 |
This Web site is a component of the SAMHSA Health Information Network. |
|
Section IV.
Population Assessments
Chapter 15
The Prevalence and Correlates of Serious Mental Illness (SMI) in the National
Comorbidity Survey Replication (NCS-R)
Ronald C. Kessler, Ph.D.
Department of Health Care Policy, Harvard Medical School
Wai Tat Chiu, A.M.
Department of Health Care Policy, Harvard Medical School
Lisa Colpe, Ph.D.
National Institute of Mental Health
Olga Demler, M.A., M.S.
Department of Health Care Policy, Harvard Medical School
Kathleen R. Merikangas, Ph.D.
National Institute of Mental Health
Ellen E. Walters, M.S.
Department of Health Care Policy, Harvard Medical School
Philip S. Wang, M.D., Dr. P.H.
Department of Health Care Policy, Harvard Medical School
Department of Psychiatry and Division of Pharmacoepidemiology and Pharmacoeconomics,
Brigham and Women’s Hospital, Harvard Medical School
Introduction
Although community epidemiological surveys estimate that as much as 30 percent
of the U.S. population has a mental or substance use disorder each year (Kessler
et al., 1994; Regier et al., 1998), it is unlikely that all these people need
treatment, as some of them almost certainly have mild or self-limiting
disorders (Narrow, Rae, Robins, &
Regier, 2002). Given this likelihood, the assessment of serious mental illness
(SMI) is in some ways more important for most policy planning purposes than
the assessment of all mental illness.
Epidemiological surveys carried out over the past two decades were unable to
provide definitive data on SMI because the main concern of these surveys was
to include questions on the full set of diagnostic criteria for the Diagnostic
and Statistical Manual of Mental Disorders
(DSM) disorders they assessed. Clinical severity of these disorders was not
a major focus. Nonetheless, post hoc
analysis of these surveys can provide some indirect information about severity.
Secondary analyses of this sort, based
both on epidemiological surveys carried out in the United States (Narrow et
al., 2002) and on comparable surveys carried out in other developed countries
(Bijl et al., 2003; Demyttenaere et al., 2004), strongly suggest that a substantial
proportion of DSM cases in the community are mild.
Existing data on clinical severity of community cases are limited by the fact
that only crude indicators of severity
were included in previous community epidemiological surveys. In an effort to
provide more direct data of this sort, the World Health Organization (WHO) recently
expanded its Composite International Diagnostic Interview (CIDI) (Robins, Wing,
Wittchen, & Helzer, 1988), the interview used in almost all major psychiatric
epidemiological surveys in the world over the past decade, to include detailed
questions about severity (Kessler & Ustun, 2004). This new version of the
CIDI has now been used in a series of community epidemiological surveys coordinated
by the WHO throughout the world. These surveys, known as the WHO World Mental
Health (WMH) Survey Initiative (Demyttenaere et al., 2004), are designed explicitly
to estimate the global burden of mental and substance
disorders in comparison to commonly occurring physical disorders. The United
States participated in the WMH Survey Initiative by carrying out a nationally
representative household survey known as the National Co-morbidity
Survey Replication (NCS-R) (Kessler
et al., 2004; Kessler & Merikangas, 2004). This chapter provides a brief
overview of the initial NCS-R
results on the prevalence and severity of mental and substance use disorders
in the United States. A more detailed presentation of these results is reported
elsewhere (Kessler, Chiu, Demler, & Walters, in press-b).
In addition, we present an overview of initial results regarding 10-year
time trends in the prevalence and severity
of mental disorders based on aggregate trend comparisons of the NCS-R
with the original National Co-morbidity
Survey (NCS) (Kessler et al., 1994). The NCS was carried out a decade before
the NCS-R. A more detailed presentation
of these results is reported elsewhere (Kessler et al., in press-a).
In the 1980s, the Epidemiologic Catchment
Area (ECA) study found that approximately 30 percent of the adult respondents
in that survey met criteria for one or
more of the 12-month DSM-III
disorders assessed (Robins & Regier, 1991). A decade later, the NCS found
that approximately 30 percent of people ages 15–54 in that survey met
criteria for one of the 12-month
DSM-III-R
disorder assessed (Kessler et al., 1994). In the past 10 years there have been
dramatic changes in the use of mental health services in the United States.
The Substance Abuse and Mental Health Services Administration (SAMSHA) documents
that annual encounters in specialty mental health treatment centers increased
by nearly 50 percent between 1992 and 2000 (Manderscheid et al., 2001). The
National Ambulatory Medical Care Survey documents that the number of people
receiving healthcare treatment for depression more than tripled between 1987
and 1997 (Olfson et al., 2002). The Robert Wood Johnson Foundation Community
Tracking Survey documents that the proportion of people with serious mental
illness who received specialty care increased by nearly 20 percent between 1997–8
and 2000–1 (Mechanic & Bilder, 2004). To the extent that these increases
in treatment were effective, we might expect that the prevalence of mental disorders
would be lower today than at the times of the ECA and NCS surveys. Comparison
of the NCS and NCS-R prevalence
data can be used to evaluate this prediction.
Finally, we review initial results on individual-level
changes in the prevalence and severity of DSM disorders assessed first in the
baseline NCS in 1990–2002 and then a second time in a re-interview
with the same respondents a decade later
(2001–03) in the NCS follow-up
survey (NCS-2). A more detailed
presentation of these results is reported elsewhere (Kessler et al., 2003).
This part of the analysis addresses a practical problem that mental health policy
analysts have wrestled with ever since the publication of the ECA prevalence
data in the mid-1980s: that the
12-month prevalence of DSM disorders
substantially exceeds the number of people who could be treated for mental or
substance use disorders with current treatment resources. In recognition of
this problem, several more restrictive definitions have been proposed to narrow
the number of people qualifying for treatment. The National Institute of Mental
Health (NIMH) National Advisory Mental Health Council (1993), for example, distinguished
people with severe and persistent mental illness (SPMI) from other mentally
ill people, while the Alcohol, Drug Abuse and Mental Health Administration (ADAMHA)
Reorganization Act stipulated that State mental health Block Grant funds can
be used only to treat people with SMI (ADAMHA, 1992). Many health plans have
followed suit by restricting mental health coverage to a subset of DSM disorders
that they consider to be "biologically based."
