Rates of Mental Illness and Addiction among High-Cost Users of Medical Services in Ontario

Abstract

Objective:

To quantify the burden of mental illness and addiction among high-costing users of medical services (HCUs) using population-level data from Ontario, and compare to a referent group of nonusers.

Method:

We conducted a population-level cohort study using health administrative data from fiscal year 2011-2012 for all Ontarians with valid health insurance as of April 1, 2011 (N = 10,909,351). Individuals were grouped based on medical costs for hospital, emergency, home, complex continuing, and rehabilitation care in 2011-2012: top 1%, top 2% to 5%, top 6% to 50%, bottom 50%, and a zero-cost nonuser group. The rate of diagnosed psychotic, major mood, and substance use disorders in each group was compared to the zero-cost referent group with adjusted odds ratios (AORs) for age, sex, and socioeconomic status. A sensitivity analysis included anxiety and other disorders.

Results:

Mental illness and addiction rates increased across cost groups affecting 17.0% of the top 1% of users versus 5.7% of the zero-cost group (AOR, 3.70; 95% confidence interval [CI], 3.59 to 3.81). This finding was most pronounced for psychotic disorders (3.7% vs. 0.7%; AOR, 5.07; 95% CI, 4.77 to 5.38) and persisted for mood disorders (10.0% vs. 3.3%; AOR, 3.52; 95% CI, 3.39 to 3.66) and substance use disorders (7.0% vs. 2.3%; AOR, 3.82; 95% CI, 3.66 to 3.99). When anxiety and other disorders were included, the rate of mental illness was 39.3% in the top 1% compared to 21.3% (AOR, 2.39; 95% CI, 2.34 to 2.45).

Conclusions:

A high burden of mental illness and addiction among HCUs warrants its consideration in the design and delivery of services targeting HCUs.

Clinical Implications

• Nearly 1 in 5 of the highest-costing users of medical services in Ontario has a diagnosed serious mental illness or addiction. When anxiety and other disorders are included, this burden rises to more than 1 in 3.

• Mental illness and addiction must be considered in the design and delivery of any programs that target high-cost users of medical services.

• Considering and addressing the prevalence of mental illness in high-cost medical users may have implications for cost savings and health outcomes.

Limitations

• Costs are based on a cross-sectional “snapshot” of medical service utilization.

• Mental illness and addiction diagnosis rates may be underestimated since definitions relied on prior health service utilization.

In most developed countries, health care spending is on the rise.¹ In 2011, Canada spent 11% of its gross domestic product (GDP) on health care, up from 9% in 2000. Similarly, over the same decade, the United States and the United Kingdom experienced a 4% and 2% rise, respectively, in the fraction of GDP being spent on health care.¹ It is being increasingly recognized that a very small proportion of individuals account for a disproportionate amount of health care spending.^2–4 In Ontario (Canada’s most populous province), the top 5% of users account for 60% of expenditures on hospital, home, and emergency health care, the services that drive spending.³ This pattern is not dissimilar to what has been reported in the United States⁴ and Australia.⁵ Ontario allocates over 40% of its program spending to publicly funded health care with evidence that this will continue to expand.⁶ Consequently, and following on recommendations made in other jurisdictions, policy and intervention initiatives in Ontario are shifting to target the highest users of health care resources in an attempt to contain and reduce growing costs.^2–5

Although mental health care is a costly resource, it accounts for a fraction of overall expenditures, which are driven by acute medical and surgical care.³ Mental illness and physical health problems, however, often co-occur with negative implications for health care utilization and outcomes.^7–12 Previous studies have shown that individuals with mental illness comorbidity use hospital and emergency medical services at a high rate and often have higher medical costs compared to those without mental illness.^13–16 To our knowledge, however, the actual rate of mental illness and addiction among high-costing users (HCUs) of medical services has not been quantified on a population level. If HCUs have significant mental illness comorbidity, then this should be accounted for in the design, organization, and implementation of supports and services that aim to improve health outcomes and reduce costs in this user group.

The aim of this study was to examine rates of mental illness and addiction among users of hospital-based, emergency, rehabilitation, and home medical services in Ontario, Canada, across the user cost gradient, with comparison to a referent group of nonusers.

