Latino Alzheimer’s Disease Caregivers and Depression

Abstract

This study used stress coping theory to examine the effects of spirituality and religion on depression among a sample of Latino family members caring for a person with Alzheimer’s disease (AD) in the United States. Participants consisted of 209 Latino caregivers (CGs) drawn from baseline data from the Resource for Enhancing Alzheimer’s Caregivers Health (REACH) II clinical trial. The findings indicate that church attendance moderates the relationship between subjective forms of stress and depression in tandem with exhibiting direct effects on depression. Consistent with the central role religion plays in Latino culture, the results imply that religious involvement may play an important role in mitigating depression through indirect and direct pathways.

Keywords

Alzheimer’s disease caregiving Hispanics Latinos spirituality

In less than a decade, by 2019, Latinos will be the largest minority group among older adults in the United States (Administration on Ageing, 2008a). The exponential growth of older Latinos with AD will result in a dramatic increase in the number of family caregivers (CGs; Coon et al., 2004). It is well established in the literature that caring for a family member with AD is stressful (Sorensen & Pinquart, 2005). Among the most common outcomes engendered by this stress is an increased likelihood of depression (Gallagher-Thompson, Gray, Dupart, Jimenez, & Thompson, 2008).

Despite the projected increase in the number of Latino CGs, surprisingly little research has examined the relationship between caregiving and depression among this population (Morano & Sanders, 2005). An extensive review of minority caregiving research revealed few studies on the topic (Dilworth-Anderson, Williams, & Gibson, 2002). Based on their review, Dilworth-Anderson et al. underscored the importance of further research using samples of Latino CGs, along with the use of conceptual models and culturally relevant constructs.

In keeping with these recommendations, this study uses Pearlin and associates’ (Pearlin, Mullan, Semple, & Skaff, 1990) stress coping model to examine the relationship between stress and depression among a sample of Latino AD CGs. In particular, we examine the effects of spirituality and religion on depression. These two overlapping constructs may reflect culturally salient mechanisms to deal with stress (Wilson, 2008). In the following section, we briefly discuss the stress coping model, distinctions and connections between spirituality and religion, and relevant empirical literature.

Literature Review

The stress coping model is widely used to conceptualize the relationship between caregiving and detrimental health outcomes such as depression (Dilworth-Anderson et al., 2002). Empirical tests of Pearlin and associates’ stress coping model indicate that it appears to fit the caregiving experience across racial and ethnic groups (Hilgeman et al., 2009). At its most basic level, this model posits that objective and subjective stressors engender deleterious outcomes (see Figure 1). Objective stressors relate to the needs and demands of the care recipient (CR; that is, CR cognitive and physical functioning and CR behavioral problems). Subjective stressors refer to strains perceived by the CG (i.e., CG perceived health status, income adequacy, and the perceived burden of caregiving; Leblanc, Driscoll, & Pearlin, 2004; Pearlin et al., 1990). These stressors interact with each other as well as background and contextual variables, leading to negative outcomes.

Figure 1.

A model adapted from Pearlin and associates’ (Pearlin, Mullan, Semple, & Skaff, 1990) stress coping model; dotted line indicates a possible moderation effect.

Resource variables are hypothesized to act on the relationship between stressors and depression, leading to differential outcomes. Lower levels of depression are fostered through various pathways. Pearlin and associates (1990) cited social support as an example of a resource that may inhibit detrimental outcomes. Spirituality and religion are two other variables that may also impact CG depression.

Spirituality and religion are commonly conceptualized as overlapping, but distinct, multidimensional constructs (Aranda, 2008; Idler et al., 2009; Miller & Thoresen, 2003). Spirituality commonly pertains to an individual’s existential relationship with, or connection to, God or the Transcendent (Hodge, 2005; Wuthnow, 2007). Religion is typically related to a shared set of beliefs and practices that have been mutually developed with people who have similar understandings of the Transcendent and is an intermediary between an individual’s relationship with God or the Transcendent (Geppert, Bogenschutz, & Miller, 2007; Koenig, McCullough, & Larson, 2001). At the risk of oversimplifying, spirituality emphasizes the individual, whereas religion emphasizes the communal.

Spirituality and religion can be operationalized individually. Religious involvement is commonly measured by assessing attendance at religious services (Koenig et al., 2001). Spirituality, for instance, can be operationalized with a measure of prayer, an activity designed to nurture one’s connection with God (Farran, Paun, & Elliott, 2003). Alternatively, it is also possible to merge various dimensions of spirituality and religion into a single measure. Pargament’s (2007) measure of spiritual/religious coping might serve as an example.

Each construct provides a somewhat different rationale for alleviating depression, although the overlapping nature of spirituality and religion in most people’s lives complicates attempts to distinguish discrete mechanisms (Musick, Traphagan, Koenig, & Larson, 2000). To follow up on the above examples, prayer might be used to alleviate the stress associated with caregiving. Incorporating the Transcendent into the stress-coping equation can help CGs cope with trying circumstances by altering perceptions about the caregiving process. In contrast to this more individually oriented mechanism, church attendance posits that the communal nature of religion acts on the relationship between stress and depression. For example, attending religious services might mitigate stress through participating in a caring community of like-minded individuals who can help share the burden associated with caregiving by providing emotional support, theological reframing, help with the caregiving process, and other forms of assistance.

Qualitative research suggests that CGs frequently resort to spirituality and religion to deal with the stress associated with caring for someone with AD (Farran et al., 2003; Nightingale, 2003; Vickrey et al., 2007). For reasons that are presently unclear, quantitative research has attempted to test the relationship between spirituality and AD CGs but has revealed inconsistent results with various health-related outcomes (Hebert, Weinstein, Martire, & Schulz, 2006; Wilks & Vonk, 2008). Similarly, studies using Pearlin and associates’ (1990) stress coping theory to model the effects of spirituality and religion on stress and depression have found, at best, mixed results.

For instance, Morano and King (2005) tested the hypothesis that a multidimensional measure of spirituality and religion would mediate the relationship between stressors and depression. This hypothesis was not confirmed, although some evidence of a direct effect on depression, independent of stress, was found. Consistent with theory, higher levels of spirituality and religion were associated with lower levels of depression. Similarly, Leblanc and associates (2004) tested mediation and moderation models, along with an examination of direct effects. Neither of the two hypothesized models was confirmed using measures of prayer, religious attendance, and self-ascribed importance of religion. The only significant relationship was a direct association between self-ascribed importance of religion and depression. In contrast to theory, however, higher self-ascribed importance of religion was associated with higher levels of depression. These results, obtained with AD CGs, are generally consistent with the results obtained with other types of CGs (Hebert et al., 2006).

