Goldman Mellor Saxton Catalano 2010 International Journal of Mental Health PDF

6 INTERNATIONAL JOURNAL OF MENTAL HEALTH
International Journal of Mental Health vol. 39, no. 2, Summer 2010, pp. 631.
2010 M.E. Sharpe, Inc. All rights reserved.
ISSN 00207411/2010 $9.50+0.00.
DOI 10.2753/IMH0020-7411390201
Sidra J. Goldman-Mellor, Katherine B.

Saxton, and Ralph C. Catalano
Economic Contraction and Mental

Health
A Review of the Evidence, 19902009
ABSTRACT: Background: Theory and empirical evidence suggest that

economic contraction predicts increased incidence of psychological disorder.
The extent to which this relation can be causally attributed to the economic
experiences of individuals remains uncertain. Methods: We critically examine
literature concerning the impact of economic contraction, measured at the
individual or ecological level, on four mental health outcomes (depression,
suicide, substance abuse, and antisocial behavior) from the past two decades.
Studies at the individual level use job loss, transition to inadequate employ-
ment, or welfare as the independent variable. Studies at the ecological level
primarily use the unemployment rate. Results: In the studies that best estab-
lish causality, research indicates a moderate but significant adverse effect
of job loss on individual depression symptoms, but the net population effect
remains speculative. For suicide and antisocial behavior, individual- and
ecological-level studies converge to suggest a moderate positive association
with economic contraction. Although some research on substance abuse
suggests procyclical effects, the majority indicate that job loss significantly
increases the risk of heavy drinking and symptoms of alcohol abuse. For all
outcomes, various characteristics of the population or the specific economic
exposure studied can modify the overall association. Conclusions: The studies
reviewed suggest that adverse economic transitions predict increased mental
Sidra J. Goldman-Mellor, MPH, is a doctoral candidate at the School of Public Health, Uni-
versity of California, Berkeley. Katherine B. Saxton, MPH, is doctoral candidate at the School
of Public Health, University of California, Berkeley. Ralph C. Catalano, Ph.D., is Professor
of Public Health at the School of Public Health, University of California, Berkeley.
6
SUMMER 2010 7
health problems, particularly depression, suicide, and substance abuse. The

strength of the association, particularly when measuring the response of
populations to contracting economies remains unclear.
The current economic downturn is among the deepest to strike the international
community since the Great Depression. With so many facing job loss and financial
reversals, we believe it useful to review the accumulated evidence concerning
the relation between economic contraction and mental health. The popular press
frequently reports individual examples of a rise in mental illness associated with
the worsening economy (e.g., [1]), but only future empirical studies can ascertain
whether incidence increased during this period. Much effort, however, has been
expended on investigating the economymental health relation throughout past
decades, and this research can help inform our understanding of what to expect
for the duration of the current recession and beyond.
Mental disorders represent a substantial burden of disease worldwide. A recent
review of the international literature on the economic consequences of mental ill-
ness concludes that direct treatment costs for these illnesses accounts for between 1
percent and 2 percent of total national health-care costs in the United States, Europe,
and the United Kingdom. Once indirect costs are added in, the estimate for total
costs is considerably higher [2]. In addition to the often devastating psychologi-
cal and social effects of mental illness, psychiatric disorders can have significant
consequences for physical health. Depression and other affective disorders have
repeatedly been shown to increase risk for cardiovascular disease and stroke [36],
while alcohol abuse is associated with liver disease and certain kinds of cancer
[7]. In 2001, the World Health Organization highlighted the urgency of finding ef-
fective prevention and treatment for mental disorders and called for a significant
increase in funding for research investigating the causes, courses, and outcomes
of these illnesses [8].
The literature includes many reports suggesting that economic adversity, at both
the individual and population levels, may contribute to the onset of mental disorders.
Although that evidence has been reviewed over the years [915], the magnitude
of the current economic contraction, a wealth of new research, and limitations of
previous reviews provide the impetus for us to comprehensively reexamine the
field. More than 80 original-research studies assessing the relation between various
measures of economic contraction and mental health have been published since
1991 (when the last comprehensive review of the field was published [9]), more
than one-half of them since 1999. In contrast to much of the earlier work, many of
these recent studies have incorporated longitudinal designs, controls for important
confounding variables, and sophisticated statistical techniques that together should
allow for improved estimates of the nature of the association.
We provide a rigorous assessment of the recent literature concerning the impact
of economic contraction on four main mental health outcomes: major depression
and anxiety disorders, suicide, substance abuse, and violence/antisocial behavior.
We focus on studies published since 1990 and primarily discuss those using a
longitudinal design, because such studies provide the best opportunity for elucidat-
ing whether economic contraction leads to poor mental health or vice versa (i.e.,
reverse causation).
We organize this article into several sections starting with a discussion of the
two principal categories into which we have separated the literature: those that use
individuals as the unit of analysis and those that use populations. We then provide
a brief overview of the theoretical models explicit or implicit in the literature. We
next describe our methods for selecting and reviewing the articles included in this
study and, in the subsequent section, we present our findings. We organize the find-
ings into four subsections corresponding to our outcomes of interest, and within
each subsection we separately discuss analyses at the individual and population
levels.
Individual-Based and Ecological Studies
Building on the example of frequently cited reviews of the health effects of

economic insecurity [9, 11], we organize our review using two broadly defined
objective measurements of economic contraction based on the level of analysis. In
this literature, individual-level analyses measure economic contraction as experi-
enced by individual people: therefore, in these studies, the unit of analysis is the
person, and economic contraction refers to job loss, change from employment to
underemployment, transfer into the welfare system, or alternatively, gain of new
employment. In contrast, ecological analyses use population aggregates to measure
the association over time between economic contraction, operationalized primarily
as the unemployment rate, and incidence of a relevant mental health outcome.
Individual and ecological approaches measure different processes, and both
have utility. Public health professionals and clinicians can use the findings from
individual-level studies to identify the mechanisms by which economic contrac-
tion affects mental health and to identify at-risk individuals. Associations found in
individual-level studies may, however, arise from reverse causation. Data typically
used in individual-level studies cannot rule out the possibility that preexisting men-
tal illness caused the adverse change in economic circumstances. Predisposition
to mental illness or abnormal behavior often begins in childhood [16], creating
antecedents that may influence an individuals probability of getting hired, fired,
or responding adversely to economic adversity.
Epidemiologists and psychologists have long been wary of ecological analysis
[17] because they fear that nave readers will infer that an association between two
characteristics of a population (e.g., the incidence of unemployment at time t and
of mental illness at t+n) implies that individuals who experienced the independent
variable (i.e., not working but looking for work) also exhibited the dependent
variable (i.e., mental illness). No fallacy, however, follows if research consumers
SUMMER 2010 9
use ecological analyses to anticipate changes in incidence that predictably follow

