The Relation Between Perceived Need for Mental

Original Research

The Relation Between Perceived Need for Mental Health Treatment, DSM Diagnosis, and Quality of Life: A Canadian Population-Based Survey Jitender Sareen, BSc, MD, FRCPC1, Murray B Stein, MD, FRCPC, MPH2, Darren W Campbell, PhD3, Thomas Hassard, PhD4, Verena Menec, PhD5 Objectives: Prevalence estimates of mental disorders were designed to provide an indirect estimate of the need for mental health services in the community. However, recent studies have demonstrated that meeting criteria for a DSM-based disorder does not necessarily equate with need for treatment. The current investigation examined the relation between self-perceived need for mental health treatment and DSM diagnosis, with respect to quality of life (QoL) and suicidal ideation. Methods: Data came from an Ontario population-based sample of 8116 residents (aged 15 to 64 years). The University of Michigan Composite International Diagnostic Interview was used to diagnose mood, anxiety, substance use, and bulimia disorder according to DSM-III-R criteria. We categorized past-year help seeking for emotional symptoms and (or) perceiving a need for treatment without seeking care as self-perceived need for treatment. We used a range of variables to measure QoL: self-perception of mental health status, a validated instrument that measured well-being, and restriction of activities (current, past 30 days, and long-term). Results: Independent of subjects’ meeting criteria for a DSM-III-R diagnosis, self-perceived need for treatment was significantly associated with poor QoL (on all measures) and past-year suicidal ideation. Conclusions: Self-perceived need for mental health treatment, in addition to DSM diagnosis, may provide valuable information for estimating the number of people in the population who need mental health services. The relation between self-perceived need for treatment and objective measures of treatment need requires future study. (Can J Psychiatry 2005;50:87–94) Information on funding and support and author affiliations appears at the end of the article.

Clinical Implications · Self-perceived need for mental health treatment is associated with poor quality of life, regardless of whether the subject meets criteria for a DSM diagnosis. · Public health strategies that improve access to treatment for individuals who perceive a need for it should be considered. · Individuals who do not meet criteria for a mental disorder but nevertheless perceive a need for mental health treatment may also benefit from treatment. Limitations · The survey provides self-reported information collected at a cross-sectional level by trained lay interviewers. · Although it includes a wide range of mental disorders, the survey did not assess the presence of several Axis I and Axis II diagnoses. · The findings may not be generalizable to individuals with psychotic disorders.

Key Words: mental health service use, mood, anxiety, substance abuse, phobia, perceived need, quality of life

Can J Psychiatry, Vol 50, No 2, February 2005 W

87

The Canadian Journal of Psychiatry—Original Research

n the last 2 decades, the main approach to estimating the number of people in the community who “need” care for mental illness has been to provide prevalence estimates of mental disorders based on the explicit criteria of an accepted diagnostic system (1). For 2 major reasons, this methodology has recently been questioned (2,3). First, several authors have suggested that the current DSM diagnostic system may overestimate the need for mental health treatment (4–7). Their arguments are based on the reasoning that most individuals diagnosed with a DSM disorder do not seek treatment because they are suffering from transient or mild symptoms that do not require treatment (4–7). Regier and others have suggested that defining “need” for treatment in community samples requires severity criteria additional to those in the DSM (8,9). Second, significant data suggest that the DSM diagnoses assessed in community surveys do not necessarily capture all the individuals who need treatment. Recent investigations of many different mental disorders (for example, depression, 10–12; posttraumatic stress disorder [PTSD], 13,14; anxiety, 15,16; bipolar disorder, 17–19; and alcohol use disorders, 20) have shown that people with subthreshold symptoms (that is, those who do not meet DSM criteria) often have levels of impairment similar to those of individuals who meet criteria for the disorder. This burgeoning evidence that subthreshold symptoms are associated with significant dysfunction has led to a conceptual shift from a dichotomous to a dimensional approach to mental disorders (3,21,22). In short, deciding who needs care for emotional symptoms solely on their meeting criteria for DSM diagnoses does not seem optimal.

I

To address this controversial issue, recent community surveys have focused on the prevalence and correlates of selfperceived need for mental health treatment (23–25). Mojtabai and others (26) analyzed data from the US National Comorbidity Survey (NCS) to investigate the correlates of perceived need for treatment (that is, help seeking or perceived need) among individuals with a past-year affective, anxiety, or substance use disorder defined according to DSM-III-R criteria. They found that, among respondents diagnosed with a DSM-III-R disorder, individuals who perceived a need for treatment had much higher levels of suicidal ideation and comorbidity than did individuals who did not perceive a need for treatment or had not sought treatment. They suggested that their findings provide “partial support” for the hypothesis that the rates of mental disorders have been exaggerated and do not reflect the true need for mental health care (26) and speculated that most individuals with less severe or impairing psychopathology might have recognized the self-limited nature of their problems and might not have perceived a need for mental health treatment. The main limitation of Mojtabai and colleagues’ study was that they examined only the NCS subsample having a DSM 88