Methods

Design and Data Sources

We conducted a population-based study using health administrative data for all residents of Ontario, Canada, who had a valid Ontario health card and were 18 years of age or older as of April 1, 2011, the index date (N = 10,909,351). The Ontario Health Insurance Plan (OHIP) provides universal coverage for all eligible hospital and home care services, along with physician care. As such, health administrative data are collected on all Ontarians who are eligible for OHIP. These data are available at the Institute for Clinical Evaluative Sciences (ICES), a not-for-profit research institute encompassing a secure and accessible array of Ontario’s health-related data. At ICES, individual-level health care and sociodemographic data from various sources are decoded and linked using an encrypted health care number (ICES Key Number).

Individual demographic data were extracted from the Registered Persons Database, which contains the sex, birth date (to derive age), postal code (to derive neighbourhood income quintile), and death date (if applicable) for all Ontarians with a valid OHIP number. Health care utilization data used to calculate costs of medical service use were extracted from health administrative databases as described in Appendix A. Inpatient service use was retrieved from the Canadian Institutes of Health Information–Discharge Abstract Database (CIHI-DAD); day surgery, emergency room, cancer, and dialysis clinic use were retrieved from the National Ambulatory Care Reporting System (NACRS); continuing care data from the Continuing Care Reporting System (CCRS); rehabilitation data from the National Rehabilitation Reporting System (NRS); and home care data from the Home Care Database (HCD). Mental illness and addiction diagnoses were derived from hospitalization databases (CIHI-DAD and the Ontario Mental Health Reporting System [OMHRS]) and outpatient physician billing data from OHIP. Around 80% of psychiatric hospital admissions in the province are recorded in OMHRS, with the remaining 20% captured in CIHI-DAD. The accuracy and reliability of these databases have been assessed.¹⁷ The overall demographic information in the databases has been shown to be complete and accurate well over 90% of the time.¹⁷ For inpatient hospitalizations, chart abstractions studies have demonstrated that coding for primary diagnoses is reliable.¹⁷ Billing claims for physician services are complete and accurate with respect to broad categories of illness (e.g., psychotic disorders), but there is variability at the diagnostic level (e.g., schizophrenia vs. schizoaffective disorder).¹⁸ This study was approved by the Research Ethics Board at Sunnybrook Health Sciences Centre (ICES logged study: 2014 0904 318 000).

Medical care cost groups

For each individual, we calculated the cost of his or her non–mental health, medical service use for the 2011-2012 fiscal year (1 year from the index date). For this calculation, we included 1) acute inpatient hospitalization costs and same-day surgeries, except for those with a primary diagnosis of mental health (International Classification of Diseases, 10th Revision, Canada [ICD-10-CA], F00-F99) and obstetrical deliveries; 2) emergency department visits, except for those with a primary diagnosis of mental health (ICD-10-CA, F00-F99) or evidence of a self-harm or suicide attempt in any diagnostic field (ICD-10-CA, X60-X84, Y10-Y19, Y28); 3) dialysis and cancer clinic visits; 4) inpatient rehabilitation; 5) complex continuing care hospitalizations; and 6) home care visits. Among the highest-costing users in Ontario, these medical services account 80% of health spending and drive pressures to shift care to the community setting where possible.¹⁹ Hospital and home care alone account for over 40% of the entire Ontario Ministry of Health and Long Term Care budget.⁶ So, in keeping with previous research focused on high-cost users, we did not include costs related to the use of outpatient physician services.³ Cost calculations took into account both consumption of resources and price, standardized to the 2011 fiscal year, using a costing approach developed at ICES.²⁰ This approach involves generating a cost/unit of service (e.g., 1 inpatient day) and multiplying by the amount of utilization (e.g., length of inpatient admission). Because not all patients are equal in terms of the severity and complexity of their needs, and accordingly the intensity of health care resources they require, a resource intensity weight (RIW) is assigned to all inpatient hospitalizations, same-day surgeries, emergency department, outpatient dialysis and cancer clinics, and inpatient rehabilitation. The RIW uses a case mix method to estimate hospital resource use for a person with a given condition relative to the average resources used by other persons.²⁰ Additional weights are applied in the calculation of continuing care to account for the fact that a patient’s resource intensity changes over time.²⁰ For the HCD, costs are calculated using an hourly rate based on allocated funds for nonphysician care (i.e., nursing) and OHIP physician-billing for care received in the home.²⁰

All individuals with medical care costs greater than zero were ranked by cost and clustered into percentiles (1-100) by cost rank with each percentile group containing approximately the same number of individuals. Percentiles were further clustered into the top 1% of users, the top 2% to 5% of users, the top 6% to 50% of users, and the bottom 50% of users. This classification is based on previous work that has shown that costs are relatively stable in the low user categories and then rise exponentially in the top 5% and 1% of health care system users.¹⁷ A zero-cost group (individuals with no medical service costs during the observation period) was also created and served as the referent group in all analyses.

