0

21

5

6.94

1.81

8.75

20–24 years

0

0

30

6

79.25

15.45

94.70

25–29 years

0

0

28

6

174.34

25.44

199.78

30–34 years

0

0

35

8

217.46

34.82

252.28

35–39 years

1

0

39

11

271.03

54.84

325.87

40–44 years

2

1

31

9

217.90

52.25

270.15

45–49 years

1

0

25

8

131.79

40.50

172.29

50–54 years

0

1

20

6

101.11

36.45

137.56

55–59 years

0

0

15

6

19.58

12.91

32.49

60–64 years

2

3

14

3

20.90

13.77

34.67

65+ years

20

38

35

9

—

—

—

Total

26

43

297

78

1240.30

288.25

1528.55

Using 1-month friction period

 

413.43

96.08

509.51

Assuming 5% of suicides attributable to schizophrenia

       

606.23

135.78

742.01

Assuming 15% of suicides attributable to schizophrenia

       

1818.70

407.34

2226.04

aSource: Statistics Canada “Causes of Death,” 1994 (ICD-195) (38).
bAssumes 10% of all suicides (ICD-E950-E959) attributable to schizophrenia (39,40).
cAssumes a 3-month friction period. See Appendix for calculation details.

Based on estimates of the number of persons currently with schizophrenia (Table 1) and employment earnings data for the general population and for persons with schizophrenia (Table 3), the value of lost productivity due to schizophrenia morbidity is presented in Table 5. Although estimates of the number of persons aged 15–64 years with schizophrenia are roughly equally distributed between males and females, the value of lost productivity due to schizophrenia morbidity was substantially higher for males; 82% of $1.23 billion. Nearly one-half of the total morbidity productivity-cost estimate was for 25–44-year-olds. Using the 95% confidence interval for schizophrenia prevalence, the morbidity productivity-cost estimate ranged from $0.64 to $1.81 billion.

Total Costs

The total productivity-cost estimate (mortality and morbidity) for Canada in 1996 was $1.23 billion. When combined with direct costs, the total financial burden of schizophrenia was estimated at $2.35 billion for Canada in 1996. For the base case estimate, direct costs represented 48% of total cost. Since the mortality productivity-cost estimate using the FC method represented only $1.53 million (0.06% of total cost), the overall cost estimate was not very sensitive to either the friction period or the percentage of suicides attributable to schizophrenia. However, the overall cost estimate was sensitive to the assumed prevalence rate of schizophrenia. The total financial burden estimate ranged from $1.17 to $2.94 billion when the lower and upper confidence intervals of prevalence were used.

Table 5. Value of production lost due to morbidity among persons with schizophrenia in Canada in 1996, by age and gender

 

Number of persons with schizophreniaa

Value of lost production due to schizophrenia morbidityb ($, millions)

Age-group

Male

Female

Male

Female

Total

15–19 years

11 743

9927

4.82

1.18

6.00

20–24 years

11 993

10 364

59.30

19.69

78.99

25–34 years

28 802

37 520

123.84

85.93

209.77

35–44 years

28 573

37 888

296.57

94.99

391.56

45–54 years

16 622

7 006

311.69

12.60

324.29

55–64 years

11 160

4 898

212.43

6.35

218.78

Total

108 894

107 662

1008.64

220.74

1229.38

Lower 95% CI of schizophrenia prevalence

   

491.21

153.21

664.43

Upper 95% CI of schizophrenia prevalence

   

1526.06

288.27

1814.33

aSources: Statistics Canada (37) and Regier (1,36).
bSee Appendix for calculation details.

Discussion

Using a 6-month prevalence definition, the age-adjusted prevalence of schizophrenia in Canada was estimated to be 9.4 per 1000 population. This yields an estimated 221 131 persons with schizophrenia in Canada, split almost equally between males and females. In addition, a total of 444 premature deaths (342 preretirement) were estimated to be a result of schizophrenia in 1996. The majority of these deaths (84%) had suicide listed as the primary cause of death.

