|
Methods
In the 1994–1995 NPHS, subjects who were age
18 years or over and resided in Alberta at the time of interview
(n = 1039) provided information about major depression and
how the subjects dealt with stress from unexpected and difficult
problems related to family and personal crises. Table
1 presents the questions related to coping strategies. Subjects
might have offered either a “Yes” or “No” answer
to each of those questions. A “Yes” answer was considered
evidence that the subject used this coping strategy to deal with
unexpected family problems and personal crises in this analysis.
In the NPHS, major depression was measured using the Composite
International Diagnostic Interview-Short Form (CIDI-SF) for major
depression, developed and validated by Kessler and colleagues (15).
Major depression, as defined in the NPHS, represents a 90% predictive
cut-point for the CIDI-SF (14). This cut-point corresponds to reporting
5 of 9 DSM-IV diagnostic criteria for major depression (the number
of symptoms required to make the diagnosis) during the same 2-week
period in the past 12 months, and at least 1 of which must be depressed
mood or loss of interest (16). Therefore, using this cut-point has
its face validity for the DSM-IV criterion A for major depression.
The sensitivity and specificity of the CIDI-SF ranged between 90%
and 94% in studies conducted by Kessler and colleagues (15). However,
the CIDI-SF does not contain probe questions to determine whether
depressive symptoms are due to substance use, physical illness,
and bereavement. The CIDI-SF development and validation documents
showed that organic exclusions were used in the empirical work to
select the scale items and were considered in generating the possibilities
of caseness (http://www.who.int/msa/cidi/cidisfscoring.pdf).
However, a recent validation study using a community sample suggested
the CIDI-SF might pick up a broader spectrum of depressive morbidity
than would major depression as strictly defined using the full version
of CIDI (17). About 75% of subjects scoring 5 or more on the Composite
International Diagnostic Interview-Short Form for Major Depression
(CIDI-SFMD) had major depression according to the full CIDI.
In this analysis, stresses
were identified by identifying those subjects who reported 1 or
more recent life events or chronic stress. The NPHS used 17 questions
to measure chronic stress in 7 dimensions: personal stress, financial
stress, relationship stress (with a partner), relationship stress
(no partner), environmental stress, child problems, and family health
stress. Tables 2 and 3
present the questions used to measure recent life events and chronic
stress. Again, subjects might have offered either a “Yes”
or a “No” answer to each of these questions. In this analysis,
logistic regression models were used to detect interaction effects
between stress and various coping strategies on major depression
prevalence. All analyses were performed in men and in women separately.
Since the NPHS used a complex sampling design, the estimates of
this analysis were weighted to account for sampling procedures.
Because we examined the impacts of 8 different
coping strategies on the association between stress and the prevalence
of major depression, multiple comparison problems presented. To
address this issue, we used Bonferroni’s correction in this
analysis. Specifically, the significance level was set at 0.006
(0.05/8 = 0.00625) in the current analysis. The results of statistical
tests, above this level were considered not significant. Although
Bonferroni’s correction is simple and applicable in any multiple
test situation, the weakness of Bonferroni’s correction is
low power (18). Therefore, Bonferroni’s correction should be
considered a conservative strategy for adjusting the multiple comparison
problem.
Results
To evaluate the moderating
effects of reported coping strategies on the associations between
stresses and major depression, interaction terms involving the specific
stress and coping responses were fitted into a series of logistic
regression models. Table
4 illustrates the significant results with Bonferroni’s
correction. As seen from this table, when exposed to financial stress,
using “pray or seek religious help” as a coping strategy
alleviated the risk of major depression for women. In addition,
when women were exposed to relationship stress (with a partner),
those who reported using “talk to others about the situation”
as a coping strategy were found to have a lower risk of major depression
than did other women.
Table 5
lists the results that were significant at the level of 0.05. The
data suggested that, when subjects were exposed to 1 or more recent
life events, women who reported using “try not to think about
it” or “expressing emotion” as a coping strategy
might have a lower risk of major depression than did other women.
The former strategy could be loosely characterized as avoidance.
The data also suggested that using “talk to others about the
situation” as a coping strategy moderates the risk of major
depression to a lower degree for men when they reported 1 or more
recent life events. Women who used “expressing emotion”
as a coping strategy to deal with personal stress appeared to be
at a lower risk of major depression. This finding was not observed
among men. When exposed to financial stress, using problem-solving
strategies (“try to do something to change the situation”)
as a coping mechanism seemed to decrease the risk of major depression
for men. When exposed to relationship stress (with and without a
partner) and environmental stress, using “expressing emotion”
might decrease the risk of major depression among women. The data
provided no evidence that the reported coping strategies moderated
the risk of major depression for men dealing with relationship and
environmental stresses. One unexpected finding of this part of the
analysis was that reported use of “try to see the situation
differently” as a coping strategy might be associated with
an increased prevalence of major depression for those women who
were not married or without partners when they dealt with relationship
stress.
Discussion
In this analysis, none of the reported coping strategies
were found to have a general moderating effect on the association
between major depression prevalence and all of the stressors evaluated.