disorder and not the whole population sample. Many community samples have demonstrated that a significant number of individuals perceive a need for treatment or seek treatment without meeting criteria for a DSM disorder (4,25,27,28). These individuals could have either subthreshold DSM symptoms or a DSM disorder that was not assessed in the community survey. Prior to concluding that need for mental health care has been exaggerated, we suggest that the relation between perceived need, diagnosis, and a range of quality of life (QoL) and suicidal ideation measures needs to be examined in a sample comprising the whole population, rather than a sample limited to those with a DSM disorder. To extend this area of inquiry, our study uses the Mental Health Supplement to the Ontario Health Supplement (OHS, 29), a community survey contemporaneous to the NCS. For our purposes, an advantage of the OHS over the NCS is that it included a comprehensive assessment of QoL, all of which could be used as indicators of need for mental health treatment (30). The main objective of this study was to examine the relation between meeting criteria for a DSM disorder and self-perceived need for treatment with QoL and suicidal ideation in a population-based sample.

Methods The Mental Health Supplement to the OHS was a general population survey of participants aged 15 to 64 years, with a response rate of 67.4% (n = 8116). For the sake of brevity, and since the details of the OHS methodology have been published in numerous articles (4,5,29,31,32,32–36), we will not provide this information here. We received access to the OHS dataset from the Ontario Ministry of Health for the specific purpose of conducting secondary data analysis. Assessment of Mental Disorders The highly reliable University of Michigan Composite International Diagnostic Interview (UM-CIDI, 37) was used to assess all respondents for the following mental disorders: major depression, dysthymia, bipolar disorder, generalized anxiety disorder, specific phobia, social phobia, agoraphobia, panic disorder, alcohol abuse or dependence, cannabis abuse or dependence, other drug (that is, cocaine, amphetamines, hallucinogens, or heroin) abuse or dependence, and bulimia nervosa. In line with the work of Kessler and others (38), social phobia was further differentiated into a speaking subtype (SP-speaking) that included those who had social fears limited to speaking in small or large groups and a complex subtype (SP-complex) that involved fear and avoidance of multiple situations (38). Respondents meeting criteria for one or more of the above diagnoses were categorized as having at least one DSM disorder in the past year. W Can J Psychiatry, Vol 50, No 2, February 2005

The Relation Between Perceived Need for Mental Health Treatment, DSM Diagnosis, and Quality of Life

Perceived Need for Mental Health Treatment and Help-Seeking Self-perceived need for mental health treatment was identified if the respondent endorsed the question “Was there ever a time during the past 12 months when you felt that you might need to see a professional because of problems with your emotions or nerves or your use of alcohol or drugs?” and (or) the respondent endorsed help seeking for emotional symptoms with any professional (that is, mental health worker, general practitioner, other physician, minister, priest, spiritualist, and herbalist), owing to emotional problems or substance use. These OHS questions assessing perceived need were the same as the questions used in the NCS (26). Measures of QoL Current or Past-30-Day QoL. The OHS determined current perception of emotional status by asking “In general, compared to other persons of your age, would you say your mental health is 1) excellent, 2) very good, 3) good, 4) fair, or 5) poor.” This was dichotomized into 1) excellent, good, very good (96.4% of respondents); and 2) poor and fair (3.6% of respondents). Current satisfaction in 6 different life domains (specifically, main activity, family relationships, friendships, leisure activities, income, and life in general) was measured. Responses were scored on a 6-point satisfaction scale (that is, “extremely satisfied,” “quite satisfied,” “fairly satisfied,” “fairly dissatisfied,” “quite dissatisfied,” or “extremely dissatisfied”). Any of the 3 negative responses to these questions (that is, fairly, quite, or extremely dissatisfied) were coded as dissatisfaction in the particular domain of life. Respondents were dichotomized into those who reported dissatisfaction in at least 1 of the 6 domains of life (25.8% of the respondents) and those who reported dissatisfaction in none of the 6 domains (74.2% of the respondents). To assess past-30-day functioning, respondents were asked to report the number of days in the past 30 in which 1) they were totally unable to do the things they normally did; 2) they had to cut down on what they did or accomplished less; or 3) they had to make an extreme effort to perform their usual level of work, because of problems with their emotions, nerves, mental health, or use of alcohol or drugs. The responses to these questions were highly skewed, with only 1.4%, 2.6%, and 2.9% of the respondents reporting any days totally disabled, days of reduced activity, or days requiring extreme effort, respectively. A dichotomous variable was created, based on all 3 questions on past-30-day functioning: zero days affected by mental illness for all 3 questions vs one or more days affected by mental illness for any of the 3 questions. Past-Year QoL. All respondents were asked about suicidal ideation in the past year and responded to questions from a Can J Psychiatry, Vol 50, No 2, February 2005 W

