Religiosity and quality of life: a dyadic perspective of individuals with dementia and their caregivers

Abstract

Objectives:

Dyadic coping theory purports the benefit of joint coping strategies within a couple, or dyad, when one dyad member is faced with illness or stress. We examine the effect of religiosity on well-being for individuals with dementia (IWDs). In particular, we look at the effect of both dyad members’ religiosity on perceptions of IWDs’ quality of life (QoL). Neither of these issues has been extensively explored.

Method:

One hundred eleven individuals with mild-to-moderate dementia and their family caregivers were interviewed to evaluate IWDs’ everyday-care values and preferences, including religious preferences. Using an actor–partner multi-level model to account for the interdependent relationship of dyads, we examined how IWD and caregiver ratings of religiosity (attendance, prayer, and subjective ratings of religiosity) influence perceptions of IWDs’ QoL.

Results:

After accounting for care-related stress, one’s own religiosity is not significantly related to IWDs’ or caregivers’ perceptions of IWD QoL. However, when modeling both actor and partner effects of religiosity on perceptions of IWDs’ QoL, caregivers’ religiosity is positively related to IWDs’ self-reports of QoL, and IWDs’ religiosity is negatively associated with caregivers’ perceptions of IWDs’ QoL.

Conclusion:

These findings suggest that religiosity of both the caregiver and the IWD affect perception of the IWD’s QoL. It is important that caregivers understand IWDs’ values concerning religion as it may serve as a coping mechanism for dealing with dementia.

Introduction

When people face situations over which they have little control, they may use spiritual or religious beliefs to manage feelings of helplessness and give meaning and order to the events of their lives (Katsuno, 2003). According to the Cognitive Theory of Stress and Coping, individuals may utilize meaning-focused coping (Lazarus, 1966) to generate positive intrinsic emotions by drawing on their beliefs or values to self-motivate and maintain well-being in difficult times. One such situation is when a family member is diagnosed with Alzheimer’s disease or another form of dementia. Both individuals with dementia (IWDs) and family caregivers may turn to religious beliefs as a way of coping with dementia; yet, religious beliefs and spirituality have not been extensively evaluated in this population. Given the dyadic nature of the relationship between caregivers and IWDs in the context of care (Berg & Upchurch, 2007), caregivers perceptions of their relatives’ religiosity likely play a role as well. Using a sample of individuals with mild-to-moderate dementia and their caregivers, we consider the implications of IWDs’ and caregivers’ religiosity on perceptions of IWDs’ quality of life (QoL).

Measuring religiosity

Religiosity is considered to have three dimensions – organizational, non-organizational, and subjective. Organizational religiosity addresses formal religious involvement, such as attending services or events at a place of religious worship. Non-organizational religiosity addresses private practices, including prayer or reading religious texts. Subjective assessment of the degree of religiosity addresses internal feelings (Leblanc, Driscoll, & Pearlin, 2004). Prior studies often focus on only one or two of these dimensions of religiosity (Krause, 1998); yet, it is important to examine religiosity as a composition of these multiple facets.

Religiosity and caregivers

Caregivers use religiosity to cope with illness and stressors in their lives. In some samples, the majority report that they pray nearly every day and perceive religion to be important (Herbert, Dang, & Schulz, 2007). However, studies examining the role of religion in the lives of caregivers yield mixed findings. Higher levels of the different dimensions of religiosity have been associated with lower depressive symptomology (Herbert et al., 2007). Reports of higher levels of spiritual well-being are associated with less burden, reflecting the idea that religion/spirituality may allow individuals to interpret the experience as less burdensome (Spurlock, 2005). Such findings are supported by qualitative findings that caregivers express the importance of religion/spiritual experiences when coping with their role, and often cite prayer and the idea that God has a plan as sources of coping and comfort (Stuckey, 2001).

Another body of literature illuminates the possibility that religiosity is associated with negative psychological, emotional, and physical outcomes. Higher levels of religiosity have been associated with feelings of role overload among spouse caregivers (Leblanc, Driscoll, & Pearlin, 2004), and increased church attendance has been associated with more psychological distress (Cohen, Teresi, & Blum, 1994) and lower self-rated health (Haley, Levine, Brown, & Bartolucci, 1987). In sum, religiosity is associated with both positive and negative outcomes for caregiver well-being. However, research has yet to address how caregivers’ religiosity impacts the QoL or well-being of their care recipients.