Mental illness and addiction

For each individual, we determined whether diagnoses for any of the following categories of conditions had been recorded in the health care databases within the 2 years prior to the index date (i.e., prior to April 1, 2011) (see Appendix B for detailed list of diagnoses and data codes): 1) psychotic disorders, 2) major mood disorders, and 3) substance use disorders, excluding nicotine dependence. We focused on these most severe and persistent mental illnesses because these are the illnesses that are most likely to require intensive psychiatric intervention and to impact most on medical health outcomes.⁷ In addition, we conducted a sensitivity analysis with an overall category that included anxiety and other disorders (see Appendix B for the complete list of included disorders) that may affect health care utilization but tend to be less rigorously defined and coded in administrative data to determine if the same pattern held.

Statistical Analysis

Mean and total costs were calculated for each cost category and descriptive variables (age, sex, and socioeconomic status approximated by neighbourhood income quintile) were compared across cost groups. We calculated the proportion of individuals in each cost category with any diagnosed major mental illness or addiction and by diagnostic subgroups (psychotic disorders, major mood disorders, and substance use disorders), presenting the results as unadjusted odds ratios (ORs) with 95% confidence intervals (CIs), and adjusted (AOR) for age, sex, and neighbourhood income quintile. For our sensitivity analysis, we report the same statistics for individuals with any mental illness, expanded to include anxiety and other disorders. All statistical analyses were conducted using SAS Statistical software for UNIX (version 9.3; SAS Institute, Cary, NC, USA).

Results

There were 10,909,351 eligible individuals in Ontario as of April 1, 2011. Most individuals did not use any of the medical services examined during the observation period and comprised the zero-cost referent group (n = 7,759,791, 71%). The remaining 3,149,560 individuals were clustered into cost groups based on their percentile rank in the population. The top 5% of medical service users (representing 1.4% of the total population of Ontarians) accounted for 56.1% of the health care costs ($8.8 billion) during the observation period, with nearly half of that attributed to the top 1% only ($3.8 billion) (see Table 1). The top 1% and top 2% to 5% of users were older than those in the other groups, and the top 1% of users were more often in the lowest neighbourhood income quintile compared to the other groups. The average cost per user was more than 3 times greater for the top 1% group compared to the top 2% to 5% group and exponentially higher than the other lower cost groups (see Figure 1).

Table 1.

Characteristics of 10,909,351 Ontarians by Cost Group.

	Cost Group
	Top 1% (n = 31,494)	Top 2%-5% (n = 125,983)	Top 6%-50% (n = 1,417,081)	Bottom 50% (n = 1,575,002)	Zero Cost (n = 7,759,791)
% of total population	0.3	1.2	13.0	14.4	71.1
Total medical cost, $CAD	3 848 523 722	4 966 028 444	6 393 481 164	496 747 304	0
% of overall costsa	24.5	31.6	40.7	3.2	0
Cost per user, $CAD, mean ± SD	122 198.6 ± 67 513.8	39 418.2 ± 12 776.1	4511.7 ± 4828.5	315.4 ± 192.1	0 ± 0
Age, y, mean ± SD	68.5 ± 18.1	69.2 ± 18.1	57.4 ± 30.4	45.2 ± 0.0	45.3 ± 0
Female sex, n (%)	14 784 (46.9)	64 573 (51.3)	805 570 (56.8)	827 136 (52.5)	3 872 465 (49.9)
Income Q1,b n (%)	7637 (24.2)	27 239 (21.6)	285 990 (20.2)	325 239 (20.7)	1 451 105 (18.7)

^aCalculated as proportion of total population’s hospital, emergency, home, continuing care, cancer and dialysis clinic visits, and rehabilitation costs for medical reasons only.

^bQ1 = Lowest neighbourhood income quintile.

Figure 1.

Average medical service costs per user and unadjusted rate of mental illness or addiction, by cost group.

