INTRODUCTION
Many studies have shown that religiosity and spirituality have positive effects on the physical health outcomes of psychological, social, and behavioral pathways [1,2]. Cognitive function can be beneficially influenced by religious variables [3]. Among community-dwelling older adults, those with higher spiritual activity have been seen to have better cognitive function than those with lower spiritual activity [4]. In a longitudinal study, slower progression of Alzheimer’s disease (AD) was related to higher spirituality and individual religious practices [5]. Another study on patients with AD found that higher levels of religiosity appeared to be associated with slower cognitive and behavioral deterioration [6]. Based on these results, religious affiliation seems to be a protective factor not only for physical health, but also for mental health, such as cognitive impairment.
Although many studies have suggested a beneficial effect of religion on cognitive function, most have only compared religiously affiliated and non-affiliated groups. Few studies have investigated the effects of different religions. In a case-control study that investigated the association between religious affiliation and dementia risk in a Chinese population, Christian affiliation showed a decrease in the risk of AD, but no significant association was observed among other religious affiliations, such as Taoism or Buddhism [7]. Additionally, another study examined whether religiosity and spirituality could affect caregiving outcomes based on the different religious affiliations of the caregivers, and the protective effect of religiosity and spirituality on the relationship between the activities of daily living (ADL) of patients with AD and depressive symptoms in caregivers differed by religious affiliation [8]. Thus, it is necessary to compare cognitive functions among different religious groups because each religion has a different religious culture (e.g., praying, meditation, worship, and doctrines) that could affect one’s cognitive function.
Furthermore, most studies have evaluated cognitive function based on the Global Cognitive Function Scale (e.g., Mini-Mental State Examination, clinical dementia rating, and ADL) [4-7,9]. A recent study reported that specific subdomains of cognitive function (memory, language, and constructive ability) are differently associated with various religiosity/spirituality domains, such as organizational religious activity (ORA), non-organizational religious activity (NORA), and intrinsic religiosity (IR) [10].
If we can clarify the subdomains of cognitive function affected by specific religious variables and different religious affiliations, we may understand the positive effects of religion on cognitive functions better.
Since cognitive impairment is becoming a serious global issue due to rapid aging, we sought to investigate the association between religious variables and cognitive function. Particularly, we aimed to understand the differences in the subdomains of cognitive function based on different religious affiliations, including the religiously non-affiliated group, and to identify which aspects of religiosity are associated with cognitive function among patients with AD.
METHODS
Participants
A total of 317 participants aged >60 years were recruited from among outpatients of the Geriatric Psychiatric Clinic of Chuncheon Sacred Hospital in Gangwon Province, Republic of Korea, between April 2013 and December 2014. A comprehensive work-up, including geriatric examinations and neurocognitive tests, was performed to exclude participants with major comorbidities (e.g., major depressive disorder, organic brain syndrome, and vascular dementia). Forty-four participants with major comorbidities were excluded. Ten participants of other religions (e.g., traditional Korean Shamanism) were excluded owing to the small sample size. Also, 16 participants who scored more than 5 on the Global Deterioration Scale (GDS) were excluded, considering the reliability of the self-report. After excluding participants with major neurocognitive disorders caused by factors other than Alzheimer’s according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, the final sample included 247 participants. The study protocol was approved by the Institutional Review Board of the Chuncheon Sacred Heart Hospital (IRB approval number: 2014-22).
Clinical evaluation
All participants underwent a structured evaluation by psychiatrists, which included medical history and neurological examination. Detailed cognitive function was tested through one-on-one interviews with each participant, and all clinical psychologists with more than 3 years of clinical experience were educated by a researcher and supervised by a senior psychologist.
Measures
Cognitive function
The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) is widely used to evaluate cognitive function in clinical and research settings [11]. It has been translated into Korean and standardized for the Korean population [12]. The CERAD-K consists of a clinical and neuropsychological assessment battery that includes the Mini-Mental State Examination in the Korean version of the CERAD (MMSE-KC), verbal fluency, the Korean version of the Boston Naming Test, word list memory, word list recall, word list recognition, constructional praxis, and constructional recognition [12]. The GDS included in this battery is widely used to assess dementia severity [13]. It has been translated into Korean and standardized for the Korean population [14]. Moreover, the CERADK showed equivalence to the original packet, confirming its reliability and validity.
Religious affiliation
Religious affiliation is defined as a self-identified association with a religion, denomination, or a subdenominational religious group [7]. In this study, participants were asked to choose religious affiliation among Christianity (i.e., Catholicism and Protestantism), Buddhism, others, and no religion. Participants who chose “no religion” were classified as the religiously non-affiliated group.
