Acculturation Stress and Depression
among Asian Immigrant Elders
Ada C. Mui and Suk-Young Kang
This study examines the association between acculturation stress and depressive symptoms in a
regional probability sample (n = 407) of six groups of Asian immigrant elders (Chinese,
Korean, Indian, Filipino, Vietnamese, and Japanese). Findings suggest that about 40 percent of
the sample were depressed, indicating higher depression rates than found in other studies of
older American or Asian elderly samples in the United States and Asia. Multiple regression
analyses indicated that acculturation stress caused by elders' perception of a cultural gap
between themselves and their adult children was associated with high depression levels. Other
predictors of depression were poor perceived health, stressful life events, religiosity, proximity
of children, assistance received from adult children, and longer residence in the United States.
Data suggest that depression is prevalent among urban Asian immigrant elders and that there is
great heterogeneity among Asian ethnic subgroups. Implications for social work practice are
discussed.
KEYWORDS: acculturation stress; Asian American elders; Asian immigrant elders; depression
D
emographically.the United States faces two
dramatic population changes: aging and
increasing ethnic and racial diversification.
The population segment age 65 and older (35 million) currently and until 2010 will comprise about
12.4 percent of the total U.S. population (Hetzel &
Smith, 2001). Between 2010 and 2030, the baby
boomers will join this older population. By 2030,
about 20 percent of the total U.S. population (69.4
million) is projected to be age 65 and older. The
non-Hispanic white share of the total population is
projected to fall steadily from 74 percent in 1995 to
53 percent by 2050. At the same time, other racial
and ethnic populations will increase in number.The
groups with the highest expected rates of increase
are Hispanic, Asian, and Pacific Islander Americans.
Data from the 2000 census indicate that the Asian
American and Asian immigrant elderly population
grew by 76 percent from 1990 to 2000 and is projected to grow by 246 percent from 2000 to 2025.
This growth is comparable with the 9.2 percent
and 73 percent growth rates in the same years among
the white elderly population (U.S. Census Bureau,
1990,2001).
CCC Code: 0037-8046/06 $3.00 O2006 National Association of Sociai Wori<ers
The increasing number of Asian and other immigrants results in greater demands for research
methodology sensitive to cross-cultural issues. Although depression is a common psychological problem among elderly populations, few researchers have
studied depression in older Asian Americans or elderly Asian immigrants. There is also a dearth of
empirical knowledge about within-group differences among the Asian subgroups. Although available national long-term care datasets have been
found large enough to conduct analysis of nonHispanic white and African American elderly populations, Asian American and Asian immigrant elderly groups are rarely included in sufficient numbers
to enable meaningful statistical analysis (LaVeist,
1995). Moreover, previous studies on Asian American and Asian immigrant elders used small local
samples and did not examine the association between specific acculturation stress and depression
of Asian elders from different nationality backgrounds. Within-group variations may reflect differential patterns of acculturation, family dynamics, family values, or expectation in terms of
intergenerational exchanges.
243
The study reported here fills this gap by using
an urban area probabihty sample of six different
Asian elderly groups (Chinese, Filipino, Indian,
Japanese, Korean, and Vietnamese). It examined
correlates of depression and group differences in
them. The primary predictor of interest was acculturation stress, defined as an acculturation gap between the elderly person and his or her adult children. The specific research question examined was
all things being equal, what is the association between acculturation stress and depression among
the Asian elders? The importance of this research
is founded in the consistent reports that depression is the risk factor most frequently associated
with suicide (Bartels et al., 2002; Lapierre,
Pronovost, Dube, & Delisle, 1992; Mui, 1996a;
2001).
ELDER DEPRESSION
Prevalence and Incidence Rates
A quarter of all late-life suicides are due to depression (American Association of Retired Persons,
1997). Depression may occur frequently in Asian
immigrant elders because they have limited resources in dealing with the multiple losses associated with the process of adaptation, acculturation,
and family disruption (Mui, 1996a; 1996b).Epidemiological studies in the United States have examined the prevalence of depressive symptoms in community samples of elders using a variety of self-rating
scales and interviews. Depending on the selected
cutoff points and instruments, estimates of the prevalence of major depression vary widely. Using DSMIV-based criteria for major depression, a one-year
prevalence rate is estimated at about 5 percent or
less among community-dwelling people age 65 and
older (Mui, Burnette, & Chen, 2001). Depressive
symptoms or syndromes are more prevalent, with
about 15 percent to 20 percent of communitydwelling elders experiencing them (Gallo &
Lebowitz, 1999). Prevalence rates of depressive
symptoms based on assessment with the Geriatric
Depression Scale (Sheikh &Yesavage, 1986; Yesavage
et al., 1983) among community-dwelling elders
from different ethnicity or nationality backgrounds ranged from 12 percent to 50 percent,
indicating high rates of possible depression (Halier,
Weggenmans, Ferry, & Guigoz, 1996; Mui, 1996b;
2001 ;Shibusawa & Mui, 2001). Prevalence rates for
ethnic minority elderly populations may be underestimated or biased because of low cultural relevance
Z44
of standardized measures (Mui et al.,2001). Along
with physical, cognitive, and functional impairment,
sociocultural factors, such as differences in acculturation, perception, interpretation, valuation, expression, and tolerance of symptoms, may contribute to this bias (Mui, 1996a, 1996b). Despite
substantial prevalence rates, symptoms of depression are often underrecognized, underdiagnosed,
and undertreated because of patient- and health
care-related barriers and problems in the organization, financing, and delivery of mental health services for older adults (Gottlieb, 1991; Mui et al.,
2001).
