Publications
Health Research Working Paper Series
Mental comorbidity and increased use of acute care hospital services
Introduction, findings, and conclusions
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
Introduction
About one in five Canadians have suffered from a mental condition at some point in their lives.1,2,3 About 14% of the global burden of disease has been attributed to neuropsychiatric conditions, mostly because of the chronically disabling nature of depression and other common mental conditions, alcohol-use and substance-use conditions, and psychoses.4 Studies by the World Bank and World Health Organization suggest that, among non-communicable diseases, depression is estimated to be the third highest reason for healthy years of life lost for women and the fifth highest cause for men.5 Organizations around the world have identified mental health as a priority area.6,7
Like other health conditions, mental conditions represent an economic burden to society with costs often comparable to physical conditions such as heart disease.8,9 Expenditures on mental conditions and addictions for Canadian provinces in 2003/2004 were $6.6 billion, of which $5.5 billion was from public sources.10 Nationally, the largest portion of expenditures was for hospitals, followed by community mental health expenses and pharmaceuticals. Even with these amounts, Canadian public spending on mental health was found to be lower than in most developed countries.10
There also appears to be an inadequate appreciation of the burden of mental conditions on physical well-being and that there can be "no health without mental health." 4 Hence, understanding the burden of mental conditions as a comorbid condition among those with physical morbidities is important. Many major psychiatric conditions are observed with physical conditions.11 The prevalence of several chronic physical conditions, namely, chronic pain, diabetes, cardiovascular disease, high blood pressure and respiratory conditions is higher among people with mental conditions.12,13,14 Reasons for the excess rates of mental conditions among those with physical morbidities are diverse, and not fully understood. Many physical health conditions increase the risk for poor mental health, and mental comorbidity often complicates help-seeking, diagnosis and treatment, which may influence prognosis.4
Regardless of the causal pathway, the combination of physical and mental conditions often results in higher rates of healthcare use. Higher rates of emergency room and outpatient services and pharmaceuticals expenditures, for example, were found among chronically ill individuals with mental conditions compared with those with physical morbidities alone.15 Specifically, among diabetics, higher health care expenditures were associated with mental conditions, including substance use disorders.16 Those with mental conditions may also be at risk because of drug interactions between general medical drugs and psychotropics.17
A recent review of mental health services in Canada conducted by the Standing Senate Committee on Social Affairs, Science and Technology concluded that Canada currently has no national picture of the status of mental health across the country. It concluded that more information is necessary to determine the extent of the problem and to plan and implement necessary services and programs.18,19 The Public Health Agency of Canada has voiced similar concerns, stating that the profile of mental conditions is very limited in Canada, and that there is a need to complement available hospitalization data with additional data, to provide a more comprehensive understanding of the patients currently accessing the healthcare system for mental health services and of the comorbidity of mental conditions with other illnesses.20,21
There has been an increase in population-based studies of mental health at the national22,23,24,25,26 and provincial level.27,28,29 However, significant gaps remain in our understanding of the extent of the burden and of the patient profile of individuals admitted to an acute-care hospital with a diagnosis of a mental condition. For example, most of these reports regarding hospitalizations focus on mental conditions as a most responsible diagnosis and provide limited information regarding the burden of mental conditions as comorbidities. Furthermore, much of the administrative health data provide limited information regarding the characteristics of individuals admitted to hospital with a diagnosis of a mental condition. This information is critical to our understanding of the burden of service use placed by these individuals.
This report builds upon the emerging picture of the burden of mental conditions using a range of data sources available to Statistics Canada. Part 1 of the report focuses on the burden of mental conditions in acute-care hospitals using administrative hospital data. We consider a more expansive role of mental health by looking at both hospital admissions where a mental condition is reported as the most responsible diagnosis as well as those admissions where a mental condition is reported as a comorbid diagnosis. This enables a more comprehensive assessment of the impact of mental conditions on hospital resource use. The number of days and the average length of stay, rather than the number of hospitalizations, are presented to look more closely at use issues.
Part 2 of the report focuses on understanding the characteristics of patients admitted to hospital with a mental condition, reported as most responsible or comorbid diagnosis. We use linked data to identify patient characteristics associated with hospitalization with a diagnosis of a mental condition. This information provides some insights into other patient factors that may be contributing to the increased use of resources.