The financial burden associated with schizophrenia has been estimated for the US (33,39,54), the United Kingdom (35,61), Australia (32), the Netherlands (62), and Puerto Rico (63). There has been no attempt to estimate the financial burden of schizophrenia in Canada. In this study, we combined secondary data sources with primary data collection to estimate costs in Canada. A considerable portion of time and effort was spent tracking down relevant national and provincial government statistics on service use by persons with schizophrenia. Most of the data collected for this study have not been reported previously in the literature and have not been assembled for Canada as a whole.

The direct health care and non-health care cost associated with schizophrenia was estimated to be $1.12 billion in 1996. The largest single category of direct cost was acute care hospitalizations (29%), followed by provincial psychiatric hospitalizations (21%). The 444 premature deaths (342 preretirement) due to schizophrenia were estimated to cost $1.53 million in lost productivity. Another $1.23 billion in lost productivity was estimated for morbidity associated with schizophrenia. Of the $2.35 billion total cost estimate due to schizophrenia in 1996, 48% was direct health care and non-health care cost and 52% the result of mortality and morbidity lost productivity. The schizophrenia burden estimate represented approximately 0.3% of Canadian gross domestic product in 1996 (64).

This estimate for Canada may be lower than that found in studies from some other countries (32–35). These differences may stem from the use of the human-capital approach, rather than FC method, in these studies. If the traditional human-capital approach is used for the 444 premature deaths (342 preretirement) due to schizophrenia, the mortality productivity-loss cost estimate increases 69-fold to $105 million. The total cost estimate of schizophrenia in 1996 would then increase to $2.46 billion.

Despite this fundamental difference in costing methodology, other factors may have contributed to a conservative estimate of the financial burden of schizophrenia in Canada. First, the estimates do not attach a dollar value to the obvious pain and suffering or impact on quality of life for persons with schizophrenia or their families and friends. Nor is there sufficient data to accurately estimate the financial costs to families. Second, because of the social stigma associated with severe mental illnesses such as schizophrenia, there may be some degree of underreporting for some cost categories. For example, since diagnosis is not a mandatory reporting item for physician billing in Canada, the estimated cost of physician billing for schizophrenia may be underestimated. Third, in estimating the cost of provincial psychiatric hospitalizations, patients discharged in 1996 with lengths of stay greater than 365 days were truncated to 1 year for analysis purposes. A more accurate cost estimate would be based on actual patient days for all inpatients with schizophrenia in 1996, rather than only on patients separated during the year. And finally, our morbidity productivity-cost estimates do not include costs for persons not in the labour force. In their estimate of schizophrenia costs in the US, Rice and Miller included housekeeping participation rates and values for housekeeping services in their analysis (39). Given the controversy surrounding the inclusion of household production in costing studies, we elected not to include these costs in our estimates.

Clinical Implications

Limitations

Acknowledgements

The productivity-cost component of this study was funded by the Mental Health Division of Health Canada. The direct-cost component was funded by Zeneca Pharma Inc. Dr O’Brien is supported by a career award in health sciences from the Medical Research Council and Prescription Drug Manufacturers of Canada. The authors thank Christine Henderson and various provincial and federal government representatives for their valuable research assistance.

References

1. Regier DA, Boyd JH, Burke JD Jr, Rae DS, Myers JK, Kramer M, and others. One-month prevalence of mental disorders in the United States. Based on five epidemiologic catchment area sites. Arch Gen Psychiatry 1988;45:977–86.

2. Murray CJL, Lopez AD. The global burden of disese. Harvard School of Public Health; 1996.

3. Warner R, de Girolamo G. Epidemiology of mental disorders and psychosocial problems. Schizophrenia. Geneva: World Health Organization; 1995.

4. Dohrenwend BP, Doll W, Link B, Neugebauer R, Wunsch-Hitzig R. Mental illness in the United States. Epidemiological estimates. New York: Praeger Publishers; 1980.

5. Von Korff M, Nestadt G, Romanoski A, Anthony J, Eaton W, Mercant A, and others. Prevalence of treated and untreated DSM-III Schizophrenia. Results of a two-stage community survey. J Nerv Ment Dis 1985;173:577–81.