Acculturation Stress
Acculturation is a process by which one cultural
group adopts the beliefs and practices of a host
culture (Mills & Henretta, 2001). The association
between acculturation and depression is intriguing.The acculturation process is multidimensional,
including physical, psychological, financial, spiritual, social, language, and family adjustment. This
process can be very stressful for immigrant elders
in the United States because they have fewer resources, such as income, education, and English
proficiency, to assist them in adapting to their new
life situation (Black, Markides, & Miller, 1998;
Casado & Leung, 2001). Research on acculturation and depression has found that less acculturated elderly Hispanic immigrants \vere more likely
to be depressed than were their more acculturated
counterparts (Black et al.; Falcon & Tucker, 2000;
Gonzalez, Haan, & Hinton, 2001; Hovey, 2000). A
few small sample studies of Asian elders also reported that immigrants who were more acculturated to the host society tended to have better mental health status than those who were less
acculturated (Pang, 1998; Stokes, Thompson,
Murphy, & Gallagher-Thompson, 2001). Other
correlates of minority elders' high depression rate
include shorter lengths of residence in the United
States, poorer health, more life stresses, more financial strain, poor English proficiency, dependence
on children, social isolation, and lack of social support (Casado & Leung; Falcon &Tucker; Gonzalez
et al.; Lee, Crittenden, &Yu, 1996; Stokes et al.;
Tran, 1992). In sum, immigrant elders' depression
may be due to migration stress and grief, adaptation difficulties, poverty, illness, and weakening family support (Gelfand &Yee, 1991; Mui, 1996b, Mui
et al.,2001).Acculturation is a psychologically and
SocialWork VOLUME 51, NUMBER 3
JULY 2006
socially complex process, and further understanding of the relationships between acculturation stress
and mental health status may be able to inform the
design of effective intervention programs for older
immigrants.
Social Support and Elder Depression
Researchers have also studied the role of social support, particularly that provided by family members,
and its impact on the psychological well-being of
elders from different ethnic backgrounds. Findings
show that fewer family contacts and a smaller social
network are associated with higher depression
(Hovey,2000;Lee et al, 1996;Pang, 1998;Stokes et
al, 2001). By and large, studies on Asian family support have shown that Asian elders receive a considerable amount of emotional and instrumental support from adult children (Mui, 2001; Shibusawa &
Mui,2001).In addition, these Asian families have a
strong sense of family obligation that often supersedes the needs of individual family members. Asian
elders expect their family to assist them in their old
age and to treat them with respect (Mui, 1996b,
2001).These studies suggest that there is a cultural
expectation factor in family support among elderly
Asian immigrants that may differ from other ethnic
groups. Despite a high level of social support exchange and reciprocity between generations of
children and grandchildren, Chinese and Korean
immigrant elders still report high levels of depression (Mui, 1996b, 2001). It is not clear whether
these respondents' depression was due to high unmet
needs or higher family expectation in
intergenerational exchange. These are unanswered
questions in the literature.
CONCEPTUAL FRAMEWORK FOR STUDYING
ACCULTURATION STRESS AND DEPRESSION
In this study ofAsian immigrant elders, a stress and
coping framework (Aldwin, 1994; Lazarus &
Folkman, 1984) was used to conceptualize and examine the relationships between life and acculturation stress, coping resources, and depressive symptoms. Ways of coping with stress are determined by
cognitive appraisal and include both cognitive and
behavioral efforts to manage stresses that are appraised as taxing. The stress and coping framework
acknowledges the importance of personal and environmental stress (stressful life events, acculturation stress, and health stress) and their effects on
elderly Asian American and Asian immigrant el-
ders' psychological well-being. Coping resources
usually include spiritual, fmancial, and social supports that are available to individuals (Burnette &
Mui, 1994; Lazarus & Folkman; Mui, 1993; 1996b).
Based on the stress and coping framework, this study
examines the effects of stress (acculturation stress,
stressful life events, poor health conditions) and
coping resources (English proficiency, social support, religiosity) on the self-reported depressive
symptoms among an elderly Asian immigrant
sample.
METHOD
Data Sources and Sample
Data analyzed here are firom the Asian American Elders
in NewYork City, 2000, a survey sponsored by the
Asian American Federation of New York (Ryan,
Mui, & Cross, 2003). The study researched Asian
Americans age 65 and older who belonged to one
of the following six ethnic groups: Chinese, Filipino, Indian, Japanese, Korean, and Vietnamese.
These six groups were selected because they represented 94.4 percent of all Asian American and Asian
immigrant elders in the United States (U.S. Census
Bureau, 2000). The survey explored how health,
mental health, fmancial well-being, informal support systems, formal service utilization, acculturation and immigration experiences, and quality-oflife issues related to depression and overall life
satisfaction.
The sample was drawn from a 1990 US. census
list of more than 5,785 block groups in the five
boroughs of NewYork City. Census block groups
were ranked according to the percentage of Asian
households recorded in the 1990 census. A cut-off
was made when the block groups represented 70
percent of all Asians 65 and older contained in the
1990 census data. The fmal sample consisted of 60
block groups or primary sample units (PSUs) that
met the inclusion criteria.This survey used an area
probability sample. Area probability samples are
designed to give each eligible population member
or household in the area a known chance of being
interviewed. Using a street map and specially designed listing sheets, trained bilingual interviewers
went to a randomly designated point in the area
and listed housing units. For each of the 60 selected block groups, 100 to 150 households were
listed and then screened for interviewing. For this
study, interviewers were required to contact a selected household up to four times to determine
M U I A N D K A N G / Acculturation Stress and Depression among Asian Immigrant Elders
eligibility for study inclusion. Because of the nature of the random sampling procedure, no household substitution was allowed.Therefore, when an
interviewer identified an eligible respondent, he
or she was required to call back up to four times
to procure an interview. One respondent per
household was interviewed. If more than one eligible respondent lived in the household, then the
person who had most recently celebrated a birthday was selected as the respondent.
A total of 407 Asian immigrants age 65 and older
were interviewed in their homes. There were no
American born Asian elders identified in this sample,
all respondents were foreign born, having immigrated to the United States in middle to late middle
age. Interviewers for each subgroup of these Asian
elders received a bilingual interviewing manual and
a set of question instructions to minimize the systematic errors and to increase the inter-rater reliabihty.The completed survey response rate was 84.3
percent of those eligible to participate. Interviews
averaged one hour and 30 minutes in length and
were conducted between February 2,2000 and May
31, 2000. Interviews were conducted in EngHsh,
Chinese,Tagalog, Hindi, EngHsh, Korean, andVietnamese.The questionnaire was prepared in English
and was translated and back-translated into different ethnic languages by panels of bilingual professionals to ascertain that the items were culturally
valid, and conceptually and linguistically consistent.