Religiosity and individuals with dementia

The religious needs and interests of IWDs have been neglected in research and in health care (Doherty, 2006). Individuals with mild-to-moderate dementia may demonstrate higher levels of religious activity and intrinsic religiosity than other older adults (Katsuno, 2003), as religion can provide a way to understand problems related to the daily struggles associated with dementia (Snyder, 2003). Other people, however, may feel conflicted over whether or not they are being punished by God and question their religious practice when their prayers go unanswered (Snyder, 2003).

Quantitative investigations with IWDs regarding religious beliefs have been rare, but show that total religiosity scores accounting for organized, non-organized, and subjective religiosity are correlated with higher QoL in IWDs (Katsuno, 2003). Meanwhile, qualitative reports have shown that IWDs look to God for strength, security, and comfort and see God as someone they can always turn to for help and protection (Katsuno, 2003). At the same time, IWDs express that private religious practice is more important, as they experience changes that limit their attendance at religious activities. Additional quantitative work is needed to examine and confirm the effects of religiosity on IWDs’ QoL.

Dyadic nature of religiosity

With the progression of dementia, IWDs will increasingly depend on their caregivers. As IWDs gradually lose the ability to express their own preferences, responsibility for sustaining religious practices may fall to caregivers (Snyder, 2003). While some IWDs will be able to attend and participate in social meetings at church, they may need support from others and assistance from family in order to do so (Katsuno, 2003). Therefore, understanding the perspectives of both individuals may be key to obtaining a holistic viewpoint of how religiosity impacts outcomes for IWDs, namely QoL. Examining perspectives of both individuals provides for the opportunity to look at the relationship as a dynamic process of bidirectional interactions that affect care outcomes (Lyons & Sayer, 2010). An actor–partner model can be used to examine the effects of one’s own religious practices on one’s own well-being, as well as the effects of one’s partner’s beliefs. This can provide for a fuller understanding of how religiosity may affect dyadic members (see Figure 1). Research has yet to be conducted to examine such an effect regarding religiosity on QoL for IWDs.

Figure 1.

Depiction of an APIM examining the effects of religiosity on perceptions of the QoL of IWDs where U and U′ represent the residual variance around each person’s reports.

The current study

In this study, we use a dyadic approach to investigate the effects of religiosity on IWDs’ QoL (Cook & Kenny, 2005). We determine if the IWDs’ own religiosity plays a unique role in predicting their own reports of QoL and whether caregivers’ religiosity affects their perception of IWDs’ QoL (i.e., actor effects). It is hypothesized that higher religiosity of the IWD or the caregiver will predict higher respective perceptions of IWD QoL. Next, we examine how the caregivers’ personal religiosity affects IWDs’ self-reports of QoL and how IWDs’ religiosity affects caregivers’ perceptions of IWDs’ QoL (i.e., partner effects). It is hypothesized that higher caregiver religiosity will predict higher QoL as reported by the IWD. The same effects will be seen for IWDs’ religiosity on caregiver perceptions of IWD QoL. An advantage of these analyses lies in utilizing both caregiver and IWD reports regarding religiosity and QoL.

Methods

Participants

Dyads (N = 111), consisting of an IWD and his or her informal spouse or child caregiver, were recruited from two service-based organizations in the San Francisco Bay Area and Cleveland, Ohio. This sample is drawn from a population of caregivers characterized as ‘help seeking’, actively seeking help and support (Aneshensel, Pearlin, Mullan, Zarit, & Whitlatch, 1995). Per the request of the Institutional Review Board, ages above 90 were recoded to 90 to minimize identification of specific individuals. See Table 1 for participant characteristics.

Table 1.

Characteristics of caregivers and IWDs.

	Caregivers		IWDs
	M (SD) or % (n)	Range	M (SD) or % (n)	Range
Age	61.20 (14.00)	30–90+	76.80 (8.90)	39–90+
Female (yes = 1)	0.81 (0.39)	0–1	0.53 (0.50)	0–1
Spouse (yes = 1)	0.41 (0.49)	0–1	–	–
African-American (yes = 1)	0.52 (0.50)	0–1	0.48 (0.50)	0–1
Education	4.00 (1.20)	1–6	3.20 (1.50)	1–6
Care-related stress	19.90 (16.00)	0–72	–	–
MMSE	–	–	20.7 (3.80)	13–26
Religious affiliation
Protestant	57.7% (64)	–	55.0% (61)	–
Catholic	10.8% (12)	–	14.4% (16)	–
Jewish	5.4% (6)	–	6.3% (7)	–
Christian	2.7% (3)	–	6.3% (7)	–
Other	22.5% (25)	–	15.3% (17)	–
Total religiosity	4.20 (1.40)	0–6	3.70 (1.50)	0–6
QoL	–	–	2.66 (0.43)	1.77–3.85

Note: Total sample N = 111; percentages for religious affiliation do not add up to 100 due to limited missing data.