Similar restrictions are being discussed to limit the number of people who
qualify for a diagnosis in DSM-IV
(Narrow et al., 2002; Regier, 2000; Regier & Narrow, 2002). The proposal
to restrict the definition of DSM cases in this way has important implications
not only for the definition of current unmet need for treatment, but also for
current research and consideration of future treatment needs. The key fact here
is that research has repeatedly shown that many syndromes currently defined
as mental disorders are extremes on continua that appear not to have meaningful
thresholds (Preisig, Merikangas, & Angst, 2001; Sullivan, Kessler, &
Kendler, 1998). This means that early interventions to prevent progression along
a given severity continuum might reduce the prevalence
of serious cases (Eaton, Badawi, & Melton, 1995). Removal of mild cases
from the DSM system would have the potential to undercut such efforts as well
as to distort the reality that mental disorders, like physical disorders, vary
widely in seriousness (Kendell, 2002; Spitzer, 1998).
To shed some light on this issue, we carried out an analysis of the NCS and
NCS-2 panel data that expanded
on a prior secondary analysis of the NCS (Narrow et al., 2002). In that study,
12-month DSM-III-R
cases in the NCS were divided into those the authors considered either clinically
significant (CSMI) or clinically nonsignificant (CNMI) based on respondent reports
about interference and treatment. Comparison of these two sub-groups
showed, not surprisingly, that various indicators of illness severity (e.g.,
days out of role, history of suicide
attempts) were higher in the former than the latter. The authors concluded from
these results that mild cases should be excluded from DSM-V.
We built on this analysis in two ways. First, we used data from the NCS-2
to examine the associations of baseline NCS 12-month
illness severity with clinically significant outcomes assessed in NCS-2.
Second, we expanded the number of illness severity categories from two to four
by dividing the cases that Narrow (2002) and Regier et al. (1998) defined as
having clinically significant mental illness into severe, serious, and moderate
cases. As described below and presented in more depth elsewhere (Kessler et
al., 2003), differences in the risk of clinically significant outcomes in NCS-2
across these severity categories are as large as, and in some cases larger than,
those between moderate and mild (i.e., CNMI) cases. We also show that the elevated
risk of the NCS-2 outcomes among
mild cases versus noncases is consistently larger than the elevated risk among
moderate cases versus mild cases. These results call into question the suggestion
that the DSM-V case threshold
should be set above CNMI rather than at any other arbitrary point on the severity
gradient.
Methods
Samples
As described in more detail elsewhere (Kessler et al., 2004), the NCS-R
interviewed 9,282 English-speaking
household residents ages 18 and older in a nationally representative sample
of the coterminous United States. Respondents were selected from a multistage
clustered area probability sample. Face-to-face
interviews were carried out between February 2001 and April 2003 by the professional
interview field staff of the Institute for Social
Research at the University of Michigan. The response rate was 70.9 percent.
The survey was administered in two parts. Part I included a core diagnostic
assessment that was administered to all respondents. Part II included
questions about risk factors, consequences, and severity. Part II was administered
to all Part I respondents who met lifetime criteria for any core disorder plus
a probability subsample of other respondents, for a total Part II sample size
of 5,692. We will focus on this Part II sample in the current chapter. This
sample was weighted to adjust for the oversampling of cases from the Part I
sample and for differential probabilities of selection
due to household size and demographic–geographic correlates of response.
We also carried out a nonrespondent survey
in which a subsample of initial nonrespondents was recruited to complete a very
brief screening survey in order to obtain
basic information on several core symptoms of common mental disorders. A final
Part II sample weight was developed based on this nonrespondent survey to adjust
for psychiatric correlates of response.
The NCS-R, as the name implies,
is a replication of the earlier National Comorbidity Survey (NCS) (Kessler et
al., 1994). The NCS was a nationally representative household survey of respondents
ages 15–54 carried out in 1990–92. The response rate was 82.4 percent,
with 8,098 completed interviews. The same two-part
interview was used in the NCS-R
as in the NCS, the main difference in the two samples being that the age ranges
differed. For purposes of trend comparison,
a consolidated data file was created
that combined cases in the overlapping
age range in the two samples (18–54). There were a total of 5,388 Part
II NCS respondents and 4,319 NCS-R
respondents in this age range.
In addition, an attempt was made to re-interview
the 5,877 respondents in the Part II NCS sample in conjunction
with the NCS-R. A total of 5,463
of these baseline respondents were successfully traced, of whom 166 were deceased
and 4,375 interviewed, for a conditional response rate of 76.6 percent. The
unconditional response rate, which takes into consideration the baseline NCS
response rate of 82.4 percent, is 63.1 percent (.766 x .826). NCS-2
respondents differ significantly from other baseline NCS respondents in having
higher probabilities of being female, well educated, and residents of rural
areas (Kessler et al., 2003). A propensity score adjustment weight (Rosenbaum
& Rubin, 1983) was used to correct the NCS-2
sample for these compositional biases.
There was remarkably little difference between NCS-2
respondents and nonrespondents in either the prevalence of baseline NCS disorders
or in the severity of these disorders once these demographic compositional
adjustments were made (Kessler et al., 2003).
Diagnostic Assessment
DSM-IV diagnoses were made in
the NCS-R using the WHO’s
WMH Survey Initiative version of the Composite
International Diagnostic Interview (CIDI) (Kessler & Ustun, 2004) a fully
structured lay-administered diagnostic
interview that generates diagnoses according to the definitions and criteria
of both the ICD-10 (World Health
Organization, 1991) and DSM-IV
(American Psychiatric Association, 1994) diagnostic systems. DSM-IV
criteria are used in the current report, and we focus on respondents with disorders
in the past 12 months. Organic exclusion
rules and diagnostic hierarchy rules were used in making all diagnoses. The
12-month disorders
considered here are anxiety disorders (panic disorder, generalized anxiety disorder,
agoraphobia without panic disorder, specific phobia, social phobia, post-traumatic
stress disorder, obsessive-compulsive
disorder, and separation anxiety disorder), mood disorders (major depressive
disorder, dysthymia, and bipolar disorder I or II), impulse-control
disorders (oppositional-defiant
disorder, conduct disorder, attention-deficit/hyperactivity
disorder, and intermittent explosive disorder), and substance use disorders
(alcohol and drug abuse and dependence).
In addition, a screen was included for non-affective
psychosis (NAP), including schizophrenia, schizophreniform disorder, schizoaffective
disorder, delusional disorder, and psychosis not otherwise specified. As described
elsewhere (Kessler et al., 2004), confirmatory interviews carried out in a probability
subsample of NCS-R respondents
by clinical interviewers found generally good concordance between DSM-IV
diagnoses based on the WMH-CIDI
and clinician assessments of anxiety, mood, and substance use disorders. WMH-CIDI
diagnoses of impulse-control disorders
were not validated because a gold standard clinical interview exists for those
disorders.