6. Bamrah JS, Freeman HL, Goldberg DP. Epidemiology of schizophrenia in Salford, 1974–84. Changes in an urban community over ten years. Br J Psychiatry 1991;159:802–10.

7. Beiser M, Iacono WG. An update on the epidemiology of schizophrenia. Can J Psychiatry 1990;35:657–68.

8. Bland RC. Prevalence of mental illness. Annals RCPSC 1988;21:89–93.

9. Eaton WW. Epidemiology of schizophrenia. Epidemiologic Reviews 1985; 7:105–26.

10. DeSalvia D, Barbato A, Salvo P, Zardo F. Prevalence and incidence of schizophrenic disorders in Portogruaro. An Italian case register study. J Nerv Ment Dis 1993;181:275–82.

11. Halldin J. Prevalence of mental disorder in an urban population in central Sweden. Acta Psychiatr Scand 1984;69:503–18.

12. Walsh D, O’Hare A, Blake B, Halpenny JV, O’Brien PF. The treated prevalence of mental illness in the Republic of Ireland — the three county case register study. Psychol Med 1980;10:465–70.

13. Weissman MM, Myers JK. Psychiatric disorders in a US community. The application of research diagnostic criteria to a resurveyed community sample. Acta Psychiatr Scand 1980;62:99–111.

14. Ryan PM. Epidemiology, etiology, diagnosis, and treatment of schizophrenia. AJHP 1991;48:1271–80.

15. Torrey EF. Prevalence studies in schizophrenia. Br J Psychiatry 1987; 150:598–608.

16. Jablensky A. Editorial. Culture and schizophrenia. Psychol Med 1975;5:113–24.

17. Bland RC, Newman SC, Orn H. Period of prevalence of psychiatric disorders in Edmonton. Acta Psychiatr Scand 1988;77:33–42.

18. Ni Nuallain M, O’Hare A, Walsh D. Incidence of schizophrenia in Ireland. Psychol Med 1987;17:943–8.

19. Roy C, Choudhuri A, Irvine D. The prevalence of mental disorders among Saskatchewan Indians. Journal of Cross-Cultural Psychology 1970;1:383–92.

20. Dembling B. Mental disorder as a contributing cause of death in the U.S. in 1992. Psychiatr Serv 1997;48:45.

21. Roy A. Depression, attempted suicide, and suicide in patients with chronic schizophrenia. Psychiatr Clin North Am 1986;9:193–206.

22. Caldwell CB, Gottesman II. Review article. Schizophrenia —  a high-risk factor for suicide: clues to risk reduction. Suicide Life Threat Behav 1992;22:479–93.

23. Kramer M, Von Korff M, Kessler LG. The lifetime prevalence of mental disorders: estimation, uses and limitations. Psychol Med 1980;10:429–35.

24. Tsuang MT, Woolson RF. Excess mortality in schizophrenia and affective disorders. Do suicides and accidental deaths solely account for this excess? Arch Gen Psychiatry 1978;35:1181–5.

25. Torrey E. Surviving schizophrenia: a family manual. New York: Harper and Row Publishers; 1988.

26. Osmond H, Hoffer A. Schizophrenia and suicide. Journal of Orthomolecular Psychiatry 1978;7:57–67.

27. Dyck RJ, Bland RC, Newman SC, Orn H. Suicide attempts and psychiatric disorders in Edmonton. Acta Psychiatr Scand 1988;77:64–71.

28. Addington DE, Addington JM. Attempted suicide and depression in schizophrenia. Acta Psychiatr Scand 1992;85:288–91.

29. Black DW, Fisher R. Mortality in DSM-IIIR schizophrenia. Schizophr Res 1992;109–16.

30. Allebeck P. Schizophrenia: a life-shortening disease. Schizophr Bull 1989;15:81–9.

31. Mortensen PB, Juel K. Mortality and causes of death in first admitted schizophrenic patients. Br J Psychiatry 1993;163:183–9.

