close

Вход

Забыли?

вход по аккаунту

?

The development of depressive symptoms among women with rheumatoid arthritis the role of function.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM Volume 38
Number 1, January 1995, pp 49-56
0 1995, American College of Rheumatology
49
THE DEVELOPMENT OF DEPRESSIVE SYMPTOMS
AMONG WOMEN WITH RHEUMATOID ARTHRITIS
The Role of Function
PATRICIA P. KATZ and EDWARD H. YELIN
Objective. To examine the role of function and
functional changes in the development of depressive
symptoms among women with rheumatoid arthritis (RA).
Methods. Data from a longitudinal panel study of
persons with RA were used. Functional changes were
assessed from 1989 to 1991, and depressive symptoms
during 1992.
Results. Loss of valued activities was a significant
risk factor for developing depressive symptoms (i.e.,
score of 2 7 on the Short Form Geriatric Depression
Scale), but overall functional decline was not.
Conclusion. The aspect of functional decline that
leads to development of depressive symptoms is the loss
of valued activities, not simply the functional impairment.
Depression is an important problem among
persons with rheumatoid arthritis (RA). It appears to
be more prevalent among persons with RA than in the
general population (1-3). Moreover, studies of persons
with RA have demonstrated cross-sectional relationships between depression or depressive symptoms and
impaired functioning and poor clinical status (4-7). For
example, we have found that, among a group of RA
patients, those with depressive symptoms had poorer
function, were more likely to have a major physical
limitation, spent more days in bed, and reported more
joints with pain as well as higher levels of pain (7).
What is it about RA that places individuals with
the disease at risk for depression? In surveys, persons
Supported by grants from the NIH (NIAMS grant AR20684) and the Arthritis Foundation.
Patricia P. Katz, PhD, Edward H. Yelin, PhD: University
of California, San Francisco.
Address reprint requests to Patricia P. Katz, PhD, Arthritis
Research Group, University of California, San Francisco, 1388
Sutter Street, Suite 700, San Francisco, CA 94109.
Submitted for publication May 9, 1994; accepted in revised
form July 14, 1994.
with RA have reported that maintaining physical function or independence is one of their greatest concerns
(8). Given these sentiments, it seems likely that physical disability might increase the likelihood of psychological distress in the form of depressive symptoms.
Indeed, large differences in functional status have
been found between individuals with RA who are
depressed and those who are not depressed (7). Some
studies have found that changes in function are correlated with changes in depressive symptoms (9,lO).
These findings suggest that perhaps worsening function is a risk factor for worsening psychological status,
but could also suggest that worsening psychological
status is a risk factor for worsening function. Because
of the methodology of these studies (correlation of
changes), we do not know whether changes in psychological status preceded changes in function or the
reverse. Other studies that have focused specifically
on the role of physical functioning or disability in the
development of depression have generally suggested
that illness or physical disability precedes the depression (1 1,12); however, these studies have not focused
on persons with RA.
Our goal was to expand on previous studies and
examine the role of function in the development of
depressive symptoms among a group of women with
RA. Specifically, we examined two types of function
and functional decline, and their roles as risk factors
for the development of depressive symptoms. We
defined a risk factor as a condition or event that clearly
preceded the development of depressive symptoms.
The identification of risk factors in this manner is more
complex than the identification of simultaneously occurring changes, but it is clinically important, since the
findings may lead to earlier detection, or perhaps even
actual prevention, of depressive symptoms.
KATZ AND YELIN
50
PATIENTS AND METHODS
This was a 4-year study (from 1989 through 1992) of
a group of women with RA. We examined the influence of
baseline characteristics (at the start of 1989), as well as
changes in these characteristics from 1989 to 1991, on the
development of depressive symptoms during the fourth year
of observation (i.e., 1992). Women who exhibited depressive
symptoms prior to 1992 were excluded from analysis.
Study sample. The study drew upon data from the
community-based Rheumatoid Arthritis Panel Study at the
University of California, San Francisco (UCSF). (Complete
details of the selection and maintenance of the RA Panel
Study are published elsewhere [13].) Panel members were
interviewed annually by telephone and were questioned
about symptoms, disease history, function, comorbid conditions, medications, and sociodemographic characteristics.
Measures of psychological status were also included. In
1992, the year in which outcomes were assessed for these
analyses, 554 panel members were interviewed.