In addition to disorder-specific
analyses, we developed a measure of overall disorder severity that expanded
on SAMHSA’s definition of SMI (Substance Abuse and Mental Health Services
Administration, 1993). Respondents with a 12-month
mental disorders were defined as having serious disorder (SMI) if they had at
least one of the following: 12-month
bipolar I disorder or NAP; a 12-month
suicide attempt; at least two areas of role functioning
with self-described "severe"
role impairment on the Sheehan Disability Scales (Leon, Olfson, Portera, Farber,
& Sheehan, 1997); or a pattern of functional impairment at a level consistent
with a Global Assessment of Functioning (GAF) (Endicott, Spitzer, Fleiss, &
Cohen, 1976) score of 50 or less. Respondents who did not meet criteria for
SMI were classified moderate if they had at least one of the following: Bipolar
II disorder; a suicide gesture, plan or ideation; or any other 12-month
DSM-IV disorder with at least
moderate role impairment in at least two areas of role functioning on the Sheehan
Disability Scales. The remaining cases of disorder did not meet the specified
impairments and were classified mild.
The four-category severity gradient was the focus of time trend analyses that
compared prevalence in the NCS-R versus the NCS. The actual prevalence of individual
disorders was not considered in the trend analysis because the NCS diagnoses
used DSM-III-R criteria, versus DSM-IV criteria in the NCS-R. To account for
these changes, a calibration process was used to create comparability in the
prevalence measures. This was done by developing a series of nested logistic
regression equations in the NCS-R that used symptom measures available in both
surveys to predict (a) serious disorder versus all others, (b) serious-moderate
disorder versus all others, and (c) any disorder versus no disorder. Prediction
accuracy was good in all three equations, with area under the receiver operator
characteristic curve of .68 for serious, .84 for serious-moderate, and .81 for
any DSM-IV disorder. The coefficients in these equations were then used to generate
predicted probabilities for each NCS and NCS-R respondent for each nested outcome.
These predicted probabilities were then used to impute dis-crete scores on the
severity gradient. As described in more detail elsewhere (Kessler et al., in
press-a), and briefly described in the next section, the method of Multiple
Imputation (MI) (Rubin, 1987) was used to adjust significance tests for the
imprecision of these imputations.
Analysis Methods
Data analysis was carried out using the Taylor series linearization method
(Wolter, 1985), as implemented in the SUDAAN software system (Research Triangle
Institute, 2002), to adjust for the weighting and clustering of the NCS-R
data. In the case of the time trend analysis, MI was used to adjust for the
imprecision of imputed disorder severity measures. This approach was implemented
by generating ten independent and representative pseudo-samples
from the original NCS-R sample,
with predicted probabilities of severity converted into dichotomous case classifications
based on probability distributions. Uncertainty in classification was reflected
in variation across the 10 imputations and was included in standard errors by
defining the estimated variance of each coefficient as the sum of the average
design-adjusted within-replicate
variance of the coefficient estimate and the variance of the estimated coefficients
across the ten replicates. In the case of logistic
regression, coefficients were exponentiated
to generate odds-ratios (OR) with
95 percent confidence intervals (CIs). Significance
of predictor sets was evaluated with Wald χ2 tests using design-adjusted
MI coefficient variance-covariance
matrices.
Results
Prevalence and Severity
Data on the 12-month prevalence
of core DSM-IV disorders in the
NCS-R, originally reported elsewhere
(Kessler et al., in press b), are presented in table
15.1. Twelve-month prevalence
of any disorder is 26.2 percent, with
somewhat more than half of these cases (14.4 percent) meeting criteria for only
one disorder and smaller proportions for two (5.9 percent) or more (5.9 percent)
disorders. Anxiety disorders are by far the most prevalent class of disorders
(18.2 percent), followed by mood disorders (9.5 percent), impulse-control
disorders (8.9 percent), and substance disorders (3.8 percent). The most common
individual disorders are specific phobia (8.7 percent), social phobia (6.8 percent),
and major depressive disorder (6.7 percent).
Twenty-two percent of respondents
with at least one 12-month disorder
are classified serious, 35.5 percent moderate, and 37.0 percent mild. The remaining
1.3 percent of 12-month cases
are unclassified, as the severity distinction was made only for respondents
with mental disorders, while the table also includes respondents with substance
use disorders. These unclassified cases consist of the respondents diagnosed
exclusively with a substance use disorder. On a base of 26.2 percent of the
population, 22.0 percent serious translates into 5.8 percent of the population
who meet criteria for SMI. Severity is strongly related to number of diagnoses,
with the proportion classified serious ranging from 9.7 percent among respondents
who meet criteria for exactly one diagnosis
to 25.6 percent among those with two diagnoses, and 48.9 percent among those
with three or more diagnoses. The distribution of severity across classes of
disorder is quite different from the distribution of prevalence,
with mood disorders having the highest percentage (44.8 percent) and anxiety
disorders the lowest (22.5 percent) of cases classified serious. Individual
disorders within each class with the highest percentage classified serious are
panic disorder (45.1 percent) among the anxiety disorders, bipolar disorder
(82.9 percent) among the mood disorders, oppositional-defiant
disorder (49.6 percent) among the impulse-control
disorders, and drug dependence (57.3 percent) among the substance use disorders.
The Implications of the Severity
Gradient for Role Functioning
In an effort to provide external validation of the severity ratings, respondents
who met criteria for a given disorder were asked how many days out of 365 in
the past 12 months they were totally unable to work or carry out their other
normal daily activities because of this disorder. To be conservative in combining
these reports across multiple disorders in the subsample of respondents who
met criteria for multiple disorders, we coded such respondents as having a score
equal to their highest score for any single disorder rather than as the sum
of their scores across disorders. A statistically significant gradient (F2,5689
= 17.7, p < .001) with substantial variation across the means was found for
the mean number of days out of role among respondents who differed
in their score on the severity gradient. Respondents classified as having SMI
reported an average of 88.3 days out of role because of their worst mental disorder
diagnosis during the 365 days before interview. This is much higher than the
averages of respondents who are classified moderate (4.7) or mild (1.9).
Sociodemographic Correlates
As shown in table 15.2, significant sociodemographic
correlates of having a core 12-month
DSM-IV disorder in the NCS-R
include young age, female gender, low education, low family income, never married,
previously married, and unemployed-disabled
(compared to the employed). Retired people have significantly lower odds of
12-month disorder than the employed.