The respondents were administered the questionnaire in either English or the ethnic language they
preferred. Most participants used the native language questionnaire consistent with their ethnicity.
The Japanese elders used the English questionnaire.
Measures
The coding scheme, mean, and standard deviation
of all the variables in the study are presented in
Table 1 .The dependent variable—depression—was
measured by the 30-item GDS. These items consist of symptoms similar to DSM-IV criteria, such
as depressed mood, feelings of hopelessness and
worthlessness, diminished interest in activities, poor
concentration,and indecisiveness (Mui et al., 2001).
The assessment of depression in an elderly population is more difficult than in a younger population because of the higher prevalence of somatic
complaints, genuine physical problems, and medication use. One of the strengths of the GDS is that
it contains no somatic items that can introduce age
246
bias into the depression screening scale and infiate
total scores among the elderly population (Kessler,
Foster,Webster, & House, 1992). Respondents were
classified as depressed on the basis of the standard
cutoff point of the total GDS score, which is a 11
(Yesavage et al., 1983). Literature suggests that the
GDS is a reliable measure of depression for Asian
American elderly groups, and it has shown adequate internal consistency reliability (Mui, 1996c;
Mui, Kang, Chen, & Domanski, 2003).The computed alpha coefficients of the GDS for different
groups in this study ranged from .85 to .92 (see
Table 1), indicating satisfactory reliabihty performance of this measure.
Life stress and acculturation stress were operationally defined by six variables: length of stay in the
United States, self-rated health (1 = poor, 5 = excellent), number of medical conditions, number of
stressful life events, family responsibility expectation, perceived cultural gap between the respondent and his or her adult children.The stressful life
events variable was a composite score of a list of up
to six stressful life events experienced by the respondent in the preceding three years. Examples of
these events included major losses in life, such as
death of spouse, family member, or good friend;
serious illness or injury; being robbed or having
home burglarized;having adult children who moved
out; and relocation.
Acculturation stress measures are found in Table
2. Family responsibility expectation was a composite score of 12 family value items, which measured
the degree of family value acculturation of the elders (a = .80).The 12 family value questions measured elders' expectations regarding extended family, nuclear family, elder care, marriage, divorce,
gender roles, and family living arrangements. Elders answered each question in terms of how they
felt about these statements on a four-point scale
ranging from 0 = not at all important to 3 = very
important. High scores indicated elders who were
less acculturated because they adhered more to their
traditional cultural beliefs. The perceived cultural
gap variable was a global measure evaluated by asking elders to assess how their view regarding these
family values differed from that of their children on
a four-point scale ranging from 0= not at all different to 3 = different in a very important way.
Indicators of coping resource variables were elders' English proficiency, religiosity, number of children living within a two-hour driving distance, and
SocialWork VOLUME 51, N U M B E R 3
JULY 2006
68
' ^-'
• -^ 0CO
0
ro
IN
_^
in
*^
"-^
^^
0
CN
0
0
CO
m
IN
00
0
s
CO
0
CN
00
.
—
1
tn
•^
0 '^
•^
CN
CO
o cs o
m o m
n
r^
—
cv
xf
80 (10
02 05)'
80 00)'
27)"
00
13)'
CN
.36
23
95
—'
o
cs
^
in
• — '
.28
.40
90
10
06
80)'
30)'
r
\^
HI-
§1
tn
en
ro
CN
-<
lA
^
vo
s
en
=(l
CO
CN
[.^ —<
CO
00 tn
-^ en
=(Z6
•^
0
V5
22- i C-
sa
CN
{:;
eS
:S-^
t~~
2.04)
1.48)
(5.4)
(.71)
13.2)
1.09)
(.98)
tl.58)
2.65)
tN
0
^^
(1.
^
CN
m
75
0
(1.
CN
28
c\
91
44 (.50)
(.39)
(.50)
(7.0)
__,C\
0
Cs
CN NT
\O ^-^
00
00
CO
tn
in
0
o
CO
CN - ^
Si
V - '
75)
36)
51)
vo
f-H
.33
.04
.52
.20
.37
.sin
' — -
.38
90
.84
(61
2(5
50)
60)
49)
— cs
90(1
91
23
18)
71
80
70
q q q
o a d
cs cn
00
CS
00
tn"
in -*—
r-<
in
00
cs
cs
--
00 —
S R S S S
^
in
' cs en
' tn in
•—" in
v_^
_H
0
0
0
00
in
en
in
>—<
CN
00
en
rs en
cs —^
in
tn
vo rs en
^-'
78 (
90 (
m
o
20(1
41
vin
^H
Cs
00
in
Is
-a
J
Icg 7
O
s
£ P. w
I
o
7 °
60
•is
I
S
Q J
-= I 2 S
- 1o
•2 °- -3
"
C
w
bO
rt
vo
u
•S
.2
ci '2
n
i
111
g" S s s ^
+:
V
IU
5£ S
S£^
< I-^
O c S
-a t«
o u .£
CN
rTi
Xf
.—I
CTi
<r\ r-- r--. h^ 00
o o o o o o o o o o
tS
r->
r—
o
fN
O
O
1
'E
1,1
o
a »
5 2
U .2
•
c
^
c -o
>?
-C
..c
'E
"Ia. OS
-C
c
; i
-a
S
a. -a "c
1
d
o 3
o.
1.1
^
1
•a
I
(2
p=
S
^
3
•«
£
o
5
•"
2 „ =
E
S
S
-
—
&•
(u
E
c
•"
"
s
^ ° "2 o
£ £ O Z
5 &S
the amount of instrumental, financial, and emotional assistance received fi-om children. English
proficiency was assessed in terms of respondents'
ability to read, write, and speak English. In this study,
English proficiency was conceptualized as a coping
resource because it represents elders' behavioral
efforts to cope with the host culture. Religiosity
assessed respondents' perception about the level of
importance of religion in their lives on a four-point
scale. Sociodemographic variables, including gender, living arrangements, marital status, age, and
Medicaid coverage, were used as controls. Because
the income measure had a lot of missing data,
Medicaid coverage was used as a proxy measure of
elders' financial status. Even though Medicaid entitlement may not be a precise measure of economic resources, it uses means testing and income
related to federal poverty guidelines as eligibility
requirements.