The rate of major mental illness or addiction increased across cost groups, being highest in the top 1% group at 17.0% and lowest at 5.7% in the zero-cost referent group (see Figure 1). Unadjusted and adjusted odds ratios for all groups compared to the zero-cost referent group were statistically significant, with the odds of mental illness comorbidity for the top 1% group being almost 4 times that of the zero-cost referent group (AOR, 3.70; 95% CI, 3.59 to 3.81) (see Figure 2). Each mental illness diagnostic category followed a similar pattern (see Table 2). When we included anxiety and other disorders, the rate of diagnosed mental illness rose to 39.3% in the top 1% compared to 21.3% in the zero-cost referent group. The same pattern of odds ratios existed, but the magnitude of the difference was attenuated (see Table 2).

Figure 2.

Adjusted odds ratios for risk of mental illness or addiction by cost, relative to the zero-cost group.

Table 2.

Odds Ratios (OR) for Mental Illness Diagnostic Categories by Cost, Relative to the Zero-Cost Group.

	n (%)	Unadjusted OR (95% CI)	Adjusted ORa (95% CI)
Psychotic disorders
Zero cost (referent)	54 751 (0.7)	1.00	1.00
Bottom 50%	18 563 (1.2)	1.68 (1.65 to 1.71)	1.66 (1.63 to 1.69)
Top 6% to 50%	25 805 (1.8)	2.61 (2.57 to 2.65)	2.63 (2.59 to 2.68)
Top 2% to 5%	2970 (2.4)	3.40 (3.27 to 3.53)	3.36 (3.23 to 3.49)
Top 1%	1150 (3.7)	5.33 (5.03 to 5.66)	5.07 (4.77 to 5.38)
Major mood disorders
Zero cost (referent)	256 781 (3.3)	1.00	1.00
Bottom 50%	91 176 (5.8)	1.80 (1.78 to 1.81)	1.77 (1.75 to 1.78)
Top 6% to 50%	104 905 (7.4)	2.34 (2.32 to 2.35)	2.36 (2.34 to 2.38)
Top 2% to 5%	9648 (7.7)	2.42 (2.37 to 2.48)	2.61 (2.56 to 2.67)
Top 1%	3138 (10.0)	3.24 (3.12 to 3.36)	3.52 (3.39 to 3.66)
Substance use disorders
Zero cost (referent)	182 092 (2.3)	1.00	1.00
Bottom 50%	67 497 (4.3)	1.86 (1.85 to 1.88)	1.86 (1.84 to 1.88)
Top 6% to 50%	66 839 (4.7)	2.06 (2.04 to 2.08)	2.39 (2.37 to 2.41)
Top 2% to 5%	6842 (5.4)	2.39 (2.33 to 2.45)	3.03 (2.96 to 3.11)
Top 1%	2213 (7.0)	3.15 (3.01 to 3.28)	3.82 (3.66 to 3.99)
Any mental illness or addictionb
Zero cost (referent)	1 654 186 (21.3)	1.00	1.00
Bottom 50%	488 512 (31.0)	1.66 (1.65 to 1.67)	1.64 (1.64 to 1.65)
Top 6% to 50%	501 839 (35.4)	2.02 (2.02 to 2.03)	1.97 (1.96 to 1.98)
Top 2% to 5%	44 653 (35.4)	2.03 (2.00 to 2.05)	2.00 (1.97 to 2.02)
Top 1%	12 385 (39.3)	2.39 (2.34 to 2.45)	2.39 (2.34 to 2.45)

CI, confidence interval; OR, odds ratio.

^aAdjusted for age, sex, and neighbourhood income quintile.

^bSensitivity analysis with diagnostic categories expanded to include anxiety and other disorders in addition to the 3 diagnostic categories above.

Discussion

Seventeen percent of the most costly users of hospital-based, emergency, home, and rehabilitation medical services in Ontario has a prior diagnosis of a psychotic, major mood, or substance use disorder. When anxiety and other disorders are included, nearly 40% of the top HCUs are affected. This is nearly a 4-fold increased rate of major mental illness or addiction among HCUs compared to nonusers. Furthermore, the rate of mental illness and addiction rises incrementally across increasing user cost categories. While the pattern is similar regardless of mental health diagnosis, the findings are most pronounced for psychotic disorders.