As is the case with AD CGs more broadly, the vast majority of the existing research on spirituality and religion among AD CGs has been conducted with Anglo-majority samples (Dilworth-Anderson et al., 2002; Hebert et al., 2006). Yet, there is reason to believe that spirituality and religion may be particularly salient mechanisms for coping with stress among Latino AD CGs. Although not every Latino is spiritually or religiously involved, spirituality and religion are woven into the fabric of Latino culture (Wilson, 2008).

Latino cultural values may influence the choice and use of strategies to deal with the challenges of caregiving (Knight & Sayegh, 2010). Consistent with this line of reasoning, research suggests that AD Latino CGs frequently use spiritual and religious (S/R) resources (Napoles, Chadiha, Eversley, & Moreno-John, 2010). For example, using data from the Resource for Enhancing Alzheimer’s Caregivers Health (REACH) I study, Coon and associates (2004) found that Latina CGs reported greater use of S/R coping than Anglo CGs. Specifically, Latina CGs reported higher levels of prayer and religious attendance and attributed greater importance to religion. Citing potential mechanisms such as enhanced social integration, social support, and a relationship with God, the authors speculated that S/R activities may be particularly important mechanisms for buffering the effects of stress.

Only a few studies have examined the relationship between various measures of spirituality and religion and depression among samples of Latino AD CGs. Using data from REACH I, Thompson and associates (Thompson, Solano, Coon, Mausbach, & Gallagher-Thompson, 2002) found that participation in spiritual activities was associated with lower levels of depression among Latina CGs, but not Anglo CGs. Using REACH II data, Lee, Czaja, and Schulz (2010) found that S/R coping moderated the effects of an intervention designed to reduce depression among African American CGs, but not Anglo or Latino CGs. A smaller study by Herrera and associates (Herrera, Lee, Nanyonjo, Laufman, & Torres-Vigil, 2009) revealed largely nonsignificant relationships between various measures of the religion and depression among a convenience sample of Mexican Americans (N = 66) caring for family members with or without dementia.

More research is clearly needed to understand the effects of spirituality and religion on depression among Latino AD CGs. Toward this end, the present study uses Pearlin and associates’ (1990) stress coping model to examine the relationship between stress and depression among a sample of Latino AD CGs. As Pearlin (2002) observes, spirituality and religion can impact depression through three pathways:

First, spirituality and religion may mediate the relationship between stressors and depression. In this pathway, the effect of stressors on depression is channeled through spirituality and religion. In essence, the stressors activate S/R responses that, in turn, reduce the effect of the stressors on depression.

Second, spirituality and religion may moderate the relationship between stressors and depression. Stress moderation occurs when CGs are exposed to a similar set of stressors and they vary in their levels of depression according to their level of spirituality and religion. In other words, spirituality and religion exert a buffering effect on stress only under certain conditions.

Third, it is also possible that spirituality and religion directly engender lower levels of depression, apart from any relationship with stress. In this pathway, spirituality and religion are not mediating channels through which the relationship between stress and depression is mitigated or a condition in which stress is buffered. Rather, stress, and spirituality and religion, both have separate and independent relationships with depression.

Given the present evidence, it seems premature to posit a specific pathway through which spirituality and religion may impact depression. Rather, in keeping with prior examinations of the relationship between stress and depression using Pearlin and associates’ model, we test all three pathways using various measures of spirituality and religion (Leblanc et al., 2004). More specifically, each pathway is tested using a measure of spirituality, religious involvement, and S/R coping.

In light of the importance attributed to spirituality and religion by Latino CGs, it seems reasonable to predict an effect on depression through at least one of the three pathways noted above. Although the largely nonsignificant quantitative findings underscore the need for caution, a significant body of research links spirituality and religion with lower depression (Koenig et al., 2001). Thus, we tentatively hypothesize that spirituality and religion will exhibit either an indirect or direct effect on depression. More specifically, we posit that higher levels of spirituality and/or religion will be associated with lower levels of depression.

Method

Participants and Procedures

To test this hypothesis, a secondary data analysis was conducted with baseline data from the REACH II project (Belle et al., 2006). REACH II is one of the few multisite clinical trials that implemented and evaluated multicomponent psychosocial interventions with CGs from diverse racial and ethnic backgrounds. Latino participants were recruited from multiple community organizations in three sites: Palo Alto, CA, Philadelphia, PA, and Miami, FL. Baseline data were collected during face-to-face interviews after obtaining informed consent from eligible participants.

To be included in the study, CGs had to meet the following criteria: be at least 21 years old, live with or share cooking facilities with the CR, provide an average of 4 or more hours of care per day, provide care in the past 6 months, and report a minimum of two distressful symptoms associated with caregiving (Belle et al., 2006). CRs had to have a diagnosis of AD or a related disorder, or a score of < 24 on the Mini-Mental State Exam (MMSE; Folstein, Folstein, & McHugh, 1975), and at least two functional impairments on the Instrumental Activities of Daily Living (IADL) Scale (Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963) or one impairment on the Activities of Daily Living (ADL) Scale (Lawton & Brody, 1969). Further information about the REACH II procedures is described elsewhere (Belle et al., 2006).

Measures

The study included a number of variables drawn from Pearlin and associates’ (Pearlin et al., 1990) stress coping model. Included among these were objective stressors (i.e., CR cognitive functioning, CR physical functioning, and CR behavior problems), subjective stressors (i.e., self-rated health, income inadequacy, and burden), potential resources (i.e., social support, spirituality and religion), and the outcome variable (i.e., depression). REACH investigators ensured all measures used in the study had good reliability and validity with Spanish-speaking CGs (Belle et al., 2006; Wisniewski, 2003).

Objective stressors

CR cognitive functioning was measured with the MMSE (Folstein et al., 1975). The MMSE measures the cognitive status of the CR by providing a brief assessment of a person’s orientation to time and place, recall ability, short-term memory, and mathematical ability. Scores range from 0 to 30, with scores equal to or below 24 indicating cognitive impairment. The average CR MMSE score of this sample was 11.33 (SD = 6.96). The Cronbach’s alpha of the MMSE was .81 in this study.