With the exception of gender and being retired, all these sociodemographic variables
are also significantly related to SMI among 12-month
cases. In addition, non-Hispanic
blacks with a 12-month disorder
have significantly elevated odds of SMI compared to non-Hispanic
whites. The odds-ratios (ORs)
of these significant sociodemographic variables in predicting SMI in the total
sample are in the range 1.4 (non-Hispanic
black compared to non-Hispanic
white) to 4.1–4.2 (ages 18–29 and 30–44 compared to 60+).
Aggregate Time Trends
Time trend analysis originally reported elsewhere (Kessler et al., in press-a)
found that 12-month prevalence
of any DSM-IV disorder does not
differ significantly between the baseline NCS (29.4 percent) and the NCS-R
(30.5 percent; z = 1.1, p = .271). Table 15.3
presents the distributions for all four categories of the summary disorder
gradient in the two surveys. The NCS-R
severity distribution in this table differs from the distribution in table
15.2 because the trend analysis was carried out only among respondents in
the common age range of the two surveys (18–54). No significant difference
exists between the two surveys in the prevalence of SMI (5.3 percent
in the NCS versus 6.3 percent in the NCS-R;
z = 1.1, p = .271), moderate disorder (12.3 percent in the NCS versus 13.5 percent
in the NCS-R; z = 1.0, p = .298),
or mild disorder (11.8 percent in the NCS versus 10.8 percent
in the NCS-R; z = -0.9,
p = .370). In addition, we carried out analyses that investigated whether significant
statistical interactions existed between time and sociodemographic variables
in predicting prevalence. The motivation
for doing this was the possibility that prevalence might have changed in some
segments of society—possibly even increasing significantly in some segments
and decreasing significantly in others—so that the population-wide
trend was insignificant even though meaningful changes were occurring in important
population segments. As shown in table 15.4,
no evidence was found for such significant subgroup differences in time trends.
The Implications of the Severity Gradient for Future Risk
As reported in more detail elsewhere (Kessler et al., 2003), a consistent
monotonic relationship was found between the illness severity categories in
the baseline NCS and a series of outcomes in the NCS-2
re-interviews that were selected
as indicators of clinically significant outcomes. These outcomes include being
hospitalized for emotional problems at any time in the decade between the two
interviews, being placed on work disability because of emotional problems at
any time in the same interval, making a suicide attempt at any time in the same
interval, and meeting criteria for SMI in the follow-up
interview. Results are reported in table 15.5.
As shown there, a more refined severity gradient was used here than in the aggregate
analyses, which divided cases classified
as having SMI into those with a severe-persistent
mental illness (SPMI) and those with less severe SMI. The operational definition
of SPMI is discussed elsewhere (Kessler et al., 2003).
The largest ORs in the table, associated with SPMI, are in the range 5.6–42.4,
while the smallest ORs, associated with
mild cases, are in the range 1.3–2.7. Three of the five ORs associated
with mild cases are statistically significant at the .05 level. As
table 15.6 shows, 10 statistically significant differences (p < .05,
two-sided tests) out of 20 comparisons
of pair-wise differences in outcomes
are found across contiguous categories of the baseline illness severity gradient.
Important for the purposes of our analysis, the differences between moderate
versus mild cases are consistently smaller than either the differences between
SPMI versus other SMI or other SMI versus moderate. The moderate versus mild
distinction is statistically significant in only one comparison
(predicting SMI in the NCS-2).
The mild versus none distinction, in comparison, is significant in three comparisons
and consistently larger than the moderate versus mild distinction.
Discussion
Several limitations of the NCS family of surveys are relevant to the results
reported in this chapter. First, the samples might underrepresent people with
mental illness either because of sample frame exclusions (e.g., failing to include
homeless people or institutionalized people in the sampling frame) or greater
reluctance of mentally ill than other people to participate in a survey about
mental illness. Evidence for bias of these types has been reported in other
community surveys of mental illness (Allgulander, 1989; Eaton, Anthony, Tepper,
& Dryman, 1992; Kessler, Little, & Groves, 1995), although no evidence
for the nonresponse bias component of this problem was found in NCS or NCS-R
nonresponse surveys (Kessler et al., 1995, 2004). To the extent that downward
bias exists, though, the NCS-R
estimates of 12-month prevalence
and severity are likely to be conservative.
Second, survey participants may underreport 12-month
prevalence. This possibility is consistent with evidence in the survey methodology
literature that embarrassing behaviors are often underreported (Cannell, Marquis,
& Laurent, 1977). Studies of experimental survey methods show that this
problem can be reduced substantially by using strategies aimed at decreasing
embarrassment (Kessler et al., 1998; Turner et al., 1998). As discussed in more
detail elsewhere (Kessler & Ustun, 2004), a number of these strategies were
used in the NCS family of surveys. To the extent that these strategies were
unsuccessful, though, the NCS-R
estimates of 12-month prevalence
and severity are likely to be biased in a conservative direction.
Third, the CIDI is a lay-administered
diagnostic interview rather than a clinician-administered
interview, introducing possible bias
into estimates of prevalence and severity. As reported elsewhere (Kessler et
al., 2004), a clinical reappraisal study in which a probability sample of NCS-R
respondents was blindly interviewed by trained clinicians with the Structured
Clinical Interview for DSM-IV
(SCID) (First, Spitzer, Gibbon, & Williams, 2002) found generally good individual-level
concordance with diagnoses based on the CIDI and also found that CIDI lifetime
prevalence estimates are, for the most part, lower than SCID prevalence estimates.
Fourth, the NCS-R included only
a screen for the diagnoses of schizophrenia and other nonaffective psychoses.
Although these are important disorders, they were excluded from the core NCS-R
assessment because previous validation studies have shown these disorders to
be dramatically overestimated in lay-administered
interviews like the CIDI (Bebbington & Nayani, 1995; Eaton, Romanoski, Anthony,
& Nestadt, 1991; Keith, Regier, & Rae, 1991; Kendler, Gallagher, Abelson,
& Kessler, 1996; Spengler & Wittchen, 1988). These same studies also
showed that the vast majority of respondents with clinician-diagnosed
NAP meet criteria for CIDI anxiety, mood, or substance disorders and are consequently
captured as cases even if NAP is not assessed. It remains possible, though,
that the severity of some such cases are underestimated in the CIDI even if
they are detected as cases, resulting in conservative bias in the estimation
of severity.
Fifth, with regard to the trend analysis, severity was assessed indirectly
with imputation due to the inconsistency of measures in the NCS-2
and NCS-R compared to the earlier
NCS. This introduces the possibility of bias in trend estimates if our assumption
of temporal consistency in the imputation equation coefficients is incorrect.