Procedures

Caregivers had to be the primary family caregiver (i.e., the person most involved in providing assistance and daily care to the IWD). The IWD had to have a confirmed diagnosis of dementia from a physician or be exhibiting symptoms of progressive memory problems typical of dementia and be living at home in the community. Cognitive impairment ranged from mild-to-moderate, as defined by scores of 13 to 26 on the Mini-Mental State Exam (MMSE, Folstein, Folstein, & McHugh, 1975). This range was selected based on previous research that found that most IWDs with scores above 13 could complete the research procedures, while scores above 26 raised the possibility that the person may not have dementia (Whitlatch, Feinberg, & Tucke, 2005). Of 173 individuals screened for the study, 62 (36%) were ineligible due to MMSE scores (see Table 2). The final sample consisted of 111 dyads of individuals with cognitive impairment and their family caregivers (total N = 222). Each member of the dyad completed two in-person interviews at home to evaluate everyday care values and preferences of the IWDs.

Table 2.

Reasons for dyad ineligibility.

Reason	N
IWD’s MMSE score below 13	18
IWD’s MMSE score above 26	37
IWD or caregiver did not successfully complete the written-consent process or declined to continue with the study after completing a partial interview	5
Dyad was used in pretesting procedures	2
Total ineligible dyads	62

Measures

Dependent measure.

Though there are many possible outcomes of religiosity, we chose to measure QoL which yields a comprehensive understanding of the IWD’s overall life evaluation, as religiosity exerts a broad influence over various aspects of life. The Quality of Life – Alzheimer’s Disease Scale (QoL-AD; Logsdon, Gibbons, McCurry, & Teri, 1999) is a 13-item questionnaire that provides a report of the IWD’s QoL. The items address Lawton’s four domains of QoL in older adults – perceived QoL, behavioral competence, psychological status, and interpersonal environment (Logsdon, Gibbons, McCurry, & Teri, 2002); they are rated on a Likert scale from 1 (poor) to 4 (excellent). Caregivers completed the measure regarding their perception of their relative’s QoL. IWDs completed the same questionnaire in regard to their own QoL. A mean score was computed as an indicator of perceived QoL, with higher scores indicating greater perceived QoL for the IWD (M = 2.66, SD = 0.43, α = 0.86).

Independent measures

Total religiosity.

Caregivers and IWDs responded independently to three questions that address the main components of religiosity: organizational religiosity (How often do you attend religious services?), non-organizational religiosity (How often do you pray or meditate on your own?), and subjective religiosity (How religious or spiritual would you say you are?). For questions pertaining to organizational and non-organizational religiosity, respondents answered on a Likert scale of 0 (never) to 6 (nearly every day/4 or more times per week). For subjective religiosity, respondents answered on a Likert scale from 1 (not at all religious/spiritual) to 4 (very religious/spiritual). We examined correlations among these three items (see Table 3) and found that they were highly correlated. We recoded items so they each had equal weight and summed them to form a total religiosity score for both the caregiver and the IWD (caregivers: α = 0.71, M = 4.17, SD = 1.36; IWDs: α = 0.66, M = 3.69, SD = 1.49). This religiosity score provides a comprehensive evaluation of religiosity addressing the various dimensions of religious practice.

Table 3.

Correlations among individual facets of religiosity.

	1	2	3	M	SD
Caregivers
1. Subjective appraisal	–			4.64	1.46
2. Frequency of prayer	0.61***	–		5.22	1.80
3. Attendance at religious services	0.32***	0.39***	–	2.67	1.88
Individuals with dementia
1. Subjective appraisal	–			3.96	1.80
2. Frequency of prayer	0.62***	–		4.74	2.09
3. Attendance at religious services	0.30**	0.28**	–	2.43	1.76

Note: **p < .01, and ***p < .001.

Covariates.

We controlled for factors likely to affect IWD QoL, including the following:

• Cognitive functioning. To assess global cognitive functioning, IWDs completed the MMSE (Folstein et al., 1975). Lower scores indicate a greater degree of cognitive impairment.