changes in unemployment. As described in the following discussion, such estimates
can inform policy formation by anticipating the net effect of mechanisms, trig-
gered by economic contraction, that either increase or decrease the incidence of
the dependent variable.
Ecological and individual studies can, when they yield similar results, comple-
ment each other in that the strength of one compensates for the weakness of the
other. Finding similar associations in individual and ecological studies increases
confidence in the inference that economic contraction causes observed increases in
incidence of illness, because reverse causation and the ecological fallacy cannot
survive both [9, p. 1148].
Theories and Pathways
Theory suggests that economic contraction may impact mental health through
several pathways at the individual level [11]. Economic contraction may directly
affect individuals by putting them out of work, causing them to lose income, and
removing work-based social networks. These losses, in turn, increase psychosocial
and material stress, and they strain nonwork-based social relationships including
those with spouses, partners, and other family. Contraction of the economy may
also have direct effects on persons remaining employed, through restructuring
of job routines and decreased job security and pay, coupled with increased hours
and workload. Economic contraction may act on those not directly affected by
unemployment through indirect effects such as loss of psychosocial assets, through
social contagion pathways, or through decreased community resources. Finally,
economic contraction may indirectly affect family members and dependents of job
and income losers through family disruption or divorce.
Stress and coping theories have often been invoked as mediators of the relation
between economic contraction and mental health. Job loss is often conceptualized
as a stressful life event, challenging an individuals ability to cope, both through
financial strain and through the loss of the psychosocial benefits of work, which
in some cases may induce negative changes in mental health. Aggregate unem-
ployment may also increase stress among those not directly affected, for example,
through changing the structure of work. However, remaining employed during a
period of layoffs may inhibit behaviors considered aberrant, thereby reducing the
risk of both layoff and mental health diagnosis.
One group of theories describes changes in the psychosocial environment associ-
ated with job loss, which then impact the mental health of job-losers. In Jahodas
[18] functional model, employment provides not only the manifest function of
income but also imparts latent functions such as structured time, social contact, and
status. The loss of a job deprives the individual of these latent functions, provoking
distress. Similarly, Warr [19] described environmental features or vitamins that
affect mental health, including social status, income, goals, and opportunity for
control, among others. He claimed that job loss affects mental health by depleting
these vitamins from an individuals environment.
Other theories attribute the effects of job loss to the affected persons understand-
ing of its cause and to subsequent coping mechanisms. For example, frustration
aggression theory [20, 21] has often been invoked to explain the link between
job loss and increased risk of violence [2224]. This theory suggests job loss,
due to slack demand for labor rather than poor performance, creates frustration,
because the experience interferes with goal attainment and may be perceived as
unfair. Such circumstances induce frustration, and a fraction of those who become
frustrated cope through aggression. Therefore, antisocial behavior can be induced
or provoked by job loss. Similarly, alcohol and other substance use can serve as a
coping mechanism in response to stressful experiences. Both initiation and relapse
of substance abuse are thought to be influenced by stress [25].
Rising unemployment also affects those who remain employed. Frustration ag-
gression theory includes inhibitory mechanisms such as expectation of punishment,
which act to reduce deviant behavior, including aggression, substance use, and other
behaviors associated with mental illness. Being laid off is one such punishment, the
threat of which increases internal regulation of behavior. Therefore, among those
remaining employed during economic contraction, the fear of layoffs may inhibit
deviant behavior, such that the incidence of violence, substance abuse, and mental
illness decreases among those who do not lose their jobs [22]. Consequently, the
net effect of economic contraction at the population level may be less than that
expected from increases among job-losers, because of the countervailing decreases
among the employed.
Persons frustrated by job or income loss need not aggress overtly against others.
They may, rather, become less tolerant of the petty or noisome deviance of others
they encounter in the community [26, 27]. For example, the functional capacity of
individuals presented for civil commitment increases when the economy contracts,
indicating a context-dependent threshold for mental health diagnosis [26]. There-
fore, the increase in diagnosed mental illness may not result entirely from increases
in incidence but also from a changing standard of diagnosis or tolerance.
We focus our review solely on the effects of economic events on mental
health, independent of the context of noneconomic stressors. However, reciprocal
relations have been identified, such that the experience of undesirable economic
events increases the likelihood of experiencing subsequent noneconomic stressful
events [28]. This effect suggests that the relation between economic contraction
and mental health may be more complex than most research indicates [28]. Ad-
ditional complications arise when considering the transmission of stress from the
job-loser to family members and spouses. For example, a husbands job loss may
increase psychological distress in his wife as severely as do adverse events that
she experiences directly. Therefore, a focus on job-losers only may underestimate
the mental health effects of economic contraction [29].
SUMMER 2010 11
Methods
The most recent comprehensive literature review in this field that we know of ap-
peared in 1991 [9]. Accordingly, we restrict our review to articles appearing since
1990 and refer the reader to Brenner [30], Catalano [9], Dooley, Fielding, and
Levi [11] and Jin, Shah, and Svoboda [13] for an overview of the research prior to
this date. We identify articles for potential inclusion in the review using reference
tracing from key studies, the Web of Knowledge online citation tracking tool, and
electronic searches of computerized databases using relevant keywords.
Major works used for reference tracing include Catalano [9], Dooley et al. [11]
and Brand, Levy, and Gallo [31]. Key word searches in the online databases Web
of Knowledge, PubMed, and PsycINFO combined the economy-related terms
economic insecurity/security, unemployment, job loss, economy, and layoff with
outcome-specific keywords, including mental health, depression, anxiety, major
depressive disorder, affective disorder, substance abuse, alcohol abuse, violence,
and antisocial behavior. All English-language titles identified during the search pe-
riod (February 2009June 2009) were reviewed. The resulting list was then filtered
through abstract and full-text review to identify the studies of most relevance.
We include studies if they use either (a) ecological-level data with a longitudinal
or repeated cross-sectional design, (b) individual-level data with a longitudinal
design, or (c) cross-sectional data, if the authors presented plausible arguments
for guarding against reverse causation and other significant confounding. Studies
meeting these criteria must have also attempted to measure either true prevalence of
disorder (via a representative sample or ecological-level data) or diagnosed but not
necessarily treated disorder. Intervention studies are only included if they analyze
policy-level interventions (e.g., statewide re-employment programs or unemploy-
ment benefits) as the treatment. We exclude studies that use socioeconomic status
as the exposure of interest, because such research does not address the question
of how contraction of the economy affects mental health. We additionally exclude
studies that do not include a comparison group, have a small sample size (<50),
or do not meet other minimum standards of internal and external validity (includ-
ing consistent and objective measurement of both exposure and outcome across
comparison groups, control for well-established confounding variables, high rate
of follow-up, and minimal selection and information bias).
Two independent reviewers assessed article titles and abstracts to select those
to be included in the review. Articles of indeterminate quality or relevance were
discussed until agreement on inclusion was reached.
Results
Approximately 1,200 studies were identified through PubMed, PsycInfo, and the
other sources described. More than 1,100 studies were excluded because they did
not meet the inclusion criteria, leaving 67 empirical studies included in the final
Table 1
Referenced Research on Economic Contraction and Mental Health
Individual-level analyses Ecological-level analyses
Depression/anxiety [32, 33, 3662]
Suicide [6772] [7380]
Substance abuse [8394] [9598]
Violence/antisocial behavior [99] [22, 23, 100104]
review: 29 on depression/anxiety, 14 on suicide, 16 on substance abuse, and 8

on violence/antisocial behavior (see Table 1). Twenty-eight percent of the stud-
ies used ecological-level longitudinal data, and 72 percent used individual-level
longitudinal data. Study sample sizes for individual-level analyses ranged from
200 to over 2 million.
Depression/Anxiety
The impact of adverse economic experiences on depressed mood is one of the most
consistent findings in the economy-health literature. Dozens of studies have found
statistically significant associations between negative economic transitions and de-
pression, although a few researchers have reported finding no effect [32, 33]. The
great majority of these studies are at the individual level; we identified no ecological-
level articles examining depression that met our inclusion criteria. While findings
from most individual-level studies suffer from the possibility of uncontrolled con-
founding by reverse causation, a few careful and rigorous studies have surmounted
these difficulties and provide reasonably unbiased estimates of effect.
Measuring severity of disorder by self-reported symptoms remains controver-
sial [9], and only psychiatric assessment instruments based on the Diagnostic and
Statistical Manual (DSM) series, such as the Composite International Diagnostic
Interview Schedule, Diagnostic Interview Schedule, and Structured Clinical Inter-
view for DSMIV Axis I Disorders, permit diagnosis of distinct mental disorders.
How conservative researchers should be in their conclusions about the nature of
the economydepression association depends on ones view of the clinical impor-
tance of nonspecific psychological morbidity [34]. While subclinical depression
and anxiety do not generally cause substantial functional impairment [35], such
demoralization is emotionally distressing and may predispose people to other
adverse physical and mental health outcomes. Accordingly, we review studies us-
ing both kinds of psychological assessment instruments and do not substantially
differentiate between the interpretations of their findings.
Individual-Level Studies
Research at the individual level has converged on the finding that economic con-
traction poses a significant risk for depression symptoms. Analyzing two large
SUMMER 2010 13
longitudinal surveys, Burgard, Brand, and House [36] used self-reported data on the
reasons for job losses, as well as the timing of the job losses and acute negative health
events, to discriminate between health-related job losses and other involuntary job
losses. They showed that losing a job for reasons unrelated to health significantly
increased participants risk for depressive symptoms even after controlling for labor
force experience, mental ability, and socioeconomic variables. In a smaller scale
study using a sample of Mexican-Americans, Catalano, Aldrete, Vega, Kolody, and
Aguilar-Gaxiola [37] found that odds of suffering a first-ever episode of clinical
depression in the six months before the survey was more than five times higher
(odds ratio [OR] = 5.7) among those who had lost jobs 7 to 12 months before the
survey. Several other studies have also found evidence of increased depression and
anxiety symptoms resulting from job loss, with ORs averaging somewhat under 2.0,
even after controlling for mental health status at previous survey waves [3843].
However, without detailed information about the chronology of symptoms and un-
employment, these researchers could not rule out the possibility that an unmeasured
intrawave onset of depression or underlying vulnerability to mood disorder caused
the subsequent job loss. While all of these studies did reaffirm previous evidence
that reverse causation is an important factor in shaping labor market engagement
[33, 44, 45], the consistency of their support for a causal relation between economic
contraction and depression is noteworthy.
A variety of more specific permutations of the economydepression relation
have also been investigated. Some researchers have focused on how one type
of economic change (usually job loss) might affect different subgroups, such as
men versus women. We designate this type of study as population characteristics
research. Another group of researchers have focused on how different types of
economic exposures, such as layoffs versus plant closings, might affect a popula-
tion. We designate this type of study exposure characteristics research.
Population characteristics. Many of the early studies in this literature focused
on men, as the researchers were assessing the effects of job losses or layoffs among
blue-collar factory workers, who were chiefly male [9, 13]. However, women are
far more likely to be diagnosed with depression in the general population [35], and
researchers have noted the value of understanding whether job loss affects men and
women differently [40]. Many studies either statistically control for sex, restrict their
analyses to either men or women, or do not examine sex differences even when
possible, apparently assuming that sex is a confounder without testing to determine
whether it does, in fact, confound the results. However, results from several studies
indicate that sex can act as an effect modifier on the effect of economic adversity.
Some authors have found evidence that men are more psychologically affected by
job loss than are women [46, 47]. Brand et al. [31] found that men but not women
were significantly more likely to report depression as a result of being laid off.
These authors suggested that differences in social and economic roles may explain
this variation: Men may have more attachment to the labor force and greater psy-
chosocial needs for re-employment, particularly as they approach retirement, than
do older women. However, the impact of sex appears to be inconsistent. Brand et