The strong relationship of imputed values to direct measures of severity in
the NCS-R and the use of MI to
adjust significance tests minimize concern
about the first limitation, but we still have to bear in mind that the trend
analyses must be considered tentative because of this indirect assessment.
Within the context of these limitations, the initial NCS-R
prevalence results reviewed in this chapter are generally
consistent with those of the two previous major psychiatric epidemiological
surveys in the United States, the ECA Survey (Robins & Regier, 1991) and
the NCS (Kessler et al., 1994), in finding that 12-month
mental disorders are highly prevalent. The estimate that 26.2 percent of the
population meets criteria for at least one 12-month
DSM-IV disorder in the NCS-R
is very close to estimates of 28.1 percent in the ECA (Regier et al., 1998)
and 29.5 percent in the NCS (Kessler et al., 1994). In addition, the three most
prevalent 12-month disorders
in the NCS-R (specific phobia,
social phobia, and major depressive disorder) are identical to the three most
prevalent in the baseline NCS. Two of these three were also most prevalent in
the ECA. The exception is social phobia, which was not comprehensively assessed
in the ECA. The findings that 12-month
anxiety disorders, as a class, are more prevalent than mood disorders and that
mood disorders are more prevalent than substance disorders are also consistent
with both the ECA and the baseline NCS.
The externalizing disorders considered in the NCS-R
have not been included in previous epidemiological surveys of adults. Some limited
information is available, however, on the prevalence of intermittent explosive
disorder in the general population (Olvera, 2002), which is consistent with
the NCS-R estimate that 2.6 percent
of the population meets criteria for this disorder in a given year. We are aware
of no independent information on the prevalence of the other impulse-control
disorders among adults—oppositional-defiant
disorder (ODD), conduct disorder (CD),
and attention-deficit/hyperactivity
disorder (ADHD)—although these disorders are routinely assessed in epidemiological
surveys of children and adolescents (Costello, Mustillo, Erkanli, Keeler, &
Angold, 2003; Lahey et al., 2000; Scahill & Schwab-Stone,
2000).
As noted in the section on measures, NCS-R respondents were retrospectively
asked about full criteria for these impulse-control disorders when they were
children and were asked only a single question about 12-month prevalence regarding
whether they still had "any" of the symptoms of the disorder during
that interval. Twelve-month prevalence estimates of these disorders are consequently
estimates of residual adult symptoms and not necessarily of full syndromes.
The 12-month prevalence estimates of ODD and CD are only a small fraction of
the estimates typically found in community epidemiological surveys of youth.
The prevalence estimate of ADHD, in comparison, is approximately 50 percent
as high as the estimates typically found in community epidemiological surveys
of youth. This finding is consistent with independent evidence from follow-up
studies of children treated for ADHD that up to half continue to have the disorder
in adulthood (Pary et al., 2002).
The NCS-R results also support
the conclusion of previous studies regarding the severity of 12-month
disorders that a large proportion of
12-month cases are mild. Indeed,
nearly twice as many 12-month
NCS-R cases are classified mild
(40.4 percent) as are classified serious (22.3 percent). Nonetheless, the 14.0
percent of the population estimated to have a 12-month
serious or moderate DSM-IV disorder
is a substantial proportion. The 5.7 percent of the population estimated to
have a serious 12-month disorder
(.223 x .262, based on results in table 15.1
that 26.2 percent of the sample meet criteria for at least one 12-month
disorder and that 22.3 percent of this 26.2 percent meet criteria for a serious
disorder) is almost identical to the estimated 12-month
prevalence of SMI, using the SAMHSA definition of that term, among 18–54-year-old
respondents in the baseline NCS (Kessler et al., 1996). The finding that mood
disorders are more likely than anxiety disorders to be classified as serious
is consistent with a cross-national
comparative analysis of five earlier CIDI surveys that used a less precise
measure of severity (Bijl et al., 2003), as well as with the result in the more
recent WHO WMH Surveys (Demyttenaere et al., 2004). It is also striking that
impulse-control disorders, which
have not been assessed in previous community epidemiological studies of adult
mental disorders, are found in over one-third
of cases and have a higher proportion classified serious than either anxiety
or substance disorders.
The results regarding sociodemographic correlates are broadly consistent with
those in previous epidemiological surveys
in finding that mental disorders are associated with a general pattern of disadvantaged
social status, including being female, unmarried, having low socioeconomic status,
and being non-Hispanic black (Bland,
Orn, & Newman, 1988; Canino et al., 1987; Demyttenaere et al., 2004; Hwu,
Yeh, & Cheng, 1989; Lee et al., 1990; Lépine et al., 1989; Wells,
Bushnell, Hornblow, Joyce, & Oakley-Browne,
1989; WHO International Consortium in Psychiatric Epidemiology, 2000; Wittchen,
Essau, von Zerssen, Krieg, & Zaudig, 1992). It is not clear whether the
associations of achieved social statuses (i.e., marital status, socioeconomic
status) with risk of disorders are due to effects of environmental experiences
on mental disorders, to effects of mental disorders on achieved social status,
to unmeasured common biological causes, or to some combination. In the case
of the ascribed social statuses (i.e., sex and race), the causal effects clearly
flow from the statuses to the disorders, although the relative importance of
environmental and biological mediators is unclear.
The finding that no change occurred either in the prevalence or in the severity
of mental disorders between the baseline NCS (1990–2) and the NCS-R
(2001–03) is striking, especially in light of independent evidence that
treatment of mental illness increased dramatically during that same period (Wang
et al., in press). Two explanations are
consistent with these results. The first is that prevalence would have been
higher in the early 2000s than the early 1990s were it not for increased treatment.
The second is that the increased treatment over the decade did not cause a decrease
in the prevalence of mental disorders. Consistent with the first possibility,
the economic recession of the early 2000s began shortly before and deepened
throughout the NCS-R field period.
In addition, the 9/11 attacks occurred in the middle of the field period. It
is plausible to think that mental disorders might have been more prevalent at
this time because of these stressors were it not for increased treatment. Consistent
with the second possibility, recent studies have shown that most patients in
treatment for mental disorders receive
treatments that are not consistent with evidence-based
guidelines (Katz, Kessler, Lin, & Wells, 1998; Wang, Berglund, & Kessler,
2000; Wang, Demler, & Kessler, 2002). In addition, as most treatment is
of fairly short duration, we might expect even effective treatment to influence
episode duration more than 12-month
prevalence. This cannot be evaluated directly, though, as episode duration was
not assessed in the NCS.