CR physical functioning was measured with the ADL (Katz et al., 1963) and IADL scales (Lawton & Brody, 1969). The 7-item ADL Scale assesses the CR’s ability to perform basic tasks of daily functioning independently (e.g., bathing, dressing, and eating). Similarly, the 8-item IADL Scale assesses the assistance needed to perform functional tasks such as shopping, operating the telephone, preparing meals, doing housework or laundry, and managing finances or medications. Total level of assistance needed for ADLs and IADLs was summed, with values ranging from 0 to 15 such that higher scores indicated more functional impairment. The Cronbach’s alpha of this scale was .86 in this study.

CR behavior problems were measured with the Revised Memory and Behavior Problem Checklist (RMBPC; Teri et al., 1992). The RMBPC assesses the presence of 24 problem behaviors that the CR may have exhibited in the past week (e.g., “CR had trouble remembering recent events”; “CR asked the same question over and over”). CGs rate the frequency of problem behaviors on a 4-point scale, ranging from not in the past week (0) to daily or more often (3). The summed score was used as a proxy to measure the presence of objective behavioral problems related to the dementia diagnosis. The scale’s Cronbach’s alpha in this study was .81.

Subjective stressors

Self-rated health and income inadequacy were assessed using single-item measures. Subjective health was assessed with an item from the caregiver health scale (Schulz et al., 1997). CGs rated their general health on a 5-point scale from excellent (0) to poor (4). Income inadequacy was assessed with an item that asked CGs to rate their perceived difficulty in paying for basic needs such as food, housing, medical care, and heating on a 4-point response key ranging from not difficult at all (0) to very difficult (3).

Burden was measured with a 12-item abbreviated version of the Zarit Caregiver Burden Inventory (ZCBI; Bedard et al., 2001; Zarit, Orr, & Zarit, 1985). The ZCBI assesses the perceived burden associated with caregiving (e.g., not enough time for oneself, not as much privacy). CGs rated each item on a 5-point scale from never (0) to nearly always (4). Total scale scores range from 0 to 48, with higher scores indicating higher burden. The scale’s Cronbach’s alpha in this study was .85.

Resources

Social support was assessed with a scale adapted from previous work on the topic (Lubben, 1988). This scale assessed the number of relatives that CGs hear from, and the number of persons CGs feel close to on a 6-point response key ranging from none (0) to nine or more (5). The summed scores range from 0 to 10, with higher scores indicating larger social networks and connectedness.

The study included measures to assess spirituality and religion, as separate and overlapping constructs. Religious involvement was measured by a single item that asks participants how often they attended church services from never (0) to nearly everyday (5). Spirituality was measured with a single item that assessed CG frequency of prayer/meditation on a 6-point response key that ranged from never (0) to nearly everyday (5).

A measure of S/R coping was also included (Pargament et al., 1998). This scale consists of two subscales: positive S/R coping (e.g., working with God as partners) and negative S/R coping (e.g., stress resulting from punishment from God). The second subscale can be reverse scored and added to the first subscale to obtain a single measure of beneficial S/R coping (Lee et al., 2010). The total score ranges from 0 to 18, with higher scores indicating higher levels of positive S/R coping. The Cronbach’s alpha in this study was .70.

Outcome

Depression was measured with the Center for Epidemiological Studies–Depression Scale (CES-D; Radloff, 1977). The 10-item CES-D Scale assesses the frequency with which depressive symptoms were experienced during the past week (e.g., “I was bothered by things that usually don’t bother me”; “I felt that everything I did was an effort”). Each item is answered on a 4-point response key that ranges from rarely or none of the time (0) to most or all of the time (3). Total scores range from 0 to 30, with higher scores indicating the presence of more depressive symptoms. The scale’s Cronbach’s alpha in this study was .85.

Data Analysis

Data analysis proceeded in three stages: descriptive, bivariate, and the multivariate analysis to determine the presence of indirect and direct effects. First, a descriptive analysis of the sample characteristics was preformed. Since the Latino population includes many relatively distinct subgroups, the one-way analysis of variance (ANOVA) and chi-square procedures were used to determine the existence of demographic differences based on nation of origin. Second, bivariate analysis was conducted to examine the relationships between the study variables and depression. Third, multivariate analysis was conducted to determine whether spirituality and religion (a) mediated, or (b) moderated the relationship between stress and depression, or (c) exhibited a direct effect on depression apart from the effects of stress. To test for mediation, we followed Baron and Kenny’s (1986) three-step procedure to identify possible mediating variables. To identify a significant mediation effect, a fourth step drawn from Kenny, Kashy, and Bolger (1998) was used.

To establish mediation, three conditions must be satisfied (Baron & Kenny, 1986). First, a significant relationship must exist between spirituality/religion and stressors (objective or subjective stressors). Second, a significant relationship must exist between the stressor and depression. Third, a significant relationship must exist between spirituality/religion and depression. Thus, in order for mediation to occur, indicators of spirituality/religion must be significantly related to depression and at least one stressor that must also be significantly related to depression. For a mediation effect to exist, the inclusion of the mediator in the multivariate model must reduce the effect of the stressor on depression (Kenny et al., 1998).

To test the first condition, bivariate correlation analyses were conducted. To test the second and third conditions, a series of individual regression analyses were conducted to control for the effects of the demographic variables and social support. For the fourth step, hierarchical regression analysis was used to identify a reduction in the effects of a given stressor on depression due to a given mediator. Hierarchical regression analyses were also used to test the existence of moderation effects and direct effects of spirituality and religion on depression.

Results

Descriptive Results

Of the 209 Latino CGs included in REACH II, 82.3% were female. The mean age was 58 years (SD = 13.7), and the average length of residence in the United States was 32 years (SD = 14.8). About 7.2% of the participants reported no religious affiliation, 69.1% identified themselves as Catholics, and 23.7 % reported belonging to other religious affiliations. In terms of national heritage, 34.6% were Cuban, 21.6% Mexican, 21.2% Puerto Rican, and 22.6 % reported other Latino backgrounds.