al. also found that among workers who lost their jobs as a result of plant closings,
only women reported significantly increased depression symptoms. Likewise,
Mandal and Roe [42], using the same population as Brand et al. but with different
methods and an updated data set, reported that women were more distressed by
any job loss than were men. Many other studies report no heterogeneity between
men and women in analyses of the effect of job loss on psychological distress or
of the effect of depressive symptoms on subsequent employment status [40, 41,
43, 48, 49]. Depressed men may also encounter more intense selection processes
than do depressed women, forcing them to depart more frequently from or remain
out of the labor force [50].
Age has also been suggested as an important factor in the relation between
employment status and mental health. As reviewed by Breslin and Mustard [38],
cross-sectional data indicate that psychological distress brought on by job loss may
be worse in middle-aged and older adults than in young adults, perhaps due to the
more serious financial and family responsibilities borne by older workers. Using
longitudinal data, Breslin and Mustard found support for this hypothesis, report-
ing that becoming unemployed was associated with significantly higher distress
levels and borderline significant levels of clinical depression among older adults.
No relation was found among young adults. A few studies have found that young
adults (generally in school-leaver samples) did not experience increased risk of
depression symptoms with unemployment (e.g., [32]), but others have found that
becoming unemployed significantly predicted symptoms in young people [41, 51,
52]. Longitudinal studies of older adults have consistently reported that losing a
job increases risk for psychological distress [31, 42, 53].
Lastly, a few studies have examined whether race interacts with the economy
depression relation. Both African-Americans and Hispanics have lower rates of
depression and most anxiety disorders compared to whites [54, 55] but are more
likely to be unemployed and to occupy low-paying positions. Catalano et al. [37]
found that Mexican-Americans in a Fresno, California sample were significantly
more likely to be clinically depressed after losing a job, while Rodriguez, Allen,
Frongillo, and Chandra [56] reported that current unemployment was not related
to depression in African-Americans, after controlling for previous depression. The
relation was significant in whites.
Exposure characteristics. Although involuntary job loss is the most widely rec-
ognized form of adverse employment change, persons with unstable or inadequate
employment (involuntary part-time or underpaid employment) make up an increas-
ingly large percentage of the workforce [40, 57] and may also be at risk for psy-
chological distress. Generally, studies have reported that the levels of psychological
distress associated with unstable/inadequate employment lie somewhere between
those linked with satisfactory employment and unemployment. Dooley, Prause, and
Ham-Rowbottom [40], who conducted some of the earliest work in this area using
the National Longitudinal Survey of Youth (NLSY) sample, reported that shifting
SUMMER 2010 15
from stable employment to inadequate employment predicted a similar level of

significantly increased risk for depression symptoms (b = 0.67) as did losing a job
(b = 0.84), even controlling for mediating variables and previous depression [40].
Likewise, Bjarnason and Sigurdardottir [48] reported that moving beyond unem-
ployment to secure employment was associated with significantly less psychological
distress, while moving to temporary employment was associated with an interme-
diate level of distress. Using the British Whitehall cohort, researchers found that
insecure re-employment after job loss predicted increased psychiatric distress (OR
= 1.4), albeit to a lesser extent than did unemployment (OR = 2.2) [58]. Exposure to
chronic job insecurity is often associated with the worst psychiatric outcomes [53,
59]. In a sample of unemployed NLSY adults, however, having found any kind of
employmentadequate or inadequateby a subsequent survey wave was associated
with less depression [50]. Re-employment has been shown to significantly reduce
depression symptoms in a number of other studies as well, including in a series of
studies using a randomized controlled jobs training intervention program [6062].
Such evidence suggests that, in general, inadequate or insecure employment is
detrimental to mental health but that having any kind of employment is protective
compared to the significantly increased risk associated with job loss.
Length of unemployment has proven to be a less consistent factor in predicting
depression symptoms. Any additive effects observed in the long-term unemployed
may result in part from confounding by health selection, as trait characteristics
predisposing an individual to depression would be expected to influence persistently
labor market engagement. Alternatively, financial strain and other stressors intensify
with duration of unemployment, potentially increasing stress-related depressive
symptoms. One study of British men reported that after excluding men with a measure
of preexisting depression tendencies from the sample, accumulated unemployment
remained a statistically significant risk factor for symptoms of anxiety and depres-
sion (risk ratio [RR] = 1.63), although recent unemployment was associated with
the highest risk (RR = 2.10) [51]. In a recent meta-analysis, McKee-Ryan, Song,
Wanberg, and Kinicki [14] found that the long-term unemployed had much lower
levels of well-being than did the short-term unemployed. However, some researchers
have found no relation between length of unemployment and depression symptoms
[38, 52], while others found a protective effect of long-term unemployment, at least
among white men [56]. Such findings may reflect the positive response [14] or even-
tual psychological adaptation [38] to unemployment in a subset of the population.
Most studies exploring the differential effects of distinct types of job displace-
ment (layoffs vs. plant closings) have restricted their analyses to economic or
somatic outcomes or have lacked appropriate control groups, undermining the
validity of their conclusions (see [13] and [31] for reviews). As discussed previous,
Brand et al. [31] examined the effects of these two job displacement types using
the Health and Retirement Survey and found strong effect modification by sex.
The conflicting theoretical rationales and paucity of empirical data on this relation
encourage further research in the area.
Ecological-Level Studies
The dearth of ecological analyses is unfortunate, as it constrains the strength of

conclusions drawn from research on depression and economic contraction. Several
studies of this type were conducted prior to 1990; some found that economic con-
ditions did not predict depressive symptomseither directly or through increased
stressful life eventsand others reporting that economic contraction did predict
symptoms of psychological distress (see [9] for a review). Such inconsistencies
and the general lack of empirical data call out for further research investigation.
The metrics used in measuring depression, however, preclude ecological analyses
of the effects on incidence in the general population, as treated disorder is only
measured at the aggregate level. The net effect of economic contraction on de-
pression symptoms remains unknown at this point.
Conclusions and Limitations
Evidence from the individual-level literature strongly indicates that economic

contraction increases risk for psychological distress, specifically symptoms of
depression. While not consistent in all groups across all exposure types, losing a
job or transitioning to employment in an insecure job appears to increase depressed
mood at twice the rate experienced by those remaining stably employed. This ef-
fect is remarkably robust across dozens of studies. Nevertheless, the magnitude
of the effect is moderate, and many studies estimates are undoubtedly inflated by
unmeasured reverse causation processes. The possibility also exists that a subset
of the population with predisposition to psychological problems is responsible for
much of the observed effect, a hypothesis generally untested.
We offer several additional caveats about findings from the individual-level
literature on depression/anxiety. Most surveys necessarily entail long time intervals
between survey waves, often of two years. As previously noted, this risks misclas-
sification of depression episodes as resulting from, rather than causing, job loss.
Nearly 20 percent of depressive episodes last longer than 24 months and close to
40 percent longer than 6 months [63]; more severe episodes are more likely to pre-
cipitate job termination. It is also possible that the intervals result in underestimates
of the effect of job losses unrelated to health, if short-term depressive episodes are
missed. Lastly, control for mediating factors varies widely among studies, and the
literature offers no consensus on which mechanisms are primarily responsible for
the pathway between economic contraction and depression.
Suicide
As with many negative health outcomes, suicide rates are consistently found to be
higher among the unemployed than among the employed [64, 65]. Unemployment
SUMMER 2010 17
may increase the likelihood of suicide because (a) it precipitates increased risk for
problems such as mental illness, marital stress, or financial strain, which can then
lead to suicide; (b) it heightens vulnerability to suicidal tendencies by magnifying
the impact of other stressful life events; or (c) it is confounded by a third variable
that independently predicts both unemployment and suicide risk. Distinguishing
among these effects has posed a challenge for researchers dating to Durkheim in
the late nineteenth century [66].
Individual-level studies
Longitudinal cohorts have been assembled in a variety of countries (including the