validated QoL scale (39). All were asked whether they had ever “seriously thought about committing suicide” or “ever tried to commit suicide” during the past 12 months. In this sample, 3.7% of the respondents stated that they had seriously thought about committing suicide and (or) attempted suicide; 0.6% attempted suicide in the past year. The General Well-Being Schedule (GWB, 39), an empirically validated scale, was used as an indicator of subjective feelings of psychological well-being and distress over the past year. It covers 7 domains: energy, state of morale, control over emotions, interest in life, perceived stress, perceived health status, and satisfaction with relationships. In this sample, Cronbach’s coefficient alpha for the GWB was 0.85. GWB scores ranged from 0 to 42. For logistic regression analyses, they were categorized (on the basis of their distribution in the sample) into states of “very low or low well-being” (that is, scores of 0 to 24, which included 17.5% of respondents) or “medium or high well-being” (that is, scores of 25 to 42, which included 82.5% of respondents) (40). Long-Term QOL. Restricted activities owing to emotional symptoms or substance use over months to years was determined in the following manner. All respondents were asked whether their ability to perform their main activity (for example, work, school, household chores, or volunteering) was limited. Questions about child care, getting around the neighbourhood, getting out of the house, and personal care were used to assess limitations in other activities. If the respondents affirmed limitations in main or other activities, they were asked whether they attributed their limitations to emotional problems or substance use. They were also asked to estimate the duration of disability in each of the domains (specifically, less than 6 months, 6 to 12 months, 1 to 2 years, more than 2 years, or duration unknown). Restricted activities attributable to emotional problems or substance use were coded if the respondents affirmed limitations in main or other activities for any period of time (2.2% of the total sample). Description of Covariates in the Analysis Sociodemographic Variables. Sex, age, marital status, education, household income, place of birth, and residence location (urban or rural) were the sociodemographic variables analyzed. Age was dichotomized into 2 groups: age 15 to 24 years and age 25 to 64 years. Marital status was also dichotomized into 2 groups: widowed, divorced, or separated; and married, common-law, or never-married. Similarly, education was divided into 2 categories: 0 to 11 years and 12 years or more. Last, respondents were dichotomized as either born in Canada or not born in Canada. For place of residence, a population of fewer than 50 000 people was considered rural, and a population of 50 000 or more was considered urban. Household income was categorized into a dichotomous variable based on 89


the poverty lines and low-income cut-off developed by the National Council of Welfare and Statistics Canada (5). Attitudes Toward Mental Health Treatment. Respondents were asked 3 hypothetical questions to assess their attitudes toward mental health help seeking. They were asked 1) whether they would seek help when faced with a serious emotional problem (response options were “definitely go,” “probably go,” “probably not go,” and “definitely not go”), 2) how comfortable they would feel talking about personal problems with a professional (response options were “very comfortable,” “somewhat comfortable,” “not very comfortable,” and “not at all comfortable”), and 3) how embarrassing it would be if friends knew about the professional help (response options were “very embarrassed,” “somewhat embarrassed,” “not very embarrassed,” and “not at all embarrassed”). For each of these questions, respondents were categorized according to positive (that is, they would definitely or probably seek help) or negative (that is, they would probably or definitely not seek help) attitude toward help seeking. A similar dichotomization strategy was used for the variables “comfort in seeking care” and “embarrassment in seeking care.” Respondents were also asked to estimate the percentage of patients who were helped by mental health treatments. They were dichotomized into those who felt that treatment was efficacious (that is, they felt that more than 50% of patients were helped) and those who felt that treatment was not efficacious (that is, they felt that fewer than 50% of patients were helped). Comorbidity of Mental and Physical Conditions. Lifetime comorbidity with mental disorders and past-year number of physical illnesses were considered as covariates in this study. The lifetime DSM-III-R mental disorders among the respondents were grouped into 0, 1, and 2 or more disorders. Respondents were also asked about specific past-year physical health problems. This list included musculoskeletal (for example, arthritis), respiratory (for example, asthma or tuberculosis), cardiovascular (for example, heart attack), gastrointestinal (for example, ulcer), endocrine (for example, diabetes), and neurological (for example, multiple sclerosis or epilepsy) disorders, as well as cancer, AIDS, and other conditions. Previous work by Mojtabai and others (26) has demonstrated that perceived need for treatment is significantly associated with having one or more physical conditions in the past year. Therefore, we dichotomized the number of past 12-month physical conditions into no conditions and one or more conditions. Statistical Analysis First, we categorized the entire sample into 4 mutually exclusive categories based on presence or absence of a past-year DSM-III-R mental disorder (Dx) and self-perceived need for mental health treatment (PN). Second, we used chi-square 90

analysis to determine differences among these 4 groups with respect to sociodemographics, type, and number of DSM-III-R mood, anxiety, and substance use disorders. Third, we used multiple logistic regression to determine the relation between PN and Dx on measures of QoL and suicidal ideation. Covariates in each of the regressions were sociodemographic variables, attitudes toward mental health treatment (that is, probability of seeking care, comfort in seeking care, and embarrassment about seeking care), and comorbidity with mental or physical conditions. Because the OHS had a multistage design that used stratification and clustering (29), we assigned sampling weights and incorporated the design effect into the analysis. We used SPSS 10.7 software to summarize and analyze the data (41). Because we made multiple comparisons, we set significance at an alpha level of P < 0.01.