Outcome variable. The presence of a high level of
depressive symptoms in 1992 was the dependent, or outcome, variable. Depressive symptoms were assessed with
the Short Form of the Geriatric Depression Scale (S-GDS)
(14,15). The GDS has been shown to be a valid and reliable
measure of depressive symptoms among younger adults as
well as older adults, for whom it was originally developed,
and has been used extensively in a variety of settings
(16-19). It has been found to have a very high correspondence to clinical diagnoses of depression (17,18). The 15item S-GDS correlates well with the original 30-item GDS,
and has also been used in younger populations as well as
among the elderly (20).
When responding to the GDS, individuals are asked
to base their answers on their feelings over the previous
week. (This reference period is the same as for other
commonly used measures of depressive symptoms, such as
the Center for Epidemiological Studies Depression Scale and
the Beck Depression Inventory [21].) The GDS has a “yes/
no” response format, which was intended to simplify responding. The GDS also contains fewer items concerning
somatic symptoms of depression than most commonly used
screening instruments. This is important in assessing depressive symptoms among persons with RA, since some of the
somatic symptoms of depression are also symptoms of RA
(22). It has been recommended that a cut-point of 7 on the
S-GDS be used to define the presence of high levels of
depressive symptoms (23). Thus, individuals whose S-GDS
scores were 7 or greater were classified as exhibiting high
levels of depressive symptoms, and were referred to as
“depressed.” (This does not constitute a clinical diagnosis
of depression.) Individuals whose scores were below 7 were
classified as not exhibiting depressive symptoms, and were
referred to as “not depressed.”
Independent variables. Functional status was the
independent variable of primary interest in these analyses.
Other independent variables tested for potential use as
covariates were sociodemographic and clinical characteristics (including baseline depressive symptoms) that have been
found to be different among individuals who are and are not
depressed or that have been shown to be risk factors for the
development of depression. Both baseline Characteristics
and changes over a 3-year period were examined. Baseline
characteristics reflected information from the 1989 interview. Changes were assessed in 1-year increments; that is,
changes from 1989 to 1990, and from 1990 to 1991.
Functional status. Function has been found to be
worse among persons with depressive symptoms, and was
the focus of these analyses. Two baseline measures of
functional status were examined. The first was the individual’s score on the Health Assessment Questionnaire (HAQ)
in 1989. The HAQ score was developed specifically to assess
basic function, similar to activities of daily living (ADL),
among persons with arthritis (24). Scores range from 0 (no
functional impairment) to 3.0 (severe impairment), in increments of 0.125.
The second measure of functional status reflected the
degree to which persons were able to perform valued activities. In 1987, a measure of life activities was developed to
determine the impact of arthritis (25). Most of these activities were things that one would do daily or frequently, such
as working, cooking, shopping for food, making minor
household repairs, visiting with family, getting around the
neighborhood, being involved with hobbies and crafts, and
going away on vacation. This measure of activity performance was incorporated into the UCSF Panel Study interview in 1989. Individuals were presented with the list of 75
life activities and asked if they had participated in each
activity in the previous 6 months. They were also asked to
rate the importance of being able to perform each activity,
according to a scale of 1 (very important) to 5 (not at all
important).
For the present study, an activity was defined as
important to the individual if she rated it as 1 or 2. We have
previously found that the most striking difference in personal
activities between persons with RA who had high levels of
depressive symptoms and those who did not was in the
proportion of activities rated as important that the respondent actually performed (26). Persons with RA who exhibited high levels of depressive symptoms performed -12%
fewer of their valued activities than did those who exhibited
low levels of depressive symptoms. Thus, for the current
study, the second baseline measure of functional status
examined was the proportion of valued activities that the
individual performed in 1989.
The changes in function we examined were a decline
in basic function and a decline in the proportion of important
activities performed. The decline in basic function was
reflected by an increase in the HAQ score. Three magnitudes
of increase in HAQ score were examined: any increase (on
the 0-3 scale), an increase of 20.25, and an increase of
20.50. A small increase in HAQ scores may simply reflect
day-to-day fluctuations in an individual’s functional status.
The 2 larger increases in HAQ would represent more substantial functional declines, which might have greater impact
on the individual.
The second change in function was defined as a
210% decrease in the proportion of activities rated as
important that the individual actually performed.