Findings
Part 1: The burden of use of hospital services with a mental diagnosis
Hospital services for those with a diagnosis of a mental condition
Rates of acute-care hospitalizations with a diagnosis of a mental condition
In 2003/2004, the total number of hospitalizations with a most responsible diagnosis of a mental condition was 155,000, representing 5.7% of all hospitalizations in that year. The number of hospitalizations with mental conditions listed as a comorbid diagnosis totalled 228,000, representing 8.4% of all hospitalizations. Looking at the percentage of hospital days, 12.4% (2,471,000) were associated with hospitalizations reporting a most responsible (MR) mental diagnosis, and 16.6% (3,306,000) with hospitalizations reporting a comorbid mental diagnosis.
In Canada, those who experience a hospitalization with a diagnosis of a mental condition make up a relatively small proportion of the overall population but use a disproportionately large share of hospital services. In 2003/2004, approximately 291,000 people, 0.9% of all Canadians, were discharged at least once from an acute-care hospital with either a most responsible or comorbid mental diagnosis. These individuals, however, represented 14.7% of all individuals who were discharged from hospital during that year.
Individuals with a mental condition used a disproportionate amount of hospital resources: 14.1% of all hospitalizations and 29.0% of hospital days (5.8 million hospital days) (Figure 1).
To obtain a comprehensive assessment of the burden of mental conditions on acute-care hospitalizations, we clearly need to consider the role of mental conditions both as a most responsible and as a comorbid diagnosis.
Findings by age and sex
Distributions by age and sex tell us more about who is using hospital days associated with diagnoses of a mental condition (figures 2a and 2b). While women use more hospital days overall (55.1%), the percentages within each sex associated with most responsible or comorbid diagnoses of a mental condition were similar (most responsible 12.1% for males vs. 12.7% females; comorbid 17.1% for males vs. 16.2% females). The overall pattern across ages is similar for males and females (Figure 2a).
As expected, the number of days associated with comorbid mental diagnosis increases with age. The number of most responsible days in hospital associated with a mental diagnosis was spread out across all ages and had a slight increase in the thirties and forties. The percentage of most responsible days in hospital associated with a mental diagnosis decreases with age (the dip in female percentages is due to childbirth). For young males, over 30% of their hospital days were associated with a most responsible mental diagnosis. The percentage of days in hospital associated with a comorbid mental diagnosis increases with age for both sexes (Figure 2b).
The average length of stay for hospitalizations with and without a mental diagnosis provides an indication of the intensity of hospital service use. As expected, the average length of stay of hospitalizations with a mental diagnosis (either most responsible or comorbid) is greater than for those hospitalizations with no mental diagnosis (Figure 3). The average length of stay is for hospitalizations with a most-responsible mental diagnosis is 15.9 days; for a comorbid mental-condition diagnosis, 14.5 days, and for no mental diagnosis, 6.1 days (Figure 3).
Figure 3 Average length of stay, by presence of mental diagnosis and age group, Canada, 2003/2004
Within each age group, the average length of stay was often more than three times as long for hospitalizations with a most responsible mental diagnosis compared with hospitalizations where no diagnoses of a mental condition was reported. The average length of stay for hospitalizations with a comorbid mental diagnosis was in between, but still more than twice as long as the average for hospitalizations with no diagnoses of a mental condition (ratio=2.3) for all ages. For example, among those aged 60 to 69 years, the average length of stay was 20.6 days for hospitalizations with a most responsible mental diagnosis, 14.2 days for a comorbid mental diagnosis and only 7.1 days for hospitalizations with no mention of a mental diagnosis.
Findings by type of mental diagnosis
It is also informative to look at the burden of hospitalization associated with each type of mental condition by various age groups (Figure 4). Based on these results, the most prevalent comorbid types of mental conditions were organic (dementia, delirium), substance-related and mood conditions. As expected, organic mental comorbidity (dementia, delirium) was most prevalent among those over 70 years of age. Substance abuse was prevalent over a wide range of ages, mostly as a mental comorbidity. Mood conditions first appear as a most responsible diagnosis at young ages and then shift to comorbid conditions as age increases. Schizophrenic and psychotic conditions occur most often as a most responsible diagnosis for those aged 20 to 60 years. Anxiety and personality conditions were found as mostly comorbid conditions across most ages.