United States, New Zealand, the United Kingdom, Denmark, Italy, and Pakistan) to
assess the role of individual-level economic factors in suicide risk. Such cohorts are
frequently drawn from random samples of government vital statistics records; the
accumulated body of results therefore includes highly representative samples from
diverse populations. Typically, these individual-level data sets include information
on participants sex, age, socio-economic factors, marital status, and employment
status, and more infrequently, history of mental health problems and exposure to
stressful life events. Because multiple measurements of the various risk factors are
only rarely available, competing tests of the three hypotheses previously discussed
are generally impossible.
These studies find that unemployment is a moderate but significant risk fac-
tor for suicide, with reported relative risk magnitudes varying between 1.35 and
3.0. We identify just three individual-level studies that adjust for previous men-
tal health status. In a population-based case-control study of suicide victims in
Denmark, Mortensen, Agerbo, Erikson, Qin, and Westergaard-Nielsen [67] found
that unemployment increased the risk of suicide by 35 percent, after controlling
for previous mental illness serious enough to necessitate hospital admission, as
well as for education, family structure, and income. The authors concluded that
unemployment accounts for 3 percent of suicides, but given that the analyses could
not control for the impact of milder mental health problems, this percentage may
be an overestimate. Blakely, Collings, and Atkinson [68] employed sensitivity
analyses to assess the extent of confounding by mental illness in their population-
based cohort of New Zealand adults. Their best estimate of the relative risk of
suicide for unemployed persons, controlling for mental health, marital status,
and socioeconomic variables, was 1.88. This figure represented a 47 percent
reduction in excess relative risk from the unadjusted estimate. Likewise, results
from a longitudinal study of psychiatric outpatients showed that unemployment
was significantly associated with suicide (adjusted hazard ratio [HR] = 2.56)
[69]. Findings from several other studies support the idea that unemployment
significantly increases risk for suicide, but did not adjust for the mental health
status of study participants [7072].
Aggregate time-series research on suicide and parasuicide (i.e., apparent attempts

to commit suicide that are intentionally or unintentionally unsuccessful) has in-
creased greatly since 1990 and now includes analyses conducted in a wide range
of populations. Improvements in data quality and statistical methods have resulted
in a number of highly rigorous studies. Most of this work adjusts for multiple
confounding variables, including rates of divorce, marriage, birth, alcohol use,
and labor force participation. The research also includes analyses assessing effects
spanning both normal economic cycles and sharp, unexpected economic shocks.
These factorsalong with the diversity of the populations examinedstrengthen
the validity and generalizability of the findings.
Suicide rates have been found to correlate positively with the unemployment rate
or other markers of economic contraction in the United Kingdom [73], the United
States [74], Russia [75, 76], Japan [77], and other Asian countries [78]. Increases in
the suicide rate have been particularly striking in the aftermath of severe economic
crises. In Japan, Hong Kong, and Korea, suicide rates jumped between 39 percent
and 45 percent after the financial crisis of 1998, apparently due in large part to the
resulting steep rise in unemployment rather than to changes in the divorce rate or
other factors [79]. Suicide rates in Russia also sharply increased in the aftermath of
the 1998 economic collapse [75, 76]. This research did not specifically analyze the
unemployment rate, instead hypothesizing that the increase in suicides and other
deaths was due in part to the general economic chaos, including loss of job security,
rapid currency devaluation, hyperinflation, and political and economic uncertainty
[76]. Coefficients for the effect size of unemployment on suicide across normal
economic cycles tend to be smaller but remain significant [73, 74, 77]. Some stud-
ies have, however, reported negative findings as well. Platt, Micciolo, and Tansella
[78] found that change in the suicide rate across 18 regions of Italy was unrelated
to change in the unemployment rate, and Hintikka, Saarinen, and Viinamaki [80]
reported that suicide mortality in both men and women was negatively related to
unemployment in Finland from 19851990. However, unemployment in the Italian
study consistently increased over the study period, leaving no empirical experi-
mentation with which to assess the effects of declining unemployment. Likewise,
the Finnish study period may not have been long enough to draw firm conclusions
about the nature of the unemploymentsuicide relation in that population. All other
studies reviewed used data spanning at least 21 years.
Population characteristics. Analyses that stratify by age and sex allow for more
in depth investigation and demographic-specific conclusions. In general, women
commit suicide less frequently than do men [81], and evidence from both ecological-
level and individual-level data suggest that unemployment is a more potent predictor
of the suicide among men [68, 7375, 77, 78], perhaps due to mens greater at-
tachment to the labor force [38]. However, other results from individual-level data
are more mixed. In Blakely et al.s [68] study, effect modification by sex occurred
SUMMER 2010 19
only among adults between the ages of 18 and 24; unemployed men and women
between the ages of 25 and 64 demonstrated similar magnitudes of increased risk.
Lewis and Sloggett [72] reported no interaction by sex in their estimates of the
unemploymentsuicide association, whereas Kposowa [71] found that unemployed
women were significantly more likely to commit suicide than were unemployed
men (although unemployment predicted suicide risk in both groups).
Additionally, theory suggests that losing a job may be more psychologically
detrimental for those with heavy financial and familial responsibilities, such as
middle-aged adults. Although some research at the ecological level has reported
that suicide rates among working-age adults are indeed more adversely affected
by unemployment than are rates among younger and elderly groups [76, 78], other
studies have found that older adults are more affected [73]. Most individual-level
studies reviewed adjusted for age in their statistical models without examining its
role as an effect modifier, leaving the evidence inconclusive.
Convergence of research at both the individual and ecological levels indicates that
economic contraction acts as a moderately sized but important causal risk factor
for suicide. The consistency and cross-cultural replication of the findings suggest
economic contraction to be nearly universal in its role as a severe psychological
stressor. The pathways between unemployment and suicide remain poorly elucidated
but implicate mental illness. To what extent such mental illness is newly incident
and directly caused by job loss is uncertain, as even the best individual-level data
typically use mental health measurements that predate employment status measure-
ments. Future work should aim to identify these mechanisms better and consider
potential intervention programs in at-risk populations.
Substance Abuse
The literature on the effect of the economy on substance use, while substantial, of-
fers conflicting findings due to different economic exposures studied and different
measures of substance use. The most commonly studied substance has been alcohol,
although several studies have examined the effect of the economy on illicit drug
use. Although alcohol consumption or drug use, particularly at young ages, may
reflect risk of future dependence [82], these outcomes do not necessarily reflect
mental health problems. Therefore, although studies of alcohol consumption are also
included in this review, we draw particular attention to the studies using psychiatric
assessment instruments such as the DSM and the Diagnostic Interview Schedule.
Exposure characteristics. The strongest evidence for the effect of the economy
on substance use comes from individual-level, prospective studies that control
for previous substance use or abuse. Several studies have examined the effect of
different economic transitions, using strong assessment instruments rather than
simply self-reported consumption. Data from the Epidemiologic Catchment Area
Study shows that job loss increases the incidence of clinically significant alcohol
abuse (DSMIII criteria for current alcohol abuse or dependence), controlling for
history of the disorder (OR = 6.05). However, people remaining employed amid
high unemployment are at reduced risk of alcohol abuse. This study found no
effect of loss of income without job loss [83]. Another longitudinal study using
the NLSY found that adverse changes in employment predicted increased heavy
drinking among previous heavy drinkers (OR > 4.0) as well as increased symptoms
of alcohol abuse or dependence (OR = 2.21) [84].
Receiving welfare while working, rather than finding sufficient employment,
may also increase the risk of substance use. Women working and receiving welfare
experience higher odds of heavy drinking, compared to full-time employed women
(OR = 3.37), adjusting for previous alcohol use [85].
Reemployment or favorable employment change has also been studied prospec-
tively. Among a community sample of unemployed adults in Norway, Claussen
[86] found that reemployment was associated with a lower risk of developing
harmful drinking habits (identified using the Alcohol Use Disorder Identification
Test), compared to those remaining unemployed. Having a DSMIII diagnosis of
alcohol disorder at baseline was associated with a strong selection effect, as none
of the individuals with alcohol dependence regained employment during the study.
Favorable employment change, either from unemployment or from underemploy-
ment, was found to have protective effects in the NLSY [87]. Among individuals
with a history of binge drinking, favorable employment change resulted in decreased
odds of binge drinking one year later. Changing from unemployment to underem-
ployment was associated with a 91 percent decreased odds of later binge drinking,
whereas change from unemployment to employment and from underemployment
to employment were associated with 87 percent and 71 percent reduction in the
odds of later binge drinking, respectively.
The relation between unemployment and substance use may not remain constant
over time following job loss. Rather, one study has found that the risk of drinking
correlated positively with time since job loss: Three or more years of involuntary
unemployment increased the risk of heavy drinking, controlling for baseline heavy
drinking and poverty status (OR = 1.47 for 34 years of involuntary unemploy-
ment; OR = 1.66 for 5 or more years) [88]. Another study found that short-term
unemployment predicted reduced alcohol use but that long-term unemployment
led to increased use, although alcohol dependence and alcohol problems did not
correlate with employment status [89].
Population characteristics. Several studies have attempted to identify the ef-
fect of the economy on substance abuse within population sub-groups. To date,
these studies have focused primarily on different age groups. Teenagers and young
adults form one subgroup of interest. Arkes [90] found an effect of the economy on
SUMMER 2010 21
teenagers in the NLSY. The state unemployment rate showed a positive associa-
tion with self-reported marijuana and cocaine use among both men and women
of white and African-American ethnicity, and some evidence supported a coun-
tercyclical effect on alcohol use as well. The economy appeared to affect women
and African Americans most strongly. Teenagers were more likely to sell drugs in
a weak economy as well. However, this study did not assess substance abuse or
dependence, focusing instead on self-reported frequency of use. A study of 17- to
20-year-old Norwegians found no effect of unemployment on volume of alcohol
consumed, adjusting for prejob loss drinking and lifestyle covariates such as friends
drinking behavior. However, frequency of marijuana use increased following un-
employment, adjusting for baseline use [91]. A nationally representative survey in
the United States found increased prevalence of use of cigarettes, marijuana, and
prescription drugs among unemployed 35 year olds, adjusting for substance use
in high school [92]. Although this study, part of the Monitoring the Future Study,
was a panel design, this analysis does not adjust for recent substance use and may
suffer from problems of selection.
Older workers have also received particular attention. Gallo, Bradley, Siegel,
and Stanislav [93], as part of the Health and Retirement Survey, found an increased
risk for the onset of drinking among older workers who suffer involuntary job loss
(OR = 2.01), with most individuals changing from no drinking to light consumption.
Controlling for baseline problem drinking and baseline consumption, Gallo et al.
found no effect of job loss on alcohol consumption. Perreira and Sloan [94], also
using the Health and Retirement Survey, found that retirement increases the risk
of heavy drinking, controlling for history of problem drinking. However, no other
type of employment transition predicted a change in alcohol use.
The ecological studies examining the effect of macroeconomic change on sub-