Results Table 1 divides the OHS sample into 4 groups that are based on possible combinations of past-year PN and that meet criteria for one or more DSM-III-R Dx. Weighted prevalences were as follows: PN without Dx, 4.5% (standard error 0.5%); Dx without PN, 11.5% (standard error 0.7%); and PN plus Dx, 4.9% (standard error 0.5%). Women were more likely to be in the PN without Dx group. Young adults (aged 15 to 24 years) were more likely to be in the Dx-only group. Respondents with low family income were more likely to be in the PN plus Dx group (Table 1). Across the 4 groups, we noted no significant differences with respect to education. The group with both PN and Dx had the greatest proportion of respondents with comorbid DSM III-R disorders (specifically, 69% had 2 or more lifetime disorders). The 4 most common mental disorders in the PN plus Dx group were major depression, SPcomplex, specific phobia, and alcohol abuse or dependence. Table 2 considers the relation between a DSM diagnosis, self-perceived need for treatment, QoL, and suicidal ideation. Individuals with a past-year DSM diagnosis were more likely to endorse both poorer QoL on all measures and past-year suicidal ideation than were individuals in the control group (that is, respondents with no PN and no Dx). More interestingly, individuals with a self-perceived need for mental health treatment who did not meet criteria for a DSM diagnosis (that is, who had a PN without Dx) also had poorer QoL and a higher likelihood of endorsing suicidal ideation than did the control group. Finally, individuals with both a PN and Dx had the greatest odds of poor QoL and suicidal ideation (adjusted odds ratios ranged from 6 to 42).

Discussion A self-perceived need for mental health treatment independent of the DSM diagnoses assessed in this survey was W Can J Psychiatry, Vol 50, No 2, February 2005


Table 1. Sociodemographic and lifetime psychopathology variables in relation to past-year DSM-III-R diagnosis (Dx) and past-year perceived need or help seeking for emotional symptoms (PN)a Neither PN nor Dx (n = 5954) %

PN no Dx (n = 340) %

Dx no PN (n = 983) %

Both PN and Dx (n = 391) %

c2

df

3

P

Sociodemographics and diagnosis Sex (female respondents)

48.0

67.1

51.9

54.4

25.49

Age (15 to 24 years)

20.3

20.9

32.4

20.0

31.7

3

< 0.001

8.7

13.7

11.5

21.3

25.24

3

< 0.001

73.5

76.7

80.6

81.6

5.43

3

0.137

Major depression

3.1

9.3

17.1

46.9

339.3

3

< 0.001

Dysthymia

1.4

1.9

6.1

21.5

130.8

3

< 0.001

Bipolar disorder

0.1

0.6

2.2

9.1

80.6

3

< 0.001

Social phobia, complex subtype

1.9

3.3

27.3

45.2

429.6

3

< 0.001

Social phobia, speaking subtype

3.9

5.2

31.8

27.3

265.2

3

< 0.001

Specific phobia

1.7

4.7

39.0

39.6

620.4

3

< 0.001

Generalized anxiety disorder

0.4

0.6

3.1

19.9

163.1

3

< 0.001

Agoraphobia with or without panic disorder

0.6

3.9

10.5

23.4

227.5

3

< 0.001

Panic disorder with or without agoraphobia

0.2

1.2

5.0

14.9

145.0

3

< 0.001

Bulimia nervosa

0.1

0.6

2.6

4.4

52.1

3

< 0.001

Alcohol abuse or dependence

6.1

9.8

27.9

30.2

217.3

3

< 0.001

Cannabis abuse or dependence

1.6

6.1

9.1

11.7

87.8

3

< 0.001

Drug abuse or dependence

0.4

2.5

3.9

7.3

61.9

3

< 0.001

1820.2

6

< 0.001

Current family income (below poverty) Education (Grade 12 or more)

< 0.001

Lifetime DSM-III-R diagnosis

Total number of lifetime disorders 0

82.2

65.0

—

1

14.0

25.2

53.3

30.4

3.8

9.8

46.7

69.6

2 or more

—

Note: Variables were dichotomized a

Past-year diagnosis denotes the presence of one or more of the DSM-III-R disorders displayed in the current table.

associated with poor QoL and suicidal ideation. First, the combination of meeting criteria for a DSM diagnosis and believing that treatment was needed (that is, help seeking and a perceived need for treatment) was associated with poorer QoL, higher levels of comorbidity, and suicidal ideation—findings that are consistent with previous surveys in the US (25) and Australia (26,42). Significant confidence in these results is possible because the OHS included a comprehensive assessment of QoL. Second, the current investigation was the first study to demonstrate that self-perceived need for treatment, even though criteria for the assessed set of DSM disorders were not met, was associated with low QoL and suicidal ideation. We speculate that this group comprised individuals with subthreshold DSM-III-R symptoms for the disorders surveyed in the OHS; alternatively, they possibly had DSM-III-R disorders not assessed in this survey (for example, adjustment disorder, bereavement, obsessive– compulsive disorder, PTSD, or Axis II pathology). Since no Can J Psychiatry, Vol 50, No 2, February 2005 W