Findings by province and health region
Hospitalizations with a mental diagnosis, by province and health region
The burden of mental conditions on acute-care hospitalizations varied across provinces (Figure 5). The proportion of hospitalizations with a most responsible mental condition ranged from 3.7% in Newfoundland and Labrador (52,580 hospitalizations) to 8.8% in Prince Edward Island (15,912 hospitalizations). The proportion of hospitalizations with comorbid mental conditions varied from less than 3.1% in Newfoundland and Labrador to a high of 15.1% in Quebec.
Variations across the provinces may reflect possible differences in either the prevalence of mental conditions within the population or in acute-care-hospital treatment practices. However, the variations may also reflect differences in coding practices specifically related to comorbidity. The average number of diagnoses per patient per hospitalization documented in the hospital data generally varies greatly from province to province: Quebec reported the highest average number of comorbid diagnoses per hospitalization; Newfoundland and Labrador and Prince Edward Island reported the lowest.39
Health care planners would find it useful to consider the burden of mental conditions in smaller areas. In Figure 6, each square represents a health region. These results reveal considerable variation across health regions in hospitalizations related to mental conditions. For example, in Alberta health regions, the percentage of hospitalizations with a mental comorbidity ranges from a low of 5.0% to a high of 12.9%. Like differences across provinces, differences across health regions may be attributable to differences in underlying health states, treatment variation or even coding practices, although within a province coding differences are expected to be less of a factor. Full health region data can be found in Appendix Table B.
Hospital days associated with a mental diagnosis, by province and health region
The use of hospital days provides a proxy measure of resource use. As expected, this also varies across provinces (Figure 7). The percentage of total hospital days associated with hospitalizations with amost responsible mental diagnosis ranges from 7.3% in Saskatchewan to 12.9% in Quebec. The percentage of days associated with hospitalizations with a comorbid mental diagnosis varied from 7.1% in Newfoundland and Labrador to 25.9% in Quebec.
The results reveal considerable variation in the use of hospital days across health regions, even within provinces (Figure 8). In Alberta, for example, the percentage of hospital days with a mental comorbidity ranges from a low of 8.9% to a high of 24.9%. To identify health regions, see Appendix Table B for health region data.
We investigated the percentage of hospital days by type of mental diagnosis to see their separate impacts on hospital utilization across provinces. Table 2 shows that higher numbers of days in hospital were associated with a most responsible diagnosis of a mood, psychotic and organic (dementia, delirium) condition. Most days in hospital associated with a comorbid mental diagnosis were for organic, substance and mood mental conditions (dementia, delirium). More that one mental condition was reported in some hospitalizations, so the total is lower than the sum. Anxiety and personality diagnoses have consistently low percentages of days across the provinces. The percentage of hospital days associated with comorbid organic diagnoses range from 2.4% in the Northwest Territories to 9.6% in Alberta. The percentage of days associated with the mood and psychotic diagnoses vary across the provinces, while those for substance abuse are higher in the territories.
Table 2 Percentage of hospital days with mental most responsible or comorbid diagnosis, by province and mental disorder, all ages, Canada, 2003/2004Length of stay of hospitalizations with a diagnosis of a mental condition, by province and health region
While the number of days is a proxy for the burden associated with diagnoses of a mental condition on the acute-care hospital system as a whole, the average number of days per hospitalization provides an indication of the intensity of hospital service use by patients. Comorbid diagnoses of a mental condition appear to have a significant impact on the use of hospital services during a hospitalization. The average length of stay with a comorbid mental diagnosis is two times the length of stay without any diagnosis of a mental condition. This finding is fairly consistent across the provinces. The ratio of the average number of days per discharge with a mental comorbidityto the average number of days of hospitalizations without a mental comorbidityranges from 1.9 in Saskatchewan to 3.0 in Newfoundland and Labrador. (Figure 9, horizontal line)
Similarly, there were variations among health regions within a province in the ratio of average number of hospital days per hospitalization with a mental comorbidity to average number of hospital days per hospitalization without a mental comorbidity (Figure 10). In British Columbia for example, the ratio varies from 1.6 to more than 3.2. To identify health regions, see Appendix Table B for health region data.