The findings regarding conditional risk of serious mental health outcomes
in NCS-2 as a function of disorder
severity in the baseline NCS are sobering in that they clearly document the
prognostic significance of mild base-line
disorders. These findings call into question the suggestion that the DSM diagnostic
system should exclude mild cases. This is not to say that more principled considerations,
based on future epidemiological, biological, or taxometric studies, might not
lead to the conclusion that diagnostic thresholds for certain DSM disorders
should be modified upward. Nor is it to say that the problem that motivated
some mental health policy analysts to propose
narrowing the DSM criteria, that the number of people who meet current criteria
is much larger than the number who can be treated with available treatment resources,
is unimportant. However, the solution of defining the problem out of existence
by excluding mild cases from the diagnostic system is ill conceived. The definition
of a case should not be considered synonymous with need for treatment any more
than with clinically significant distress or impairment (Spitzer & Wakefield,
1999). Instead, the problem of unmet need for treatment should be addressed
by developing comprehensive triage rules that allocate available resources based
on evidence-based assessments
of the cost-effectiveness of available
treatments across the severity threshold of the disorder. Severity
gradients are widely used in this way in other branches of medicine (NCEP Expert
Panel, 1993). In the absence if such rules, which currently do not exist, ad
hoc decision-making is inevitable
(Mechanic, 2003). In developing these rules for mental disorders, consideration
should be given not only to current distress and impairment, but also to risk
of progression from mild to more severe disorder. It is unclear whether these
rules, once they are developed, would define treatment of mild cases as cost-effective.
Even if they did not, though, mild cases should be retained in the definition
of disorders both to acknowledge that mental disorders, like physical disorders,
vary in severity and to remind us that the development of cost-effective
treatments for mild disorders might prevent a substantial proportion of future
serious disorders.
Acknowledgements
The National Comorbidity Survey Replication is supported by the National Institute
of Mental Health (U01-MH60220)
with supplemental support from the National Institute on Drug Abuse, the Substance
Abuse and Mental Health Services Administration,
the Robert Wood Johnson Foundation (Grant 044708), and the John W. Alden Trust.
Collaborating NCS-R investigators
include Ronald C. Kessler (Principal Investigator, Harvard Medical School),
Kathleen Merikangas (Co-Principal
Investigator, NIMH), James Anthony (Michigan State University), William Eaton
(The Johns Hopkins University), Meyer Glantz (NIDA), Doreen Koretz (Harvard
University), Jane McLeod (Indiana University), Mark Olfson (New York State Psychiatric
Institute, College of Physicians and Surgeons of Columbia University), Harold
Pincus (University of Pittsburgh), Greg Simon (Group Health Cooperative),
Michael Von Korff (Group Health Cooperative), Philip Wang (Harvard Medical School),
Kenneth Wells (UCLA), Elaine Wethington (Cornell University), and Hans-Ulrich
Wittchen (Max Planck Institute of Psychiatry; Technical University of Dresden).
The views and opinions expressed in this report are those of the authors and
should not be construed to represent the views of any of the sponsoring organizations,
agencies, or U.S. Government. A complete
list of NCS publications and the full text of all NCS-R
instruments can be found at http://www.hcp.med.harvard.edu/ncs.
Send correspondence to ncs@hcp.med.harvard.edu.
The NCS-R is carried out in conjunction with the World Health Organization
World Mental Health (WMH) Survey Initiative. We thank the staff of the WMH Data
Collection and Data Analysis Coordination Centres for assistance with instrumentation,
fieldwork, and consultation on data analysis. These activities were supported
by the National Institute of Mental Health (R01 MH070884), the John D. and Catherine
T. MacArthur Foundation, the Pfizer Foundation, the U.S. Public Health Service
(R13-MH066849, R01-MH069864, and R01-DA016558), the Fogarty International Center
(FIRCA R01-TW006481), the Pan American Health Organization, Eli Lilly and Company,
Ortho-McNeil Pharmaceutical, Inc., GlaxoSmithKline, and Bristol-Myers Squibb.
A complete list of WMH publications can be found at http://www.hcp.med.harvard.edu/wmh/.
The authors appreciate the helpful comments on earlier drafts of William Eaton,
Michael Von Korff, and Hans-Ulrich
Wittchen. Corresponding author and reprints: R. C. Kessler, Ph.D., Department
of Health Care Policy, Harvard Medical
School, 180 Longwood Avenue, Boston, MA 02115 USA. Voice: 617-432-3587;
Fax: 617-432-3588;
E-mail: kessler@hcp.med.harvard.edu.
References
Alcohol, Drug Abuse, and Mental Health Administration. (1992). Alcohol, Drug
Abuse and Mental Health Administration Reorganization Act. PL 102-321.
Allgulander, C. (1989). Psychoactive drug use in a general population sample,
Sweden: Correlates with perceived health, psychiatric diagnoses, and mortality
in an automated record-linkage study. American Journal of Public Health,
79, 1006–1010.
American Psychiatric Association. (1994). Diagnostic and Statistical Manual
of Mental Disorders, 4th ed. Washington, DC: Author.
Bebbington, P. E., & Nayani, T. (1995). The psychosis screening questionnaire.
International Journal of Methods in Psychiatric
Research, 5, 11–19.
Bijl, R. V., de Graaf, R., Hiripi, E., Kessler, R. C., Kohn, R., Offord, D.
R., et al. (2003). The prevalence of treated and untreated
mental disorders in five countries. Health Affairs, 22, 122–133.
Bland, R. C., Orn, H., & Newman, S. C. (1988). Lifetime prevalence of
psychiatric disorders in Edmonton. Acta Psychiatrica Scandinavica, 77,
24–32.
Canino, G. J., Bird, H. R., Shrout, P. E., Rubio-Stipec, M., Bravo, M., Martinez,
R., et al. (1987). The prevalence of specific psychiatric disorders in Puerto
Rico. Archives of General Psychiatry, 44, 727–735.
Cannell, C. F., Marquis, K. H., & Laurent, A. (1977). A summary of studies
of interviewing methodology: 1959–1970. In Vital Health Statistics,
2nd ed., Series 2(69).
Costello, E. J., Mustillo, S., Erkanli, A., Keeler, G., & Angold, A. (2003).
Prevalence and development of psychiatric disorders in childhood and adolescence.
Archives of General Psychiatry, 60, 837–844.
Demyttenaere, K., Bruffaerts, R., Posada-Villa,
J., Gasquet, I., Kovess, V., Lepine, J.P., et al. (2004). Prevalence, severity
and unmet need for treatment of mental disorders in the World Health Organization
World Mental Health (WMH) surveys. Journal of the American Medical Association,
291, 2581–2590.