large-scale community survey has ever been able to assess all the DSM disorders, we suggest that estimates of need for treatment should include respondents who perceived a need for treatment (that is, those who sought help and perceived a need for help). In summary, the results from this investigation suggest, first, that persons with more severe disorders (such as major depression, mania, and generalized social phobia) are more likely to perceive a need for treatment and, second, that not all perceptions of low QoL and suicidal ideation are accounted for by the set of disorders studied in the OHS. The evidence on this topic underscores the importance of assessing selfperceived need for treatment in addition to assessing DSM disorders when estimating the number of people in the community who require treatment for mental disorders. Our data do not support the hypothesis that the current DSM diagnostic system overestimates the need for treatment in the community (8). Our findings imply that people who meet criteria for a DSM diagnosis but who do not perceive a need for 91


Table 2. The associations of quality of life, disability, and suicidal ideation variables with the 4 groups of past-year perceived need (PN) and past year one or more DSM-III-R mental disorder diagnosis (Dx). Dependent variables

Neither PN nor Dx PN no Dx (n = 5954) (n = 340)

Dx no PN (n = 983)

Both PN and Dx (n = 391)

A. Current perception of emotional status compared with others (poor–fair vs good–excellent) % poor–faira b

AOR (99%CI)

1.7

5.5

5.3

21.7

1.00

1.66 (0.45–6.19)

1.39 (0.51–3.78)

7.77(2.88–20.96)**

B. Current dissatisfaction in at least 1 of the 6 domains of life (yes vs no) % dissatisfied

18.6

46.6

40.2

70.6

AOR (99%CI)

1.00

3.22 (1.95–5.31)**

1.53 (1.00–2.35)**

6.05 (3.40–10.77)**

C. Past 30 days dysfunction owing to emotions or substance use (1 or more day loss vs none) % 1 or more day loss

1.8

14.2

6.7

32.5

AOR (99%CI)

1.00

6.79 (2.99–15.40)**

1.37 (0.57–3.26)

9.45 (4.05–22.01)**

D. Past-year suicidal ideation or attempt (yes vs no) % suicidal

1.6

9.4

5.6

19.3

AOR (99%CI)

1.00

8.12 (2.87–22.97)**

2.18 (0.63–7.56)

11.94 (3.45–41.24)**

E. Past-year General Well Being Schedule score (poor < 24 vs good 24–42) % poor score

11.2

24.6

27.9

60.2

AOR (99%CI)

1.00

2.57 (1.35–4.91)**

1.97 (1.10–3.54)**

9.00 (4.43–18.30)**

F. Self-reported disability in main or other activities (months to years) owing to emotional or substance problems (yes vs no) % disability

0.3

11.0

2.9

AOR (99%CI)

1.00

30.02(7.57–119.00)**

5.87 (1.03–33.57)** 41.73 (8.30–210.0)**

20.0

**P < 0.01 a

Percentages are weighted

b

Adjusted odds ratios (AOR) include the following covariates in the multiple logistic regression: sex; age 15–24 years or 25–64 years; education 12 or more years vs 11 or less; current family income below poverty vs not; attitudes toward mental health treatment seeking favourable vs not favourable; past-year physical health problems 1 or more vs none; lifetime DSM-III-R disorders 0, 1, 2, or more. The following findings were noted after separate logistic regression analyses (data not shown, but available on request). Individuals with both PN and Dx had a significantly poorer quality of life and endorsement of suicidal ideation, compared with those with PN no Dx and those with Dx no PN. The latter 2 groups did not differ significantly with respect to most of the indicators of need for treatment.

care have, nonetheless, more distress than do undiagnosed individuals who do not perceive a need and who might also benefit from treatment. These findings are also consistent with Mojtabai’s study of the NCS (26), which showed that 5% of individuals with a DSM diagnosis who did not perceive a need for help nonetheless displayed suicidal ideation and that 26% of this group had severe impairment. However, cross-sectional studies are not able to ascertain the proportion and characteristics of the individuals in this group who experience mild and transient symptoms that resolve over time, compared with those who have symptoms that will become more severe and lead to further disability. Recent evidence from the US National Comorbidity Survey (43) demonstrated that “mild” disorders in respondents (that is, those meeting DSM criteria for a mood or anxiety disorder not associated with serious impairment in the form of suicidality, disability, marital violence, or extreme social isolation) at Time 1 were associated with increased levels of disability, distress, and suicidal ideation 10 years later. The study authors concluded that cost-effective treatment of “mild” disorders might prevent a substantial proportion of future serious mental disorder (43). Public education programs designed to increase awareness of mental illness symptoms and available treatments are likely the best method of approaching individuals who meet criteria for a DSM disorder but do not perceive a need for care. 92