Overall burden of mental conditions as a comorbid diagnosis, by major disease type
To better understand the burden of mental conditions as a comorbid condition, we looked at differences in the length of hospitalizations with and without a comorbid mental diagnosis, by major disease types. The major disease types were identified by the most responsible diagnoses for each hospitalization as defined by the ICD-10-CA chapters (Table 3). As no good crossovers exist between ICD-9 and ICD-10-CA chapters, this analysis excludes Manitoba and Quebec, as hospitalizations in these provinces were coded in ICD-9 and ICD-9-CM in 2003/2004.
We present the percentage of hospital days with a comorbid mental diagnosis, the mean number of days per hospitalization with and without a comorbid mental diagnosis, and the ratio of length of stay (LOS) for hospitalizations with and without a comorbid mental diagnosis (Table 3). These results are presented by age group because, given the differences that we saw earlier in this report, LOS is very dependent on age. (Similar analyses using hospitalizations rather than hospital days are given in Appendix Table C.)
The share of hospital days associated with a comorbid mental diagnosis averaged 15.1% among the provinces included in this analysis. The share of hospital days associated with it ranged from 2.0% for hospitalizations related to pregnancy, to 29.2% for hospitalizations for nervous diseases. Other major disease types with a comorbid diagnosis of a mental condition of note were infectious/parasitic (20.7%), metabolic (22.4%), respiratory (18.7%), skin (18.1%) and injury/poison (21.7%).
The share of days within an ICD-10-CA chapter associated with a mental comorbidity diagnosis increased from 4.4% at age 0 to 19 to 18.5% at 65 and older. Some chapter hospitalizations were more likely to have a mental comorbidity in some age groups than others. For example, for respiratory disorders, the share of days associated with a mental diagnosis increased with age from 3.0% to 20.8%. However, within skin disorders the percentage of chapter days associated with a mental comorbidity was similar from age 20 to 65 and older. The percentage of chapter days associated with a mental comorbidity was generally low in the 0 to 19 age group.
To summarize the differences in length of stay, we present a ratio between hospitalizations with and without a mental diagnosis (Table 3). The highest ratios were found for hospitalizations with most responsible conditions related to the eye (3.5) and congenital conditions (3.5). Hospitalizations with a comorbid mental condition were generally longer than those without one across all age groups. A noted exception was the eye and ear disease chapters at the youngest age grouping. Overall, there was a 2.5-fold difference in the average length of stay from a hospitalization with a mental comorbidity compared with one without. The impact of mental comorbidity was fairly consistent across age groups, with ratios ranging from 1.8 to 2.1.
Provincial breakdowns by ICD-10-CA are provided in Appendix tables D, E and F.
To further understand the burden of mental conditions as a comorbid diagnosis, we looked at the specific disease chapters by major mental diagnoses (Table 4). The percentage of hospital days for different mental conditions varies between chapters. For example, 4.9% of hospital days for skin diseases list substance abuse as comorbidity, compared with only 1.9% of hospital days for circulatory diseases. The disease chapters having more than 18% of their hospital days with a mental comorbidity were infection, metabolic, mental, nervous, respiratory, skin, injury/poison, and ill-defined. Note that the total mental is lower than the sum of the seven categories, as some hospitalizations reported more than one mental comorbidity. Similar data on hospitalizations rather than hospital days are given in Appendix Table G.
Summary
To briefly summarize Part 1, those who were discharged at least once from an acute-care hospital with a mental condition represent slightly less than 1% of Canadians but make up a disproportionate share. 12.8%, of hospitalized individuals. Twenty-nine percent of all hospital days, or 3.3 million hospital days, were used in hospitalizations that reported a mental diagnosis. More than one-half of these days were from hospitalizations associated with a comorbid mental diagnosis. The average length of stay for those hospitalizations with a mental comorbidity is two times the length of stay without one. This finding was fairly consistent across age, most responsible chapter designation and provinces. Thus, people with mental conditions put a heavy burden on the health care system.