stance use most commonly identify procyclical effects of the economy on alcohol
consumption, such that a contracting economy results in reduced consumption at
the ecological level. Ruhm [95] found a negative relation between the state-level
unemployment rate and both per capita liquor consumption and vehicle fatalities
over the time period 197588, with hard liquor responding most strongly to the
macro economy. Using improved statistical methods and a time span including
four economic cycles (197095), Freeman [96] reproduced these results several
years later for the outcome of state-level alcohol consumption. These authors
concluded that substance use decreases during economic contraction because of
lowered income.
Two ecological studies have used data from the Behavioral Risk Factor Sur-
veillance System (BRFSS) to examine the relation between economic cycles and
the prevalence of drinking behaviors over time. Because the BRFSS consists of
repeated cross-sectional assessments, these studies can only draw inference at the
ecological level. Ruhm [97] attempted to replicate his earlier findings using the
BRFSS (198799). Despite finding no effect of the macro economy on drinking
participation, Ruhm found that among drinkers the number of drinks consumed
and the probability of drunk-driving showed procyclical variation. Heavy drink-
ers decreased to moderate use during economic downturns, presumably due to an
income effect. Adding unemployment to the model did not change the results. In an
exception to the procyclical results, Dee [98] used BRFSS data from 198495 and
found that economic downturn is associated with a decrease in overall consumption
but an increase in prevalence of binge drinking, especially among those who remain
employed. Dee observed this same effect in all sex, race, and age strata, but the ef-
fect was strongest among men and women ages 1824. The counter-cyclical binge
drinking finding supports the stress mechanism, rather than the income effect.
Economic downturn and job loss often lead to increased substance use at the indi-
vidual level, although some evidence exists of procyclical effects at the population
level. Future studies attempting to provide evidence for an effect of macroeconomic
change on individual behavior should evaluate the effect of changes in the eco-
nomic climate, while holding constant individual-level economic transitions such
as job loss. However, the studies of macroeconomic effects reviewed lack strong
measurements of individual-level economic events and do not follow individuals
over time. In addition, these studies do not explicitly measure alcohol dependence
or symptoms of alcohol abuse, so they may not reflect mental disorder.
The evidence from well-controlled individual-level prospective studies indicates
that individuals suffering involuntary job loss or layoff are at increased risk of
alcohol abuse or increased heavy drinking. These studies also support a selection
effect, in which those with alcohol problems are more likely to be laid off or less
likely to become reemployed. However, favorable employment change shows
promise for prevention of alcohol-related disorders.
Although studies attempt to control for history of substance use and abuse, past
behavior may not fully capture an underlying propensity for substance abuse. Even
some individual-level studies cannot always identify the exact temporal sequence
of job loss and alcohol behavior changefor instance, if both changes took place
between waves of the survey. Therefore, the temporal ordering of exposure (i.e.,
job loss) and changes in substance use must be specified clearly.
Finally, the findings at the individual level and the population level at first glance
appear to contradict one another. However, these studies do not measure the same
outcome: Individual-level studies mostly examine clinically diagnosed alcohol
abuse, while the existent population-level studies typically measure total consump-
tion (which includes nonproblem drinking). Further ecological studies of alcohol
dependence and substance abuse would be useful to ensure comparable outcomes
at both levels of analysis. The countervailing effects of economic contraction on
SUMMER 2010 23
those who lose their jobs versus those who remain employed help to explain this
discrepancy. Although the risk of alcohol use increases among job losers, the
decreased risk among the large proportion of the population remaining employed
would lead to a net decrease in risk at the ecological level. Future studies should
attempt to integrate state- or national-level economic change with individual-level
job transitions for a more robust picture of these relations. In addition, the lack
of focus on mechanisms such as stress, coping, or change in the nature of social
interactions may lead to the lack of consistency even within the individual-level
studies. Therefore, further attempts to measure and identify the mechanisms by
which economic contraction affects substance use will help to clarify the risks.
Violence/Antisocial behavior
Among studies that have investigated the effect of job loss on violence, only one
meets our standard of longitudinal data. Using longitudinal data from the Epide-
miologic Catchment Area (ECA) survey and measuring the outcome with the DIS,
Catalano, Dooley, Novaco, Wilson, and Hough [99] found that suffering a layoff
increased the likelihood of violence among people not violent at initial interview
(OR = 5.97). However, remaining employed in an industry experiencing high
rates of layoffs reduced violent behavior. Finally, they found strong evidence of
selection, as violent people were nearly 16 times more likely to be laid off than
nonviolent people.
Several ecological studies have used time-series methods to explore the relation
between weekly unemployment insurance claims and antisocial behavior, using civil
commitment for danger to others as the outcome. The results indicate a quadratic
relation, providing evidence of provocation effects (i.e., an increase in the incidence
of violence following increased layoffs) and then inhibition effects (i.e., reduced
incidence of violence as layoffs continue to rise) for both men and women in San
Francisco, as well as men in Pennsylvania, with a three-week lag [22, 23]. These
findings were replicated in Florida for men with a one- to three-week lag [100].
Researchers have also hypothesized that child maltreatment responds to eco-
nomic conditions. Ecological studies have found associations between increased
child abuse, neglect, and foster home placements and economic conditions. Cata-
lano, Lind, Rosenblatt, and Attkisson [101] conducted a time-series study of the
relation between the seasonally adjusted unemployment rate in California and the
monthly count of foster home placements in the state and identified both provocation
and inhibition effects. These results imply that rising unemployment increases the
likelihood of foster home placement, but then the likelihood peaks and declines as
unemployment continues to rise. Using state-level fixed effects methods, Paxson

[102] found that a decrease in state welfare benefit levels was associated with an
increased risk of foster home placements. Gillam [103] found positive correlations
between male unemployment and physical abuse of children at the local level, using
an individual-level case-cohort study. Another study also found that children whose
mothers were unemployed for at least 21 weeks in the previous year experienced
increased odds of being hospitalized for abuse or neglect (OR = 1.3) [104].
Both the individual and ecological studies suggest that economic contraction in-
creases the risk of violence among those becoming unemployed. However, evidence
also exists that the inhibition effect may also be at work at high levels of labor
market contraction. Therefore, the net effect of economic contraction on violence
appears smaller than the extrapolated effect on job losers. Future studies should
investigate the effects of economic contraction on other types of violence, such
as domestic violence or arrests for assault, using longitudinal designs. Additional
clarification of the mediators (e.g., stress, changes in social networks or family
structure, alcohol use) of this relation would help advance the field.
Conclusion
We conclude that the contraction of an economy has a detrimental effect on the