• Care-related stress. As a measure of current stress, caregivers completed a 32 item checklist adapted from the Revised Memory and Behavior Problems Checklist (Teri et al., 1992) and the Weekly Record of Behavior (Fauth, Zarit, Femia, Hofer, & Stephens, 2006) to evaluate the stress appraisal associated with dementia-related problems. If a behavior occurred in the past week, caregivers rated how bothersome or upsetting it was on a scale from 0 (not at all) to 4 (extremely). A stress sum score was computed (α = 0.86, M = 19.9, SD = 16.0; N = 110).

• Demographics. We considered five demographic characteristics: age, race, relationship, highest level of education, and income. Race was coded 1 for African-American and 0 for others. The relationship between the caregiver and IWD was coded 1 as spouse and 0 for others. Education ranged from 1 (less than high school) to 6 (postgraduate degree). Preliminary analyses suggested large correlations (p < 0.05) between caregiver education and income and relationship and caregiver age. To avoid multicollinearity we eliminated age and income.

Data analysis

First, we ran t-tests to compare caregiver and IWD reports on religiosity and QoL. We then ran means and correlations to examine the distributions in the data, the relationship between the religiosity of IWDs and their caregivers, demographic characteristics and covariates, and perceptions of IWDs’ QoL. Second, to examine the relationship between IWDs’ and caregivers’ religiosity and perceptions of IWDs’ QoL, we ran an actor–partner interdependence multi-level model (APIM) (SAS PROC MIXED; Littell, Miliken, Stroup, & Wolfinger, 1996). Multi-level modeling uses the dyad as the unit of analysis and accounts for the fact that individual observations by caregivers and IWDs are nested within dyads. It models the variance at two levels: Level 1 within-dyad differences and Level 2 between-dyad differences. More specifically, we utilized a two-intercept APIM to account for the effects for each dyad member (Cook & Kenny, 2005).

We first ran a base model (Model 1) to account for the intercept effects for each dyad member on the outcome variable of reports of IWDs’ QoL (i.e., the caregiver’s perception of IWD’s QoL and the IWD’s self-report of QoL). In Model 2 we controlled for between-dyad variables at Level 2 of the model, including relationship (spouse or other relation), race (African-American or other), cognitive ability of the IWD, and caregivers’ experience of stress related to behavior problems. We also included a within-dyad variable of education at Level 1. To simplify the model, only control variables that were significant were retained (e.g., care-related stress).

Model 3 then included the actor effects of each individual’s own total religiosity on their perceptions of IWDs’ QoL, entering the religiosity variables at Level 1 of the multi-level model. The actor effects in this model measure the effect a person’s religiosity has on his or her own perceptions of IWD QoL (i.e., the effect of IWD’s own characteristics on his or her self-reported QoL and caregiver’s characteristics on his or her perception of the IWD’s QoL). Model 4 then added in the partner effects at Level 1, which measure the effect of each dyad member’s characteristics on the outcome of his or her partner (i.e., IWD’s religiosity on caregiver’s perception of IWD QoL and caregiver’s religiosity on the IWD’s self-reported QoL). This step addressed the following questions: What are the effects of the partner’s reports of religiosity on one’s own perception of IWD QoL? What are the partner effects of religiosity on IWD QoL reports after accounting for one’s own reports of religiosity on perception of IWD QoL?

Results

Paired sample t-tests.

Total extent of religiosity was not significantly different between caregivers and IWDs when examined by paired t-tests (t (110) = 1.59, p = 0.12). We did find that caregiver perceptions of IWD QoL significantly differ from IWD reports of their own QoL (t (109) = −5.83, p = 0.00). Caregivers report lower perceptions of IWD QoL (M = 2.37, SD = 0.46) than IWDs self-report (M = 2.66, SD = 0.43).

Correlations.

Correlations among variables of interest for caregivers and IWDs indicated that religiosity was positively correlated with being African-American and negatively correlated with being a spouse and having a higher level of education (see Table 4).

Table 4.

Correlations among demographic and predictor variables for caregivers and IWDs.

	1	2	3	4	5	6	7	8	9	10
1. Race (African-American)	–
2. Relationship (spouse)	−0.50***	–
3. IWD education	−0.53***	0.50***	–
4. CG education	−0.38***	0.18†	0.41***	–
5. Care stress	−0.21*	0.11	0.19†	0.08	–
6. MMSE	−0.05	0.18†	0.11	0.20*	0.18†	–
7. IWD total religiosity	0.29**	−0.27**	−0.24*	−0.22*	0.02	0.04	–
8. CG total religiosity	0.39***	−0.30**	−0.38***	−0.33***	−0.06	0.11	0.06	–
9. IWD QoL (self-report)	−0.26**	0.28**	0.33***	0.26***	−0.07	0.11	0.06	−0.20*	–
10. IWD QoL (CG report)	−0.04	0.06	0.11	0.06	−0.32***	0.08	0.17†	−0.04	0.30***	–

Note: N = 111 dyads; ^† p < .10, *p < .05, **p < .01, ***p < .001; CG = caregiver, IWD = individual with dementia.