Part 2: Characteristics of individuals hospitalized with a diagnosis of a mental condition
Results presented thus far clearly highlight the heavy burden of mental conditions on acute-care hospitalizations, whether as most responsible or comorbid diagnoses. Hence, it is important to understand this patient population and the factors that may be contributing to the additional burden on hospital services. Evidence to date suggests that various risk factors are often more prevalent among individuals with mental conditions. Compared with those without a mental condition, individuals with a current mental condition are generally less likely to be married,26,40,41 more likely to have low income or be unemployed,28,29,40,41 more likely to be smokers,12,29,42,43,44,45,46 and, in some cases, more likely to be overweight or obese.43,44,47,48,49,50 Individuals with mental conditions have also been found to have higher prevalence of several chronic physical conditions, in particular, chronic pain, diabetes, cardiovascular disease, high blood pressure and respiratory conditions.12,13,14 These factors are also often associated with increased use of hospital services, and so they may be associated with the increased use of services among this patient population.
In this second section, we used linked survey and hospital data to identify those factors associated with future hospitalizations with a diagnosis of a mental condition. The Canadian Community Health Survey (CCHS) was linked to the hospital administrative data. (See Data source text box for more details.) The linked survey and hospital admission data provided a unique opportunity to consider a comprehensive set of characteristics that may be associated with hospitalizations with a mental diagnosis. Selected demographic, socioeconomic, health status, health behaviour and health care experience characteristics were identified for survey respondents who did and did not experience a hospitalization related to mental health within the next four years. Tables 5 to 8 summarize all the variables examined.
Findings
Individuals hospitalized at least once with a diagnosis of a mental condition tended to be in lower socioeconomic groups compared with those hospitalized with other conditions
Lower educational attainment was more common among people hospitalized with a diagnosis of a mental condition than those without this condition. Considering hospitalizations over the four years after the CCHS was administered, 42.8% of those admitted with a comorbid diagnosis of a mental condition had a secondary education or less as their highest educational attainment (Table 5). This compared with 29.2% of those admitted without a diagnosis of a mental condition.
There was a similar story for income. Lower incomes were more common among people hospitalized with a diagnosis of a mental condition than was the case for those without this condition. Approximately 8.7% of those hospitalized with a most responsible diagnosis of a mental condition, and 7.0% with a comorbid diagnosis were in the lowest income quintile compared with 3.7% in the comparison group (Table 5).
Individuals who experienced a hospitalization with a diagnosis of a mental condition were less likely to be married or common-law (43.2% MR, 47.3% comorbid) compared with the comparison group (61.8%).
Hospitalizations with a most responsible diagnosis of a mental condition and with a comorbid diagnosis featured a higher percentage of Aboriginal people (3.6% and 4.4%) than hospitalizations without a diagnosis of a mental condition (1.5%) (Table 5). People that lived in a rural area were less likely to have a diagnosis of a mental condition in a hospitalization compared with those from urban areas. Immigrant status showed no effect.
Lifestyle factors such as smoking were associated with hospitalizations with a mental diagnosis
A higher percentage of individuals hospitalized with a diagnosis of a mental condition reported being a current smoker than those hospitalized without a diagnosis of a mental condition. Approximately one-half of those who were admitted to the hospital with either a most responsible or comorbid diagnosis of a mental condition reported being current smokers compared with only 28.4% who entered the hospital without a diagnosis of a mental condition (Table 6).
A higher percentage of those hospitalized with a most responsible diagnosis of a mental condition were non-drinkers (31.2%) compared with those who went to the hospital for other causes (23.2%) (Table 6).
Forty-eight point five percent (48.5%) of those hospitalized with a most responsible diagnosis of a mental condition had a weak sense of community belonging compared with those who went to the hospital for other causes (37.2%) (Table 6).
Body mass distribution appears to be similar across the different hospitalization groups (Table 6).
Individuals who experienced at least one hospitalization with a diagnosis of a mental condition tend to report lower health status compared with those hospitalized with other conditions
Individuals who were hospitalized with a most responsible or comorbid diagnosis of a mental condition reported poorer health status across several measures, including self-reported fair/poor health and disability (as obtained from the Health Utility Index), compared with those hospitalized with no mental conditions (Table 7).