The Latino rubric includes many subcultures, which may hold relatively distinctive value systems (Hilgeman et al., 2009). As can be seen in Table 1, significant differences emerged between the four Latino groups in this study. Compared to the other three Latino groups, Cuban CGs were significantly older, more depressed, prayed less often and, with the exception of Mexicans or Puerto Ricans, attended religious services less frequently and reported lower S/R coping. Accordingly, due to distinctive nature of the Cuban subsample relative to other three groups, CGs were recoded into Cuban and non-Cuban Latinos for subsequent analyses.

Table 1.

Characteristics of Latino Caregivers From Different National Origins.

	Caregiver nation of origin mean (standard deviation)/percentage
	Cuban, n = 72	Mexican, n = 45	Puerto Rican, n = 44	Other regions, n = 48
Demographics
Age^a	66.58* (9.49)	49.41 (13.41)	54.08 (13.78)	55.71 (12.51)
Female	73.6%	84.4%	90.9%	87.2%
Years in the United States	34.17 (10.80)	30.47 (17.68)	34.14 (17.03)	28.83 (14.26)
Catholic	70.8%	77.8%	63.6%	61.7%
Objective stressors
MMSE	11.67 (7.30)	11.44 (6.90)	9.64 (6.71)	12.21 (6.72)
ADL/IADL	9.93 (3.40)	11.18 (2.43)	9.98 (3.85)	9.18 (3.37)
Behavior problems	25.50 (12.15)	26.31 (10.78)	23.89 (10.19)	22.18 (7.44)
Subjective stressors
Self-rate health	2.64 (1.27)	2.40 (0.94)	2.80 (0.90)	2.15 (1.3)
Income inadequacy	1.64 (0.91)	1.82 (0.98)	1.98 (0.95)	1.55 (0.93)
Burden	18.32 (9.74)	14.58 (9.27)	17.05 (10.48)	18.28 (9.72)
Resources
Social support network	5.71 (2.27)	5.87 (2.37)	5.23 (2.34)	5.57 (2.41)
Spirituality and religion
Church attendance^b	1.65* (1.73)	2.44 (1.63)	2.00 (1.78)	2.95 (1.62)
Prayer^a	3.97* (1.74)	4.61 (0.97)	4.81 (0.58)	4.74 (0.82)
S/R coping^b	13.89* (3.74)	14.36 (3.52)	15.18 (2.63)	15.58(2.55)
Outcome
Depression^a	13.40* (7.06)	7.33 (5.81)	9.66 (5.94)	9.57 (7.69)

Note: MMSE = Mini-Mental State Exam; ADL = activities of daily living; IADL = instrumental activities of daily living; S/R coping = spiritual and religious coping.

Indicates a significant difference between Cubans and all other Latino groups.

Indicates a significance difference between Cubans and Latinos from other regions, but not Mexicans or Puerto Ricans.

p < .05.

Bivariate Results

At bivariate level, being female was related to higher levels of depression. Similarly, those who reported no religious affiliation had higher levels of depression (M = 15.27) than Catholics (M = 9.96) and non-Catholics (M = 10.37). No significant differences emerged between Catholics and non-Catholics. As can be seen in Table 2, age was significantly related to depression. Conversely, the number of years CGs reported living in the United States—a rough proxy for acculturation—was unrelated to depression.

Table 2.

Correlations Between Depression and Other Variables.

	2	3	4	5	6	7	8	9	10	11	12	13	CESD
1. Age	.18*	.47**	.02	−.10	−.07	−.17	−.01	−.04	.06	−.16*	−.17*	−.14*	.18*
2. Years in the United States	—	.10	−.06	−.05	.10	−.04	−.20**	.06	−.06	−.05	−.05	−.06	.08
3. Cuban	—	—	.04	−.02	.06	.08	.07	.08	.03	−.22*	−.29*	−.17*	.31**
4. MMSE	—	—	—_	−.49**	−.09	−.02	−.03	−.09	.10	.02	−.04	.01	−.02
5. ADL/IADL	—	—	—	—	.05	.06	.07	.13	−.11	−.06	.14*	.05	.09
6. Behavior problems	—	—	—	—	—	.12	−.02	.52**	−.10	−.13	−.13	−.20*	.36**
7. Self-rated health	—	—	—	—	—	—	.21**	.27**	−.12	−.01	.05	−.01	.41**
8. Income inadequacy	—	—	—	—	—	—	—	.03	−.21**	.07	.05	−.07	.09
9. Burden	—	—	—	—	—	—	—	—	−.23**	−.03	−.11	−.13	.57**
10. Social support network	—	—	—	—	—	—	—	—	—	.20**	.10	.16*	−.35**
11. Church attendance	—	—	—	—	—	—	—	—	—	—	.37**	.39**	−.25**
12. Prayer	—	—	—	—	—	—	—	—	—	—	—	.66**	−.18*
13. S/R coping	—	—	—	—	—	—	—	—	—	—	—	—	−.24**

Note: MMSE = Mini-Mental State Exam; ADL = activities of daily living; IADL = instrumental activities of daily living; CES-D = Center for Epidemiologic Studies–Depression Scale.

p <. 05. **p < .01 (two-tailed).

Possible Mediators

As noted earlier, three conditions must be satisfied for meditation to occur, the first of which is a significant relationship between the mediator and the stressor. As can be seen in Table 2, prayer and S/R coping met this first condition. Prayer was positively associated with ADL/IADL or CR physical functioning. S/R coping was inversely related to CR behavioral problems. Church attendance was not related to any of the stressors. Thus, church attendance does not meet the first condition for mediation.

The second condition that must be satisfied for meditation to occur is that a significant relationship must exist between the stressor and the outcome variable. To determine the presence of significant relationships between stressors and depression, individual regression analyses were conducted for each stressor and depression, controlling for the effects of demographic variables and social support. As can be seen in Table 3, three stressors satisfied this second condition. Among objective stressors, higher levels of CR behavior problems were related to higher levels of depression. Among subjective stressors, poorer self-rated health and greater perceived burden were both associated with higher levels of depression.

Table 3.

Individual Regression Analyses Controlling for Demographics and Social Support.