Eaton, W. W., Anthony, J. C., Tepper, S., & Dryman, A. (1992). Psychopathology
and attrition in the Epidemiologic Catchment
Area Study. American Journal of Epidemiology, 135, 1051–1059.
Eaton, W. W., Badawi, M., & Melton, B. (1995). Prodromes and precursors:
Epidemiologic data for primary prevention of disorders with slow onset. American
Journal of Psychiatry, 152, 967–972.
Eaton, W. W., Romanoski, A., Anthony, J. C., & Nestadt, G. (1991). Screening
for psychosis in the general population with a self-report interview. Journal
of Nervous and Mental Disease, 179, 689–693.
Endicott, J., Spitzer, R. L., Fleiss, J. L., & Cohen, J. (1976). The global
assessment scale. A procedure for measuring overall severity of psychiatric
disturbance. Archives of General Psychiatry, 33, 766–771.
First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2002).
Structured Clinical Interview for DSM-IV Axis I Disorders, Research Version,
Non-patient Edition (SCID-I/NP). New York: Biometrics Research, New York
State Psychiatric Institute.
Hwu, H. G., Yeh, E. K., & Cheng, L. Y. (1989). Prevalence of psychiatric
disorders in Taiwan defined by the Chinese diagnostic
interview schedule. Acta Psychiatrica Scandinavica, 79, 136–147.
Katz, S. J., Kessler, R. C., Lin, E., & Wells, K. B. (1998). Medication
management of depression in the United States and Ontario. Journal of General
Internal Medicine, 13, 137–139.
Keith, S. J., Regier, D. A., & Rae, D. S. (1991). Schizophrenic disorders.
In L. N. Robins & D. A. Regier (Eds.), Psychiatric Disorders in America:
The Epidemiologic Catchment Area Study (pp. 33–52).: New York: Free
Press.
Kendell, R. E. (2002). Five criteria for an improved taxonomy of mental disorders.
In E. Helzer & J. J. Hudziak (Eds.), Defining Psychopathology in the
21st Century DSM-V and Beyond
( pp. 3–17). Washington, DC: American Psychiatric Publishing.
Kendler, K. S., Gallagher, T. J., Abelson, J. M., & Kessler, R. C. (1996).
Lifetime prevalence, demographic risk factors, and diagnostic validity of nonaffective
psychosis as assessed in a US community sample. The National Comorbidity Survey.
Archives of General Psychiatry, 53, 1022–1031.
Kessler, R. C., Berglund, P., Chiu, W. T., Demler, O., Heeringa, S., Hiripi,
E., et al. (2004). The US National Comorbidity Survey Replication (NCS-R): Design
and field procedures. International Journal of Methods in Psychiatric Research,
13, 69–92.
Kessler, R. C., Berglund, P. A., Zhao, S., Leaf, P. J., Kouzis, A. C., Bruce,
M. L., et al. (1996). The 12-month prevalence and correlates of serious mental
illness (SMI). In R. W. Manderscheid & M. A. Sonnenschein (Eds.), Mental
Health, United States, 1996 (pp. 59–70). Washington, DC: U.S. Government
Printing Office.
Kessler, R. C., Chiu, W. T., Demler, O., & Walters, E. E. (in press-b).
Prevalence, severity, and comorbidity of twelve-month DSM-IV disorders in the
National Comorbidity Survey Replication (NCS-R). Archives of General Psychiatry.
Kessler, R. C., Demler, O., Frank, R. G., Olfson, M., Pincus, H. A., Walters,
E. E., et al. (in press-a). US prevalence and treatment of mental disorders:
1990–2003. New England Journal of Medicine.
Kessler, R. C., Little, R. J. A., & Groves, R. M. (1995). Advances in
strategies for minimizing and adjusting for survey nonresponse.
Epidemiologic Reviews, 17, 192–204.
Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman,
S., et al. (1994). Lifetime and 12-month
prevalence of DSM-III-R
psychiatric disorders in the United States. Results from the National Comorbidity
Survey. Archives of General Psychiatry,
51, 8–19.
Kessler, R. C., & Merikangas, K. R. (2004). The National Comorbidity Survey
Replication (NCS-R). International Journal of Methods in Psychiatric Research,
13, 60–68.
Kessler, R. C., Merikangas, K. R., Berglund, P. A., Eaton, W. W., Koretz,
D., & Walters, E. E. (2003). Mild disorders should not be eliminated from
the DSM-V. Archives of General Psychiatry, 60, 1117–1122.
Kessler, R. C., & Ustun, T. B. (2004). The World Mental Health (WMH) Survey
Initiative Version of the World Health Organization
(WHO) Composite International Diagnostic Interview (CIDI). International
Journal of Methods in Psychiatric Research,
13, 93–121.
Kessler, R. C., Wittchen, H.U., Abelson,
J. M., McGonagle, K. A., Schwarz, N., Kendler, K. S., et al. (1998). Methodological
studies of the Composite International Diagnostic Interview (CIDI) in the US
National Comorbidity Survey. International Journal of Methods in Psychiatric
Research, 7, 33–55.
Lahey, B. B., Schwab-Stone, M., Goodman, S. H., Waldman, I. D., Canino, G.,
Rathouz, P. J. et al. (2000). Age and gender differences in oppositional behavior
and conduct problems: a cross-sectional household study of middle childhood
and adolescence. Journal of Abnormal Psychology, 109, 488–503.
Lee, C. K., Kwak, Y. S., Yamamoto, J., Rhee, H., Kim, Y. S., Han, J. H., et
al. (1990). Psychiatric epidemiology in Korea: Part I. Gender and age differences
in Seoul. Journal of Nervous and Mental Disease, 178, 242–246.
Leon, A. C., Olfson, M., Portera, L., Farber, L., & Sheehan, D. V. (1997).
Assessing psychiatric impairment in primary care with the Sheehan Disability
Scale. International Journal of Psychiatry in Medicine, 27, 93–105.
Lépine, J. P., Lellouch, J., Lovell, A., Teherani, M., Ha, C., Verdier-Taillefer,
M. G., et al. (1989). Anxiety and depressive disorders
in a French population: Methodology and preliminary results. Psychiatric
Psychobiology, 4, 267–274.
Manderscheid, R. W., Atay, J. E., Hernandez-Cartagana, M. R., Edmond, P. Y.,
Male, E., Parker, A. C. E., et al. (2001). Highlights of organized mental health
services in 1998 and major national and state trends. In R. W. Manderscheid
& M. J. Henderson (Eds.), Mental Health, United States, 2000 (
pp. 135–171). Washington, DC: U.S. Government Printing Office.