Individuals who hospitalized with a comorbid mental diagnosis were more likely to have three or more chronic conditions compared with those hospitalized with no diagnosis of a mental condition (21.3% vs. 11.8%) (Table 7). The percentages for those with particular diseases—diabetes, arthritis, chronic obstructive pulmonary disease (COPD), asthma, heart disease, and stroke—were also higher among those hospitalized with a comorbid diagnosis of a mental condition (Table 7). Those with a hospitalization related to a mental condition were also more likely to report pain (36.5% most responsible; 35.6% comorbid), compared with those who hospitalized for other types of experienced other types of hospitalizations (25.6%) (Table 7).
As would be expected, individuals who experienced a hospitalization related to a mental condition were also more likely to report being depressed, taking antidepressants or being under "high" stress at the time of the survey when compared with those who experienced a hospitalization with no diagnosed mental condition.
Individuals experiencing a hospitalization with a mental diagnosis were more likely to report unmet health care needs
Individuals hospitalized with a diagnosis of a mental condition were more likely to have seen a mental health professional compared with those with a non-mental hospitalization (44.7% most responsible mental and 21.9% comorbid mental compared with 10.1% non-mental) (Table 8). This was not unexpected. However, similar percentages of individuals hospitalized with a diagnosis of a mental condition had a regular medical doctor as those with those hospitalized without a diagnosis of a mental condition (90.5% most responsible and 87.4% comorbid compared with 91.7% without diagnosis of mental condition) (Table 8).
Individuals hospitalized with a diagnosis of a mental condition were more likely to report unmet health care needs. A higher percentage of individuals hospitalized with a mental condition reported unmet health care needs compared with those hospitalized without a mental condition (25.9% most responsible; 23.1% comorbid and 16.5% without diagnosis of mental condition) (Table 8). The percentages who reported unmet mental health care needs were also higher for those hospitalized with a most responsible diagnosis of a mental condition (10.7% most responsible vs. 1.3% without diagnosis of mental condition). These differences in unmet health care needs translate into a significantly higher likelihood of a hospitalization with a diagnosis of a mental condition.
Summary
These results reflect the prevalence of risk factors reported in the literature, such as lower socioeconomic status, single marital status, smoking, and higher prevalence of chronic physical conditions, among people with chronic mental conditions. The aim of this study, however, is to identify factors that are associated with future hospitalization with a diagnosis of a mental condition. In summary, based on the age/sex adjusted results (Tables 5 to 8), individuals who had a high school education or less, were in the lower household income quintiles, were not married, were Aboriginal, were current smokers, reported being depressed or using antidepressants, reported fair/poor health, had physical diseases such as diabetes, arthritis, COPD, heart disease, asthma, previous stroke or pain, and who reported unmet health care needs were more likely to be admitted to hospital within the next four years with a mental health condition compared with those who were admitted to hospital within the next four years without a mental health condition.
As most of these factors could also contribute to the higher use of hospital resources, future work is needed to study how these factors are associated with longer average length of hospital stay among those with a diagnosed mental condition.
Discussion and conclusions
People with mental conditions place a high burden on the health care system. While those who are hospitalized for reasons related to mental health—with either a most responsible or comorbid diagnosis—make up a small proportion of the overall population (1%), they consume a significantly larger proportion of hospital services. Twenty-nine percent of all hospital days, or 5.8 million hospital days, were used in hospitalizations that recorded a mental condition. More than one-half of these days were from hospitalizations associated with a comorbid mental diagnosis. Most previous reports regarding hospitalizations focus on mental conditions as a most responsible diagnosis and provide only limited information regarding the burden of mental conditions as comorbidity.21,22,23 Thus, very little information has been previously available across Canada on mental conditions as a comorbid condition. Our report emphasizes comorbid mental conditions.
The average length of stay for those hospitalizations with a mental comorbidity is two times the length of stay for hospitalizations without one. This finding was fairly consistent across age, most responsible chapter designation and provinces. Some early U.S. studies used retrospective record review and prospective diagnosis to compare the average length of stay of patients with diagnosed or suspected mental comorbidity to comparison groups. These studies also found longer hospitalizations for those with a mental condition.53,54,55,56 Results of these retrospective analyses led to prospective studies that identified patients with psychiatric conditions during hospitalization, and that also found longer lengths of stay with mental comorbidity.57,58,59,60 The majority of these studies failed to adequately control for other variables affecting length of stay and, in addition, these studies were largely carried out within only one or two sites. The analysis by Bressi (2006)61 addresses these concerns by using a nationally (United States) representative sample of hospitalizations and controlling for patient and hospital characteristics. Their analyses showed that the existence of a psychiatric comorbidity predicted longer hospitalizations for medical inpatients.