Sociodemograpic features. Depression has been
found under some conditions to be more common among
individuals who are older, non-white, less educated, living
51
DEPRESSIVE SYMPTOMS IN RA PATIENTS
alone, have lower incomes, or who have experienced job
loss or the loss of a spouse (27-29). Therefore, baseline (in
1989) sociodemographic characteristics examined for this
study were: (a) age, (b) race (white versus non-white), (c)
years of formal education, (d) marital status (married versus
not married), (d) living arrangements (living alone versus
living with someone), (e) family income, and (0 work status
(working versus not working).
We examined the following changes in sociodemographic characteristics: (a) change in marital status, from
married to not married; (b) change in living arrangements,
from living with someone to living alone; (c) decrease in
family income; and (d) change in work status, from working
to not working.
Clinical features. Individuals who have depressive
symptoms report greater pain, more severe RA, and RA of
longer duration (7). Thus, the baseline clinical characteristics examined were the duration of RA, the number of
painful and swollen joints (by self-report, using a checklist
for the specific joints or joint groups with pain or swelling
during the previous year; data were summed), and overall
RA pain rating (from the HAQ [24]; 0 = no pain and 100 =
very severe pain).
The clinical changes tested for their predictive ability
were: an increase in the number of painful or swollen joints
(1989 versus 1990 and 1990 versus 1991); an increase in
overall RA pain rating (HAQ scores in 1989 versus 1990 and
in 1990 versus 1991); and patients' assessments in 1990 and
1991 of whether their RA pain, joint swelling, or overall RA
was worse than the previous year.
Inclusion criteria. Only individuals whose S-GDS
scores were below 7 (low levels of depressive symptoms) in
all of the study years prior to 1992 (1989, 1990, and 1991)
were considered for this study. Most research has found
depression to be a rather stable phenomenon. By excluding
individuals with high levels of depressive symptoms in the 3
years preceding 1992, it was more likely that high levels of
depressive symptoms in 1992 were of relatively new origin
and not a recurrence.
Among this group, the number of men whose S-GDS
scores were above 7 in 1992 (n = 4) was too small for
analysis; thus, our analyses include only women who exhibited low levels of depressive symptoms in the 3 years prior to
1992 (n = 319).
Statistical analysis. The purpose of the analysis was
to assess the role of function and functional change in the
development of depressive symptoms, controlling for appropriate sociodemographic and clinical characteristics. Functional characteristics were assessed at baseline (in 1989) and
yearly thereafter, and changes in function were assessed
from 1989 to 1991. Outcome-the presence of a high level of
depressive symptoms-was assessed in 1992. Women with
depressive symptoms in any of the study years preceding
1992 were excluded from analysis. This method permitted
the clear identification of the time-ordering of functional
changes and the development of depressive symptoms.
Because the number of women who developed depressive symptoms in 1992 was relatively small, statistical
considerations required a relatively small number of covariates for the multivariate analysis. Bivariate logistic regression analyses of each sociodemographic and clinical charac-
Table 1. Baseline characteristics of the 319 women with rheumatoid arthritis (RA)
Characteristic
Sociodemographic
Age, years
% white
% mamed
% living alone
% working
Education, years
Family income, $
Clinical
Disease duration, years
No. of comorbidities
No. of painful joints (maximum 18)
No. of swollen joints (maximum 14)
Overall RA pain rating (maximum LOO)
Function
Health Assessessment
Questionnaire score (maximum 3)
% important activities
performed (see ref. 25)
Short Form Geriatric Depression Scale
score (maximum 15)
Mean
?
SD, or %
56.5 t 13.0
80.6
64.1
22.5
60.0
13.1 2 2.7
46,254 t 38,420
16.6 ?
0.4 ?
9.5 ?
5.0 ?
36.2 ?
9.3
0.7
4.5
3.0
27.0
1.09
?
0.61
83.9
5
12.4
1.7
?
1.7
teristic (baseline and changes) were conducted, assessing the
characteristic's association with the outcome. The sociodemographic and clinical characteristics that were significantly associated with the outcome were identified from the
bivariate analyses. To further reduce the number of covariates, a multivariate logistic regression was performed, regressing the outcome on the 9 variables for which significant
bivariate results had been noted. The chi-square statistic for
6 of these variables was very low in the multivariate results
(below 2.0), and these variables were excluded from further
analysis. The 3 sociodemographic and clinical variables
remaining, which were later used as covariates in the analyses of function, were disease duration, baseline number of
painful joints, and change in marital status between 1989 and
1990.