RESULTS
Univariate and Bivariate Analysis
Table 1 presents a descriptive profile of the six groups
of Asian immigrant elderly respondents {n = 407).
Respondents were 25.8 percent Chinese (M = 105),
12.8 percent Filipino (w = 52), 24.6 percent Indian
(n = 100), 6.1 percent Japanese (n = 25), 24.6 percent Korean (w = 100), and 6.1 percentVietnamese
(n = 25). Data show that the entire sample were
born in Asian countries and typically immigrated
to the United States in middle age. The data also
confirmed that this Asian immigrant elderly sample
was not a homogeneous group. Group differences
included gender composition, living arrangements,
marital status, age, time in the United States, stress,
coping resources, and acculturation experiences. Of
the total sample, 44 percent were men, 19 percent
lived alone, and 49 percent were married. Compared with the general older population and with
other Asian groups in the study, there were more
men in the Indian andVietnamese groups (62 percent and 64 percent, respectively) .Those in the Filipino and Indian groups were less likely to live alone
(only 4 percent and 3 percent, respectively) than
w^ere those in the other groups. Compared with
other groups,Japanese and Korean elders were less
likely to be married (40 percent and 29 percent,
respectively). The ages of the sample ranged from
65 to 96 (M = 72.4 years, SD = 6.2), with the
Vietnamese group being significantly younger than
those in the other groups (M = 68.8 years).The
average length of residence in the United States for
all the respondents was 20.8 years {SD = 13.2,range
= 1 to 72 years), with the Vietnamese group reporting the shortest U S. residence (seven years only).
The whole sample had some high school education in their home country. About 41 percent {n =
167) of the total sample received Medicaid.
In terms of stress and coping resources, the Vietnamese participants were in the poorest condition
compared with others in the sample.The Vietnamese elders reported significantly poorer health, more
medical conditions, a greater perceived cultural gap
between themselves and their adult children, and
poorer English proficiency. On the other hand, they
had more children living in proximity and considered religion as more important in their lives.They
were also more likely than their counterparts to
receive Medicaid (91 percent).
Table 1 also shows the GDS mean scores for the
total sample (M = 10.1; 5D = 7.0) and for each
group. About 40 percent of the total sample scored
11 or higher on the GDS, indicating possible depressive symptomatology.There were within-group
differences experienced by members of the six
groups.Japanese respondents reported significantly
higher risk of depression than the other groups (M
= 15.0; SD = 7.7), followed by Vietnamese (M =
12.4; SD = 5.7), Chinese (M= 11.4; SD = 7.7), and
Indian (M = 11.1; SD = 6.8) respondents. Korean
and Filipino respondents reported significantly
lower levels of depressive symptoms (M =7.8; SD
= 5.8 and M = 6.2; SD = 5.0, respectively). The
incidence rates of depression (those who scored 11
points or more on the GDS) in this sample were
Japanese (76 percent), followed by Vietnamese (64
percent), Indians (50 percent), Chinese (45.7 percent), Koreans (24 percent), and Filipinos (15.4
percent).These findings all exceed the 15 percent
to 20 percent prevalence rates of depressive symptoms or syndromes among community-dwelling
elders as reported by Gallo and Lebowitz (1999).
Multivariate Analysis
Hierarchical linear regression analyses were used
to examine significant correlates of depression. First,
a simple additive model was used to test the additive effects of demographic, coping resources, and
acculturation stress on depression levels. Then, an
additive and interactive model was used to test the
interaction effect between acculturation stress
variables and dummy-coded nationality variables.
M U I AND KANG / Acculturation Stress and Depression among Asian Immigrant Elders
249
Statistically, the interaction model tested whether
the impact of acculturation stress on depression
levels would vary by nationality groups. Only Korean, Chinese, and Indian elders' data were used to
create dummy variables with Chinese as the reference group because that group is more known,
relatively speaking, in the literature (Mui, 1996b).
The Vietnamese, Filipino, and Indian groups were
too small for meaningful comparison in multivariate testing. Normality of variables and regression
assumptions were checked.There was no violation
of regression assumptions. Zero-order correlations
among independent variables ranged from .01 to
.37, indicating that multicollinearity was not a
problem.
Table 3 presents results of the hierarchical regression model ofthe GDS for the whole sample.
Findings indicate that demographic variables as a
set explained 5 percent of the variance in depression scores when entered first. Entered second, the
coping resources as a set accounted for an additional 6 percent ofthe variance in the model.The
acculturation stress variables as a set added another
36 percent variance in the model after demographic
and coping resources variables were taken into account. In the interaction model, the two dummycoded nationality variables (Korean versus Chinese;
Indian versus Chinese) and all the two-way interaction terms between nationality dummy-coded
variables and stress variables were entered. None of
Table 3: Linear Regression Model of Depressive Symptoms
by Six Groups of Elderly Asian Immigrants
Independent variable
Final Step
Step 1
Step 2
Unstandardized Unstandardized Unstandardized Standardized
Pearson r* Coefficient (5£) Coefficient (S£) Coefficient (S£)
Coefficient
Sociodemographics
Gender
Living arrangement
Marital status
Age
Medicaid coverage
-.02
,15****
1.23 (.85)
1.62(1.06)
1.45 (.97)
.26 (.78)
.02
2.10(1.22)
.78(1.02)
-.10
1.21 (.91)
-1.72(1.00)
-.07
.95 (.83)
1.52 (.75)*
-1.01 (.91)
-.48 (.79)
-.74 (.76)
.04
.04
1.70 (.87)
-.25 (.72)
-.72 (.42)
-.06 (.37)
-.88 (.37)*
.13*
/P at this step
.05
-.02
.05*
Coping resources
English proficiency
-.18***
Religiosity
-.06
No. of children living within two
hours driving distance
-.15**
Assistance from children
.02
Increment to i? at this step
.03 (.43)
-.80 (.25)**
-.70 (.20)***
.28 (.17)
.06***
.38 (.13)**
.01
-.13*
.17***
.14**
Life stress and acculturation stress
Length of stay
Perceived health
No. of medical conditions
.16**
-.56****
,38****
No. of stressful life events
Family responsibility expectation
Perceived cultural gap
.06 (.03)*
-1.05 (.12)****
.12*
.50****
3g****
.23 (.19)
1.24 (.23)****
.07
,27****
.03
28****
.10 (.07)
1.24 (.48)**
.08
Increment to R^ at this step
.13***
.36****
7^5, 319) = 3.03* 7='(9, 260) = 3.61***
F(X5,247) = 14.86****
7?2 Total
4-7****
Adjusted R^ Total
44****
•Zero-order correlation coefficients between GDS and independent variables.