Study Limitations Four important limitations of our study should be considered. First, the OHS survey relied on lay interviewers who conducted fully structured diagnostic interviews. Although the UM-CIDI yields acceptable reliability and validity (37), it is unrealistic to expect that these interviews will match the accuracy of diagnostic classifications made by trained clinicians who interview patients repeatedly, who interview family, and who use clinical data to make diagnostic decisions. Because the OHS did not include a clinician assessment of the need for mental health services, as was done in 2 earlier surveys (44,24), we cannot comment on the relation between selfperceived need for treatment and clinician-based assessment of the need for services. To the best of our knowledge, only one investigation, in a Finnish community sample (24), has examined the relation between self-perceived need for treatment and clinician-assessed need for treatment. This study found that self-reported perceived need for treatment (6.4% in men and 8.2% in women) was much lower than clinicianassessed need for treatment (14.5% in men and 19.6% in women). Future work should examine the relation between self-perceived need for treatment, clinician-assessed need for treatment, and meeting lay interviewer–based DSM criteria in the same community sample. W Can J Psychiatry, Vol 50, No 2, February 2005


Second, the question of perceived need for care was asked over a 12-month period, which could be associated with recall bias. A shorter recall period (for example, 30 days) and specific questions regarding perceived need in relation to each mental disorder would have provided more precise information. Meadows and others have validated an instrument to assess perceived need (the Perceived Need for Care Questionnaire, 45) and have used this instrument to estimate perceived need in an Australian community survey (25,42). We suggest that future community surveys should include a comprehensive assessment of perceived need in addition to an assessment of DSM diagnoses. Third, it has been approximately 10 years since the collection of the OHS data. During this time, there have been notable advances in pharmacologic and psychosocial treatments and a dramatic increase in public awareness regarding mental illness. It is possible that, with increased awareness of mental disorders, there may be an increase in the prevalence of perceived need for treatment. Data from more recent surveys, such as The National Comorbidity Survey Replication survey (46) and the Canadian Community Health Survey Cycle 1.2 (47), may answer the question of whether rates of perceived need have changed over the past decade. Finally, we would like to underscore that the findings of this investigation only apply to individuals suffering from nonpsychotic disorders.

Conclusions Self-perceived need for mental health treatment, in addition to DSM diagnoses, provides valuable information with regard to estimating the number of people in the population who need mental health services in the community. Future examination of the relation between self-perceived need for treatment and objective clinician assessment of need for treatment is required. Public health strategies to improve the quality of care and decrease the barriers to care for individuals with self-perceived need for treatment are likely to be among the most efficient strategies for decreasing the burden of mental illness in the community (36,46–50). Funding and Support The University of Manitoba Paul Thorlakson Fund and the Manitoba Health Research Council Establishment Grant (to Dr Sareen) provided grant support for this investigation. Acknowledgements We thank Mr Stephen T Hassard and Ms Shay-Lee Belik for their work on this project.

References 1. Spitzer RL, Endicott J, Robins E. Research diagnostic criteria: rationale and reliability. Arch Gen Psychiatry 1978;35:773– 82. 2. Spitzer RL. Diagnosis and need for treatment are not the same. Arch Gen Psychiatry 1998;55:120. 3. Goldberg D. Plato versus Aristotle: categorical and dimensional models for common mental disorders. Compr Psychiatry 2000;41:8–13. 4. Kessler RC, Frank RG, Edlund M, Katz SJ, Lin E, Leaf P. Differences in the use of psychiatric outpatient services between the United States and Ontario. N Engl J Med 1997;336:551–7. 5. Lin E, Goering P, Offord DR, Campbell D, Boyle MH. The use of mental health services in Ontario: epidemiologic findings. Can J Psychiatry 1996;41:572–7.