For the primary analyses, functional characteristics
were added to the covariates in 3 ways, in different regression models, to assess the differential importance of the 2
types of functional variables as risk factors for depressive
symptoms. In the first model, the functional variables based
on the HAQ score (baseline HAQ and changes in HAQ)
were added. In the second model, the functional variables
based on valued activities (proportion of valued activities
performed at baseline and loss of valued activities) were
added. In the third model, both HAQ and valued activity
variables were added to determine the relative importance of
the 2 types of functional assessments.
RESULTS
Table 1 summarizes of the characteristics of the
study group. The mean age of the group was 56.5
years, and the majority were white (80.6%), married
(64.1%), and working (60.0%). Relatively few were
KATZ AND YELIN
52
Table 2. S-GDS scores for depressive symptoms over the 4-year
study period*
S-GDS score
Year
1989
1990
1991
1992
Change from
1991 to 1992
Low level of depressive High level of depressive
symptoms in 1992
symptoms in 1992
(n = 298)
(n = 21)
1.58 I
1.64
1.54 ? 1.64
1.45 f 1.53
1.57 ? 1.52
0.12 f 1.49
2.57 2 2.01
2.81 f 1.83
3.00 ? 1.70
8.24 f 1.41
5.24 ? 2.12
* Values are the mean f SD. S-GDS = Short Form Geriatric
Depression Scale (maximum score 15).
living alone (22.5%). The average duration of RA was
>I6 years, and the group had few comorbidities.
On average, the women reported 9.5 painful
joints (from a possible 18) and 5.0 swollen joints (from
a possible 14). The overall RA pain rating was moderate (36.2 from a possible 100). Baseline HAQ scores
suggested a moderate level of functional impairment.
In general, the women were able to perform a relatively high proportion of the activities they rated as
important to them (83.9%). Baseline S-GDS scores
were low, with a mean score of 1.7.
Twenty-one (6.6%) of the women developed
high levels of depressive symptoms in 1992. Although
the depressed group had significantly higher S-GDS
scores during the first 3 years, -1-1.5 points higher,
their scores were still considerably below the cut-point
of 7 during that time (Table 2). In 1992, however, the
difference in the scores of the 2 groups was close to 7
points. No specific depressive symptoms appeared to
be more prevalent among the depressed group; significant differences were seen between the 2 groups on 14
of the 15 S-GDS items. Except for the dramatic
increase in the depressed group’s mean score in 1992,
S-GDS scores were relatively stable over time.
Among the women who were not in the depressed group in 1992, there were no significant yearto-year changes in S-GDS scores over the 4 years.
Among those in the depressed group, there were no
significant year-to-year changes over the first 3 years.
However, from 1991 to 1992, the average increase in
S-GDS score was >5 points (P< 0.0001). Twenty of
the 21 women in the depressed group had increases of
at least 3 points.
At baseline, women who subsequently developed depressive symptoms had significantly higher
HAQ scores, indicating poorer ADL-type function
(Table 3). There were no significant differences, however, in the proportion of valued activities performed
at baseline by the women who did and did not develop
depressive symptoms. Functional declines, measured
both by increases in HAQ scores and by losses of
valued activities, were more common among the
women who became depressed in 1992 than among
those who did not.
Of the 21 women who developed high levels of
depressive symptoms, 14.3% (n = 3) had experienced
a preceding increase in HAQ score of 20.50, and
38.1% (n = 8) had experienced a preceding loss of at
least 10% of their valued activities. In contrast, among
the women who did not develop depressive symptoms
in 1992,5.7%had experienced an increase in HAQ and
8.7% had experienced a loss of valued activities. (The
proportions of women with and without depressive
symptoms whose HAQ scores increased by <0.50
were not sufficiently different to warrant further analysis. For all analyses shown, an increase of 20.50 in the
HAQ score was used to define “decline in basic
function. ”)
The association of functional characteristics
and the development of depressive symptoms was
examined in 3 multivariate models. Each model consisted of the 3 covariates described above (disease
duration, number of painful joints at baseline, and
change in marital status between 1989 and 1990), plus
terms describing function and functional decline. In
the first, the functional variables related to the HAQ
score were added. The resulting model (Model 1) is
shown in Table 4. The baseline HAQ score was not
significant in the model (odds ratio [OR] = 1.84, 95%
confidence interval [95% CI] 0.80-4.24, P = 0.15);
however, an increase in the HAQ score of at least 0.50
between 1990 and 1991 was significant. After controlTable 3. Functional variables at baseline (1989) and changes in
functional variables in rheumatoid arthritis patients with and without depressive symptoms in 1992*
Depressive symptoms in 1992
No
(n = 298)
*
1.06 ? 0.60
Baseline HAQ score, mean SD
% with an increase in HAQ score of
5.7
20.50 between 1991 and 1992
Baseline proportion of valued
85.3 ? 11.4
activities performed, mean ? SD
% with a loss of 210% of valued
8.7
activities
* HAQ = Health Assessment Questionnaire.