	Depression
	B		SE (B)
Objective stressors
MMSE	−0.01	0.06
ADL/IADL	0.14	0.13
Behavior problems	0.21**	0.04
Subjective stressors
Self-rated health	2.19**	0.35
Income inadequacy	0.36	0.47
Burden	0.36**	0.04
Religion and spirituality
Church attendance	−0.55*	0.27
Prayer	−0.37	0.40
S/R coping	−0.32*	0.14

Note: MMSE = Mini-Mental State Exam; ADL = activities of daily living; IADL = instrumental activities of daily living; S/R coping = spiritual and religious coping. B indicates the estimated coefficient of each predictor when controlling for the effects of demographics (i.e., age, gender, years in the United States, country of origin, and religious affiliation) and social support network.

p < .05. **p < .01.

The third condition that must be satisfied for meditation to occur is that a significant relationship must exist between the mediator and the outcome variable. As was the case directly above, a series of individual regression analyses were conducted to determine the existence of relationships between spirituality/religion and depression while controlling for the effects of demographic variables and social support. Church attendance and S/R coping were both related to depression (although, as noted above, attendance did not satisfy the first condition).

Thus, of the three spirituality and religion variables, only S/R coping met the criteria as a possible mediator. More specifically, S/R coping may mediate the relationship between behavioral problems and depression.

Results of Mediation and Moderation Analysis

For an actual mediation effect to exist, the inclusion of the mediator in the model must reduce the effect of the stressor on depression in the complete multivariate model. The fourth step of identifying whether S/R coping and social support actually mediated the effects of stress was achieved in the following hierarchical regression analysis. In addition, hierarchical regression was also used to examine the existence of moderating and direct effects of spirituality and religion on depression.

A three-step hierarchical regression approach was used. The first model included demographics, social support, and objective and subjective stressors to predict depression (see Model 1 in Table 4). Higher levels of depression was associated with participants who were female, Cuban, as well as participants who reported poorer health status, greater levels of burden, and smaller social support networks. The first model explained 51% of the variance in depression.

Table 4.

Hierarchical Regression Results.

	Model 1		Model 2		Model 3
	B	SE (B)	B	SE (B)	B	SE (B)
Demographics
Age	0.04	0.03	0.03	0.03	0.03	0.03
Female	2.81**	0.97	3.08**	0.99	3.01**	0.98
Years in the United States	−0.02	0.03	−0.02	0.03	−0.02	0.03
Cuban	3.16**	0.89	3.07**	0.92	3.29**	0.92
No religious affiliation^a	2.76	1.46	2.52	1.79	1.81	1.80
Non-Catholic affiliation^a	0.14	0.89	0.41	0.90	0.84	0.91
Objective stressors
MMSE	0.03	0.06	0.05	0.06	0.03	0.06
ADL/IADL	0.07	0.13	0.07	0.13	0.08	0.13
Behavior problems	0.07	0.04	0.05	0.04	0.01	0.04
Subjective stressors
Self-rated health	1.45**	0.41	1.44**	0.40	1.56**	0.40
Income inadequacy	−0.19	0.40	−0.01	0.40	0.01	0.40
Burden	2.76**	0.45	2.82**	0.45	2.68**	0.45
Resources
Social support network	−0.76**	0.17	−0.67**	0.17	−0.68**	0.17
Spirituality and religion
Church attendance	—	—	−0.86*	0.42	−1.03*	0.42
Prayer	—	—	0.35	0.45	0.38	0.44
S/R coping	—	—	−0.18	0.15	−0.15	0.15
Interaction effects
Church Attendance × Health	—	—	—	—	0.78*	0.39
Church Attendance × Burden	—	—	—	—	−0.79*	0.40
Model adjusted R^2b		0.51		0.52		0.53

Note: MMSE = Mini-Mental State Exam; ADL = activities of daily living; IADL = instrumental activities of daily living; S/R coping = spiritual and religious coping.

Reference group = Catholic.

R² changes from Model 1 to Model 2 and from Model 2 to Model 3 were significant at the nontraditional .10 level.

p < .05. **p < .01.

To test the presence of direct and indirect effects, the second model incorporated the three spirituality/religion measures into the model (see Model 2 in Table 4). S/R coping was not significantly related to depression in the multivariate model that controlled for other theoretically relevant variables. Thus, S/R coping does not exhibit indirect or direct effects on depression, at least among the CGs surveyed in this study.

Prayer was not significantly related to depression in the multivariate model. Thus, prayer did not exhibit direct or indirect effects on depression after controlling for other relevant variables.

Conversely, church attendance remained statistically significant at the multivariate level. In keeping with theory, those who attended church more frequently reported lower levels of depression. Thus, church attendance exhibited a direct effect on depression, independent of stress, in the expected direction. The second model explained 52% of the variance in depression.

To test for the presence of moderating effects, the third model added interaction terms comprised of church attendance and the two significant subjective stressors: self-rated health and burden. The third model explained 53% of the variance in depression. As indicated in the Model 3 in Table 4, both interaction terms were significant. Thus, church attendance moderated the relationship between burden and depression and between self-rated health and depression.

In order to illustrate the moderation effect, Aiken and West’s (1991) method was used in tandem with the following interaction program (http://www.danielsoper.com/Interaction/) to graph the moderation effect. Figure 2 presents three estimated regression lines of depression scores depending on the levels of church attendance (i.e., mean score, 1 standard deviation above the mean, and 1 standard deviation below the mean). As illustrated in Figure 2, the effects of church attendance on depression were more powerful when CGs had better health status. In other words, when self-rated health status deteriorated, the differences in depression between frequent church attendees and less frequent church attendees tended to diminish.

Figure 2.

Interaction effect between self-rated health and church attendance on depression.

As illustrated in Figure 3, the relationship between perceived burden and depression was more substantial among less frequent attendees than among more high-frequency attendees. In other words, as CGs attended church more often, the effect of burden on depression tended to decrease. Or stated conversely, when burden was high, church attendance had more effect on depression.

Figure 3.

Interaction effect between caregiver burden and church attendance on depression.

Discussion

This study examined the impact of spirituality and religion on depression among a sample of Latino AD CGs. Drawing upon Pearlin and associates’ (1990) stress coping model, we sought to determine whether spirituality and religion directly affected depression and mediated or moderated the relationship between stressors and depression. Analysis revealed the existence of direct and indirect effects with one of the three measures of spirituality and religion used in the study. Church attendance exhibited direct effects on depression. In keeping with our expectations, higher levels of attendance were associated with lower levels of depression.