*p < .05. •*p < .01. ***p < .001. ****p < .0001.
250
SocialWork VOLUME 51, NUMBER 3
JULY 2006
the nationality dummy variables or the interaction
term between nationality and stress variables were
statistically significant (they are, therefore, not shown
in the table).These results mean that the impact of
acculturation stress on depression did not vary by
national origin among these Asian elders and that
the additive model fits the data better than the interactive model. On the other hand, the lack of
significant results in the interaction model may be
due to insufficient power in the model.
As noted inTable 3, three coping resources variables and four stress variables were significant correlates in explaining GDS depression scores, other
things being equal. Stress variables as a group had
the strongest impact on depression scores. Specifically, poor perceived health had the strongest effect
((3 = -.50), followed by the number of stressful life
events ((3 = .27), number of children living in proximity (P = -17), assistance from adult children (P =
.14), perceived cultural gap (P = .13), religiosity (P
= -.13), and length of stay (P = .12). The family
responsibility expectation factor did not make any
significant impact on depression levels, but the perception of these family value differences between
generations did.
DISCUSSION AND IMPLICATIONS
The findings support the authors'primary research
question that higher acculturation stress is associated with higher depression levels among the six
subgroups ofAsian immigrant elders in the sample.
The findings suggest that these Asian immigrant
elders, like other elderly groups, are vulnerable to
psychological distress in the form of depressive
symptoms. Our findings also confirm that Asian
immigrant elders are not a homogenous group as
observed in the significant differences in depression experienced among the six Asian ethnic groups.
Compared with other studies in the literature, the
rates and incidence of depression in these six groups
ofAsian elders were higher than observed in most
other community ethnic elderly samples. Statistically, survey results based on subgroups of small
sample size can be subject to large sampling error
(Kerlinger & Lee, 1999). Readers should, therefore,
be cautioned that the higher rate of depression
observed in this study might be a statistical artifact
of the limited size of each ethnic group. Nevertheless, depression has been documented as a risk factor associated with suicide and is more likely to be
undetected or misdiagnosed in ethnic minority el-
ders, partly because of cultural barriers to diagnosis
and treatment (Mui, 1993,1996a, 1996b, 2001).The
findings of this study suggest that health and social
service agencies need to set as a priority examination of the depression among these six subgroups
ofAsian immigrant elders and development of culturally appropriate interventions that address their
mental health needs. Depression is a treatable disease. As such, these findings urge social workers to
engage in early screening and intervention of depression among elders in these Asian subgroups.
Where screening depression is concerned, practitioners should err on the side of caution.
Relatively speaking, the data suggested that these
Filipino and Korean immigrant elders seemed to
have adjusted better than other groups because they
admitted to a much lower rate of depression in the
study. On the other hand, the data also suggested
that the incidence of depression among Japanese,
Vietnamese, Chinese, and Indian immigrant elders
was higher than other elderly American populations in the literature (Gallo & Lebowitz, 1999;
Hazuda,Wood, Lichtenstein, & Espino, 1998). Elders in these Asian subgroups varied in their emotional well-being. It is difficult to speculate why
this variance was found. It may be based on variance in the total life experiences of this sample.
Based on a life course perspective, people's early
life experiences affect their worldviews and development of adaptive resources (Stroller & Gibson,
1999).Exploring the diversity in the social and historical contexts in which today's cohort ofAsian
American and Asian immigrant elders experienced
significant life course events is essential in understanding their social reality and personal struggles
with aging. Each group ofAsian immigrant elders
may have been through war-related trauma or political turmoil in their early lives.These life experiences may become painful memories, especially if
they had hardship, regrets, accumulated stresses, or
unresolved family conflicts (Mui, 1996b; 2001). For
example, Japanese American and Japanese immigrant elders in the United States during and after
World War II may have experienced more prejudice and discrimination (Shibusawa & Mui, 2001).
The Vietnamese elders may have suffered from a
higher degree of migration grief than other groups
because they immigrated to the United States involuntarily after the Vietnam War (Ngo,Tran, Gibbons, & Oliver, 2001).Elders from China may have
experienced the horrible pohtical turmoil during
M U I A N D K A N G / Acculturation Stress and Depression among Asian Immigrant Elders
251
WWII as well as the Cultural Revolution.The Partition of India may have left Indian immigrants
devastated because of the loss of many lives in the
riots (Keen, 1998). As immigrants in the United
States, these six subgroups ofAsian elders are disadvantaged and underprivileged by age, ethnicity, class,
and language (Mui, 1996a;2001; StroOer & Gibson,
1999).These human dimensions cannot be understood in additive terms because they interlock to
shape the social reality and emotional responses of
these Asian elders. They do, however, add to our
understanding of the potential that the unique life
events experienced by these elders may have on
the risk of the higher incidence of depression observed within the Asian ethnic subgroups studied
here.
Specifically, the results from the multivariate
analyses confirmed some existing notions about
correlates of depression among elders ofAsian heritage. The findings suggest that life stress and acculturation stress are major concerns of Asian immigrant elders. Consistent with the literature (Black
et al., 1998; Falcon & Tucker, 2000; Mills &
Henretta, 2001), self-rated poor health had the
strongest effect on depression among these elders.
The comorbidity and coexistence of depression
with poor health is a very complex issue (Mui,
1996b). It reflects the "which comes first" causal
phenomenon: Does physical illness result in depression or vice versa? Methodologically, longitudinal research is needed to examine the causal relationship between health status and depression and
how depression changes over time. Poor perceived
health may be an indication of lack of access to
health care. Coping with physical and psychological losses associated with illness and disability can
be extremely challenging.This is especially true for
Asian elders who may lack the financial resources
and language proficiency to negotiate the United
States health care system.This finding points, therefore, to the importance of designing culturally sensitive health maintenance and health promotion
programs to meet Asian elders' health care needs.