Yes
(n = 21)
1.49 f 0.73
14.3
78.8
?
18.2
38.1
DEPRESSIVE SYMPTOMS IN RA PATIENTS
53
Table 4. Logistic regression models examining the effect of decline in basic function and loss of
valued activities on the development of depressive symptoms*
Variable
~
Standard
error
of beta
Odds
ratio
95% CI
0.4258
0.7012
1.84
3.96
0.804.24
1.00-15.67
0.15
0.05
-0.328 1
0.1614
0.72
0.53-0.99
0.04
1.9128
0.5408
6.77
2.35-19.55
0.0004
0.3894
1.0055
0.4687
0.7680
1.48
2.73
0.59-3.70
0.61-12.32
0.41
0.19
0.1728
0.74
0.53-1.04
0.08
0.5571
5.81
1.95-17.32
0.0016
Beta
P
~
Model 1
HAQ score in 1989
Increase in HAQ of 20.50
(1990 to 1991)
Model 2
Proportion of valued activities
performed in 1989
(per 10% of activities)
Decrease of 210% in the
proportion of valued activities
performed (1990 to 1991)
Model 3
HAQ score in 1989
Increase in HAQ of 20.50
(1990 to 1991)
Proportion of valued activities
performed in 1989
(per 10% of activities)
Decrease of 210% in the
proportion of valued activities
performed (1990 to 1991)
0.6100
1.3772
-0.3010
1.7598
* 95% CI = 95% confidence interval; HAQ = Health Assessment Questionnaire; Beta = parameter
estimate from multiple logistic regression analysis
ling for the covariates and baseline HAQ, individuals
whose HAQ scores increased by this amount were
more likely to develop depressive symptoms (OR =
3.96, 95% CI 1.00-15.67, P = 0.05).
Next, the activity variables were added to the
covariates (Table 4,Model 2). The loss of at least 10%
of one’s valued activities was a strong risk factor
(OR = 6.77, 95% CI 2.35-19.55, P = 0.0004). In
addition, individuals who, at baseline, performed a
higher proportion of their valued activities were less
likely to develop depressive symptoms (OR = 0.72,
95% GI 0.53-0.99, P = 0.04).
Finally, both the HAQ and activity variables
were added to the covariates (Table 4, Model 3). The
only functional variable that was significant in this
model was the loss of at least 10% of one’s valued
activities (OR = 5.81,95% CI 1.95-17.32, P = 0.0016).
Neither of the baseline activity variables nor an increase in HAQ score reached statistical significance.
Taking into account the covariates and other function variables, women who lost valued activities were
approximately 5 times more likely to develop depressive symptoms than those who did not lose such
activities (relative risk = 5.10).
To shed further light on these results, we explored the relationship between decline in basic func-
tion, measured by HAQ score, and the loss of activities. Forty-seven of the 319 women experienced an
increase in HAQ score and/or a loss of activities.
Interestingly, the overlap between decline in basic
function measured by HAQ and the loss of valued
activities was minimal. Only 6 women (13%) experienced both an increase in HAQ score and a loss of
activities. Fourteen women had an increase in HAQ
score and no loss of activities, and 27 women had a
loss of activities and no increase in HAQ score (Table 5).
The frequency of depressive symptoms was
Table 5. Interaction of the effects of loss of valued activities and
increase in HAQ score on the development of depressive
symptoms*
Neither increase in HAQ score
nor activity loss
Increase in HAQ score, no
activity loss
Activity loss, no increase in
HAQ score
Both increase in HAQ score and
activity loss
No. of
patients
No. (%) depressed
in 1992
272
13 (4.1)
14
1 (7.1)
27
5 (18.5)
6
2 (33.3)
* HAQ = Health Assessment Questionnaire.