Although no mediation was observed, church attendance moderated the relationship between two subjective stressors (i.e., self-rated health and perceived CG burden) and depression. Of the six stressors operationalized in this study, only two were significantly related to depression at the multivariate level. Thus, church attendance exhibited moderating effects for all forms of stress that were associated with depression in this study.

The moderation effects differed for self-rated health and CG burden. The effects of church attendance on depression were more powerful when CGs reported higher levels of self-rated health. In other words, higher levels of attendance did not mitigate the effects of poor health on depression. This result may be due to the fact that self-rated health often serves as a rough proxy for physical health (Shooshtari, Menec, & Lambert, 2007). Poor physical health may inhibit the ability of CGs to attend religious services, thus limiting their exposure to the protective effects of religious involvement.

Conversely, church attendance buffered the effects of CG burden on depression. Higher levels of attendance mitigated the effects of perceived CG burden on depression. Perceived burden is a more psychologically oriented construct relative to self-rated health. As such, burden may not inhibit attendance in the same manner as physical health limitations. In keeping with this perspective, the protective effects of religious involvement were most pronounced when CGs’ perceived burden was higher in a manner reflecting our expectations.

The moderating effects observed rather than mediation was perhaps due to a possibility that many of these individuals reported preexisting S/R resources that they operationalized when faced with life challenges. Such explanation is consistent with prior qualitative research. A number of investigators have found that religious resources often precede the experience of caregiving (Nightingale, 2003; Stuckey, 2001; Vickrey et al., 2007). These reports reflect the underlying theory associated with moderation; that is, religion exerts a buffering effect on stress only under certain conditions (Pearlin, 2002).

Conversely, the results obtained in this study conflict with some prior related quantitative research. Studies using Pearlin and associates’ (1990) stress coping theory to examine the effects of spirituality and religion on stress and depression have generally failed to exhibit either direct or indirect effects on depression among the various samples of CGs examined. Largely nonsignificant results have been obtained with Anglo-majority samples (Leblanc et al., 2004; Morano & King, 2005), at least one Latino sample (Lee et al., 2010), as well as with much of the larger literature on non-AD CGs (Hebert et al., 2006).

The differential findings may be explained, at least in part, by the multidimensional nature of spirituality and religion, and the intersection of various dimensions with cultural factors. Spirituality and religion are multidimensional constructs (Idler et al., 2009; Miller & Thoresen, 2003). Accordingly, findings may vary depending on which dimension of spirituality and religion researchers chose to operationalize in a given study, as illustrated by the differential findings reported in this study.

In addition, particular dimensions may be more salient among Latino CGs. For instance, church attendance is often a central component of Latino life (Sarkisian, Gerena, & Gerstel, 2007; Wilson, 2008). The importance of religious involvement in the Latino community may help account for the significant effects that emerged in this study.

Implications for Practice

The results have important practice implications. Participation in religious communities may help alleviate depression. CGs may rely on religious involvement as a way to buffer the effects of stress associated with AD caregiving (Coon et al., 2004). It should be noted that the findings are consistent with a relatively large body of research on religion and depression that suggests that religious attendance is inversely related to depression (Koenig et al., 2001). Although the results point to a positive effect of religious attendance on depression, the mechanisms thorough which any salutary effect is produced cannot be determined in the present study (Aranda, 2008). Nevertheless, it is interesting that church attendance was significantly associated with depression even after adjusting for the effects of social support. This raises the possibility that the protective effects of religion cannot be attributed just to the social networks and connectedness within those networks that often exist in religious communities. Rather, religious attendance may provide an additional protective effect, perhaps due to some combination of spiritual, emotional, and/or material support provided by church members (Thompson et al., 2002).

Practitioners working with Latino CGs might conduct an assessment to determine the existence of pertinent S/R assets (Hodge, 2006). Such assessments are required in many Joint Commission–accredited agencies, including those providing home care (Koenig, 2007). A spiritual assessment can provide insights into the role that spirituality and religion play in overall service provision.

Thompson and associates (2002) recommend encouraging, if appropriate, CGs to increase their level of S/R involvement. Religious participation, however, is often not a simple process for CGs. CRs may be homebound requiring constant supervision; CGs’ poor health may pose a mobility barrier. In addition, having a loved one with AD can be a significant source of embarrassment to CGs in certain situations (Montoro-Rodriguez, Kosloski, Kercher, & Montgomery, 2009).

To the extent possible, practitioners might network with relevant clergy to ensure that the barriers associated with caregiving do not prevent CGs from attending religious services and other, related community functions (Coon et al., 2004). Clergy may be able to broker caregiving arrangements that allow CGs to participate in functions that are perceived to be meaningful. Clergy can also be a valuable resource to help practitioners integrate religious themes into interventions as a way that enhances effectiveness (Thompson et al., 2002).

Incorporating religious themes and exhibiting sensitivity toward Latino AD CGs’ religious beliefs and values may help improve service utilization and retention (Mausbach, Coon, Cardeanas, & Thompson, 2003). Mausbach and associates suggest that some Latino AD CGs avoid seeking professional assistance due to concerns that practitioners will fail to treat their S/R values with dignity and respect. Developing awareness of common spiritually animated beliefs and practices among Latinos can increase practitioners’ levels of spiritual competency.

Finally, practitioners should be aware of the cultural differences that may exist among different subgroups of Latino CGs. Prior qualitative research has found more similarities than differences among Latino subgroups (Borrayo et al., 2007), suggesting underlying cultural values may be more informative of the Latino AD caregiving experience than interethnic differences (Neary & Mahoney, 2005). Nevertheless, the present study suggests that Cuban CGs may differ from other Latino CGs. Consequently, practitioners should be sensitive to the possibility that the caregiving experience differs across Latino subgroups and should be ready to adjust their interactions accordingly.

Limitations

The cross-sectional findings preclude any assessment of causality, and many potentially confounding variables were not included in the analysis. Constructs such as acculturation, familism, education, and social integration are potentially relevant variables that were not included in our models (Aranda & Knight, 1997). In addition, the study did not employ a probability sample. Thus, the results should not be generalized to other samples of Latino CGs (Hilgeman et al., 2009). It should also be noted that the CGs in this study were part of a clinical trial. Consequently, it is possible that the CGs who consented to enroll in such trial may systematically differ from those who do not enroll in such studies. These limitations serve to emphasize the importance of conducting additional research on Latino AD CGs.