32. Andrews G, Hall W, Goldstein G, Lapsley H, Bartels R, Silove D. The economic costs of schizophrenia. Implications for public policy. Arch Gen Psychiatry 1985;42:537–43.

33. Gunderson JG, Mosher LR. The cost of schizophrenia. Am J Psychiatry 1975;32:901–6.

34. Rupp A, Keith SJ. The costs of schizophrenia: assessing the burden. Schizophrenia 1993;16:413–23.

35. Davies LM, Drummond MF. The economic burden of schizophrenia. Psychiatric Bulletin 1997;14:522–5.

36. Regier DA, Farmer ME, Rae DS, Myers J, Kramer M, Robins L, and others. One-month prevalence of mental disorders in the United States and sociodemographic characteristics: the epidemiologic catchment area study. Acta Psychiatr Scand 1993;88:35–47.

37. Statistics Canada. Population projections for Canada, provinces and territories, 1993–2016. Ottawa: Minister of Supply and Services Canada; 1995.

38. Statistics Canada. Causes of death. Ottawa: Statistics Canada; 1994.

39. Rice DP, Miller LS. The economic burden of schizophrenia: conceptual and methodological issues, and cost estimates. In: Moscarelli M, Rupp A, Sartorius N, editors. West Sussex: John Wiley & Sons; 1996.

40. Goodwin FK. Manic depressive illness. New York: Oxford University Press; 1990.

41. Lesage A, Boyer R, Grunberg F, Vanier C, Morissette R, Menar-Buteau C, and others. Suicide and mental disorders: a case-control study of young men. Am J Psychiatry 1994;151:1063–8.

42. Canadian Institute for Health Information (CIHI). Discharge abstract database (DAD) of public acute hospitalizations. Toronto: CIHI; 1996.

43. Gouvernement du Québec. Medecho database of public acute hospitalizations. Quebec City: Ministère de la Santé et des Services sociaux; 1996.

44. Ontario Case Costing Project. Management information system database, Ontario. Toronto: Ontario Case Costing Project (OCCP); 1996.

45. Statistics Canada. Mental health statistics 1993–94. Ottawa: Minister of Industry; 1993.

46. Michael Loyd & Associates. Psychiatric specialty hospital cost study. Edmonton: Michael Loyd & Associates; 1995.

47. Intercontinental Medical Statistics (IMS) Canada. Canadian Disease and Therapeutic Index (CDTI). Toronto: IMS Canada; 1996.

48. Arboledo-Florez J, Holley HL, Crisanti A, Calgary World Health Organization Collaborating Centre for Research and Training in Mental Health. Mental illness and violence: proof or stereotype? Ottawa: Health Promotion and Programs Branch, Health Canada; 1996.

49. Monahan J, Arnold J. Violence by people with mental illness: a consensus statement by advocates and researchers. Psychiatric Rehabilitation Journal 1996;19:67–70.

50. Blanchette K. The relationship between criminal history, mental disorder and decidivism among federally sentenced female offenders. Carleton University. 1–197. 1996.

51. Motiuk L, Porporino FJ. The prevalence, nature and severity of mental health problems among federal male inmates in Canadian penitentiaries. Ottawa: Correctional Service Canada; 1991.

52. Ontario Ministry of Health. Mental health survey of community correctional clients in Ontario. Toronto: Correctional Services Division; 1994.

53. Wormith JS, Borzecki M. Mental disorder in the criminal justice system. Ottawa: Ministry of the Solicitor General of Canada; 1985.

54. Wyatt RJ, Henter I, Leary MC, Taylor E. An economic evaluation of schizophrenia—1991. [Review]. Soc Psychiatry Psychiatr Epidemiol 1995;30:196–205.

55. Statistics Canada. Adult correctional services in Canada 1994/95. Ottawa: Minister of Industry; 1996.

56. Koopmanschap MA, Rutten FFH, van Ivenveld BM, van Roijen L. The friction cost method for measuring indirect cost of disease. Journal of Health Economics 1995;14:171–89.