The second-most-powerful correlate of depression was the stressful life events variable. This is a
well-documented relationship among the general
older population as well as ethnic elderly populations (Casado & Leung, 2001; Swenson, Baxtor,
Shetterly, Scarbro, & Hamman, 2000).The stressful
life events variable measured major emotional, psychological, relationship, and social losses through
252
death, illnesses, victimization, relocation, or separation from children. Coping with multiple losses
and victimization (from robbery, burglary, microaggression, prejudice, racial discrimination, or racial oppression) may be extremely difficult for these
Asian immigrant elderly subgroups, especially if
they lack proficient language skills and other resources to deal with their social realities. Some of
the life events in this composite variable may have
profound cultural impacts on Asian elders. Social
work practitioners need to be sensitive to these
cultural differences. For example, death of a significant person in an elder's life may be more difficult for the Asian elders because they lost not only
the loved one, but also a significant source of support and hope (Mui, 1996b, Shibusawa & Mui,
2001).
This study's findings also point to the need to do
more research to understand intergenerational family relationships and the way these family ties can
lead to emotional distress and family conflict instead of support and comfort (Mills & Henretta,
2001). Another stressful life event of great emotional impact for Asian elders is the split of household between the elderly parent and his or her adult
children.This is especially so for those elders who
may still have high expectations of family solidarity
and coresidence (Mui, 1996b). Culturally, the split
of family household may be another indication of
intergenerational conflicts (Falcon &Tucker, 2000;
Lee et ah, 1996). The Asian parent may be confronted with the loss of power and respect because
his or her role as cultural conservator and family
decision maker may be undermined. Social workers need to be sensitive to the cultural meanings of
these changes within a family system and be able to
provide services to support Asian American
intergenerational families to deal with the disjunctions between expectations of family support and
changing social reality.
Other stress-related predictors of depression
were a perceived cultural gap and a longer length
of residence in the United States The association
of acculturation stress and depression is consistent
with the literature (Falcon & Tucker, 2000; Hovey,
2000). It is interesting that the measures of elder's
levels of acculturation in terms of family responsibility expectations had no effect on depression, but
the perception of these value differences between
generations did matter. This seems to indicate that
among these elders, intergenerational agreement
SocialWork VOLUME 51, NUMBER 3
JULY 2006
on family expectations was of greater importance
in preventing depression than what the exact expectations were. Longer residence in the United
States usually is associated with lower levels of depression (Lam, Pacala, & Smith, 1997; Mills &
Henretta, 2001). In this study, the direction of this
relationship is reversed. A possible cultural explanation for this finding is that the longer the elder
had lived in the United States the more likely he
or she was to have American-born children and
grandchildren whose acculturation and family expectations differed from the foreign-born elder.The
perceived cultural gap between generations may
reflect elders' anxiety and worries about the future. They may not be sure whether children will
be there to help when they need them.
ing" as a care recipient was associated with increased
distress (Liang,Krause,&Bennett,2001).This study's
findings provided more specificity in the relationship between social support and well-being in these
subgroups ofAsian immigrant elders. Social workers should not assume more assistance from family
to elders is always better. It is possible that more
assistance from children and frequent encounters
between the two generations may jeopardize the
quality of their relationships. Social work intervention needs to be sensitive to this association and to
assess whether the frequency of intergeneration
contact is a cause ofthe elder's depression or is the
result of the depression. Interventions should be
targeted based on whether and which causal relationship exists.
Acculturation is a multidimensional process (Mills
& Henretta, 2001).Perceived intergenerational differences in cultural values may manifest in such
areas as sense of ethnic identity, communication
style, family values, family role and gender role
expectations, or hfestyle choices. When working
with Asian American clients, social workers must
be sensitive to assessing intergenerational dynamics. In families with high levels of intergenerational
tension, social workers are advised to provide family counseling to enable these Asian families to work
through the cultural value and expectation differences. Such intervention may result in reduced
depression among the elderly parents and less stress
among their children.
Coping resources variables (fewer children living in proximity, more assistance from children, and
being less religious) also had a significant impact on
levels of depression among this study sample. The
association between fewer children Hving in proximity and depression is consistent with research in
the United States and Asia (Chi & Chou, 2001;
Hovey,2000;Lee et al., 1996). Culturally, Asian elderly parents may still expect their adult children to
live with them or to live in proximity so that they
will be available to provide support to them. Fewer
accessible children may lead to the elder's sense of
social isolation and insecurity. Additional research
on this association is advised to develop culturally
meaningful social support and intervention.
In terms ofthe effects of receiving assistance on
psychological well-being, these Asian elders reported
higher levels of depression when they received more
assistance from children. This is consistent with a
national study that also found that "over-benefit-
The link between religiosity and depression
noted here is also consistent with research on religion and health (Ai, Dunkle, Peterson, & Boiling,
1998; Hovey, 2000). The literature suggests that
most immigrants continue or rebuild their spiritual lives in the new land (Min & Kim, 2002). In
this study, high religiosity was assumed when respondents reported that they considered religion
to be very important to them. Religious interpretation of stressful life events may have power to
bring believers to a state of inner peace or acceptance of a situation that is beyond their control
(Idler, 2002). Religion is a powerful spiritual coping resource (Ai et al.).The finding observed here
on the association of higher religiosity and less depression points to the importance of collaboration
between religious or faith-based organizations and
the health care systems so that holistic mental
health care to Asian immigrant elders and to elderly populations in general is possible.
Overall, evidence from the data confirms that
there is diversity within diversity among the Asian
immigrant subgroups in the study. This study data
suggest that there is substantial variation in the
depression experienced among Asian immigrant
elderly subpopulations. Social workers need to note
the ethnic variations observed here and be careful
not to assess Asian elders as a single group. The
study found that these groups differ in sociodemographic backgrounds, financial status, English pro.ficiency,level of religiosity, support from children,
length of residence in the United States, health
conditions, life stresses, acculturation level, perceived
cultural gap, and sense of psychological well-being.