The results of this study indicate that a mental diagnosis is a common comorbidity for many physical conditions. This agrees with current evidence that suggests significant comorbidity between chronic physical conditions and mental conditions.11 Several studies have illustrated poorer physical health status and higher prevalence of chronic physical conditions, disease, and chronic disease risk factors among people with mental conditions.12,13,14,48,49,50,62 Research has frequently also found that the prevalence of mental conditions is higher in people with chronic physical disorders than in people without physical disorders.63,64,65,66 Thus, contemporary epidemiological data shows the frequent co-occurrence of certain physical diseases and certain mental conditions. In particular, cardiovascular disease, hypertension, respiratory disorders, diabetes mellitus, and other metabolic disorders have been found alongside mental conditions.67
This combination of mental and physical conditions leads to two streams in the literature. First, from the patient side, it seems that patients with a serious mental condition are less aware of co-occurring medical conditions. For example, patients from the Department of Veterans Affairs (VA) National Psychosis Registry with a serious mental condition were less likely than those without a serious mental condition to self-report having a medical condition (such as heart disease, arthritis, cancer, diabetes, back pain, congestive heart failure, or hypertension) that was recorded in their medical record.68 Patients with mental comorbidity also introduce another level of complexity of care. Mental comorbidity complicates help-seeking, diagnosis, and treatment, and influences prognosis.4
Second, from the health care side, it has been reported that health services are not provided equitably to people with mental conditions, and that the quality of care for both mental and physical health conditions for these people could be improved.4,69,70,71 Traditional reactive consultation services are likely to see only a small percentage of the general-hospital patients who could benefit from their care.72 These patients can easily fall into the crack in between different levels of disjointed health services.18 In this report, we found that those with a mental condition are more likely to report unmet health care needs compared with those who were hospitalized for a non-mental health diagnosis. In one example from the literature, Frayne SM et al. (2005) found that failure to meet diabetes performance measures was more common in patients with mental conditions, and that the percentage not meeting diabetes care standards increased with a greater number of mental conditions.73
On a more positive note, Krein S.L. et al. (2006),74 using data from the Department of Veterans Affairs (VA) health care system, found that quality-of-care measures and intermediate outcomes were comparable for diabetes patients with and without serious mental conditions. This paper suggests that care is similar "possibly because of increased levels of contact with the health system and the VA's integration of medical and mental health services."
Several other papers have suggested that it would be advantageous to integrate physical and mental health care. Kathol et al. (2008)75 describes the potential advantages if assessment and treatment of mental and substance use conditions became a clinical, administrative, and financial part of physical health care with common provider networks, the ability to combine service locations (integrated clinics and inpatient units), similar coding and billing procedures, and a single funding pool. Others have said that mental health awareness needs to be integrated into all aspects of health and social policy, health-system planning, and delivery of primary and secondary general health care.4,76,77 The challenge is creating a system in which people with coexisting mental health and/or substance abuse problems as well as physical disorders find 'no wrong door' when they seek help.78 A recent Standing Senate Committee Report18 reported that patients with mental conditions confront a confusing, fragmented, under-resourced system and that a more coordinated system would be of use.
As an example of the benefits of an integrated physical and mental health care system, there would be more awareness that many medications prescribed for serious mental conditions have significant metabolic and cardiovascular adverse effects.79 For example, Kiraly et al (2008)79 suggest that patients treated with second-generation antipsychotics should receive preventive counselling and treatment for cardiovascular disease. Drug interactions can also lead to increased adverse effects, increased or decreased drug levels, toxicity, or treatment failure. Patients might also 'self-medicate' using licit or illicit drugs.80 If patient care is co-ordinated between primary care physicians and mental health professionals, there is likely less chance of serious adverse effects.