KATZ AND YELIN
54
Table 6. Logistic regression model examining the effects of the interaction of loss of valued activities
and decline in basic function on the development of depressive symptoms*
Variable
Beta
Standard
error
of beta
Increase in HAQ score alone
Loss of valued activities alone
Loss of valued activities plus
increase in HAQ score
0.8203
1.3994
2.4272
1.1069
0.6141
0.9704
Odds
ratio
95% CI
P
2.27
3.96
11.33
0.26-1 9.88
1.22-1 3.5 1
1.65L75.88
0.46
0.02
0.01
* See Table 4 for definitions.
very different among the 3 groups. Approximately 7%
of those who experienced only an increase in HAQ
score developed depressive symptoms, compared with
18.5% of those who lost valued activities only. Onethird of those who experienced both an increase in
HAQ score and a loss of activities developed depressive symptoms.
A final multivariate logistic regression was performed to examine the effect on depressive symptoms
of an increase in HAQ score alone, a loss of valued
activities alone, and an increase in HAQ score plus a
loss of valued activities. The results of this analysis are
shown in Table 6. An increase in HAQ alone (no
activity loss) was not a signficant risk factor for the
development of depressive symptoms. As in the other
analyses, activity loss was a strong risk factor for
developing depressive symptoms, with or without a
concomitant decline in basic function. Women who
experienced an activity loss alone (no increase in HAQ
score) were about 4 times more likely to develop
depressive symptoms than those who experienced
neither an increase in HAQ score nor a loss of activities. The combination of activity loss plus an increase
in HAQ score was an even stronger risk factor. Among
women who experienced both, the risk of depressive
symptoms was increased 8-fold (relative risk; data not
shown).
DISCUSSION
The results from this study suggest that the
component of functional decline that was actually the
risk factor for depressive symptoms was loss of the
ability to perform valued activities. The loss of valued
activities, regardless of whether associated with a
decline in basic function, was strongly linked to the
development of depressive symptoms. In contrast,
decline in basic function alone was only moderately
linked to the development of depressive symptoms,
and was not at all linked when activity loss was also
considered. In fact, decline in basic functioning
seemed to play a strong role only when it occurred in
combination with activity loss, suggesting perhaps that
activity loss may have made individuals more sensitive
to overall decreases in function.
The S-GDS scores were relatively stable over
the 4 years. Among the group that did not develop high
levels of depressive symptoms in 1992, there were no
significant year-to-year differences in S-GDS scores.
Among the group that did develop depressive symptoms, there were no significant year-to-year changes
over the first 3 years. Only between the third and
fourth years was a significant increase in the S-GDS
score noted. Although the S-GDS scores were higher
among this group in all years and there were differences between the groups in other baseline characteristics, this pattern of S-GDS scores suggests that
among the group who became depressed, something
triggered the dramatic increase in depressive symptoms. Such a trigger might be some change in the
individual’s life that preceded the increase in depressive symptoms.
The functional changes that occurred more frequently among the group who developed depressive
symptoms occurred during the year preceding the
assessment of outcome (between 1990 and 1991). In
the multivariate models, these functional changes
were clearly better predictors of the development of
depressive symptoms than the baseline functional
characteristics, and the timing of these functional
changes fits the pattern of changes in the S-GDS
scores. Thus, the functional decline that occurred
between 1990 and 1991 could well have been the
trigger for the increase in depressive symptoms that
was seen the following year.
The results of the current study are consistent
with the literature, although there are few published
data concerning the influence of activity loss on the
55
DEPRESSIVE SYMPTOMS IN RA PATIENTS
development of psychological distress or depression.
For example, we have previously found that RA
patients who had high levels of depressive symptoms
performed 12% fewer valued activities than those with
low levels of depressive symptoms (26), and Williamson and Schulz found that activity restriction was
highly correlated with scores on the Center for Epidemiological Studies Depression Scale (30). A logical
extension of the current study is the identification of
reasons for activity losses. Such research is likely to
produce information that will be useful in preventing
or at least decreasing activity loss.
This study was limited to women, which makes
generalization of the results to men questionable.
However, this limitation may actually strengthen the
findings. Much previous work concerning depression
has not differentiated between correlations or risk
factors for men and women. Given the differences
between social and economic roles and expectations of
men and women in general, as well as clinical and
functional differences among men and women with
RA, it is likely that risk factors for the development of
depressive symptoms would be different for men and
women. We cannot examine this question, however,
due to the limits of our sample.