mental health of the population it supports. The evidence is particularly robust for
suicide and symptoms of depression, while remaining strong for substance abuse
and moderate for antisocial behavior. Economic contraction appears to significantly
predict risk of these outcomes independent of hypothesized mediators such as
financial insecurity and marital problems, although more research is needed to elu-
cidate mediation pathways. These effects appear mostly stable across demographic
groups, although some indication exists that among job losers, men are at higher
risk for suicide and older adults are at higher risk for depression. Research on all
four outcomes also strongly supports the existence of significant reverse causation
effects, and we caution that many of the studies reviewed may suffer from uncon-
trolled residual confounding from such processes. Findings from a few studies
of depression and substance abuse also suggest that individuals with underlying
predispositions to mental illness are those most likely to report symptoms after
experiencing economic contraction, but additional research that separately examines
these effects in psychiatrically vulnerable and resilient groups is needed.
Despite the overall strength of the studies reviewed in this article, the research
on economic contraction and mental health suffers from notable gaps. The near
total absence of longitudinal ecological-level studies on depression is one such
deficiency. The literature on substance abuse could also benefit from further ecologi-
cal-level studies, while that on antisocial behavior needs additional individual-level
SUMMER 2010 25
research. Additional characteristics, such as baseline socioeconomic status, age,

sex, and availability of unemployment insurance and job training, may modify the
relation between economic contraction and mental health but have not been widely
studied in the literature to date. The common practice of statistically adjusting for
these variables, without examining their impact, has resulted in a lack of theoreti-
cal and empirical evidence on their role in moderating the economymental health
association. Likewise, most reviewed studies did not examine the factors that may
mediate the pathways between economic contraction and mental health outcomes.
In general, those analyses that controlled for other stressful life events that may
result from economic downturn (either on the individual or ecological level),
such as divorce or changes in income, found that job loss or unemployment rates
remained a statistically significant predictor of increased mental health problems,
independent of these other events.
The net effect of economic contraction on populations cannot be reliably esti-
mated from the individual-level findings currently available, and findings at the
ecological level fail to agree on effects because, in part, researchers do not agree on
methods. We, therefore, suggest that more individual-level research focus on how
job loss affects the family members and former coworkers of the job loser. Such
work would allow for better estimation of the net psychological effects of economic
contraction on communities, if not labor markets or political jurisdictions.
Differences in country- or region-specific social and economic programs (e.g.,
adequacy of unemployment insurance, universal access to health insurance), as well
as cultural norms regarding both responses to economic adversity and mental illness,
will inevitably play a role in the impact of economic contraction on mental health.
Such resources and cultural norms unquestionably affect both actual incidence of
mental health outcomes in a population and the formation of relevant policy deci-
sions. Very few of the reviewed studies discuss the generalizability of their findings
given these differences in social and structural resources. Until more research is
devoted to understanding how these resources and cultural values moderate the
effect of economic adversity on mental health, any conclusions as to their impact
must remain speculative.
Importantly, very few studies investigate the psychobiological mechanisms
through which unemployment influences mental health [3]. The consistency of an
observable effect over multiple mental health outcomes suggests that exposure to
economic contraction may precipitate a generalized psychological vulnerability,
which then interacts with individuals specific socio-environmental circumstances
and biological propensities to give rise to a diagnosable condition. A better under-
standing of the physiological basis for economic contractions role in mental health
would assist in the development of interventions and policies designed to alleviate
the psychological suffering associated with job loss and other economic shocks.
The findings reviewed may have important implications for policymakers, par-
ticularly in light of the current economic circumstances. Unemployment rates have
risen considerably during the latest economic crisis and will likely remain high.
The four primary mental health outcomes discussed here are serious conditions that
constitute a significant economic burden, both in terms of health-care costsmore
of which will be shifted onto publicly financed providers as employer-sponsored
health insurance is lost along with employmentand in terms of lost productivity.
Additional research into the effectiveness of existing interventions and programs
would be useful [61, 62].
References
1. Belluck, P. (2009, April 9) Recession anxiety seeps into everyday lives. New York Times.
Available at www.nytimes.com/2009/04/09/health/09stress.html, accessed March 11,
2010.
2. Hu, T.W. (2006) Perspectives: an international review of the national cost estimates
of mental illness, 19902003. Journal of Mental Health Policy and Economics, 9(1),
313.
3. Arnetz, B.B.; Brenner, S.O.; Levi, L.; Hjelm, R.; Petterson, I.L.; Wasserman, J.; et al.
(1991) Neuroendocrine and immunological effects of unemployment and job insecurity.
Psychotherapy and Psychosomatics, 55(24), 7680.
4. Irwin, M., & Miller, A. (2007) Depressive disorders and immunity: 20 years of progress
and discovery. Brain Behavior and Immunity, 21(4), 374383.
5. Kubzansky, L.D.; Cole, S.R.; Kawachi, I.; Vokonas, P.; & Sparrow, D. (2006) Shared
and unique contributions of anger, anxiety, and depression to coronary heart disease:
A prospective study in the Normative Aging Study. Annals of Behavioral Medicine,
31, 2129.
6. Steptoe, A. (2006) Depression and the development of coronary heart disease. In A.
Steptoe (Ed.), Depression and physical illness (pp. 5386). Cambridge: Cambridge
University Press.
7. Pschl, G.; Stickel, F.; Wang, X.D.; & Seitz, H.K. (2004) Alcohol and cancer: Genetic
and nutritional aspects. Proceedings of the Nutrition Society, 63(1), 6571.
8. World Health Organization. (2001) The world health report 2001. Mental health: New
understanding, new hope. Geneva. Available at www.who.int/whr/2001/en/index.html,
accessed March 25, 2010.
9. Catalano, R.C. (1991) The health effects of economic insecurity. American Journal of
Public Health, 81(9), 11481152.
10. Dooley, D., & Catalano, R. (1980) Economic change as a cause of behavioral disorder.
Psychology Bulletin, 87(3), 450468.
11. Dooley, D.; Fielding, J.; & Levi, L. (1996) Health and unemployment. Annual Reviews
of Public Health, 17, 449465.
12. Ezzy, D. (1993) Unemployment and mental health: A critical review. Social Science
and Medicine, 37(1), 4152.
13. Jin, R.; Shah, C.; & Svoboda, T. (1995) The impact of unemployment on health: A review
of the evidence. Canadian Medical Association Journal, 153(5), 529540.
14. McKee-Ryan, F.; Song, Z.; Wanberg, C.; & Kinicki, A. (2005) Psychological and
physical well-being during unemployment: A meta-analytic study. Journal of Applied
Psychology, 90(1), 5376.
15. Murphy, G., & Athanasou, J. (1999) The effect of unemployment on mental health.
Journal of Occupational and Organizational Psychology, 72(1), 8399.
16. Kessler, R.C.; Aguilar-Gaxiola, S.; Alonso, J.; Chatterji, S.; Lee, S.; Ormel, J.; et al.
(2009) The global burden of mental disorders: An update from the WHO World Mental
Health Surveys. Epidemiologia e Psichiatria Sociale, 18(1), 2333.
SUMMER 2010 27
17. Morgenstern, H. (1995) Ecologic studies in epidemiology: Concepts, principles, and