Findings point to the need to equip social workers'
M U I AND K A N G / Acculturation Stress and Depression among Asian Immigrant Elders
253
cultural competence with evidence-based knowledge about these culturally diverse subgroups of
Asian immigrant elders.
Although this may be the first study with an area
probability sample of Asian American elders, its
generalizability is somewhat limited because of the
design and small size of the subgroups. It is also
limited by the fact that all respondents were foreign born, so differences between native and foreign-born Asian elders were not studied. Readers
need to be cautious in drawing inferences about
the wider population of Asian American elders,
Asian immigrant elders, and specific Asian American nationality subgroups. Because the study was
based on a sample of Asian elders living in an urban setting densely populated by people of Asian
cultural heritage, the findings may be more indicative of characteristics of Asian elders who live in
urban ethnic enclaves than it is of those who are
more dispersed. Despite such limitations, the findings of this study provide extremely important
empirical information for social workers and other
helping professionals working with Asian immigrant elders and their families. Findings also have
substantial implications for mental health programs
and practice. Data point to the need to design culture-sensitive, bilingual depression screening, prevention and intervention programs for elderly
Asian populations. The within-group differences
in this study confirm the importance of not grouping all Asian elders into one cultural group when
screening, assessing, and treating depression. This
study reinforces that more research is needed to
understand the internal heterogeneity within this
population.
REFERENCES
Ai, A. L., Dunkle, R. E., Peterson, C , & Boiling, S. (1998).
The role of private prayer in psychological recovery
among midlife and aged patients following cardiac
surgery. CewtitologisI, 38{5), 591-601.
Aldwin, C. M. (1994). Stress, coping and development. New
York: Guilford Press.
American Association of Retired Persons. (1997).
Depression in /a(er/i/c.Washington, DC: Author.
Bartels, S.J., Coakley, E., Oxman,T. E., Constantino, G.,
Oslin, D., Chen, H., Zubritsky, C , Cheal, K., Durai,
U.N.B., Gallo,J.J., Llorente, M., & Sanchez, H.
(2002). Suicidal and death ideation in older primary
care patients with depression, anxiety, and at-risk
alcohol use. American Journal of Geriatric Psychiatry,
r 0,417-427.
Black, S. A., Markides, K. S., & Miller,T. Q. (1998).
Correlates of depressive symptomatology among
older community-dwelling Mexican Americans: The
Hispanic EPESE.JoHmak of Gerontology: Series B:
254
Psychological Sciences and Social Sciences,
53B{4),
S198-S208.
Burnette, D., & Mui, A. C. (1994). Determinants of selfreported depressive symptoms by frail elderly
persons living 3\one.Journal of Gerontological Social
l%rfe, 22(1/2), 3-19.
Casado, B. L., & Leung, P. (2001). Migratory grief and
depression among elderly Chinese American
immigrants.Jowrna/ of Gerontological Social Work,
3^(1/2), 5-26.
Chi, I., & Chou, K.-L. (2001). Social support and
depression among elderly Chinese people in Hong
Kong. International fournal ofAging and Human
Development, 52(3), 231-252.
Falcon, L. M., & Tucker, K. L. (2000). Prevalence and
correlates of depressive symptoms among Hispanic
elders in Massachusetts. Jowrnab of Gerontology: Series
B: Psychological Sciences and Social Sciences,
55B{2),
S108-S116.
Gallo,J.J.,& Lebowitz, B. D. (1999).The epidemiology of
common late-life mental disorders in the community:Themes for the new century. Psychiatric Services,
50,1158-1166.
Gelfand, D., & Yee, B. W. K. (1991). Influence of
immigration, migration, and acculturation on the
fabric of aging in America. Generations, 75(4), 7—10.
Gonzalez, H. M., Haan, M. N., & Hinton, L. (2001).
Acculturation and the prevalence of depression in
older Mexican Americans: Baseline results of the
Sacramento Area Latino Study on Aging._/oMr«a/ of
the American Geriatrics Society, 45^,948-953.
Gottlieb, G. L. (1991, November). Barriers to care for older
adults with depression. Paper presented at the National
Institutes of Health Consensus Development
Conference on diagnosis and treatment of depression in later life, Bethesda, MD.
Halier, J.,Weggenmans, R. M., Ferry, M., & Guigoz,Y.
(1996). Mental health: Mini-mental state examination and geriatric depression score of elderly
Europeans in the SENECA study of 1993. European
fournal ofGUnical Nutrition, 50(Suppl. 2), SI 12-Sl 16.
Hazuda, H. P.,Wood, R. C , Lichtenstein, M. J., & Espino,
D.V. (1998). Sociocultural status, psychosocial
factors, and cognitive functional limitation in elderly
Mexican Americans: Findings from the San Antonio
Longitudinal Study of Aging. Joi/rna/ of Gerontological
SodalWork, 30{\/2),99-\2\.
Hetzel, L., & Smith, A. (2001). Vie 65 years and over
population: 2000 (Census 2000 Brief C 2 K : B R / O 1 10). Retrieved March 19, 2005, from http://www.
census.gov/prod/2001pubs/c2kbr01-10.pdf
Hovey,J. D. (2000). Acculturative stress, depression, and
suicidal ideation in Mexican immigrants. Guttural
Diversity and Ethnic Minority Psychology, 6{2), 134—
151.
Idler, E. L. (2002).The many causal pathways linking
religion to health. Public Policy and Aging Report,
12(4), 7-12.
Keen, S. (Spring 1998).The impact and aftermath of the
Partition of India. Retrieved March 19, 2005, from
http://www.emoryedu/ENGLISH/Bahri/
Part.html
Kerlinger, F N, & Lee, H. B. (1999). Foundations of
behavioral research. New York: Wadsworth.
Kessler, R. C , Foster, C,Webster, P. S., & House,J. S.
(1992).The relationship between age and depressive
symptoms in two national surveys. Psychology and
Aging, 7{\),n9-\26.
Lam, R. E., Pacala,J.T., & Smith, S. L. (1997). Factors
related to depressive symptoms in an elderly
Chinese American sample. Clinical Gerontologist,
J 7(4), 57-70.