A high rate of mental illness and addiction among users of high-cost medical services compared to nonusers is supported by existing literature demonstrating several mechanisms through which medical service utilization and costs are increased.^7,8,11,12 The presence of depression has been associated with higher use of medical services, including inpatient care (both occurrence and duration of stay) and outpatient visits, even after adjusting for medical disease severity.^14,21,22 Underrecognition of mental illness and its impact during medical admissions,²³ along with variables such as increased pain²⁴ and multimorbidity,¹⁴ have been cited as possible factors mediating the higher service use and costs among individuals with depression. In a sample of disabled adults receiving mandatory Medicaid managed care in the United States who had been hospitalized,¹² 70% of those identified to be at the highest risk for all-cause readmission had serious mental illness (major mood or psychotic disorder) or substance disorder comorbidity. Despite this, 70% of the readmissions within 1 year were for medical reasons. Similarly, a recent study from the United States found that individuals with serious mental illness were more likely to have 30-day readmissions for diagnoses of pneumonia, heart failure, and myocardial infarction.²⁵ Consistent with these studies, a report from the United States showed that having a severe persistent mental health diagnosis is associated with increased total monthly health care spending, with the largest increase in costs for medical rather than behavioural (i.e., psychiatric) service use.²⁶

A key finding from our study was the observation that individuals in the top 1% of medical service users were over 5 times more likely to have had a recorded diagnosis of a psychotic disorder in the preceding 2 years compared to nonusers of medical services. Due to a number of illness and system-level factors, preventative and maintenance medical care among individuals with psychosis is often suboptimal.⁸ Therefore, they often present with more advanced disease,⁸ are more likely to have other illnesses that are poorly controlled, and are at a substantially higher risk of medical and surgical complications during hospital admissions.²⁷ The prevalence of a psychotic disorder in the top costing percentile of medical care users was 3.2%, which is substantially higher than what has been observed in Canadian estimates of the population prevalence of psychosis (1.1%).²⁸ This suggests that services and supports should be designed appropriately to account for the disproportionate number of high medical users with psychosis and to manage the complexity of medical and severe psychiatric comorbidity.

To our knowledge, no studies have actually quantified the prevalence of mental illness among high-cost medical care users; as such, this is the unique contribution of our study. Moreover, in contrast to other studies that have examined the prevalence of mental illness in only a single user group (e.g., medical inpatients²² or high-risk disabled adults¹²), we examined users of medical services on a population level. Ontario’s universal health insurance coverage and the availability of health administrative data from multiple services permitted the linkage of population data. We also used a person-centred costing algorithm based on weighted estimates to calculate individual health care expenditures, and thus the prevalence of mental illness and addiction could be examined in relationship to actual health care spending.

It is important to note that we could be underestimating the prevalence of mental illness and addiction because our definition of mental illness relies on a hospital or physician visit over the 2-year period examined and does not include outpatient visits with non-OHIP covered practitioners (e.g., nurse practitioners, social workers, psychologists) or the small number of physicians operating in alternative settings or under alternative payment models who do not bill OHIP. This would include community health centres, which serve less than 1% of Ontario’s population, and some primary care practices that operate through capitation payment models (although many visits would be captured through shadow billing as physicians receive partial payment or other incentives).²⁹ It is possible that this limitation might affect the zero-cost group more (i.e., zero-cost users are less likely to use any health care services), but this difference in underrecognition would have to be greater than 3 times to nullify our findings. Our results are also limited to the cross-sectional “snapshot” of service users in 1 fiscal year and do not reflect the total medical costs for individuals with and without mental illness or addiction over time. Finally, the findings of this study are based on Ontario data and may not be completely generalisable to other health care systems, although data from the United States suggest similar trends.^4,12

Conclusions

Nearly 1 in 5 of HCUs in Ontario has a diagnosed major mood, psychotic, or substance use disorder. When anxiety and other disorders are included, this burden rises to more than 1 in 3. Effective interventions will likely require a comprehensive multidisciplinary approach that attends to the social, medical, and psychiatric needs of these complex patients. Considering and addressing mental illness and addiction in HCUs may result in cost savings and improved health outcomes.

Appendix

Appendix A.

Detailed Description of Data Sources.