Conclusion

Over the next few decades, millions of Latino American older adults will require care due to AD and other forms of dementia (Administration on Ageing, 2008b; Coon et al., 2004). As a result, many Latino family members will find themselves caring for a family member with AD. Yet, despite the projected growth in the number of Latino CGs, little research has explored the cultural resources that might assist CGs in dealing with the deleterious health outcomes associated with caregiving, such as depression (Dilworth-Anderson et al., 2002).

Drawing from Pearlin and associates’ (1990) stress coping model, this study examined the effects of spirituality and religion on depression. The findings suggest that church attendance moderates the relationship between stress and depression in tandem with exhibiting direct effects on depression. The results imply that involvement in religious communities may be an important resource for Latino American CGs as they face the challenges associated with caring for a family member wrestling with AD.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Fei Sun is an assistant professor with the School of Social Work at Arizona State University. He was awarded as a Hartford Geriatric Social Work Scholar in 2010. His area of research interest is mental health and aging, with a particularly focus on Alzheimer’s disease caregiving in ethnic minority families.

David R. Hodge is an assistant professor at Arizona State University’s School of Social Work and senior nonresident fellow at the University of Pennsylvania’s Program for Research on Religion and Urban Civil Society.

References

Administration on Ageing. (2008a). A statistical profile of Hispanic older Americans aged 65+. Retrieved from http://www.aoa.gov/AoARoot/Aging_Statistics/Minority_Aging/Facts-on-Hispanic-Elderly.aspx

Administration on Ageing. (2008b). Population projections by race and Hispanic origin for persons 60 and over: 2000 to 2050. Retrieved from http://www.aoa.gov/AoARoot/Aging_Statistics/future_growth/future_growth.aspx

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: SAGE.

Aranda

M. P.

(2008). Relationship between religious involvement and psychological well-being: A social justice perspective. Health & Social Work, 33(1), 9-21.

Aranda

M. P.

Knight

B. G.

(1997). The influence of ethnicity and culture on the caregiver stress and coping process: A sociocultural review and analysis. The Gerontologist, 37(3), 342-354.

Baron

R. M.

Kenny

D. A.

(1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182.

Bedard

Molloy

D. W.

Squire

Dubois

Lever

J. A.

O’Donnell

(2001). The Zarit Burden Interview: A new short version and screening version. The Gerontologist, 41, 652-657.

Belle

S. H.

Burgio

Burns

Coon

Czaja

S. C.

Gallagher-Thompson

. . . Zhang

(2006). Enhancing the quality of life of dementia caregivers from different ethnic or racial groups: A randomized, controlled trial. Annals of Internal Medicine, 145, 727-738.

Coon

D. W.

Rubert

Solano

Mausbach

Kraemer

Arguelles

. . . Gallagher-Thompson

(2004). Well-being, appraisal, and coping in Latina and Caucasian female dementia caregivers: Findings from the REACH study. Ageing & Mental Health, 8(4), 330-345.

10.

Dilworth-Anderson

Williams

I. C.

Gibson

B. E.

(2002). Issues of race, ethnicity, and culture in caregiving research: A 20-year review (1980-2000). The Gerontologist, 42(2), 237-272.

11.

Farran

C. J.

Paun

Elliott

M. H.

(2003). Spirituality in multicultural caregivers of persons with dementia. Dementia, 2(3), 353-377.

12.

Folstein

McHugh

(1975). Mini-Mental State: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189-198.

13.

Gallagher-Thompson

Gray

H. L.

Dupart

Jimenez

Thompson

L. W.

(2008). Effectiveness of cognitive/behavioral small group intervention for reduction of depression and stress in non-Hispanic white and Hispanic/Latino women dementia family caregivers: Outcomes and mediators of change. Journal of Rational Emotive Cognitive Behavioral Therapy, 26, 286-303.

14.

Geppert

Bogenschutz

M. P.

Miller

W. R.

(2007). Development of a bibliography on religion, spirituality, and addiction. Drug and Alcohol Review, 26(4), 389-395.

15.

Hebert

R. S.

Weinstein

Martire

L. M.

Schulz

(2006). Religion, spirituality and well-being of informal caregivers: A review, critique, and research prospectus. Ageing & Mental Health, 10(5), 497-520.

16.

Herrera

A. P.

Lee

J. W.

Nanyonjo

R. D.

Laufman

L. E.

Torres-Vigil

(2009). Religious coping and caregiver well-being in Mexican-American families. Ageing & Mental Health, 13(1), 84-91.

17.

Hilgeman

M. M.

Durkin

D. W.

Sun

DeCoster

Allen

R. S.

Gallagher-Thompson

Burgio

L. D.

(2009). Testing a theoretical model of the stress process in Alzheimer’s caregivers with race as a moderator. The Gerontologist, 49(2), 248-261.

18.

Hodge

D. R.

(2005). Spiritual life maps: A client-centered pictorial instrument for spiritual assessment, planning, and intervention. Social Work, 50(1), 77-87.

19.

Hodge

D. R.

(2006). A template for spiritual assessment: A review of the JCAHO requirements and guidelines for implementation. Social Work, 51(4), 317-326.

20.

Idler

E. L.

Boulifard

D. A.

Labouvie

Chen

Y. Y.

Krause

T. J.

Contrada

R. J.

(2009). Looking inside the black box of “attendance at services”: New measures for exploring an old dimension in religion and health research. The International Journal for the Psychology of Religion, 19, 1-20.

21.

Katz

Ford

Moskowitz

Jackson

Jaffe

(1963). Studies of illness and the aged: The index of ADL, a standardized measure of biological and psychological function. Journal of the American Medical Association, 185, 914-919.

22.

Kenny

D. A.

Kashy

D. A.

Bolger

(1998). Data analysis in social psychology. In Gilbert

Fiske

S. T.

Lindzey

(Eds.), Handbook of social psychology (4th ed., Vol. 1, pp. 233-265). New York, NY: McGraw-Hill.

23.

Knight

B. G.

Sayegh

(2010). Cultural values and caregiving: The updated sociocultural stress and coping model. Journal of Gerontology: Psychological Sciences, 65B(1), 5-13.

24.