It should be noted that this study assesses
depressive symptoms and not depression. However, a
person who has depressive symptoms is generally
thought to be at risk of developing clinical depression
(21), and scores on the GDS have been found to have
a high correspondence to clinical diagnoses of depression. These data were also collected at 1-year intervals. This time period may be too long to detect other
relevant changes that may have been related to the
development of depressive symptoms. However, such
a strong relationship between the loss of valued activities and the development of depressive symptomseven over this time period-suggests that the risk
conferred by this loss is indeed great.
It should also be noted that, while activity loss
was a strong risk factor for the development of depressive symptoms, the majority of individuals who developed depressive symptoms did not experience an
activity loss. Clearly, other factors, not assessed in
this study, can make important contributions to the
development of depressive symptoms among women
with RA.
Finally, this study points out the value of longterm longitudinal databases for studying issues such as
causal or time-ordered relationships. While correlations of changes can be assessed with 2 data points,
one needs a minimum of 3 data points to assess a
time-ordered relationship. To answer some questions,
particularly about a long-term chronic disease such as
RA, even more data collection points may be necessary. Although such databases are difficult to construct in a representative manner and difficult to
maintain without excessive attrition, the answers to
many important questions about the course of a
chronic disease such as RA can only be answered with
long-term longitudinal studies.
In summary, it appears that while a decline in
basic, ADL-type function is a risk factor for the
development of depressive symptoms in women with
RA, the aspect of function that is most crucial to
psychological status is the ability to perform valued
activities. There appears to be a time lag between the
time of the activity loss and the development of
depressive symptoms, making it clear that the loss of
valued activities precedes, and is thus a risk factor for,
development of depressive symptoms. Given these
apparently time-ordered events, it may be possible to
avoid the development of depressive symptoms altogether through intervention at an early stage of decline. For this strategy to work, though, providers may
need to initiate a different type of interaction with their
patients than that to which they may be accustomed.
In addition to questions about clinical symptoms such
as pain and fatigue, practitioners would need to also
ask, “Can you do the things you want to do?”
Depression is a costly illness, both in economic
and personal terms, and its combination with a physical illness compounds the costs and disability associated with the illness. Therapies or programs designed
to help persons with RA maintain activities that are
important to them, or perhaps find adequate alternative activities, may be very effective in helping them
maintain psychological well-being. Anticipating the
need for such programs by noting critical events in
individuals’ lives may even help them avoid depressive episodes altogether, as well as help them maintain
better physical status.
REFERENCES
1. Anderson KO, Bradley LA, Young LD, McDaniel LK, Wise
CM: Rheumatoid arthritis: review of psychological factors
related to etiology, effects, and treatment. Psycho1 Bull 98:358387, 1984
2. Frank RG, Beck NC, Parker JC, Kashani JH, Elliott TR, Haut
AE, Smith E, Atwood C, Brownlee-Duffeck M, Kay DR:
Depression in rheumatoid arthritis. J Rheumatol 15:920-925,
1988
3. Palinkas LA, Wingard DL, Barrett-Connor E: Chronic illness
KATZ AND YELIN
56
and depressive symptoms in the elderly: a population-based
study. J Clin Epidemiol 43:1131-1141, 1990
4. Hawley F, Wolfe DJ: Anxiety and depression in patients with
rheumatoid arthritis: a prospective study of 400 patients. J
Rheumatol 15:932-941, 1988
5 . Newman SP, Fitzpatrick R, Lamb R, Shipley M: The origins of
depressed mood in rheumatoid arthritis. J Rheumatol 16:740744, 1989
6. Peck JR, Smith TW, Ward JR, Milano R: Disability and
depression in rheumatoid arthritis: a multi-trait, multi-method