Can J Psychiatry, Vol 50, No 2, February 2005 W

6. Bijl RV, Ravelli A, van Zessen G. Prevalence of psychiatric disorder in the general population: results of the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Soc Psychiatry Psychiatr Epidemiol 1998;587–95. 7. Bijl RV, Ravelli A. Psychiatric morbidity, service use, and need for care in the general population: results of the Netherlands Mental Health Survey and Incidence Survey. Am J Public Health 2000;90:602–7. 8. Regier DA, Kaelber CT, Rae DS, Farmer ME, Knauper B, Kessler RC, and others. Limitations of diagnostic criteria and assessment instruments for mental disorders: implications for research and policy. Arch Gen Psychiatry 1998;55:109–15. 9. Narrow WE, Rae DS, Robins LN, Regier DA. Revised prevalence estimates of mental disorders in the United States: using a clinical significance criterion to reconcile 2 surveys’ estimates. Arch Gen Psychiatry 2002;59:115–23. 10. Preisig M, Merikangas KR, Angst J. Clinical significance and comorbidity of subthreshold depression and anxiety in the community. Acta Psychiatr Scand 2001;104:96–103. 11. Kessler RC, Zhao S, Blazer DG, Swartz M. Prevalence, correlates, and course of minor depression and major depression in the National Comorbidity Survey. J Affect Disord 1997;45:19–30. 12. Chen LS, Eaton WW, Gallo JJ, Nestadt G, Crum RM. Empirical examination of current depression categories in a population-based study: symptoms, course, and risk factors. Am J Psychiatry 2000;157:573–80. 13. Stein MB, Walker JR, Hazen AL, Forde DR. Full and partial posttraumatic stress disorder: findings from a community survey. Am J Psychiatry 1997;154:1114–9. 14. Marshall RD, Olfson M, Hellman F, Blanco C, Guardino M, Struening EL. Comorbidity, impairment, and suicidality in subthreshold PTSD. Am J Psychiatry 2001;158:1467–73. 15. Stein MB, Kirk P, Prabhu V, Grott M, Terepa M. Mixed anxiety-depression in a primary-care clinic. J Affect Disord 1995;34:79–84. 16. Stein MB. How shy is too shy? Lancet North Am Ed 1996;347:1131. 17. Akiskal HS. The bipolar spectrum—the shaping of a new paradigm in psychiatry. Curr Psychiatry Rep 2002;4:1–3. 18. Akiskal HS, Pinto O. The evolving bipolar spectrum. Prototypes I, II, III, and IV. Psychiatr Clin North Am 1999;22:517–34, vii. 19. Akiskal HS, Weise RE. The clinical spectrum of so-called “minor” depressions. Am J Psychother 1992;46:9–22. 20. Saunders JB, Lee NK. Hazardous alcohol use: its delineation as a subthreshold disorder and approaches to its diagnosis and management. Compr Psychiatry 2000;41:95–103. 21. Helmchen H, Linden M. Subthreshold disorders in psychiatry: clinical reality, methodological artifact, and the double-threshold problem. Compr Psychiatry 2000;41:1–7. 22. Magruder KM, Calderone GE. Public health consequences of different thresholds for the diagnosis of mental disorders. Compr Psychiatry 2000;41:14–8. 23. Katz SJ, Kessler RC, Frank RG, Leaf P, Lin E, Edlund M. The use of outpatient mental health services in the US and Ontario: the impact of mental morbidity and perceived need for care. Am J Public Health 1997;87:1136–43. 24. Lehtinen V, Joukamaa M, Jyrkinen E, Lahtela K, Raitasalo R, Maatela J, and others. Need for mental health services of the adult population in Finland: results from the Mini Finland Health Survey. Acta Psychiatr Scand 1990;81:426–31. 25. Meadows G, Burgess P, Fossey E, Harvey C. Perceived need for mental health care, findings from the Australian National Survey of Mental Health and Wellbeing. Psychol Med 2000;30:645–56. 26. Mojtabai R, Olfson M, Mechanic D. perceived need and help-seeking in adults with mood, anxiety or substance use disorders. Arch Gen Psychiatry 2002;59:77–84. 27. Angold A, Costello EJ, Farmer EM, Burns BJ, Erkanli A. Impaired but undiagnosed. J Am Acad Child Adolesc Psychiatry 1999;38:129–37. 28. Issakidis C, Andrews G. Service utilisation for anxiety in an Australian community sample. Soc Psychiatry Psychiatr Epidemiol 2002;37:153–63. 29. Boyle MH, Offord DR, Campbell D, Catlin G, Goering P, Lin E, and others. Mental health supplement to the Ontario Health Survey: methodology. Can J Psychiatry 1996;41:549–58. 30. Goering P, Lin E, Campbell D, Boyle MH, Offord DR. Psychiatric disability in Ontario. Can J Psychiatry 1996;41:564–71. 31. Offord DR, Boyle MH, Campbell D, Goering P, Lin E, Wong M, and others. One-year prevalence of psychiatric disorder in Ontarians 15 to 64 years of age. Can J Psychiatry 1996;41:559–63. 32. Macmillan HL, Fleming JE, Trocme N, Boyle MH, Wong M, Racine YA, and others. Prevalence of child physical and sexual abuse in the community: results from the Ontario Health Supplement. JAMA 1997;278:131–5. 33. Macmillan HL, Fleming JE, Streiner DL, Lin E, Boyle MH, Jamieson E, and others. Childhood abuse and lifetime psychopathology in a community sample. Am J Psychiatry 2001;158:1878–83. 34. Macmillan HL, Boyle MH, Wong MY, Duku EK, Fleming JE, Walsh CA. Slapping and spanking in childhood and its association with lifetime prevalence of psychiatric disorders in a general population sample. CMAJ 1999;161:805–9. 35. Alegria M, Bijl RV, Lin E, Walters EE, Kessler RC. Income differences in persons seeking outpatient treatment for mental disorders. Arch Gen Psychiatry 2000;57:383–91.