Koenig

H. G.

(2007). Spirituality in patient care (2nd ed.). Philadelphia: Templeton Foundation Press.

25.

Koenig

H. G.

McCullough

M. E.

Larson

D. B.

(2001). Handbook of religion and health. New York, NY: Oxford University Press.

26.

Lawton

Brody

(1969). Assessment of older people: Self-maintaining and instrumental activities of daily living. The Gerontologist, 9, 179-186.

27.

Leblanc

A. J.

Driscoll

A. K.

Pearlin

L. I.

(2004). Religiosity and the expansion of care giver stress. Ageing & Mental Health, 8(5), 410-421.

28.

Lee

C. C.

Czaja

S. J.

Schulz

(2010). The moderating influence of demographic characteristics, social support, and religious coping on the effectiveness of a multicomponent psychosocial caregiver intervention in three racial ethnic groups. Journal of Gerontology: Psychological Sciences, 65B(2), 185-194.

29.

Lubben

J. E.

(1988). Assessing social networks among elderly populations. Family and Community Health, 11, 42-52.

30.

Mausbach

B. T.

Coon

D. W.

Cardeanas

Thompson

L. W.

(2003). Religious coping among Caucasian and Latina dementia caregivers. Journal of Mental Health and Ageing, 9(2), 97-110.

31.

Miller

W. R.

Thoresen

C. E.

(2003). Spirituality, religion, and health: An emerging research field. American Psychologist, 58(1), 24-35.

32.

Montoro-Rodriguez

Kosloski

Kercher

Montgomery

R. J. V.

(2009). The impact of social embarrassment on caregiving distress in a multicultural sample of caregivers. Journal of Applied Gerontology, 28(2), 195-217.

33.

Morano

C. L.

King

(2005). Religiosity as a mediator of caregiver well-being: Does ethnicity make a difference? Journal of Gerontololgical Social Work, 45(1/2), 69-84.

34.

Morano

C. L.

Sanders

(2005). Exploring differences in depression, role captivity, and self-acceptance in Hispanic and non-Hispanic adult children caregivers. Journal of Ethnic & Cultural Diversity in Social Work, 14(1/2), 27-46.

35.

Musick

M. A.

Traphagan

J. W.

Koenig

H. G.

Larson

D. B.

(2000). Spirituality in physical health and ageing. Journal of Adult Development, 7(2), 73-86.

36.

Napoles

A. M.

Chadiha

Eversley

Moreno-John

(2010). Developing culturally sensitive dementia caregiver interventions: Are we there yet? American Journal of Alzheimer’s Disease & Other Dementias, 25(5), 389-406.

37.

Nightingale

M. C.

(2003). Religion, spirituality, and ethnicity: What it means for caregivers of persons with Alzheimer’s disease and related disorders. Dementia, 2(3), 379-391.

38.

Pargament

K. I.

(2007). Spiritually integrated psychotherapy: Understanding and addressing the sacred. New York, NY: Guilford.

39.

Pargament

K. I.

Zinnbauer

B. J.

Scott

A. B.

Butter

E. M.

Zerowin

Stanik

(1998). Red flags and religious coping: Identifying some religious warning signs among people in crisis. Journal of Clinical Psychology, 54, 77-89.

40.

Pearlin

(2002). Some institutional and stress process perspectives on religion and health. Psychological Inquiry, 13(3), 201-238.

41.

Pearlin

L. I.

Mullan

J. T.

Semple

S. J.

Skaff

M. M.

(1990). Caregiving and the stress process: An overview of concepts and their measures. The Gerontologist, 30, 583-594.

42.

Radloff

L. S.

(1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401.

43.

Sarkisian

Gerena

Gerstel

(2007). Extended family integration among Euro and Mexican Americans: Ethnicity, gender, and class. Journal of Marriage and Family, 69, 40-54.

44.

Schulz

Newsom

J. T.

Mittlemark

Burton

Hirsh

Jackson

(1997). Health effects of caregiving: The caregiver health effects study. Annals of Behavioral Medicine, 12, 110-116.

45.

Shooshtari

Menec

V. H.

Lambert

(2007). Ethnic differences in self-rated health among older adults. Journal of Ageing and Health, 19(1), 62-86.

46.

Sorensen

Pinquart

(2005). Racial and ethnic differences in the relationship of caregiving stressors, resources, and sociodemographic variables to caregiver depression and perceived health. Ageing & Mental Health, 9(5), 482-495.

47.

Stuckey

J. C.

(2001). Blessed assurance: The role of religion and spirituality in Alzheimer’s disease caregiving and other significant life events. Journal of Ageing Studies, 15, 69-84.

48.

Teri

Traux

Logsdon

Uomoto

Zarit

Vitaliano

P. P.

(1992). Assessment of behavioral problems in dementia: The Revised Memory and Behavior Problems Checklist (RMBPC). Psychology and Ageing, 18, 375-384.

49.

Thompson

L. W.

Solano

Coon

D. W.

Mausbach

Gallagher-Thompson

(2002). Pleasurable activities and mood: Differences between Latina and Caucasian dementia family caregivers. Journal of Mental Health and Ageing, 8(3), 211-224.

50.

Vickrey

B. G.

Strickland

T. L.

Fitten

L. J.

Adams

G. R.

Ortiz

Hays

R. D.

(2007). Ethnic variations in dementia caregiving experiences: Insights from focus groups. Journal of Human Behavior in the Social Environment, 15(2/3), 233-249.

51.

Wilks

S. E.

Vonk

M. E.

(2008). Private prayer among Alzheimer’s caregivers: Mediating burden and resiliency. Journal of Gerontological Social Work, 50(3/4), 113-131.

52.

Wilson

C. E.

(2008). The politics of Latino faith: Religion, identity, and urban community. New York, NY: New York University Press.

53.

Wisniewski

S. R.

(2003). Resources for Enhancing Alzheimer’s Caregiver Health (REACH): Project design and baseline characteristics. Psychology and Ageing, 18(3), 375-384.

54.

Wuthnow

(2007). After the baby boomers: How twenty- and thirty-somethings are shaping the future of American religion. Princeton, NJ: Princeton University Press.

55.

Zarit

S. H.

Orr

N. K.

Zarit

J. M.

(1985). The hidden victims of Alzheimer’s disease: Families under stress. New York, NY: New York University Press.