Mechanic, D. (2003). Is prevalence of mental disorder a good measure of need?
Health Affairs, 22, 8–20.
Mechanic, D., & Bilder, S. (2004). Treatment of people with mental illness:
A decade-long perspective. Health
Affairs, 23, 84–95.
Narrow, W. E., Rae, D. S., Robins, L. N., & Regier, D. A. (2002). Revised
prevalence estimates of mental disorders in the United States: Using a clinical
significance criterion to reconcile 2 surveys’ estimates. Archives
of General Psychiatry, 59, 115–123.
National Advisory Mental Health Council. (1993). Health care reform for Americans
with severe mental illnesses: Report of the National Advisory Mental Health
Council. American Journal of Psychiatry, 150, 1447–1465.
NCEP Expert Panel. (1993). Summary of the second report of the National Cholesterol
Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment
of High Blood Cholesterol in Adults (Adult Treatment Panel II). Journal
of the American Medical Association, 269, 3015–3023.
Olfson, M., Marcus, S. C., Druss, B., Elinson, L., Tanielian, T., & Pincus,
H. A. (2002). National trends in the outpatient treatment of depression. Journal
of the American Medical Association, 287, 203–209.
Olvera, R. L. (2002). Intermittent explosive disorder: Epidemiology, diagnosis
and management. CNS Drugs, 16, 517–526.
Pary, R., Lewis, S., Matuschka, P. R., Rudzinskiy, P., Safi, M., & Lippmann,
S. (2002). Attention deficit disorder in adults. Annals of Clinical Psychiatry,
14, 105–111.
Preisig, M., Merikangas, K. R., & Angst, J. (2001). Clinical significance
and comorbidity of subthreshold depression and anxiety
in the community. Acta Psychiatrica Scandinavica, 104, 96–103.
Regier, D. A. (2000). Community diagnosis counts [Commentary]. Archives
of General Psychiatry, 57, 223–224.
Regier, D. A., Kaelber, C. T., Rae, D. S., Farmer, M. E., Knauper, B., Kessler,
R. C., et al. (1998). Limitations of diagnostic criteria and assessment instruments
for mental disorders. Implications for research and policy. Archives of General
Psychiatry, 55, 109–115.
Regier, D. A., & Narrow, W. E. (2002). Defining clinically significant
psychopathology with epidemiologic data. In J. E. Helzer & J. J. Hudziak
(Eds.), Defining Psycho-pathology in the 21st Century: DSM-V and Beyond
(pp. 19–30). Washington, DC: American Psychiatric Publishing.
Research Triangle Institute. (2002). SUDAAN: Professional Software for Survey
Data Analysis [computer program] (Version 8.01). Research Triangle Park, NC:
Author.
Robins, L. N., & Regier, D. A. (Eds.). (1991). Psychiatric disorders
in America: The Epidemiologic Catchment Area Study. New York: Free Press.
Robins, L. N., Wing, J., Wittchen, H.U.,
& Helzer, J. E. (1988). An epidemiologic instrument suitable for use in
conjunction with different diagnostic systems and in different cultures. Archives
of General Psychiatry, 45, 1069–1077.
Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity
score in observational studies for causal effects. Biometrika, 70,
41–55.
Rubin, D. B. (1987). Multiple Imputation for Nonresponse in Surveys.
New York: John Wiley & Sons.
Scahill, L., & Schwab-Stone, M. (2000). Epidemiology of ADHD in school-age
children. Child and Adolescent Psychiatric Clinics of North America, 9,
541–555.
Spengler, P. A., & Wittchen, H.U.
(1988). Procedural validity of standardized symptom questions for the assessment
of psychotic symptoms: A comparison of
the DIS with two clinical methods. Comprehensive Psychiatry, 29, 309–322.
Spitzer, R. L. (1998). Diagnosis and need for treatment are not the same.
Archives of General Psychiatry, 55, 120.
Spitzer, R. L., & Wakefield, J. C. (1999). DSM-IV
diagnostic criterion for clinical significance: Does it help solve the false
positive problem? American Journal of Psychiatry, 156, 1856–1864.
Substance Abuse and Mental Health Services Administration. (1993). Final notice
establishing definitions for (1) children with a serious emotional disturbance,
and (2) adults with a serious mental illness. Federal Register, 58,
29422–29425.
Sullivan, P. F., Kessler, R. C., & Kendler, K. S. (1998). Latent class
analysis of lifetime depressive symptoms in the national comorbidity survey.
American Journal of Psychiatry, 155, 1398–1406.
Turner, C. F., Ku, L., Rogers, S. M., Lindberg, L. D., Pleck, J. H., &
Sonenstein, F. L. (1998). Adolescent sexual behavior, drug use, and violence:
Increased reporting with computer survey technology. Science, 280,
867–873.
Wang, P. S., Berglund, P., & Kessler, R. C. (2000). Recent care of common
mental disorders in the United States: Prevalence
and conformance with evidence-based
recommendations. Journal of General Internal Medicine, 15, 284–292.
Wang, P. S., Demler, O., & Kessler, R. C. (2002). Adequacy of treatment
for serious mental illness in the United States. American Journal of Public
Health, 92, 92–98.
Wang, P. S., Lane, M., Olfson, M., Pincus, H. A., Wells, K. B., & Kessler,
R. C. (in press). Twelve-month use of mental health services in the U.S.: Results
from the National Comorbidity Survey Replication (NCS-R). Archives of General
Psychiatry.
Wells, J. E., Bushnell, J. A., Hornblow, A. R., Joyce, P. R., & Oakley-Browne,
M. A. (1989). Christchurch Psychiatric Epidemiology
Study, part I: Methodology and lifetime prevalence for specific psychiatric
disorders. Australian and New Zealand Journal of Psychiatry, 23, 315–326.
WHO International Consortium in Psychiatric Epidemiology. (2000). Cross-national
comparisons of the prevalences and correlates
of mental disorders. Bulletin of the World Health Organization, 78,
413–426.
Wittchen, H.U., Essau, C. A., von Zerssen,
D., Krieg, C. J., & Zaudig, M. (1992). Lifetime and six-month
prevalence of mental disorders in the Munich Follow-up
Study. European Archives of Psychiatry and Clinical Neuroscience, 241,
247–258.
Wolter, K. M. (1985). Introduction to variance estimation. New York: Springer-Verlag.
World Health Organization. (1991). International classification of diseases
(ICD-10). Geneva, Switzerland:
World Health Organization.
|
|