methods. Annual Review of Public Health, 16, 6181.
18. Jahoda, M. (1982) Employment and unemployment: A socialpsychological analysis.
Cambridge: Cambridge University Press.
19. Warr, P. (1994) A conceptual framework for the study of work and mental health. Work
and Stress, 8(2), 8497.
20. Berkowitz, L. (1989) Frustration-aggression hypothesis: Examination and reformulation.
Psychological Bulletin, 106(1), 5973.
21. Dollard, J.; Dobb, L.; Miller, N.E.; Mowrer, O.H.; & Sears, R. (1939) Frustration and
aggression. New Haven: Yale University Press.
22. Catalano, R.; Novaco, R.; & McConnell, W. (1997) A model of the net effect of job loss
on violence. Journal of Personality and Social Psychology, 72(6), 14401447.
23. Catalano, R.; Novaco, R.W.; & McConnell, W. (2002) Layoffs and violence revisited.
Aggressive Behavior, 28(3), 233247.
24. Catalano, R., & Hartig, T. (2004) Economic predictors of admissions to inpatient psy-
chiatric treatment in Sweden. Social Psychiatry and Psychiatric Epidemiology, 39(4),
305310.
25. Brady, K. (1999) The role of stress in alcohol use, alcoholism treatment, and relapse.
Alcohol Res Health, 23(4), 263271.
26. Catalano, R.; Snowden, L.; Shumway, M.; & Kessell, E. (2007) Unemployment and
civil commitment: A test of the intolerance hypothesis. Aggressive Behavior, 33(3),
272280.
27. Hovland, C.I., & Sears, R.R. (1940) Minor studies of aggression. VI: Correlation of
lynchings with economic indices. Journal of Psychology, 9(April), 301310.
28. Catalano, R.; Dooley, D.; & Rook, K. (1987) A test of reciprocal risk between undesirable
economic and noneconomic life events. American Journal of Community Psychology,
15(5), 633651.
29. Rook, K. (1991) Stress transmission: The effects of husbands job stressors on the emo-
tional health of their wives. Journal of Marriage and the Family, 53(1), 165177.
30. Brenner, M. (1973) Mental illness and the economy. Cambridge, MA: Harvard Uni-
versity Press.
31. Brand, J.; Levy, B.; & Gallo, W. (2008) Effects of layoffs and plant closings on subsequent
depression among older workers. Research on Aging, 30(6), 701722.
32. Schaufeli, W., & Vanyperen, N. (1993) Success and failure in the labor market. Journal
of Organizational Behavior, 14(6), 559572.
33. Tolman, R.; Himle, J.; Bybee, D.; Abelson, J.; Hoffman, J.; & Etten-Lee, M. (2009)
Impact of social anxiety disorder on employment among women receiving welfare
benefits. Psychiatric Services, 60(1), 6166.
34. Dohrenwend, B.P.; Shrout, P.E.; Egri, G.; & Mendelsohn, F.S. (1980) Nonspecific psy-
chological distress and other dimensions of psychopathology: Measures for use in the
general population. Archives of General Psychiatry, 37(11), 12291236.
35. Kessler, R.C.; Chiu, W.T.; Demler, O.; & Walters, E.E. (2005) Prevalence, severity,
and comorbidity of 12-Month DSM-IV disorders in the National Comorbidity Survey
Replication. Archives of General Psychiatry, 62(6), 617627.
36. Burgard, S.; Brand, J.; & House, J. (2007) Toward a better estimation of the effect of job
loss on health. Journal of Health and Social Behavior, 48(4), 369384.
37. Catalano, R.; Aldrete, E.; Vega, W.; Kolody, B.; & Aguilar-Gaxiola, S. (2000) Job loss
and major depression among Mexican Americans. Social Science Quarterly, 81(1),
477487.
38. Breslin, F., & Mustard, C. (2003) Factors influencing the impact of unemployment on
mental health among young and older adults in a longitudinal, population-based survey.
Scandinavian Journal of Work, Environment, and Health, 29(1), 514.
39. Dooley, D.; Catalano, R.; & Wilson, G. (1994) Depression and unemployment: Panel
findings from the Epidemiologic Catchment Area Study. American Journal of Community
Psychology, 22(6), 745765.
40. Dooley, D.; Prause, J.; & Ham-Rowbottom, K. (2000) Underemployment and depression:
Longitudinal relationships. Journal of Health and Social Behavior, 41(4), 421436.
41. Hammarstrom, A., & Janlert, U. (1997) Nervous and depressive symptoms in a longi-
tudinal study of youth unemployment: Selection or exposure? Journal of Adolescence,
20(3), 293305.
42. Mandal, B., & Roe, B. (2008) Job loss, retirement, and the mental health of older Ameri-
cans. Journal of Mental Health Policy and Economics, 11(4), 167176.
43. Thomas, C.; Benseval, M.; & Stansfeld, S. (2005) Employment transitions and mental
health: An analysis from the British household survey. Journal of Epidemiology and
Community Health, 59(3), 243249.
44. Mastekaasa, A. (1996) Unemployment and health: Selection effects. Journal of Com-
munity and Applied Social Psychology, 6(3), 189205.
45. Whooley, M.A.; Kiefe, C.I.; Chesney, M.A.; Markovitz, J.H.; Matthews, K.; & Hulley,
S.B. (2002) Depressive symptoms, unemployment, and loss of income: The CARDIA
study. Archives of Internal Medicine, 162(2), 26142620.
46. Graetz, B. (1993) Health consequences of employment and unemployment: Longitudinal
evidence for young men and women. Social Science and Medicine, 36(6), 715724.
47. Virtanen, M. (2008) Temporary employment and antidepressant medication: A register
linkage study. Journal of Psychiatric Research, 42(3), 221229.
48. Bjarnason, T., & Sigurdardottir, T. (2003) Psychological distress during unemployment
and beyond: Social support and material deprivation among youth in six northern Eu-
ropean countries. Social Science and Medicine, 56(5), 973985.
49. Fergusson, D.; Boden, J.; & Horwood, L.J. (2007) Recurrence of major depression in
adolescence and early adulthood, and later mental health, educational and economic
outcomes. Social Science and Medicine, 53(3), 305320.
50. Prause, J., & Dooley, D. (2001) Favourable employment status change and psychologi-
cal depression: A two-year follow-up analysis of the National Longitudinal Survey of
Youth. Applied Psychology: An International Review, 50(2), 282304.
51. Montgomery, S.; Cook, D.; & Wadsworth, M. (1999) Unemployment pre-dates symptoms
of depression and anxiety resulting in medical consultation in young men. International
Journal of Epidemiology, 28(1), 95100.
52. Morrell, S.; Taylor, R.; Quine, S.; Kerr, C.; & Western, J. (1994) A cohort study of un-
employment as a cause of psychological disturbance in Australian youth. Social Science
and Medicine, 38(11), 15531564.
53. Gallo, W.; Bradley, E.; Teng, H.; & Kasl, S. (2006) The effect of recurrent involuntary
job loss on the depressive symptoms of older US workers. International Archives of
Occupational and Environ Health, 80(2), 109116.
54. Himle, J.A.; Baser, R.E.; Taylor, R.J.; Campbell, R.D.; & Jackson, J.S. (2009) Anxiety
disorders among African Americans, blacks of Caribbean descent, and non-Hispanic
whites in the United States. Journal of Anxiety Disorders, 23(5), 578590.
55. Smith, S.M.; Stinson, F.S.; Dawson, D.A.; Goldstein, R.; Huang, B.; & Grant, B. (2006)
Race/ethnic differences in the prevalence and co-occurrence of substance use disorders
and independent mood and anxiety disorders: Results from the National Epidemiologic
Survey on Alcohol and Related Conditions. Psychological Medicine, 36(7), 987998.
56. Rodriguez, E.; Allen, J.; Frongillo, E.; & Chandra, P. (1999) Unemployment, depression,
and health: A look at the African-American community. Journal of Epidemiology and
Community Health, 53(6), 335342.
57. Roberts, B., & Povich, P. (2008) Still working hard, still falling short. Working Poor
SUMMER 2010 29
Families Project. Available at www.workingpoorfamilies.org/still_working.html, ac-