SocialWork VOLUME 51, NUMBER 3
JULY 2006
Lapierre, S., PronovostJ., Dube, M., & Delisle, I. (1992,
September). Risk factors associated with suicide in
elderly persons living in the community. Canada
Mental Health, pp. 8-12.
LaVeist.T. A. (1995). Data sources for aging research on
racial and ethnic groups. Cerontotogist, 35, 328—339.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal and
coping. NewYork: Springer Verlag.
Lee, M. S., Crittenden, K. S., &Yu, E. (1996). Social
support and depression among elderly Korean
immigrants in the United States. International Journal
ofAging and Human Development, 42,313—327.
Liang,J., Krause, N. M., & Bennett, J.M. (2001). Social
exchange and well-being: Is giving better than
receiving? Psychology and Aging, J6,511—523.
Mills,T. L., & Henretta,J. C. (2001). Racial, ethnic, and
sociodemographic differences in the level of
psychosocial distress among older Americans.
Research on Aging, 23(2), 131-152.
Min, P. C , & Kim,J. H. (Eds.). (2002). Religions in Asian
America: Building faith communities.Wn\nut Creek,
CA: AltaMira Press.
Mui, A. C. (1993). Self-reported depressive symptoms
among black and Hispanic frail elders: A sociocultural perspective.Journal ofApplied Gerontology, J2(2),
170-187.
Mui, A. C. (1996a). Correlates of psychological distress
among Mexican, Cuban, and Puerto Rican elders
living in the USA. Journal of Cross-Culturat
Gerontology, ^ ( 2 ) , 131-147.
Mui, A. C. (1996b). Depression among elderly Chinese
immigrants: An exploratory study. SocialWork, 4i,
633-645.
Mui, A. C. (1996c). Geriatric Depression Scale as a
community screening instrument for elderly
Chinese immigrants. International Psychogeriatrics, 8,
445-458.
Mui, A. C. (2001). Stress, coping and depression among
elderly Korean immigrants.Jowrnfl/ of Human
Behavior in the Social Environment, 3(3/4), 281-299.
Mui, A. C , Burnette, D., & Chen, L. M. (2001). Crosscultural assessment of geriatric depression: A review
of the CES-D and the GDS.Journal of Mental Health
and Aging, 7(1), 137-164.
Mui, A. C , Kang, S.Y., Chen, L.M., & Domanski, M. D.
(2003). Reliability of the Geriatric Depression Scale
for use among elderly Asian immigrants in the USA.
International Psychogeriatric, ?5(3), 253-273.
Ngo, D.,Tran,T.V., Cibbons,J. L., & Oliver,J. M. (2001).
Acculturation, premigration traumatic experiences,
and depression among Vietnamese Americans. In N.
G. Choi (Ed.), Psychosocial aspects of the AsianAmerican experience: Diversity within diversity (pp. 225—
242). NewYork: Haworth Press
Pang, K.Y.C. (1998). Symptoms of depression in elderly
Korean immigrants: Narration and the healing
process. Culture, Medicine and Psychiatry, 22, 93-122.
Ryan, A. S., Mui, A., & Cross, R (2003). Asian American
elders in NewYork Gity:A study of health, social needs,
quality of life and quality of care. New York: Asian
American Federation of NewYork.
Sheikh,J. 1., & Yesavage, J. A. (1986). Geriatric Depression
Scale: Recent evidence and development of a
shorter version. Glinical Gerontology, 5, 165-172.
Shibusawa,T, & Mui, A. C. (2001). Stress, coping, and
depression among Japanese American elders. JoHma/
of Gerontological Social Work, 36(1/2), 63-81.
Stokes, S. C,Thompson, L.W., Murphy, S., & GallagherThompson, D. (2001). Screening for depression in
immigrant Chinese-American elders: Results of a
pilot study._/our«a/ of Gerontological Social Work, 36{\/
2), 27-44.
Stroller, E. P, & Gibson, R. C. (1999). Worlds of differences:
Inequality in the aging experience (3rd ed.).Thousand
Oaks, CA: Pine Forge Press.
Swenson, C.J., Baxtor,J., Shetterly, S. M., Scarbro, S. L., &
Hamman, R. F. (2000). Depressive symptoms in
Hispanic and non-Hispanic white rural elderly:The
San Luis Valley Health and Aging Study. American
Journal of Epidemiology, 152,1048-1055.
Tran,T. V. (1992). Psychological traumas and depression in
a sample of Vietnamese people in the United States.
Health and Social Work, 18,184-194.
U.S. Census Bureau. (1990). J990 census ofpopulation:
Asian and Pacific Islanders in the United States (1990
CP-3-5). Retrieved March 19, 2005, from http://
www.census.gov/prod/cenl990/cp3/cp-3-5.pdf
U.S. Census Bureau. (2000). Census 2000 Summary File 2
(SF2) WO-Percent Data. Retrieved August 30,2004
from http://factfmder.census.gov/servlet/
DTGeoSearchByListServlet?ds_name=DEC_2000_
SF2_U&_lang=en&_ts=112893145481
U.S. Census Bureau. (2001). Total population by age, race and
Hispanic or Latino origin for the United States: 2000
(Census 2000 Summary File 1). Retrieved March
19,2005, from http://www.census.gov/population/
cen2000/phc-t9/tab01 .pdf
Yesavage,J. A.,Brink,T. L., Rose,T. L., Lum, O., Huang,
V.,Adey, M., & Leirer,V. O. (1983). Development
and validation of a screening scale: A preliminary
report. Jowrna/ of Psychiatric Research, / 7(1), 37-49.
Ada C. Mui, PhD,ACSW, is associate professor,
Columbia University School of SocialWork, 1255
Amsterdam Avenue, NewYork, NY 10027, e-mail: acm5@
columbia.edu. Suk-Youitg Kang, PhD, is assistant
professor. School of SocialWork, Arizona State University,
Tempe. The authors would like to thank the Asian American
Federation of NewYork and Brookdale Center on Aging,
Hunter College for their collaboration in this research. The
opinions expressed in this article are those of the authors and
should not attribute to the collaborating organizations.
Original manuscript received January 8, 2004
Final revision received March 25, 2005
Accepted June 1, 2005
M U I AND KANG / Acculturation Stress and Depression among Asian Immigrant Elden