57. Koopmanschap MA, Rutten FFH. A practical guide for calculating indirect costs of disease. Pharmacoeconomics 1996;10:460–6.

58. Bland RC, Stebelsky G, Orn H, Newman SC. Psychiatric disorders and unemployment in Edmonton. Acta Psychiatr Scand 1988;77:72–80.

59. Statistics Canada. The labour force. Ottawa: Statistics Canada; 1996.

60. Statistics Canada. Earnings of men and women, 1995. Ottawa: Statistics Canada; 1995.

61. Davies LM, Drummond MF. Economics of schizophrenia: the real cost. Br J Psychiatry 1994;165:18–21.

62. Evers SM, Ament AJ. Economic impact of schizophrenia on the Dutch society. West Sussex: John Wiley & Sons; 1996.

63. Rubio-Stipec M, Stipec B, Canino G. The costs of schizophrenia in Puerto Rico. The Journal of Mental Health Administration 1994;21:136–44.

64. Statistics Canada. CANSIM Matrix 6548 online statistical database. Ottawa, 1997.

Résumé

Objectif : Estimer le fardeau financier de la schizophrénie au Canada en 1996.

Méthode : Au moyen d’une approche fondée sur la prévalence, tous les coûts directs des soins de  santé, les frais administratifs des régimes d’aide au revenu et les coûts d’incarcération attribuable à la schizophrénie ont été établis. On a également inclus la valeur des pertes de productivité associées aux décès prématurés et à la morbidité. Outre le recours aux articles et documents publiés, on a communiqué directement avec des représentants de divers programmes fédéraux et provinciaux pour des estimations des coûts de santé directs et autres.

Résultats : On estimait à 221 000 le nombre de personnes souffrant de schizophrénie au Canada en 1996, également réparties entre hommes et femmes. Les coûts de santé directs et les coûts autres que de santé ont été estimés à 1,12 milliard de dollars en 1996. En outre, on attribuait à la schizophrénie un autre 1,23 milliard de dollars en pertes de productivité associées aux décès prématurés et à la morbidité.

Conclusions : Le fardeau financier total de la schizophrénie au Canada en 1996 a été estimé à 2,35 milliards de dollars. La catégorie la plus importante des coûts était la morbidité (52 %), suivie des soins actifs et des hospitalisations psychiatriques (14 % et 10 % respectivement). Étant donné l’ampleur de ces estimations, on pourrait réaliser des épargnes éventuelles grâce à une gestion plus efficace et à un meilleur contrôle de cette maladie débilitante.

Appendix. The algebraic formulas for estimating the productivity cost associated with premature mortality and morbidity. This appendix is not available online. Please refer to the print copy.

Manuscript received April 1998, revised, and accepted September 1998.

1Research Coordinator, Department of Clinical Epidemiology and Biostatistics, McMaster University; Centre for Evaluation of Medicines, St Joseph’s Hospital, Hamilton, Ontario.

2Associate Professor, Department of Clinical Epidemiology and Biostatistics, McMaster University; Centre for Evaluation of Medicines, St Joseph’s Hospital, Hamilton, Ontario.

3Director, Health Systems Research Unit, Centre for Addiction and Mental Health, Clarke Division, Toronto, Ontario.

4Cost Analyst, Department of Clinical Epidemiology and Biostatistics, McMaster University; Centre for Evaluation of Medicines, St Joseph’s Hospital, Hamilton, Ontario.

5Research Fellow, Centre for Evaluation of Medicines, St Joseph’s Hospital, Hamilton, Ontario.

6Research Scientist, St Michael’s Hospital, Toronto, Ontario.

7Research Assistant, Department of Clinical Epidemiology and Biostatistics, McMaster University; Centre for Evaluation of Medicines, St Joseph’s Hospital, Hamilton, Ontario.

Address for correspondence: R Goeree, Centre for Evaluation of Medicines, St Joseph’s Hospital, 50 Charlton Avenue East, H-304, Hamilton, ON  L8N 4A6

email: goereer@fhs.csu.mcmaster.ca

Can J Psychiatry, Vol 44, June 1999