investigation. Arthritis Rheum 32:1100-1 106, 1989
7. Katz PP, Yelin EH: Prevalence and correlates of depressive
symptoms among persons with rheumatoid arthritis. J Rheumato1 20:79@796, 1993
8. Chamberlain M, Buchanan J, Hanks H: The arthritic in an urban
environment. Ann Rheum Dis 3851-56, 1979
9. Turner RJ, Noh S: Physical disability and depression: a longitudinal analysis. J Health SOCBehav 29:23-37, 1988
10. Wolfe F, Hawley DJ: The relationship between clinical activity
and depression in rheumatoid arthritis. J Rheumatol 20:20322037, 1993
1 1 . Aneshensel CS, Frerichs RR, Huba GJ: Depression and physical illness: a multiwave, nonrecursive causal model. J Health
SOCBehav 25:350-371, 1984
12. Leino P, Magni G: Depressive and distress symptoms as predictors of low back pain, neck-shoulder pain, and other musculoskeletal morbidity: a 10-year follow-up of metal industry
employees. Pain 53:89-94, 1993
13. Yelin EH, Henke CJ, Kramer JS, Nevitt MC, Shearn M,
Epstein WV: A comparison of the treatment of rheumatoid
arthritis in health maintenance organizations and fee-for-service
practices. N Engl J Med 312:962-967, 1985
14. Brink TL, Yesavage JA, Lum 0, Heersema PH, Adey M, Rose
TL: Screening tests for geriatric depression. Clin Gerontol
1:37-43, 1982
15, Sheikh JI, Yesavage JA: Geriatric depression scale (GDS):
recent evidence and development of a shorter version. Clin
Gerontol 5: 165-173, 1986
16. Rule BG, Harvey HZ, Dobbs AR: Reliability of the geriatric
depression scale for younger adults. Clin Gerontol9:3743, 1989
I 1 . Dunn VK, Sacco WP: Psychometric evaluation of the geriatric
depression scale and the Zung self-rating depression scale using
an elderly community sample. Psychol Aging 4: 125-126, 1989
18. Olin JT, Schneider LS, Eaton EM, Zemansky MF, Pollock VE:
The geriatric depression scale and the Beck depression inven-
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
4-
30.
tory as screening instruments in an older adult outpatient
population. Psychological Assessment 4: 190-192, 1992
Harper RG, Kotik-Harper D, Kirby H: Psychometric assessment of depression in an elderly generally medical population. J
Nerv Ment Dis 178:113-119, 1990
Alden D, Austin C, Sturgeon R: A correlation between the
geriatric depression scale long and short forms. J Gerontol
44:P124-P125, 1989
Shaver PR, Brennan KA: Measures of depression and loneliness, Measures of Personality and Social Psychological Attitudes: Volume 1 of the Measures of Social Psychological
Attitudes Series. Edited by J P Robinson, PR Shaver, LS
Wrightman. San Diego, CA, Academic Press, 1991
Pincus T, Callahan LF, Bradley LA, Vaughn WR, Wolfe F:
Elevated MMPI scores for hypochondriasis, depression, and
hysteria in patients with rheumatoid arthritis reflect disease
rather than psychosocial status. Arthritis Rheum 29:14561466,
1986
Cwikel J, Ritchie K: Screening for depression among the elderly
in Israel: an assessment of the short geriatric depression scale
(S-GDS). Isr J Med Sci 25:131-137, 1989
Fries JF, Spitz P, Kraines RG, Holman H: Measurement of
patient outcomes in arthritis. Arthritis Rheum 23: 137-145, 1980
Yelin E, Lubeck D, Holman H, Epstein W: The impact of
rheumatoid arthritis and osteoarthritis: the activities of patients
with rheumatoid arthritis and osteoarthritis compared to controls. J Rheumatol 14:710-717, 1987
Katz PP, Yelin EH: Life activities of persons with rheumatoid
arthritis with and without depressive symptoms. Arthritis Care
Res 7:69-77, 1994
Kaplan GA, Roberts RE, Camacho TC, Coyne JC: Psychosocia1 predictors of depression: prospective evidence from the
human population laboratory studies. Am J Epidemiol 125:20&
220, 1987
Kessler RC, Foster C, Webster PS, House JS: The relationship
between age and depressive symptoms in two national surveys.
Psychol Aging 7:119-126, 1992
Kennedy GJ, Kelman HR, Thomas C, Wisniewski W, Metz H,
Bijur PE: Hierarchy of characteristics associated with depressive symptoms in an urban elderly sample. Am J Psychiatry
146:220-225, 1989
Williamson GM, Schulz R: Pain, activity restriction, and symptoms of depression among community-residing elderly adults. J
Gerontol47:P367-P372, 1992
Документ
Категория
Без категории
Просмотров
2
Размер файла
802 Кб
Теги
development, women, arthritis, role, symptom, among, function, depression, rheumatoid
1/--страниц
Пожаловаться на содержимое документа