Multi-level models.

Accounting for the dyadic nature of the data with a two-intercept APIM, we found that care-related stress is significantly associated with the reports of QoL by caregivers. Caregivers that report higher care-related stress report lower levels of QoL for the IWD (see Table 5, Models 1 and 2). Second, in examining actor effects in Model 3 (effects of one’s own religiosity on one’s own report of IWD QoL), we see that one’s own reports of total religiosity are not predictive of one’s own perceptions of IWD’s QoL, for reports given by both caregivers and IWDs (Table 5, Model 3). However, when we add in partner effects, we find a significant partner effect of caregivers’ religiosity on IWDs’ perceptions of their own QoL and for IWDs’ religiosity on caregivers’ perceptions of IWDs QoL. For each one unit increase in caregivers’ religiosity, IWDs report a 0.08 unit increase in their QoL. Simultaneously, for each one unit increase in IWDs’ religiosity, caregivers report a 0.09 unit decrease in IWDs’ QoL.

Table 5.

The influence of religiosity on caregivers’ perceptions and IWDs’ self-reports of IWDs’ QoL.

	Model 1		Model 2		Model 3		Model 4
	B	SE	B	SE	B	SE	B	SE
Fixed effect
Intercept for CG	2.37***	0.04	2.36***	0.04	2.36***	0.04	2.36***	0.04
Intercept for IWD	2.66***	0.04	2.66***	0.04	2.66***	0.04	2.66***	0.04
Actor effects
Total_Religiosity for CGa	–	–	–	–	−0.001	0.03	−0.06†	0.04
Total_Religiosity for IWD	–	–	–	–	−0.002	0.03	0.06†	0.03
Partner effects
Total_Religiosity for CG	–	–	–	–	–	–	0.08*	0.03
Total_Religiosity for IWD	–	–	–	–	–	–	−0.09**	0.03
Control variables
Caregiver stress*CG	–	–	−0.01***	0.003	−0.01***	0.003	−0.01***	0.003
Caregiver stress*IWD			−0.02	0.003	−0.002**	0.003	−0.002	0.003
Random effect (CSH)
Var(1)	0.21***	0.03	0.19***	0.03	0.19***	0.03	0.18***	0.03
Var(2)	0.19***	0.03	0.19***	0.03	0.19***	0.03	0.18***	0.02
Residuals	0.30***	0.09	0.30	0.09	0.30***	0.09	0.27**	0.09
−2 log likelihood	266.1		272.3		282.9		281.0
AIC	272.1		278.3		288.9		287.0

Notes: Dyad N = 111. †p <.10, *p < .05, **p < .01, and ***p < .001. Model 2 also tested the effects of Relation (spouse = 1), race (African-American = 1), IWD MMSE, and education (reported by each individual) but all were non-significant and were dropped from further model building.

^aTotal religiosity represents a compilation of three items for each person (attendance, prayer, and subjective ratings of religiosity) reflecting one’s total subjective practice/importance of religion.

Post-hoc analyses.

The two-intercept model estimates coefficients for caregivers and IWDs separately, providing four actor and partner effects per one predictor (e.g., actor effect for caregiver, partner effect for caregiver, actor effect for IWD, and partner effect for IWD). However, this analysis does not test the difference between coefficients for caregivers and IWDs. As such, we ran a posthoc one-intercept model to determine if actor and partner effects differ between caregivers and IWDs by looking at the significance of interaction terms that multiply a single variable of relation (caregiver = 1, IWD = 0) by each variable. In examining actor effects, we found no significant interactions (i.e., differences between caregivers and IWDs) on the key religiosity variable. After adding in partner effects, we find that the effect of one’s own religiosity on perceptions of IWD QoL significantly differed for IWDs and caregivers (B = −0.06, SE = 0.02 p < .05; caregivers: B = −0.06, IWDs: B = 0.11). In addition, there was a significant difference in the partner effects between IWDs and caregivers (B = 0.08, SE = 0.02 p < .001; caregivers: B = 0.08, IWDs: B = −0.09).