As well as the integration of physical and mental care, other systematic approaches have been suggested in the literature, including up-stream prevention strategies81,82 and support programs to help such individuals stay in the community.83 Thus, hospitalization is only one of a number of approaches to mental health care in Canada today.20 With advances in psychiatric medications, the development of a spectrum of community-based services, and the high costs associated with institutionalization, some mental health care has been moved to outside the hospital setting. Other reports caution, however, that community care may not necessarily reduce the need for hospital care.82,84,85,86,87,88,89
We found variations across provinces and health regions in the percentage of hospitalizations and hospital days and in the mean lengths of stay of hospitalizations with a mental diagnosis. The variations across health regions are associated with many factors, such as the number and type of facilities, which may vary from region to region. They could reflect potential differences in the prevalence of mental conditions within the population, in physicians' professional diagnostic and practice styles, in acute-care-hospital treatment practices or in differences in coding practices, specifically as related to comorbidity. Regarding the last point, the average number of documented diagnoses per hospitalization varies greatly from province to province: Quebec reported the highest average number of comorbid diagnoses per hospitalization; Newfoundland and Labrador and Prince Edward Island reported the lowest number.39
From the hospital data, we found that those with a mental condition are more likely to have many common types of comorbidity. They are more vulnerable in other ways, too. Using the linked survey-hospital database, we found that they were more likely to have had a weak sense of community belonging, more likely to be single rather than married, have less than high school education, are less likely to be in an upper income quintile and are more likely to have reported fair/poor health, have pain, and report disability and unmet health care needs compared with those who experienced hospitalization not related to mental health.
This reinforces and adds to the evidence reported to date that, compared with those without a mental condition, individuals with a current mental condition are generally less likely to be married,26,40,41 more likely to have low income or be unemployed,28,29,40,41 and more likely to be smokers.12,29,42,43,44,45,46 In some recent reports, administrative data have been supplemented with data from other sources for a more comprehensive picture of individuals with mental illness. For example, a recent report produced jointly by Statistics Canada and the Canadian Institute for Health Information (CIHI) uses linked survey and administrative data to identify the patient characteristics associated with hospitalization among those with depression.26 A recent Ontario-based report from CIHI uses the new Ontario Mental Health Reporting System to report on both the hospitalizations for mental illness as well as key characteristics of mental health patients including patient characteristics.29 The results revealed that those admitted to a mental health bed had lower rates of labour force participation, lower rates of being married or living with a partner, and lower levels of education compared with the overall Ontario population 15 and older. A report from Manitoba provided an overview of the prevalence of mental conditions. It compared five neighbourhood income groupings (called 'income quintiles') based on the average household income of the area, and reported that highest rates of mental illness were in the lowest income neighbourhoods.28
A key point from our study is that those with a mental condition are more likely to report unmet health care needs compared with those who experienced a non-mental health related hospitalization. This is echoed by the references above that reported that health services are not provided equitably to people with mental conditions. Other common characteristics of those with a mental condition can play a part in leading to unmet needs. For example, looking at education levels, adult respondents from the Canadian Community Health Survey Cycle 1.2 who had an anxiety or depressive disorder were more likely to see a psychiatrist, family doctor, psychologist or social worker the higher their education.90 Thus, socially disadvantaged individuals were at high risk for having their mental health service needs unmet. Steele (2007) suggested that programs for targeted services be developed and evaluated for consumers who have not completed high school.91
In summary, mental conditions can affect any age group, and are a common comorbidity for physical disorders. Our report emphasizes comorbid mental conditions. We found those hospitalized with a diagnosis of a mental condition used 29% of hospital days in 2003/2004, and that more than half of these days were associated with a mental comorbidity. Also, the average length of stay for those hospitalizations with a mental comorbidity was two times the length of stay for hospitalizations without one. A linked survey/hospital study identified a set of characteristics associated with those hospitalized with a comorbid diagnosis of a mental condition. The characteristics, such as smoking, low income, low education, and the presence of physical conditions, are also often associated with increased use of hospital services: they may be associated with greater use of services by the mental patient population. Future work plans to bring together the two sides of this study by examining to what extent factors associated with a diagnosed mental condition could contribute to longer average lengths of hospital stay.
- Date modified: