The Gerontologist Advance Access published July 25, 2014

The Gerontologist, 2014, Vol. 00, No. 00, 112

doi:10.1093/geront/gnu073
Research Article

The Protective Effects of Religiosity on

Depression: A2-Year Prospective Study
Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

Corina R.Ronneberg, MS,* Edward AlanMiller, PhD, MPA,

ElizabethDugan, PhD, and FrankPorell, PhD
Department of Gerontology, John E.McCormack Graduate School of Policy & Global Studies, University
of Massachusetts Boston.
*Address correspondence to Corina R.Ronneberg, MS, Department of Gerontology, John E.McCormack Graduate School
of Policy & Global Studies, University of Massachusetts Boston, 100 Morrissey Blvd., Boston, MA 02125. E-mail: Corina.
Ronneberg@gmail.com
Received October 29 2013; Accepted June 3 2014.

Decision Editor: Rachel Pruchno, PhD

Purpose of the Study: Approximately 20% of older adults are diagnosed with depression
in the United States. Extant research suggests that engagement in religious activity, or
religiosity, may serve as a protective factor against depression. This prospective study
examines whether religiosity protects against depression and/or aids in recovery.
Design and Methods: Study data are drawn from the 2006 and 2008 waves of the Health
and Retirement Study. The sample consists of 1,992 depressed and 5,740 nondepressed
older adults (mean age = 68.12 years), at baseline (2006), for an overall sample size
of 7,732. Logistic regressions analyzed the relationship between organizational (service
attendance), nonorganizational (private prayer), and intrinsic measures of religiosity and
depression onset (in the baseline nondepressed group) and depression recovery (in the
baseline depressed group) at follow-up (2008), controlling for other baseline factors.
Results: Religiosity was found to both protect against and help individuals recover from
depression. Individuals not depressed at baseline remained nondepressed 2years later
if they frequently attended religious services, whereas those depressed at baseline were
less likely to be depressed at follow-up if they more frequently engaged in private prayer.
Implications: Findings suggest that both organizational and nonorganizational forms of
religiosity affect depression outcomes in different circumstances (i.e., onset and recovery, respectively). Important strategies to prevent and relieve depression among older
adults may include improving access and transportation to places of worship among
those interested in attending services and facilitating discussions about religious activities and beliefs with clinicians.
Key words: Organizational religiosity, Public religiosity, Nonorganizational religiosity, Private religiosity, Intrinsic
religiosity, Religion, Social support, Mental health

Depression is a major concern in the United States, as

more than 5% of the general population over 12years old
reports being depressed at any given time (Pratt & Brody,

2008). The prevalence of depression becomes even more

alarming at older ages as 20% of those 65years and older
report being depressed (Hurst, Williams, King, & Viken,

The Author 2014. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved.
For permissions, please e-mail: journals.permissions@oup.com.

Page 1 of 12

The Gerontologist, 2014, Vol. 00, No. 00

sizes also tend to be small and focus exclusively on, for

example, the effects of religiosity on depression recovery,
rather than both depression onset and recovery (Bosworth,
Park, McQuoid, Hays, & Steffens, 2003; Hayward etal.,
2012; Koenig etal., 1998). There is a lack of consistency
in measurement as well, with one or more religiosity measures tending to be employed operationalizing such concepts
as organizational religiosity, nonorganizational religiosity,
intrinsic religiosity, religious salience, religious affiliation,
religious orthodoxy, and religious coping, though indicators of the former three domains are most commonly used
(Blay etal., 2008; Bosworth etal., 2003; Braam, Beekman,
Deeg, Smit, & Tilburh, 1997; Braam etal., 2004; Branco,
2000; Hayward et al., 2012; Idler & Kasl, 1992; King etal.,
2007; Koenig et al., 1997, 1998; Levin, 2010; Schnittker,
2001; Sun etal., 2012).
The primary goals of this study are to assess depression levels both at baseline and 2 years later, in order to
determine (a) whether religiosity protects against depression and (b) whether religiosity aids in the recovery from
depression. Shortcomings in extant research are addressed
in several ways. First, we use a larger, more representative
sample than the previous work, focusing expressly on older
adults (residing in the United States), and employ a longitudinal perspective. Second, we employ a comprehensive
set of religiosity indicators, including organizational, nonorganizational, and intrinsic measures, as well as religious
salience, affiliation, and presence of both friends and relatives at ones place of worship.

Religiosity, Depression, and OlderAdults

The biopsychosocial diathesis-stress model of depression
(BPDS) posits that there are certain interconnected biological, psychological, and social factors that can affect an
individuals predisposition to depression (Schotte, Bossche,
Doncker, Claes, & Cosyns, 2006; Stein, 2005). These
include risk factors that can serve both as potential precursors of depression and potential protective factors that
can act as buffers against depression. Some individuals are
more vulnerable to depression due to biological factors
(e.g., age and gender), somatic factors (e.g., health status,
chronic conditions, disability, or recent medical setbacks),
psychological factors (e.g., mental illness and serious alcoholic use), and social influences (e.g., marital status, education, income, social support, volunteerism, and adverse
life events). In this study, religiosity is conceptualized as a
protective factor, which may help buffer against ones total
vulnerability for depression.
Religiosity typically refers not only to a belief in a higher
entity or something greater than oneself but also formal
involvement in organized religious activities and specific,

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

2008; Paukert et al., 2008). Because depression is one of

the most common mental health issues facing older adults
(Administration on Aging, 2012), the study of depression
and its correlates has become a priority among scholars
looking to improve the quality of life among this population (Administration on Aging, 2012).
Although conflicting results have been reported (e.g., positive, negative, or no relationship), recent investigations suggest
that involvement in religious activity may serve as a protective
factor against depression (Blay, Batista, Andreoli, & Gastal,
2008; Blazer, 2010; Hayward, Owen, Koenig, Steffens, &
Payne, 2012; King et al., 2007; Koenig, 2007, 2009; Law
& Sbarra, 2009; Smith, McCullough, & Poll, 2003). These
investigations also suggest that individuals who are more
religious may not only be more likely to recover from certain ailments such as acute myocardial infarction (Martin &
Levy, 2006) and severe mental illness (Webb, Charbonneau,
McCann, & Gayle, 2011), but do so more quickly, while
experiencing shorter hospitalization stays (Contrada et al.,
2004). Together, these findings suggest a potentially important avenue for preventing and/or promoting recovery from
depression, especially given the large role that religion plays
in the lives of most Americans, 90% of whom report believing in God or a universal spirit (Gallup, 2013) and 90% of
whom report engaging in prayer (Hill etal., 2000).
The role that religion plays in peoples lives becomes
more pronounced with age. One national survey, for
example, found that nearly 70% of adults 50 years or
older reported that religion is very important in their lives
compared with 44% of adults under 30years old (Cohen
& Koenig, 2003). Older adults also exhibit higher levels
of religiosity or actual involvement in religious activities
(Boswell, Kahana, & Dilworth-Anderson, 2006). The fact
that religiosity appears to increase with age coupled with
the high prevalence of depression among older adults suggests the need to further study the effects of religious beliefs
and activities on depression among the elderly.
The need to further study the effects of religiosity on
depression is also suggested by current research. Most existing research in this area has been cross-sectional (Blay etal.,
2008; Branco, 2000; Lawler-Row & Elliott, 2009; Waddell
& Jacobs-Lawson, 2010; Yohannes, Koenig, Baldwin, &
Connolly, 2008). That which is longitudinal has focused on
limited population subsets (e.g., African Americans elders
and adolescents with psychiatric conditions) (Dew et al.,
2010; Ellison & Flannelly, 2009), local or regional populations (Idler & Kasl, 1992; King etal., 2007; Koenig etal.,
1997; Koenig, George, & Peterson, 1998; Sun etal., 2012),
and non-U.S. samples (e.g., Australian, Lebanese, Israeli,
European elders, or Brazilian) (Braam etal., 2001; Chaaya,
Sibai, Fayad, & El-Roueiheb, 2007; Iecovich, 2001; Law &
Sbarra, 2009; Payman, George, & Ryburn, 2008). Sample

Page 2 of 12

The Gerontologist, 2014, Vol. 00, No. 00

Page 3 of 12

between nonorganizational religiosity and depression

cross-sectionally, but a U-shaped association longitudinally
(King etal., 2007). Yet another study found nonorganizational religiosity to be associated with lower depression
severity after 3 months (Hayward et al., 2012). Findings
around intrinsic religiosity are also inconsistent. Parker and
coworkers (2003) found no relationship between intrinsic
religiosity and depression, King and coworkers (2007) a
positive relationship, and Koenig and coworkers (1998)
quicker remission from depression.
Evidence suggests that the impact of religiosity on depression is stronger among women who also tend to be more
active participants in both organizational and nonorganizational religious activities than men, including, for example, religious affiliation and private prayer (Wink & Dillon,
2002; Yohannes etal., 2008). This tendency is reflected in
a 2002 Health and Retirement Study (HRS) of older adults
60 years or older, which found higher ratings of religious
importance to be a protective factor against depression in
womenbut not men (Waddell & Jacobs-Lawson, 2010).
Based on previous research and according to the BPDS
model of depression, this study hypothesizes that (a)
higher religiosity will be associated with a lower likelihood
of depression 2 years later among older adults without
depression at baseline (i.e., religiosity will protect against
depression onset) and (b) higher religiosity will be associated with a lower likelihood of depression 2years later for
respondents depressed at baseline (i.e., religiosity will aid in
depression recovery).

Methods
DataSource
The sample in this study was drawn from the 2006 and
2008 waves of the HRS; the goal was to model depression in 2008 based on respondent characteristics in 2006.
The HRS is sponsored by the National Institute on Aging
(grant number NIA U01AG009740) and is conducted by
the University of Michigan (Health and Retirement Study,
2006/2008). The HRS began collecting data in 1992 and
continues to do so every 2years. The HRS is a nationally
representative study that contains rich information on
more than 22,000 community-dwelling older adults aged
50 or older, with respect to respondent health, functional
status, cognition, living arrangements, retirement, religious
affiliation, and involvement in assorted activities.

Sample
The sample utilized in this study includes the subset of
HRS respondents who completed the 2006 Leave-Behind
Questionnaire (n = 7,732). The rationale for utilizing the

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

measurable acts such as prayer, meditation, service attendance, religious readings, and affiliation with a particular religion or place of worship (Hill etal., 2000; Iecovich, 2001;
Yohannes etal., 2008). Akey characteristic of religion is that
it is organized in a hierarchical fashion with an identified
authority figure such as a priest, pastor, or rabbi presiding.
Although religion refers to someones belief system, religiosity is the actual application of such beliefs in dailylife.
The literature distinguishes between three general types
of religiosity: organizational, nonorganizational, and intrinsic. Organizational religiosity typically involves public or
group activities and is most commonly measured by ones
religious service attendance (Koenig etal., 1998; Sun etal.,
2012). Nonorganizational religiosity, by contrast, is more
private and typically occurs on a persons own time, alone,
encompassing activities such as reading religious texts,
praying, and/or meditating (Koenig etal., 1998; Sun etal.,
2012). Intrinsic religiosity is concerned with individuals
subjective meaning of religiosity and how religious beliefs
affect everyday life (Sun etal., 2012). Studies have shown
that as individuals age, they are more likely to engage in
nonorganizational activities as opposed to organizational
modes of religious expression (Yohannes etal., 2008), possibly shifting to more private religious activities, perhaps
due to physical decline, rather than giving up on religious
involvement altogether.
Nearly 75% of older people who suffer from depression or anxiety partake in some kind of religious activity at
least monthly (Paukert etal., 2009). Ameta-analysis of 147
studies found higher religiosity to be associated with fewer
depressive symptoms or indicators in more than three quarters of the studies analyzed (Smith etal., 2003). Furthermore,
individuals who regularly attend religious services display
lower rates of depression when compared with individuals who either do not attend services or do attend services
but on a more sporadic basis (Blazer, 2010; Braam et al.,
2004; Koenig etal., 1997). In a prospective study focusing
on African Americans over the age of 55, it was found that
individuals who received guidance from their religion on a
regular basis were less likely to suffer from major depression
34years later (Ellison & Flannelly, 2009).
Extant research also demonstrates that religious involvement may benefit clinically depressed individuals (Koenig
etal., 1998; Murphy & Fitchett, 2009). Depressive symptoms have been shown to decrease across time in persons
engaged in organizational religiosity (Braam et al., 2004;
Koenig, 2007; Law & Sbarra, 2009; Levin, 2010; Smith
et al., 2003). However, mixed results abound regarding
nonorganizational modes of religious involvement. For
instance, one cross-sectional study found nonorganizational religiosity unrelated to depression (Koenig et al.,
1997), whereas another found an inverse relationship

The Gerontologist, 2014, Vol. 00, No. 00

Leave-Behind Questionnaire is that it contains much more

comprehensive measures of religiosity than the basic HRS
survey. In all, 944, or 12% of respondents, were missing
information on at least one HRS item. Thus, for purposes
of our analyses, we employed multiple imputation of missing data to fill in the missing values. Twenty imputations
were conducted and pooled results were used in the analyses reported.

Dependent Variable (Depression)

Depression status (in 2008) is the outcome variable in
the present study. Depression is assessed with Center for
Epidemiological Studies Depression scale (CESD-8). The
presence of three or more symptoms, out of eight, indicates
a higher likelihood of being clinically depressed (Steffick,
2000). Therefore, respondents reporting three or more
depressive symptoms were coded 1=depressed; those with
zero, one, or two symptoms as 0=not depressed.
Independent Variables (Religiosity)
Five religiosity questions are asked in the basic HRS.
Religious affiliation was self-reported as Protestant,
Catholic, Jewish, or none/other religion. Respondents were
asked about organizational religiosity, via the frequency
of attendance at religious services: high (more than once
a week or once a week), moderate (two to three times a
month), and low/none (one or more times a year or not at
all). Additionally, respondents were asked about the presence of both friends and relatives in ones congregation (yes
or no). Lastly, respondents were asked to rate the importance of religiosity: very, somewhat, or not important.
Each of these items was coded as a series of dichotomous
variables.
The Leave-Behind Questionnaire includes two additional measures of religiosity. The first is an index of
religiosity, an intrinsic measure, composed of four items
( = .92)believe God watches over me, events unfold
according to a divine/greater plan, carry religious beliefs
into all dealings in life, find strength and comfort in religion. Possible scores range from 1 (strongly disagree) to
6 (strongly agree) (averaged across the four items) where
higher scores indicate higher religiosity levels. The second
Leave-Behind Questionnaire item measured the frequency
of prayer in private contexts, a nonorganizational measure. The scores (18) on this item were reverse coded
so that higher scores denote higher frequency of private
prayer.
The potential for multicollinearity was examined
in several ways. Neither variance inflation factors
(all<4.5) nor correlations (all < 0.62) among the seven

religiosity measures revealed problematic multicollinearity. Moreover, each of the seven religiosity variables
was entered one by one into the model and as a block,
both with and without covariates, both for the depressed
and nondepressed samples. Results on the religiosity
variables largely remained consistent across these various specifications. The final model therefore includes all
seven religiosity variables described previously along
with covariates.
Covariates
This study controls for biological, somatic, psychological, and social factors that have been found to be associated with depression. Prior research suggests that older
adults, females, and non-Hispanic Blacks exhibit higher
rates of depression compared with their younger, male,
and white counterparts (Law & Sbarra, 2009; Pratt &
Brody, 2008). Age is measured as a continuous variable (number of years); gender as a dichotomous variable, with female = 1 and male = 0; and race/ethnicity
as a series of dichotomous variables for white, black,
Hispanic, andother.
In addition, somatic or health conditions may be
related to depression status (Blazer, 2010; Centers for
Disease Control and Prevention, 2011a; Koenig, 1999; Lo
etal., 2010; Schotte etal., 2006). At baseline, respondents
were asked whether they had ever been diagnosed with
each of seven chronic ailments: high blood pressure, diabetes, cancer, lung disease, heart conditions, stroke, and
arthritis. The number of chronic conditions was summed
(07) with a higher count indicating greater comorbidity.
Respondents were also asked whether they had recently
experienced each of three somatic life events in the last
two years (since baseline): stroke, heart attack, and/or
cancer. It is important to account for the onset of illnesses
such as these, as recently experiencing a negative event has
been found to be associated with depression (Schnittker,
2001). A count (03) of somatic events was developed,
with a higher number indicating greater comorbidity.
Self-reported health status was measured using a series
of dichotomous indicator variables: excellent, very good,
good, fair, or poor. Functional limitations were measured
using counts of both instrumental activities of daily living
(IADLs) (06) and basic activities of daily living (ADLs)
(05).
Alcohol abuse appears to be associated with depression (Blay et al., 2008; Braam et al., 2004; Centers
for Disease Control and Prevention, 2011a; Idler &
Kasl, 1992; Rodriguez, Schonfeld, King-Kallimanis, &
Amber, 2010). Consistent with previous research (Satre,
Gordon, & Weisner, 2007), respondents were identified
as a serious drinker, or abuser of alcohol, if they were

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

Measurement

Page 4 of 12

The Gerontologist, 2014, Vol. 00, No. 00

Page 5 of 12

AnalyticalPlan
Basic descriptive statistics are reported, followed by bivariate analyses comparing the baseline characteristics of the
depressed and nondepressed samples (Table 1). Results
from multivariate analyses are presented next, utilizing logistic regression to model the relationship between
depression and religiosity, controlling for other baseline
factors. Two logistic regressions models were employed:
one with baseline depressed respondents and the second
with baseline nondepressed respondents (Table2).

Results
Table 1 reports descriptive statistics for the entire sample, as well as differentiated by depression status in 2006.
Depressed and nondepressed samples had similar religious
affiliations (2=3.94 (3), p > .05). At 45% and 37%, respectively, high frequency of religious service attendance was
more likely to be reported by nondepressed than depressed
respondents, whereas depressed respondents were more
likely to report low or no service attendance than their nondepressed counterparts (51% vs. 43%) (2 = 43.19 (2), p
< .001). Ahigher proportion of nondepressed respondents
(59%) reported having friends at their congregation than
depressed respondents (50%) (2 = 44.42 (1), p < .001);
approximately one quarter of each group reported sharing their congregation with family (2=0.01 (1), p > .05).
There was a small but significant difference assigned to the
importance of religion, with depressed respondents being

Table1. Descriptive and Bivariate Statistics for Baseline (2006) Depressed and Nondepressed Samples
Covariates

Religiosity factors
Religiosity (basic HRS questions)
Religious affiliation
Protestant
Catholica
Jewish
None/other
Service attendance
High
Moderatea
Low/none
Friends in congregation
Relatives in congregation
Importance of religion
Very important
Somewhat important
Not importanta

Entire sample (n=7,732)

Depressed (n=1,992)

Nondepressed (n=5,740)

# (%)/mean (SD)

# (%)/mean (SD)

# (%)/mean (SD)

4,892 (63.3%)
2,048 (26.5%)
158 (2.0%)
614 (7.9%)

1,276 (64.1%)
501 (25.2%)
48 (2.4%)
162 (8.1%)

3,616 (70.0%)
1,547 (27.0%)
110 (1.9%)
452 (7.9)

3,324 (43.0%)
917 (11.9%)
3,493 (45.2%)
4,367 (56.5%)
1,891 (24.5%)

740 (37.2%)
232 (11.7%)
1,021 (51.3%)
998 (50.1%)
489 (26.6%)

2,584 (45.0%)
685 (12.0%)
2,472 (43.1%)
3,369 (58.7%)
1,402 (24.4%)

5,273 (68.2%)
1,591 (20.6%)
871 (11.3%)

1,390 (69.8%)
422 (21.2%)
183 (9.2%)

3,883 (67.7%)
1,170 (20.4%)
688 (12.0%)

2(df)/t

3.94 (3)

43.19 (2)***

44.42 (1)***
0.01 (1)
10.60 (2)**

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

(a) a woman that consumes more than two drinks per

occasion or (b) a man that consumes more than three
drinks per occasion. Also included is a dichotomous
variable indicating whether or not an individual had
ever been diagnosed with any emotional or psychiatric
condition(s) (Koenig et al., 1998). Both variables have
been shown to increase the risk for, or coexist with,
depression (Aina & Suman, 2006; Centers for Disease
Control and Prevention, 2011b) and have been used as
covariates in similar studies.
Social factors have the potential to predispose individuals toward depression (Schotte etal., 2006). The influences
of social and economic considerations are reflected, in part,
in sociodemographic variables, including marital status
(married, divorced/separated, widowed, or never married),
education (in years), and household income (in quartile
earnings). Social considerations are also reflected, in part,
in living alone, volunteerism, and having family and friends
nearby. Certain recent adverse life events such as serious
losses, threatening occurrences, or difficulties in life are
also predictive of depression (Schnittker, 2001; Schotte
et al., 2006, p. 314). These include experiencing divorce/
separation, death of a spouse/partner, a nursing home stay,
and residential move in the last 2years. Acount has been
created (04), where a higher number indicates experiencing more recent adverse social events. Lastly, whether
respondents were living in a nursing home as opposed to a
community setting at the time of the survey was recorded,
in addition to whether survey responses were provided by
a proxy or not.

The Gerontologist, 2014, Vol. 00, No. 00

Page 6 of 12

Table1. Continued
Covariates

Depressed (n=1,992)

Nondepressed (n=5,740)

# (%)/mean (SD)

# (%)/mean (SD)

# (%)/mean (SD)

2(df)/t

4.99 (1.39)
6.11 (2.32)

5.02 (1.33)
6.25 (2.26)

4.98 (1.40)
6.06 (2.34)

1.26
3.17**

68.11 (1.39)
4,543 (58.8%)

68.80 (11.44)
1,317 (66.1%)

67.87 (10.40)
3,226 (56.2%)

3.19***
59.96 (1)***

6,009 (77.7%)
1,004 (13.0%)
601 (7.7%)
118 (1.5%)

1,462 (73.4%)
292 (14.8%)
204 (10.2%)
34 (1.7%)

4,547 (79.2%)
712 (12.4%)
397 (6.9%)
84 (1.5%)

33.84 (3)***

1.92 (1.33)

2.39 (1.37)

1.75 (1.27)

18.46***

904 (11.7%)
2,350 (30.4%)
2,395 (31.0%)
1,554 (20.1%)
543 (7.0%)
0.34 (.82)
0.34 (.94)
0.07 (0.27)

70 (3.5%)
310 (15.6%)
547 (27.5%)
702 (35.2%)
369 (18.5%)
0.63 (1.07)
0.80 (1.38)
0.09 (0.30)

834 (14.5%)
2,040 (35.5%)
1,848 (32.2%)
851 (14.8%)
174 (3.0%)
0.24 (0.70)
0.18 (0.66)
0.07 (0.26)

1174.17 (4)***

351 (4.5%)
1,244 (16.1%)

101 (5.2%)
655 (32.9%)

250 (4.4%)
589 (10.3%)

1.74 (1)
558.35 (1)***

5,039 (65.2%)
966 (12.5%)
1,504 (19.5%)
223 (2.9%)
12.58 (3.11)

1,029 (51.7%)
342 (17.2%)
542 (27.2%)
79 (4.0%)
11.78 (3.30)

4,010 (70.0%)
624 (10.9%)
962 (16.8%)
144 (2.5%)
12.86 (2.99)

216.04 (3)***

1,747 (22.6%)
1,922 (25.4%)
1,985 (25.7%)
2,078 (26.9%)
0.12 (0.12)

722 (36.2%)
528 (26.5%)
391 (19.6%)
351 (17.6%)
0.12 (0.33)

1,025 (17.9%)
1,394 (24.3%)
1,594 (27.8%)
1,727 (30.1%)
0.12 (0.33)

347.93 (3)***

1,673 (21.6%)
2,786 (36.0%)
2,157 (27.9%)
5,026 (65.0%)
283 (3.7%)
99 (1.3%)

614 (30.8%)
475 (23.8%)
577 (29.0%)
1,241 (62.3%)
63 (3.2%)
48 (2.4%)

1,059 (18.4%)
2,311 (40.2%)
1,580 (27.5%)
3,784 (70.0%)
220 (3.8%)
52 (0.9%)

133.54 (1)***
172.90 (1)***
0.94 (1)
11.80 (1)***
1.88 (1)
18.96 (1)***

15.01***
19.48***
2.73**

13.53***

0.45

Notes: ADLs=activities of daily living; HRS=Health and Retirement Study; IADLs=instrumental activities of daily living; LBQ=Leave-Behind Questionnaire.
a
Denotes reference groups.
*p < .05. **p < .01. ***p < .001.

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

Religiosity (LBQ)
Index of religiosity
Frequency private prayer
Biological variables
Demographic variables
Age
Female
Race/ethnicity
Whitea
Black
Hispanic
Other
Somatic variables
Health and functional limitation
Chronic conditions
Self-reported health
Excellent
Very good
Gooda
Fair
Poor
IADLs
ADLs
Somatic adverse life events
Psychological variables
High alcohol use
Psychological issues
Social variables
Sociodemographic variables
Marital status
Marrieda
Divorced/separated
Widowed
Never married
Education level
Household income
Quartile 1a
Quartile 2
Quartile 3
Quartile 4
Social adverse life events
Social support variables
Living alone
Volunteer status
Relatives live near
Friends live near
Have proxy respondent
Live in nursing home

Entire sample (n=7,732)

The Gerontologist, 2014, Vol. 00, No. 00

Page 7 of 12

Table2. Logistic Regressions Modeling the Relationship Between Follow-Up (2008) Depression and Religiosity and Other
Covariates Among Baseline (2006) Depressed and Nondepressed Samples
Covariates

Nondepressed at baseline
(n=5,740)

Odds ratio

p value

Odds ratio

p value

0.91

2.05
0.907

0.465

0.040*
0.644

1.03

1.30
1.19

0.768

0.382
0.325

1.36

1.18
0.92
1.13

0.062

0.330
0.498
0.336

0.65

0.75
0.95
0.92

0.001***

0.035*
0.659
0.432

0.81
1.00

0.332
0.995

1.23
1.01

0.275
0.977

1.10
0.93

0.052
0.015*

1.00
0.98

0.949
0.476

0.99
1.174

0.099
0.513

1.00
1.44

0.319
0.000***

0.81
0.95
1.04

0.158
0.785
0.971

0.873
1.02
0.67

0.304
0.889
0.304

1.06

0.127

1.09

0.020*

0.43
0.80

1.32
1.33
1.03
1.04
1.57

0.004**
0.140

0.029*
0.086
0.640
0.412
0.008**

0.54
0.72

1.77
2.69
0.896
1.13
1.16

0.000***
0.002**

0.000***
0.000***
0.113
0.058
0.106

0.803
1.607

0.329
0.000***

1.03
1.93

0.877
0.000***

0.99
0.72
0.70

0.972
0.048*
0.197

0.88
0.88
0.62

0.407
0.390
0.103

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

Religiosity factors
General HRS religiosity items
Religious affiliation
Protestant
Catholica
Jewish
None/other
Frequency of attendance at religious services
High
Moderatea
Low/none
Friends in congregation
Relatives in congregation
Importance of religion
Very important
Somewhat important
Not importanta
Religiosity items (LBQ)
Index of religiosity
Frequency of private prayer
Biological variables
Demographic variables
Age
Female
Race/ethnicity
Whitea
Black
Hispanic
Other
Somatic variables
Health and functional limitation variables
Self-reported chronic conditions
Self-reported health
Excellent
Very good
Gooda
Fair
Poor
IADLs
ADLs
Somatic adverse life events
Psychological variables
Serious alcohol use
Psychological issues
Social variables
Sociodemographic variables
Marital status
Marrieda
Divorced/separated
Widowed
Never married

Depressed at baseline
(n=1,992)

The Gerontologist, 2014, Vol. 00, No. 00

Page 8 of 12

Table2. Continued
Covariates

Nondepressed at baseline
(n=5,740)

Odds ratio

p value

Odds ratio

p value

0.98

0.155

0.99

0.333

0.73
0.81
0.66
1.461

0.017*
0.189
0.023*
0.011*

0.93
0.82
0.69
1.42

0.528
0.143
0.017*
0.002**

1.12
0.421
1.16
0.227
1.31
0.018*
0.954
0.659
0.000
1.000
0.411
0.049*
2 log likelihood: 1941.1***

1.35
0.035*
0.90
0.275
1.23
0.025*
0.91
0.291
0.236
0.000***
0.741
0.538
2 log likelihood: 3648.2***

Notes: ADLs=activities of daily living; HRS=Health and Retirement Study; IADLs=instrumental activities of daily living; LBQ=Leave-Behind Questionnaire.
a
Denotes reference groups.
*p < .05. **p < .01. ***p < .001.

more likely to report very important (70% vs. 68%)

and depressed respondents being less likely to report not
important (9% vs. 12%) (2 = 10.60 (2), p < .01). No
significant difference could be discerned between the two
groups on the religiosity index (t=1.26, p > .05), but there
was a significant difference in regard to the frequency of private prayer (t=3.17, p < .01) with depressed respondents
reporting higher frequency (6.25 vs. 6.06 [out of8]).
Significant differences could also be discerned with
respect to all covariates but high alcohol use (2=1.74 (1),
p > .05), social adverse life events (r=.45, p > .05), the presence of relatives living nearby (2=0.94 (1), p > .05), and
having a proxy respondent (2=1.88 (1), p > .05). Thus,
nondepressed respondents were more likely than depressed
respondents to be younger (r = 3.19, p < .001), male
(2=59.96, p < .001), non-Hispanic white (2=33.84, p <
.001), in excellent very good/good health (2=1174.17, (4),
p < .001), married (2=216.04, p < .001), have higher education (t=13.53, p < .001), have higher income (2=347.93
(3), p < .001), volunteer (2=172.90, p < .001), and have
friends living nearby (2 = 11.80, p < .001). By contrast,
depressed respondents were more likely than nondepressed
respondents to be chronically ill (t = 18.46, p < .001),
IADL (t=15.01, p < .001) and ADL (t=19.48, p < .001)
impaired, suffer from adverse somatic life events (t=2.73,
p < .01), have psychological issues (2=558.35, p < .001),
live alone (2=133.54 (1), p < .001), or live in a nursing
home (2=18.96 (1), p <.001).

Two main logistic regression models were estimated. The

first model is composed of individuals who were depressed
at baseline (n=1,992) (Table2). Two religiosity variables
were significant. The odds of being depressed at follow-up
were two times higher among depressed respondents with
a Jewish affiliation (odds ratio [OR]=2.05, p < .05) but
lower for those with more frequent engagement in private
prayer (OR=0.93, p <.05).
The first model also indicates that depressed individuals
who were in excellent health (OR = 0.43, p < .01), widowed (OR=0.72, p < .05), had higher household income
(OR = 0.73, p < .05; OR = 0.66, p < .05), or who lived
in a nursing home (OR=0.41, p < .05) had a decreased
likelihood of being depressed at follow-up. In contrast,
depressed persons who reported more somatic life events
(OR=1.57, p < .01), had psychological issues (OR=1.61,
p < .001), reported more social adverse life events
(OR = 1.46, p < .05), and lived closer to their relatives
(OR=1.31, p < .05) were more likely to remain depressed
2yearslater.
The second model is composed of individuals who were
not depressed at baseline (n = 5,740) (Table 2). Service
attendance was the only religiosity variable to prove significant. In particular, nondepressed respondents with high
service attendance were 35% less likely to be depressed
at follow-up (OR = 0.65, p < .01), and respondents with
low/no service attendance were 25% less likely to become
depressed (OR=0.75, p < .05), in comparison to those with

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

Education level
Household income
Quartile 1a
Quartile 2
Quartile 3
Quartile 4
Social adverse life events
Social support variables
Living alone
Volunteer status
Relatives live near
Friends live near
Have proxy respondent
Live in a nursing home

Depressed at baseline
(n=1,992)

Page 9 of 12

Discussion
This study sought to understand whether religiosity (a)
protects against future depression and (b) plays a role in
depression recovery, among older adults. Consistent with
prior research (Blazer, 2010; Koenig, 2007; Smith et al.,
2003), results provide evidence supporting both of these
expectations, though the specific aspect of religiosity found
to protect against depression (frequent service attendance)
was different from the component found to aid in depression recovery (private prayer frequency). Relative to those
with moderate service attendance, individuals who were
not depressed at baseline (in 2006)were less likely to be
depressed 2 years later if they frequently attended religious services. It was expected that high service attendance would protect against depression, perhaps due to the
availability, promotion, or benefits of social support found
in ones place of worship. In particular, the protective pathway stemming from service attendance may derive from
the comparatively higher levels of social capital resulting
from engagement in public modes of behavior, specifically
the interpersonal relationships formed and sustained by
active participation in a religious congregation. The presence of more social connections, in turn, may reduce the
likelihood of isolation and loneliness, two factors associated with depression.
Counterintuitively, individuals with low service attendance who were not depressed at baseline were also less
likely to be depressed 2 years later relative to those with
moderate service attendance. It is possible that persons
with low or no service attendance may be less likely to
be depressed at follow-up because they are more likely to
engage in other, less public forms of religiosity that, in turn,
provide protective benefits from depression and other ailments. That this may be the case is suggested by the moderate, significant inverse correlation between private prayer
frequency and level of service attendance (r = .496, p <
.05). Thus, whereas the high service attendance group may
be disproportionately devoted to organizational forms of

religiosity, the low service attendance group may be disproportionately devoted to nonorganizational forms. In
contrast, the moderate service attendance group may not
be disproportionately devoted to either form of religious
behavior and, as such, may be less likely to experience the
benefits that derive fromeach.
Consistent with expectations, persons who started
out depressed at baseline were less likely to be depressed
2 years later if they more frequently engaged in private
prayer. This finding suggests that persons who become
depressed may turn to their faith for support and as a
means of coping from adverse life eventsfinancial,
health, social, or otherwise. Subsequent engagement in
private prayer may serve, in part, to cultivate hope for
the future, potentially activating cognitive resources that
eventually counter depression.
Interestingly, Jewish respondents were much more likely
to remain depressed at follow-up than other respondents.
One possible explanation for this finding could be the
long-term, negative implications that belonging to a religious minority has on mental health (Berger, 1977). This
may be particularly important for the population surveyed
because anti-Semitism was much more visible and prevalent during our respondents formative years than it is
today. Another possible explanation could be that Jewish
elders may not benefit in the same way from religious
involvement as members of other religious affiliations.
Take Christian doctrine, for example, which emphasizes
the afterlife or Heaven at which point the body may be
restored and a reunion takes place with long deceased
loved ones (Gillman, 2007). This belief can be great source
of solace, hope, and comfort for those going through hard
times (e.g., depression), which may, in turn, support coping and recovery. This is in contrast to Jewish doctrine,
which in downplaying the hereafter in favor of the here
and now, may not provide the same level of solace, hope,
and comfort (Gillman, 2007).
One interesting yet surprising finding emerging from
this study is that having relatives living nearby was associated with depression at follow-up among both the baseline
depressed and nondepressed samples analyzed. It is plausible that there are certain unwanted expectations inherent
when family members live closerwhether, for example,
caring for a frail and disabled parent or other relative in
need of long-term care or watching a young grandchild
in need of after school care, that results in burdens and
stresses that might not otherwise exist if family lived further away. Simply measuring proximity, moreover, does
not account for the frequency or quality of the interactions
that take place. For example, some relationships, even with
relatives, may not be pursued no matter how convenient, if
those relationships are not fulfilling.

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

moderate service attendance. Other factors associated with

increased likelihood of depression onset included female
gender (OR=1.44, p < .001), chronic illness (OR=1.09,
p < .05), fair or poor health (OR=1.77, OR=2.69, both
p < .001), psychological issues (OR = 1.93, p < .001),
adverse social events (OR = 1.42, p < .01), living alone
(OR = 1.35, p < .05), and having relatives live nearby
(OR=1.23, p < .05). By contrast, factors that appeared to
guard against depression onset included reporting excellent
or very good health (OR=0.54, p < .001; OR=0.72, p
< .01), higher income (OR=0.69, p < .05), and having a
proxy respondent (OR=0.24, p < .001).

The Gerontologist, 2014, Vol. 00, No. 00

The Gerontologist, 2014, Vol. 00, No. 00

Conclusion
Several implications for policy and practice follow from
the results of this study. One is related to transportation
availability and the provision of better access to places of
worship so that older adults who are interested in religious
services are able to attend and subsequently benefit from
organizational, or public, forms of religiosity. Moreover,
given the high prevalence of depression among older
adults, clinicians should be cognizant of the benefits associated with both religious service attendance and involvement in private prayer, assess individuals religious needs
and involvement, and determine whether their clients face
any barriers to attending services or pursuing their faith if
they so desire. Through these assessments, clinicians could
help connect interested clients to such services within their

communities or help them overcome any barriers they may

be experiencing, hindering involvement private prayer. Care
plans or therapy goals can be developed, which address
these issues as well.

References
Administration on Aging. (2012). Older Americans behavioral health. Retrieved from http://www.aoa.gov/AoARoot/
AoA_Programs/HPW/Behavioral/docs2/Issue%20Brief%20
Series%20Overview.pdf
Aina, Y., & Suman, J. L. (2006). Understanding comorbidity with
depression and anxiety disorders. Journal of the American
Osteopathic Association, 106, S9S14. Retrieved from http://
www.jaoa.org/content/106/5_suppl_2/S9.full.pdf
Berger, D. M. (1977). The survivor syndrome: Aproblem of nosology and treatment. American Journal of Psychotherapy, 31,
238251. Retrieved from http://www.ajp.org
Blay, S. L., Batista, A. D., Andreoli, S. B., & Gastal, F. L. (2008). The
relationship between religiosity and tobacco, alcohol use, and
depression in the elderly community population. The American
Journal of Geriatric Psychiatry, 16, 934943. doi:10.1097/
JGP.0b013e3181871392
Blazer, D. G. (2010). The origins of late-life depression. Psychiatric
Annals, 40, 1318. doi:10.3928/00485718-20091229-01
Boswell, G. H., Kahana, E., & Dilworth-Anderson, P. (2006).
Spirituality and healthy lifestyle behaviors: Stress counter-balancing effects on the well-being of older adults. Journal of Religion
and Health, 45, 587602. doi:10.1007/s10943-006-9060-7
Bosworth, H. B., Park, K. S., McQuoid, D. R., Hays, J. C., & Steffens,
D. C. (2003). The impact of religious practice and religious coping on geriatric depression. International Journal of Geriatric
Psychiatry, 18, 905914. doi:10.1002/gps.945
Braam, A. W., Beekman, A. T.F., Deeg, J. J.H., Smit, J. H., & Tilburh,
W. V. (1997). Religiosity as a protective or prognostic factor of
depression in later life; Results from a community survey in
The Netherlands. Acta Psychiatrica Scandinavia, 96, 199205.
doi:10.1111/j.1600-0447.1997.tb10151.x
Braam, A. W., Hein, E., Deeg, D. J.H., Twisk, J. W.R., Beekman, A.
T.F., & Tilburg, W. V. (2004). Religious involvement and 6-year
course of depressive symptoms in older Dutch citizens: Results
from the Longitudinal Aging Study Amsterdam. Journal of Aging
and Health, 16, 467489. doi:10.1177/0898264304265765
Braam, A. W., Van Den Eeden, P., Prince, M. J., Beekman, A. T.F.,
Kivela, S. L., Lawlor, B. A., Copeland, J. R. M. (2001).
Religion as a cross-cultural determinant of depression in
elderly Europeans: Results from the EURODEP collaboration. Psychological Medicine, 31, 803814. doi:http://dx.doi.
org/10.1017/S0033291701003956
Branco, K. J. (2000). Religiosity and depression among nursing home
residents: Results of a survey of ten states. Journal of Religious
Gerontology, 12, 4361. doi:10.1300/J078v12n01_05
Centers for Disease Control and Prevention. (2011a). Mental health
basics. Retrieved from http://www.cdc.gov/mentalhealth/basics.htm
Centers for Disease Control and Prevention. (2011b). Depression.
Retrieved from http://www.cdc.gov/mentalhealth/basics/mentalillness/depression.htm

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

There are several limitations worth noting. First, depression is measured using the CESD-8, a self-report tool,
rather than using the clinical diagnosis of depression by a
health or mental health professional. Although the latter
may be the gold standard, the CESD-8 is a commonly used
and accepted measure of depression in studies such as this
one where clinical diagnosis was not possible (e.g., Steffick,
2000). Second, data from the CESD-8 were not utilized to
develop a measure of depression based on a continuous
count of depressive symptoms but instead used to place
individuals into depressed or nondepressed categories based
on the presence of three or more of the eight symptoms
assessed. One implication is that potentially useful variation may have been missing. This cutoff approach, however, is typically employed in studies utilizing the CESD-8
(Steffick, 2000). Athird limitation is related to the length of
the study. Given the episodic nature of depression, a 2-year
longitudinal study may miss signs of depression occurring after the time period analyzed. Future research should
extend the follow-up period studied over a longer period
of time as additional waves of the HRS become available.
Last, the Health and Retirement Study only includes measures of religiosity but not spirituality. Thus, this study is
focused exclusively on the former but not the latter. This
limitation is important to point out because extant research
suggests that spirituality may be associated with lower
rates of depression and mental illness as well (Skarupski,
Fitchett, Evans, & Mendes de Leon, 2010). Moreover, a
growing body of research suggests that spirituality to be a
unique construct, though related to religiosity (Underwood
& Teresi, 2002). Beyond this understanding experts hold
differing views regarding the distinction between religiosity
and spirituality, some maintaining that religiosity may be a
part of spirituality, whereas others viewing spirituality as a
part of religiosity (Hill etal., 2000; MacKinlay, 2006).

Page 10 of 12

Page 11 of 12

Koenig, H. G. (2007). Spirituality and depression: Alook at the evidence. Southern Medical Journal, 100, 737739. doi:10.1097/
SMJ.0b013e318073c68c
Koenig, H. G. (2009). Research on religion, spirituality, and mental
health: Areview. The Canadian Journal of Psychiatry, 54, 283
291. Retrieved from http://publications.cpa-apc.org/browse/
documents/464
Koenig, H. G., George, L. K., & Peterson, B. L. (1998). Religiosity
and remission of depression in medically ill older patients.
American Journal of Psychiatry, 155, 536542. Retrieved from
http://ajp.psychiatryonline.org/article.aspx?articleid=172792#A
bstract
Koenig, H. G., Hays, J. C., George, L. K., Blazer, D. G., Larson, D. B., &
Landerman, L. R. (1997). Modeling the cross-sectional relationships between religion, physical health, social support, and depressive symptoms. The American Journal of Geriatric Psychiatry, 5,
131144. doi:10.1097/00019442-199700520-00006
Law, R. W., & Sbarra, D. A. (2009). The effects of church attendance
and marital status on the longitudinal trajectories of depressed
mood among older adults. Journal of Aging and Health, 21,
803823. doi:10.1177/0898264309338300
Lawler-Row, K. A., & Elliott, J. (2009). The role of religious
activity and spirituality in the health and well-being of
older adults. Journal of Health Psychology, 14, 4352.
doi:10.1177/1359105308097944
Levin, J. (2010). Religion and mental health: Theory and research.
International Journal of Applied Psychoanalytic Studies, 7, 102
115. doi:10.1002/aps.240
Lo, C., Lin, J., Gagliese, L., Zimmermann, C., Mikulincer, M., &
Rodin, G. (2010). Age and depression in patients with metastatic cancer: The protective effects of attachment security and
spiritual wellbeing. Aging & Society, 30, 325336. doi:10.1017/
S0144686X09990201
MacKinlay, E. (2006). Spiritual care: Recognizing spiritual needs
of older adults. Journal of Religion, Spirituality & Aging, 18,
5971. doi:10.1300/J496v18n02_05
Martin, K. R., & Levy, B. R. (2006). Opposing trends of religious
attendance and religiosity in predicting elders functional recovery after an acute myocardial infarction. Journal of Religion &
Health, 45, 440451. doi:10.1007/s10943-006-9037-6
Murphy, P. E., & Fitchett, G. (2009). Belief in a concerned God predicts
response to treatment for adults with clinical depression. Journal
of Clinical Psychology, 65, 10001008. doi:10.1002/jclp.20598
Parker, M., Roff, L. L., Klemmack, D. L., Koenig, H. G., Baker, P., &
Allman, R. M. (2003). Religiosity and mental health in southern,
community-dwelling older adults. Aging & Mental Health, 7,
390397. doi:10.1080/1360786031000150667
Paukert, A. L., Phillips, L., Cully, J. A., Loboprabhu, S. M., Lomax,
J. W., & Stanley, M. A. (2009). Integration of religion into cognitive-behavioral therapy for geriatric anxiety and depression.
Journal of Psychiatric Practice, 15, 103112. doi:10.1097/01.
pra.0000348363.88676.4d
Payman, V., George, K., & Ryburn, B. (2008). Religiosity of
elderly depressed inpatients. International Journal of Geriatric
Psychiatry, 23, 1621. doi:10.1002/gps.1827
Pratt, L. A., & Brody, D. J. (2008). Depression in the United States
Household Population, 20052006. Centers for Disease Control
and Prevention: U.S. Department of Health and Human Services.
Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/19389321

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

Chaaya, M., Sibai A. M., Fayad, R., & El-Roueiheb, Z. (2007).

Religiosity and depression in older people: Evidence from underprivileged refugee and non-refugee communities in Lebanon. Aging
& Mental Health, 11, 3744. doi:10.1080/13607860600735812
Cohen, A. B., & Koenig, H. G. (2003). Religion, religiosity, and spirituality in the biopsychosocial model of health and ageing. Ageing
International, 28, 215241. doi:10.1007/s12126-002-1005-1
Contrada, R. J., Goyal, T. M., Cather, C., Rafalson, L., Idler,
E. L., & Krause, T. J. (2004). Psychosocial factors in outcomes of heart surgery: The impact of religious involvement
and depressive symptoms. Health Psychology, 23, 227238.
doi:10.1037/0278-6133.23.3.227
Dew, R. E., Daniel, S. S., Goldston, D. B., McCall, W. V., Kuchibhatla,
M., Schleifer, C., Koenig, H. G. (2010). A prospective study
of religion/spirituality and depressive symptoms among adolescent psychiatric patients. Journal of Affective Disorders, 120,
149157. doi:10.1016/j.jad.2009.04.029
Ellison, C. G., & Flannelly, K. J. (2009). Religious involvement and
risk of major depression in a prospective nationwide study
of African American adults. Journal of Nervous and Mental
Disease, 197, 568573. doi:10.1097/NMD.0b013e3181b08f45
Gallup. (2013). Religion. Retrieved from http://www.gallup.com/
poll/1690/Religion.aspx#1
Gillman, N. (2007). How will it all end? Eschatology in science and
religion. Cross Currents, 57, 3850. Retrieved from http://www.
crosscurrents.org/GillmanSpring07.pdf
Hayward, R. D., Owen, A. D., Koenig, H. G., Steffens, D. C., &
Payne, M. E. (2012). Longitudinal relationships of religion with
posttreatment depression severity in older psychiatric patients:
Evidence of direct and indirect effects. Depression Research and
Treatment, 2012, 18. doi:10.1155/2012/745970
Health and Retirement Study (HRS 2006 Core; HRS 2008 Core)
public use dataset. (2006/2008). Produced and distributed by
the University of Michigan with funding from the National
Institute on Aging (grant number NIA U01AG009740). Ann
Arbor, MI. Retrieved from http://hrsonline.isr.umich.edu/index.
php?p=regcou.
Hill, P. C., Pargament, K. I., Hood, R. W., Jr., McCollough, M.
E., Swyers, J. P., Larson, D. B., & Zinnbauer, B. J. (2000).
Conceptualizing religion and spirituality: Point of commonality,
points of departure. Journal for the Theory of Social Behaviour,
30, 5177. doi:10.1111/14685914.00119
Hurst, G. A., Williams, M. G., King, J. E., & Viken, R. (2008).
Faith-based intervention in depression, anxiety, and other mental disturbances. Southern Medical Journal, 101, 388392.
doi:10.1097/SMJ.0b013e318167a97a
Idler, E. L., & Kasl, S. V. (1992). Religion, disability, depression, and
the timing of death. American Journal of Sociology, 97, 1052
1079. doi:10.1086/229861
Iecovich, E. (2001). Religiousness and subjective well-being among
Jewish female residents of old age homes in Israel. Journal of
Religious Gerontology, 13, 3146. doi:10.1300/J078v13n01_04
King, D. A., Lyness, J. M., Duberstein, P. R., He, H., Tu, X.
M., & Seaburn, D. B. (2007). Religious involvement and
depressive symptoms in primary care elders. Psychological
Medicine, 37, 18071815. doi:http://dx.doi.org/10.1017/
S0033291707000591
Koenig, H. G. (1999). The healing power of faith. New York: Simon
& Schuster.

The Gerontologist, 2014, Vol. 00, No. 00

The Gerontologist, 2014, Vol. 00, No. 00

Stein, H. F. (2005). It aint necessarily so: The many faces of the

biopsychosocial model. Families, Systems, & Health, 23, 440
443. doi:10.1037/1091-7527.23.4.440
Sun, F., Park, N. S., Roff, L. L., Klemmack, D. L., Parker, M., Koenig,
H. G., & Allman, R. M. (2012). Predicting the trajectories of
depressive symptoms among southern community-dwelling
older adults: The role of religiosity. Aging & Mental Health, 16,
189198. doi:10.1080/13607863.2011.602959
Underwood, L. G., & Teresi, J. A. (2002). The daily spiritual experience scale: Development, theoretical description, reliability,
exploratory factor analysis, and preliminary construct validity
using health related data. Annals of Behavioral Medicine, 24,
2233. doi:10.1207/S15324796ABM2401_04
Waddell, E. L., & Jacobs-Lawson, J. M. (2010). Predicting positive wellbeing in older men and women. International Journal of Aging and
Human Development, 70, 181197. doi:10.2190/AG.70.3.a
Webb, M., Charbonneau, A. M., McCann, R. A., & Gayle, K. R.
(2011). Struggling and enduring with God, religious support,
and recovery from severe mental illness. Journal of Clinical
Psychology, 67, 11611176. doi:10.1002/jclp.20838
Wink, P., & Dillon, M. (2002). Spiritual development across
the adult life course: Findings from a longitudinal study.
Journal of Adult Development, 9, 7994. doi:10.102
3/A:1013833419122
Yohannes, A. M., Koenig, H. G., Baldwin, R. C., & Connolly, M. J.
(2008). Health behaviours, depression and religiosity in older
patients admitted to intermediate care. International Journal
of Geriatric Psychiatry, 23, 735740. doi:10.1002/gps.1968

Downloaded from http://gerontologist.oxfordjournals.org/ at University Library, University of Illinois at Chicago on July 31, 2014

Rodriguez, C. A., Schonfeld, L., King-Kallimanis, B., & Amber,

M. G. (2010). Depressive symptoms and alcohol abuse/misuse in older adults: Results from the Florida BRITE project.
Best Practice in Mental Health: An International Journal, 6,
90102. Retrieved from http://essential.metapress.com/content/
r1k65231237068k1/
Satre, D., Gordon, N. P., & Weisner, C. (2007). Alcohol consumption, medical conditions and health behavior in older adults.
American Journal of Health Behavior, 31, 238248. doi:http://
dx.doi.org/10.5993/AJHB.31.3.2
Schnittker, J. (2001). When is faith enough? The effects of religious
involvement on depression. Journal for the Scientific Study of
Religion, 40, 393411. doi:10.1111/0021-8294.00065
Schotte, C. K. W., Bossche, B. V. D., Doncker, D. D., Claes, S., &
Cosyns, P. (2006). A biopsychosocial model as a guide for psychoeducation and treatment of depression. Depression and
Anxiety, 23, 321324. doi:10.1002/da.20177
Skarupski, K. A., Fitchett, G., Evans, D. A., & Mendes de Leon, C.
F. (2010). Daily spiritual experiences in a biracial, communitybased population of older adults. Aging & Mental Health, 14,
779789. doi:10.1080/13607861003713265
Smith, T. B., McCullough, M. E., & Poll, J. (2003). Religiousness
and depression: Evidence for a main effect and the moderating
influence of stressful life events. Psychological Bulletin, 129,
614636. doi:10.1037/0033-2909.129.4.614
Steffick, D. E. (2000). HRS/AHEAD documentation report.
Retrieved March 7, 2011, from http://hrsonline.isr.umich.edu/
sitedocs/userg/dr-005.pdf

Page 12 of 12