cessed March 25, 2010.
58. Ferrie, J.; Martikainen, M.; Shipley, M.; Marmot, M.; Stansfeld, S.; & Davey Smith, G.
(2001) Employment status and health after privatisation in white collar civil servants:
Prospective cohort study. British Medical Journal, 322(1), 17.
59. Ferrie, J.; Shipley, M.; Stansfeld, S.; & Marmot, M. (2002) Effects of chronic job inse-
curity and change in job security on self-reported health, minor psychiatric morbidity,
physiological measures, and health related behaviors in British civil servants: The
Whitehall II study. Journal of Epidemiology and Community Health, 56(6), 450454.
60. Danziger, S.; Carlson, M.; & Henly, J. (2001) Post-welfare employment and psychologi-
cal well-being. Women and Health, 32(12), 4778.
61. Vinokur, A.D.; Price, R.; & Schul, Y. (1995) Impact of the JOBS intervention on un-
employed workers varying in risk for depression. American Journal of Community
Psychology, 23(1), 3974.
62. Vinokur, A.; Schul, Y.; Vuori, J.; & Price, R.H. (2000) Two years after a job loss: Long-
term impact of the JOBS program on reemployment and mental health. Journal of
Occupational Health Psychology, 5(1), 3247.
63. Spijker, J.; De Graaf, R.; Bijl, R.V.; Beekman, A.T.F.; Ormel, J.; & Nolen, W.A. (2002)
Duration of major depressive episodes in the general population: results from The
Netherlands Mental Health Survey and Incidence Study (NEMESIS). British Journal
of Psychiatry, 181(3), 208213.
64. Moscicki, E.K. (1997) Identification of suicide risk factors using epidemiologic studies.
Psychiatric Clinics of North America, 20(3), 499517.
65. Preti, A., & Miotto, P. (1999) Suicide and unemployment in Italy, 19821994. Journal
of Epidemiology and Community Health, 53(11), 694701.
66. Durkheim, E. (1952) Suicide. London: Routledge and Kegan. (Originally published
1897)
67. Mortensen, P.B.; Agerbo, E.; Erikson, T.; Qin, P.; & Westergaard-Nielsen, N. (2000)
Psychiatric illness and risk factors for suicide in Denmark. Lancet, 355(1), 912.
68. Blakely, T.A.; Collings, S.C.; & Atkinson, J. (2003) Unemployment and suicide. Evi-
dence for a causal association? Journal of Epidemiology and Community Health, 57(8),
594600.
69. Brown, G.; Beck, A.; Steer, R.; & Grisham, J. (2000) Risk factors for suicide in psy-
chiatric outpatients: A 20-year prospective study. Journal of Counseling and Clinical
Psychology, 68(3), 371377.
70. Khan, M.; Mahmud, S.; Karim, M.; Zaman, M.; & Prince, M. (2008) Case-control study
of suicide in Karachi, Pakistan. British Journal of Psychiatry, 193(5), 402405.
71. Kposowa, A. (2001) Unemployment and suicide: A cohort analysis of social factors
predicting suicide in the US National Longitudinal Mortality Study. Psychological
Medicine, 31(1), 127138.
72. Lewis, G., & Sloggett, A. (1998) Suicide, deprivation, and unemployment: record linkage
study. British Medical Journal, 317(7168), 12831286.
73. Gunnell, D.; Middleton, N.; Whitley, E.; Dorling, D.; & Frankel, S. (2003) Why are suicide
rates rising in young men but falling in the elderly? A time-series analysis of trends in
England and Wales, 19501998. Social Science & Medicine, 57(4), 595611.
74. Yang, B. (1992) The economy and suicide: A time-series study of the U.S.A. American
Journal of Economics and Sociology, 51(1), 8799.
75. Gavrilova, N.S.; Semyonova, V.G.; Evdokushkina, G.N.; & Gavrilov, L.A. (2000) The
response of violent mortality to economic crisis in Russia. Population Research and
Policy Review, 19(5), 397419.
76. Men, T.; Brennan, P.; Boffetta, P.; & Zaridze, D. (2003) Russian mortality trends for 1991
2001: Analysis by cause and region. British Medical Journal, 327(7421), 964969.
77. Koo, J., & Cox, W.M. (2008) An economic interpretation of suicide cycles in Japan.
Contemporary Economic Policy, 26(1), 162174.
78. Chang, S.; Gunnell, D.; Sterne, J.; Lu, T.H.; & Cheng, A. (2009) Was the economic
crisis 19971998 responsible for rising suicide rates in East/Southest Asia? A time-trend
analysis for Japan, Hong Kong, South Korea, Taiwan, Singapore and Thailand. Social
Science & Medicine, 68(7), 13221331.
79. Platt, S.; Micciolo, R.; & Tansella, M. (1992) Suicide and unemployment in Italy:
Description, analysis and interpretation of recent trends. Social Science & Medicine,
34(11), 11911201.
80. Hintikka, J.; Saarinen, P.I.; & Viinamaki, H. (1999) Suicide mortality in Finland during an
economic cycle, 19851995. Scandinavian Journal of Public Health, 27(2), 8588.
81. Centers for Disease Control and Prevention, National Center for Injury Prevention and
Control. (2005) Web-based injury statistics query and reporting system (WISQARS).
Atlanta, GA. www.cdc.gov/injury/wisqars/index.html, accessed March 25, 2010.
82. Grant, B.F.; Stinson, F.S.; & Harford, T.C. (2001) Age at onset of alcohol use and DSM-
IV alcohol abuse and dependence: A 12-year follow-up. Journal of Substance Abuse,
13(4), 493504.
83. Catalano, R.; Dooley, D.; Wilson, G.; & Hough, R. (1993) Job loss and alcohol-abuse:
A test using data from the Epidemiologic Catchment Area project. Journal of Health
and Social Behavior, 34(3), 215225.
84. Dooley, D., & Prause, J. (1998) Underemployment and alcohol misuse in the National
Longitudinal Survey of Youth. Journal of Studies on Alcohol, 59(6), 669680.
85. Rodriguez, E., & Chandra, P. (2006) Alcohol, employment status, and social benefits:
One more piece of the puzzle. American Journal of Drug and Alcohol Abuse, 32(2),
237259.
86. Claussen, B. (1999) Alcohol disorders and re-employment in a 5-year follow-up of
long-term unemployed. Addiction, 94(1), 133138.
87. Dooley, D., & Prause, J. (1997) Effect of favorable employment change on alcohol abuse:
One- and five-year follow-ups in the National Longitudinal Survey of Youth. American
Journal of Community Psychology, 25(6), 787807.
88. Mossakowski, K.N. (2008) Is the duration of poverty and unemployment a risk factor
for heavy drinking? Social Science and Medicine, 67(6), 947955.
89. Khan, S.; Murray, R.P.; & Barnes, G.E. (2002) A structural equation model of the effect of
poverty and unemployment on alcohol abuse. Addictive Behaviors, 27(3), 405423.
90. Arkes, J. (2007) Does the economy affect teenage substance use? Health Economics,
16(1), 1936.
91. Hammer, T. (1992) Unemployment and use of drug and alcohol among young people:
A longitudinal study in the general population. British Journal of Addiction, 87(11),
15711581.
92. Merline, A.C.; OMalley, P.M.; Schulenberg, J.E.; Bachman, J.G.; & Johnston, L.D.
(2004) Substance use among adults 35 years of age: Prevalence, adulthood predictors,
and impact of adolescent substance use. American Journal of Public Health, 94(1),
96102.
93. Gallo, W.T.; Bradley, E.H.; Siegel, M.; & Stanislav, V.K. (2001) The impact of involuntary
job loss on subsequent alcohol consumption by older workers: Findings from the Health
and Retirement Survey. Journal of Gerontology: Social Sciences, 56(1), S3S9.
94. Perreira, K.M., & Sloan, F.A. (2001) Life events and alcohol consumption among mature
adults: A longitudinal analysis. Journal of Studies on Alcohol, 62(4), 501508.
95. Ruhm, C.J. (1995) Economic conditions and alcohol problems. Journal of Health Eco-
nomics, 14(5), 583603.
SUMMER 2010 31
96. Freeman, D. (1999) A note on Economic conditions and alcohol problems. Journal
of Health Economics, 18(5), 661670.
97. Ruhm, C.J., & Black, W.E. (2002) Does drinking really decrease in bad times? Journal
of Health Economics, 21(4), 659678.
98. Dee, T.S. (2001) Alcohol abuse and economic conditions: Evidence from repeated
cross-sections of individual-level data. Health Economics, 10(3), 257270.
99. Catalano, R.; Dooley, D.; Novaco, R.W.; Wilson, G.; & Hough, R. (1993) Using ECA
survey data to examine the effect of job layoffs on violent behavior. Hospital and
Community Psychiatry, 44(9), 874879.
100. Kessell, E.R.; Catalano, R.; Christy, A.; & Monahan, J. (2006) Rates of unemploy-
ment and incidence of police-initiated examinations for involuntary hospitalization in
Florida. Psychiatric Services, 57(10), 14351439.
101. Catalano, R.; Lind, S.L.; Rosenblatt, A.B.; & Attkisson, C.C. (1999) Unemployment
and foster home placements: Estimating the net effect of provocation and inhibition.
American Journal of Public Health, 89(6), 851855.
102. Paxson, C., & Waldfogel, J. (2002) Work, welfare, and child maltreatment. Journal of
Labor Economics, 20(3), 435474.
103. Gillham, B.; Tanner, G.; Cheyne, B.; Freeman, I.; Rooney, M.; & Lambie, A. (1998)
Unemployment rates, single parent density, and indices of child poverty: Their relation-
ship to different categories of child abuse and neglect. Child Abuse & Neglect, 22(2),
7990.
104. Christoffersen, M.N. (2000) Growing up with unemployment: A study of parental
unemployment and childrens risk of abuse and neglect based on national longitudinal
1973 birth cohorts in Denmark. Childhood, 7(4), 421438.
To order reprints, call 1-800-352-2210; outside the United States, call 717-632-3535.
Copyright of International Journal of Mental Health is the property of M.E. Sharpe Inc. and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.

Goldman Mellor Saxton Catalano 2010 International Journal of Mental Health PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Goldman Mellor Saxton Catalano 2010 International Journal of Mental Health PDF

Uploaded by

Copyright:

Available Formats

6 INTERNATIONAL JOURNAL OF MENTAL HEALTH

Sidra J. Goldman-Mellor, Katherine B.

Economic Contraction and Mental

ABSTRACT: Background: Theory and empirical evidence suggest that

health problems, particularly depression, suicide, and substance abuse. The

Individual-Based and Ecological Studies

Building on the example of frequently cited reviews of the health effects of

use ecological analyses to anticipate changes in incidence that predictably follow

Theories and Pathways

review: 29 on depression/anxiety, 14 on suicide, 16 on substance abuse, and 8

do older women. However, the impact of sex appears to be inconsistent. Brand et

from stable employment to inadequate employment predicted a similar level of

The dearth of ecological analyses is unfortunate, as it constrains the strength of

Conclusions and Limitations

Evidence from the individual-level literature strongly indicates that economic

Longitudinal cohorts have been assembled in a variety of countries (including the

Aggregate time-series research on suicide and parasuicide (i.e., apparent attempts

Conclusions and Limitations

The ecological studies examining the effect of macroeconomic change on sub-

Conclusions and Limitations

unemployment continues to rise. Using state-level fixed effects methods, Paxson

Conclusions and Limitations

We conclude that the contraction of an economy has a detrimental effect on the

research. Additional characteristics, such as baseline socioeconomic status, age,

17. Morgenstern, H. (1995) Ecologic studies in epidemiology: Concepts, principles, and

Families Project. Available at www.workingpoorfamilies.org/still_working.html, ac-

You might also like