Discussion

Overall, we find that religiosity plays an important role in the lives of IWDs when analyzed jointly with their caregiver. Utilizing a composite scale of multiple dimensions of religiosity, we found that there is an interdependent effect of dyadic members’ religiosity on perceptions of QoL for the IWD. The results highlight that the caregivers’ religiosity affects IWDs’ own self-perceptions of QoL. Meanwhile, IWDs’ own religiosity also is associated with caregivers’ perceptions of IWDs’ QoL. This study is one of few to address religiosity of IWDs (Snyder, 2003). Moreover, religiosity has not been examined previously at the level of the dyad. The findings emphasize the importance of a sense of understanding between the caregiver and IWD regarding religiosity to affect QoL for IWDs.

Contrary to expectations, one’s own religiosity for both caregivers and IWDs was not associated with perceptions of QoL for the IWD, despite a significant difference in the effects between caregivers and IWDs. These findings suggest that perhaps the personal experience of religion alone is not meaningful enough to have substantial effects on perceptions in the caregiving experience. However, there is a significant effect of each partner’s religiosity affecting their own perceptions of the IWD’s QoL, and there are significant differences between caregivers and IWDs. We find that the caregivers’ religiosity is associated with IWDs’ self-reported QoL, such that higher levels of caregiver religiosity predicted higher self-reported IWD QoL. These findings highlight the interdependence of the dyad. Caregivers who are more religious perhaps have a positive influence on IWD’s QoL through their actions and communication. These more religious caregivers may also have a more positive life perspective on the caregiving experience, supporting such frameworks as existentialism, which posit that individuals can attribute meaning and find positive aspects in difficult situations (Farran, 1997). Such a perspective could affect the way they respond to the IWD, thus improving the experience for the IWD. The IWD also likely relies on the caregiver for aspects of religious practice. Previous research shows that increasing impairment limits IWD’s ability to participate in formal church services (Katsuno, 2003); a caregiver that is more religious and more concerned with making religious practice a part of life may be more likely to include the IWD in different aspects of religiosity. We also find that caregivers report lower QoL for the IWD when the IWD reports higher religiosity. Perhaps this effect stems from the caregiver’s awareness that while religion is of importance to their relative, their ability to practice is diminished to some extent by their growing disability.

These findings regarding the effects of religiosity on QoL for IWDs emphasize the importance of understanding IWDs’ values. Religion is just another aspect of value-focused care that should be addressed in the effort to provide person-centered care (Doherty, 2006). Caregivers should discuss these issues with their care recipients early on in care, given that abilities will change with time. Caregivers can then honor these wishes and support IWDs when they are no longer able to articulate their specific desires. The level of dependence of IWDs on their caregivers makes it crucial for caregivers to aid IWDs in ensuring their religious practices.

This study is strengthened by the use of a religiosity construct that addresses multiple aspects of religious practice; however, it is not without limitation. Our measure of religiosity used single items to assess each aspect of religious practice, as the purpose of the original study was to measure IWDs’ values and preferences. We did not have detailed religiosity scales addressing specific questions about whether religion is used to deal with difficult times and stress (Krause, 1998; Schaefer & Gorsuch, 1993). More extensive studies should focus on using more robust measures of religiosity to further our understanding of this issue. Second, high religiosity could be linked to various behaviors that are related to higher overall well-being (Herbert et al., 2007; Stuckey, 2001). We did not have baseline data on health habits or lifestyle practices (refraining from smoking, drinking, etc.) that could play such a role. Third, given that we utilized a cross-sectional study design, we are also unable to establish directionality of the effects found. Fourth, attending religious services may be less about personal beliefs and more about a desire for social integration and contact (Herbert et al., 2007). Similarly, it may be that the social integration and subsequent interest in ensuring a certain quality of care is what drives the observed relationship. Such results may also be found among people who hold strong affiliations with other groups that are not necessarily religiously affiliated; a relationship these data cannot test. Fifth, it should be noted that this sample was primarily white and Protestant. It would be useful to have a sample that is more diverse, as groups tend to differ in overarching beliefs and theology. Finally, it is important to note that we did not have data related to stressors in the lives of the IWDs, as we had for the caregiver. Including such measures would give a more complete picture of the effects of religiosity on QoL after accounting for stressful life experiences.

In conclusion, religion is of great importance to both caregivers and IWDs, and it is evident that beliefs of the caregiver have a salient effect on reports by the IWD. Communication and understanding of the IWD’s religious beliefs and desires could have marked results in improving QoL for the IWD.