Data Source	Abbreviation	Description of Dataa	Data Extracted	Yearsb
Registered Persons Database	RPD	Sex, birth date, postal code, and death date (if applicable) for all Ontarians with valid health insurance	Sex Birth date (to derive age) Postal code (to derive neighbourhood income quintile)	2011-2012
Canadian Institutes of Health Information–Discharge Abstract Database	CIHI-DAD	Demographic, administrative, and clinical data for all acute care inpatient admissionsc	Primary diagnosis Health care utilization and diagnostic data to calculate medical costs	2009-2012 2011-2012
National Ambulatory Care Reporting System	NACRS	Demographic, administrative, and clinical data for emergency room visits, outpatient dialysis and cancer clinics, and day surgeries	Primary diagnosis Health care utilization and diagnostic data to calculate medical costs	2009-2012 2011-2012
Continuing Care Reporting System	CCRS	Clinical data from centres offering continuing care. Continuing care is nonacute care provided to patients who are not ready for discharge. Services are provided in a freestanding facility or co-located in an acute care or rehabilitation program.	Primary diagnosis Health care utilization and diagnostic data to calculate medical costs	2009-2012 2011-2012
National Rehabilitation Reporting System	NRS	Clinical data from inpatient rehabilitation facilities	Primary diagnosis Health care utilization and diagnostic data to calculate medical costs	2009-2012 2011-2012
Home Care Database	HCD	Clinical, administrative, and resource utilization data for all publicly funded home care services (e.g., nursing, occupational therapy)	Primary diagnosis Health care utilization and diagnostic data to calculate medical costs	2009-2012 2011-2012
Ontario Health Insurance Plan	OHIP	Data from all physician billings in Ontario, including diagnosis and care provided	Mental health diagnosis	2009-2011
Ontario Mental Health Reporting System	OMHRS	Demographic, clinical and administrative data on inpatient mental health admissionsa	Mental health diagnosis	2009-2011

^aAll data source descriptions were obtained from the Ontario Population Health Index of Databases, which can be accessed at http://ophid.scholarsportal.info/.

^bAll years reported refer to fiscal years starting April 1 and ending March 31.

^cApproximately 80% of mental health admissions are captured in OMHRS. The other 20% are captured in CIHI-DAD and represent admissions in hospitals that do not have a dedicated mental health inpatient unit or when beds on other units are occupied because of space constraints.

Appendix B.

Detailed List of Diagnoses and Data Codes Used to Define Mental Illness and Addiction.

Mental Illness Category	Diagnoses Included	Codes
Major mood disorders	Depressive or manic episodes Recurrent depressive disorder Bipolar affective disorder Depressive disorder not otherwise specified	CIHI-DAD ICD-10-CA: F30-34, F38-39 or OHIP: 296, 311 or OMHRS diagnosis
Substance use disorders	Any of intoxication, withdrawal, psychosis, abuse, dependence, or other mental and behavioural disorder due to: Alcohol Opioids Cannabinoids Sedatives or hypnotics Cocaine Stimulants Hallucinogens Volatile solvents Other psychoactive substances	CIHI-DAD ICD-10-CA: F10-F16, F18-19 or OHIP: 291, 292, 303, 304, 305 or OMHRS diagnosis
Psychotic disorder	Schizophrenia Schizotypal disorder Persistent delusional disorders Transient psychotic disorders Schizoaffective disorder Psychotic disorder not otherwise specified	CIHI-DAD ICD-10-CA: F20, F21, F22, F23, F24, F25, F28, F29 or OHIP: 295, 297, 298 or OMHRS diagnosis
Anxiety and other disorders	Anxiety disorders Adjustment disorders Somatoform disorders Dissociative disorders Psychosomatic disorders Sexual disorders Personality disorders Disorders of conduct and impulsivity Sleep disorders Eating disorders Other nonspecific psychological and behavioural disorders	CIHI-DAD ICD-10-CA: F40-F42, F43.0, F43.1, F43.2, F43.8, F43.9, F44-F45, F50-F55, F59, F60-62, F63-66, F68-F69, F90-F95, F98 or OHIP: 300, 301, 302, 306, 307, 308, 309, 312 or OMHRS diagnosis

CIHI-DAD, Canadian Institutes of Health Information–Discharge Abstract Database; ICD-10-CA, International Classification of Diseases, 10th Revision, Canada; OHIP, Ontario Health Insurance Plan; OMHRS, Ontario Mental Health Reporting System.