93


36. Edlund MJ, Wang PS, Berglund PA, Katz SJ, Lin E, Kessler RC. Dropping out of mental health treatment: patterns and predictors among epidemiological survey respondents in the United States and Ontario. Am J Psychiatry 2002;159:845–51. 37. Kessler RC, Wittchen H-U, Abelson JM, McGonagle KA, Schwarz N, Kendler KS, and others. Methodological studies of the Composite International Diagnostic Interview (CIDI) in the US National Comorbidity Survey (NCS). Int J Methods Psychiatr Res 1998;7:33–55. 38. Kessler, RC, Stein MB, Berglund PA. Social phobia subtypes in the National Comorbidity Survey. Am J Psychiatry 1998;155:613–9. 39. McDowell I, Newell C. Measuring health: a guide to rating scales and questionnaires. New York: Oxford University Press; 1987. p 125–33. 40. Stein MB, Kean YM. Disability and quality of life in social phobia: epidemiologic findings. Am J Psychiatry 2000;157:1606–13. 41. SPSS Inc. Statistical Package for the Social Sciences. Version 10. Chicago (IL): SPSS Inc; 1999. 42. Meadows G, Burgess P, Bobevski I, Fossey E, Harvey C, Liaw ST. Perceived need for mental health care: influence of diagnosis, demography and disability. Psychol Med 2002;32:299–309. 43. Kessler RC, Merikangas KR, Berglund P, Eaton WW, Koretz DS, Walters EE. Mild disorders should not be eliminated from the DSM-V. Arch Gen Psychiatry 2003;60:1117–22. 44. Bebbington PE, Marsden L, Brewin CR. The need for psychiatric treatment in the general population: the Camberwell Needs for Care Survey. Psychol Med 1997;27:821–34. 45. Meadows G, Harvey C, Fossey E, Burgess P. Assessing perceived need for mental health care in a community survey: development of the Perceived Need for Care Questionnaire (PNCQ). Soc Psychiatry Psychiatr Epidemiol 2000;35:427–35. 46. Kessler RC, Berglund PC, Demler O, Jin R, Koretz D, Merikangas KR, and others. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). JAMA 2003;289:3095–105.

47. Beland Y, Dufour J, Hamel M. Sample design of the Canadian Mental Health and Wellbeing Survey. Proceedings of the Survey Methods Section, Statistics Canada 2001. 48. Wang PS, Demler O, Kessler RC. Adequacy of treatment for serious mental illness in the United States. Am J Public Health 2002;92:92–8. 49. Sherbourne CD, Dwight-Johnson M, Klap R. Psychological distress, unmet need, and barriers to mental health care for women. Womens Health Issues 2001;11:231–43. 50. Olfson M, Guardino M, Struening E, Schneier FR, Hellman F, Klein DF. Barriers to the treatment of social anxiety. Am J Psychiatry 2000;157:521–7.

Manuscript received February 2004, revised, and accepted August 2004. Previously presented as an abstract at the Annual General Meeting of the Canadian Psychiatric Association; October 2002; Banff (AB). 1 Assistant Professor, Department of Psychiatry, University of Manitoba, Winnipeg, Manitoba. 2 Professor-in-Residence, Department of Psychiatry, University of California, San Diego, California. 3 Postdoctoral Fellow, Department of Psychiatry, University of Manitoba, Winnipeg, Manitoba. 4 Professor, Department of Community Health Sciences, University of Manitoba, Winnipeg, Manitoba. 5 Associate Professor, Department of Community Health Sciences, University of Manitoba, Winnipeg, Manitoba. Address for correspondence: Dr J Sareen, PZ430–771 Bannatyne Avenue, Winnipeg, MB R3E 3N4 e-mail: [email protected]

Résumé : La relation entre le besoin perçu de traitement de santé mentale, le diagnostic du DSM et la qualité de vie : un sondage de la population canadienne Objectifs : Les estimations de la prévalence des troubles mentaux sont conçues pour fournir une estimation indirecte du besoin de services de santé mentale dans la collectivité. Toutefois, des études récentes ont démontré que satisfaire aux critères d’un trouble du DSM n’équivaut pas nécessairement à un besoin de traitement. La présente enquête a examiné la relation entre le besoin auto-perçu d’un traitement de santé mentale et le diagnostic du DSM, en ce qui concerne la qualité de vie (QdV) et l’idéation suicidaire. Méthodes : Les données proviennent d’un échantillon de la population ontarienne de 8 116 résidents (de 15 à 64 ans). L’entrevue de diagnostic composée internationale de l’Université du Michigan a servi à diagnostiquer les troubles de l’humeur, anxieux, induits par une substance, et boulimiques, selon les critères du DSM-III-R. Nous avons catégorisé « ayant cherché de l’aide dans l’année écoulée » pour les symptômes émotionnels, et/ou « percevant un besoin de traitement sans obtenir de soins » pour le traitement. Nous avons utilisé une gamme de variables pour mesurer la QdV : l’auto-perception de l’état de santé mentale, un instrument validé qui mesurait le bien-être, et la restriction des activités (actuelle, les 30 derniers jours, et long terme). Résultats : Indépendamment de la satisfaction des sujets aux critères diagnostiques du DSM-III-R, le besoin de traitement auto-perçu était significativement associé à une piètre QdV (à toutes les mesures) et à une idéation suicidaire dans l’année écoulée. Conclusions : Le besoin de traitement de santé mentale auto-perçu, conjointement avec le diagnostic du DSM, peut fournir des renseignements valables pour estimer le nombre de personnes dans la population qui ont besoin de services de santé mentale, dans la collectivité. La relation entre le besoin de traitement auto-perçu et les mesures objectives du besoin de traitement nécessite d’autres recherches.

94

W Can J Psychiatry, Vol 50, No 2, February 2005