This finding suggests people’s coping strategies may differ
with stressful situations, and certain coping strategies may moderate
the effect of specific types of life stress on major depression
risk or prognosis—but not necessarily to the same extent for
both sex groups. This finding is consistent with the current view
that coping strategies are situation- -specific. That is, the same
coping strategy may be helpful or counterproductive, depending on
the dimensions of challenge (19).
|
|
Some individuals may use maladaptive coping styles
to deal with a life event or hassle, which may intensify the depressive
symptoms (3,20).
Using “pray or seek religious help” may
decrease the risk of major depression due to financial stress, and
“talk to others about the situation” regarding relationship
stress may reduce the risk of major depression for women. So far,
only 1 study has investigated the impact of religiosity on the associations
between depressive symptoms and various life stressors (21). In
that study, Strawbridge and others reported that involvement in
organized and nonorganized religious activities buffered the association
between depressive symptoms and nonfamily stressors such as financial
and health problems among elderly people (aged 50 to 102 years)
(21). It is possible that women who have financial stress and who
are religious may not only receive spiritual comforts but may also
receive tangible help from others in the religious group, thus leading
to a decreased risk of major depression. With respect to the effect
of using “talk to others about the situation” as a coping
strategy, Fuhrer and others (22) reported that women have more close
personal relationships than do men, although men have larger social
networks in the Whitehall II study. Therefore, in the NPHS, women
who reported relationship stress with a partner and who use “talk
to others about the situation” as a coping strategy may more
likely share their problems with close friends and be comforted
by these friends.
The mechanisms underlying the sex-specific effects of reported
coping strategies such as “talk to others about the situation”
and “pray and seek religious help” on major depression
prevalence were not clear. Men were less likely than women to report
using these 2 coping strategies in the NPHS. Detailed studies using
more highly developed methods evaluating coping are necessary to
replicate these findings and to delineate why certain coping strategies
affect men and women to different extents in reducing the effects
of life stressors on major depression.
Previous studies have reported that expressing emotion was associated
with an increased level of depressive symptoms for women (9,23).
In a Finnish adolescent sample, Hanninen and Aro (24), found that
using emotional discharge as a coping strategy increased the levels
of depressive symptom (measured by Beck Depression Inventory [25])
for young girls when exposed to life stress. In our analysis, it
was found that using “expressing emotion” as a coping
strategy might moderate the effects of recent life events, personal
stress, relationship stress (with a partner), and environmental
stress on major depression for women, but not for men. These differences
might be due to the instruments used to measure depression, the
definition of emotional expression, and the samples used in the
different studies. In Hanninen and Aro’s study, emotional discharge
included venting anger, self-blaming, smoking, and drinking—and
these items were positively associated with depressive symptoms
(25). Further, we considered that these differences could also be
due to the buffering effect of social support because women were
more likely to use “talk to others about the situation”
as a coping strategy. In fact, a longitudinal study using university
students (26) reported findings consistent with ours, that is, emotion-focused
coping including emotional expression benefitted women in that it
reduced depressive symptom levels when they were exposed to life
stress. Some researchers have concluded that emotional expression
may enhance adjustment when exposed to life stress by facilitating
change in cognitive-affective schema, releasing inhibited physiological
and psychological tension, and motivating adaptive behaviour (27,28).
The moderating effect of “expressing emotion” on the association
between various stresses and major depression was considered not
statistically significant after Bonferroni’s correction in
this analysis. We need to confirm these findings in future studies.
Because the subjects in this analysis consisted exclusively of
those who resided in Alberta at the time of interview, specific
social context might impact on the associations observed in this
analysis. Thus, these findings should be extrapolated with caution.
Moreover, this analysis consisted of subjects who were exclusively
age 18 years and over at the time of the interview. Hence, the results
might not apply to the younger population.
One of the limitations, however, was that this analysis relied
on self-reported information. Subjects might either overreport or
underreport their coping strategies. Nevertheless, the impact of
reporting bias on the results of this analysis could not be examined
directly using the existing data. Another limitation of this analysis
was that the CIDI-SFMD is a brief indicator of major depression
that may not be as sensitive as the full version of CIDI. Unlike
previous studies asking detailed questions about coping strategies,
the NPHS questions about coping strategies are rudimentary in nature.
These questions are relatively crude and could not necessarily capture
or measure all aspects of coping. Consequently, the findings of
this analysis should be considered preliminary.
This analysis was cross-sectional. Causal relations could not be
drawn from this analysis because having a depressive disorder could
impact upon self-reported coping strategies. Longitudinal studies
will be necessary to clarify this and other temporal effects. Although
NPHS contained a longitudinal component, due to a small sample size,
the estimates based on this longitudinal component were not sufficiently
precise to support this type of analysis. Further, the NPHS is a
general-population health survey, and measures sufficiently accurate
to fully address such issues were not included in the interview.
The advantage of reporting the NPHS data is that it is population-based,
providing some insights into the experiences of members of the general
population.
Major depression is a heterogeneous condition influenced by biological,
psychological, and social factors. Recent studies of the efficacy
of brief counselling in the primary care context—particularly
those forms of counselling that focus on enhanced coping—suggest
that these approaches are efficacious, at least for mild depressive
episodes (29–32). We need detailed clinical studies to develop
and refine these strategies. Nevertheless, the data presented here
provide some epidemiological confirmation of the importance of coping
strategies in relation to stress as determinants of mood disturbances
in the population.
|