Religiosity
The Duke University Religion Index (DUREL) was developed by Koenig et al. [15], and it was adapted for the Korean population [16]. The 5-item DUREL had a Cronbach’s alpha of 0.80 in this study. In another study, the 5-item scale had a Cronbach’s alpha of 0.75 and was strongly correlated (r=0.85) with the original 10-item scale [17]. The DUREL is composed of five items and three subscales: ORA (the frequency of attending religious meetings), NORA (the frequency of private religious activities), and IR (the subjective experience of the absolute, the importance of one’s belief, and the strong will to carry on one’s belief). The ORA and NORA were scored from never (6) to more than once a day (1), and 3-items on IR were scored from definitely not true (6) to definitely true (1). We then reversed each score and added the score of the 3-items to the IR to calculate the IR score. Higher scores on the three subscales indicate a higher level of religiosity and spirituality.
Statistical analyses
Demographic data (e.g., age, educational duration, and sex) were analyzed using one-way analysis of variance and chi-square tests. We compared ORA, NORA, IR, MMSE-KC, and GDS among Christians, Buddhists, and religiously non-affiliated groups using an analysis of covariance after controlling for age, sex, and educational duration. Furthermore, the subdomains of cognitive function were compared among the groups after controlling for demographic variables. We performed a post hoc comparison using the Bonferroni method. We analyzed the correlations between the subdomains of cognitive function and religious variables. Finally, we conducted a stepwise hierarchical multiple regression analysis of the cognitive function subdomains. First, we entered the controlled variables (i.e., demographic data) into a regression model. Then, the controlled and independent variables (i.e., ORA, NORA, and IR) were entered. The independent variables were selected using an automatic procedure. At each step, a variable was added or subtracted from the set of independent variables based on the adjusted R2 [18,19]. Data were analyzed using SPSS software (version 19.0.0, IBM Corp.).
RESULTS
Group characteristics
A total of 247 participants (religiously affiliated group=133; religiously non-affiliated group=114) participated in this study. The mean age was 79.31±6.33 years, with 182 (73.7%) female, and the average educational duration was 3.23±3.56 years. The average age of the Christian, Buddhist, and religiously non-affiliated groups was 79.38±6.43, 78.35±5.85, and 79.50±6.40, respectively. The average educational duration (in years) of the Christian, Buddhist, and religiously non-affiliated groups was 4.04±3.93, 2.71±2.81, and 2.65±3.27. No significant differences were found among the groups in terms of age; however, differences were found in terms of sex (χ2=13.609, p=0.001) and educational duration (F=4.615, p=0.011). In the post hoc analysis, the Christian group had higher ORA, NORA, and IR than the Buddhist and religiously non-affiliated groups. However, no difference was observed in the GDS score. Among all participants, religious non-affiliation was the most common category (46.2%), followed by Christianity (41.3%) and Buddhism (12.6%) (Table 1).
Comparison of cognitive function domains among Christian, Buddhist, and the religiously non-affiliated group
As shown in Table 2, the Christian group showed a better level of MMSE-KC, word-list memory, word-list recall, and constructional praxis after controlling for age, sex, and educational duration (F[2,241]=6.825, p=0.001; F[2,241]=4.524, p=0.012; F[2,241]=3.170, p=0.044; F[2,241]=5.039, p=0.007). In the post hoc analysis, differences in these subdomains of cognitive function were observed due to differences between Christian and religiously non-affiliated groups, but not between Buddhist and religiously non-affiliated groups. Figure 1 shows the differences in the MMSE-KC, word list memory, word list recall, and constructional praxis among Christian, Buddhist, and religiously non-affiliated groups.
Correlation among sub-domains of cognitive function and religious variables
Correlation analysis among the sub-domains of cognitive function and religious variables showed that ORA, NORA, and IR had significantly positive associations with the MMSE-KC (r=0.191, p=0.003; r=0.217, p=0.001; r=0.192, p=0.002), word list memory (r=0.219, p=0.001; r=0.191, p=0.003; r=0.181, p=0.004), and constructional ability (r=0.211, p=0.001; r=0.222, p=0.000; r=0.171, p=0.007). Additionally, ORA showed a significant relationship with verbal fluency (r=0.166, p=0.009) and IR had a significantly positive association with word list recall (r=0.153, p=0.016) (Table 3).
Hierarchical stepwise multiple regression analysis of cognitive functions with religious variables
Step 1 of the hierarchical stepwise multiple regression analysis that set the MMSE-KC, word list memory, word list recall, and constructional praxis as dependent variables were significant in Step 1, [F(3,243)=23.449, p=0.000, R2=0.225; F(3,243)=18.655; p=0.000, R2=0.187; F(3,243)=2.717, p=0.045, R2=0.032; F(3,243)=19.886, p=0.000, R2=0.197], respectively. After step 1, the addition of ORA, NORA, and IR, as determined by a stepwise process, was performed in step 2. The addition of ORA to explain word list memory and constructional praxis significantly increased the model’s explanatory power [F(4,242)=15.823, p=0.000, R2=0.207, ΔR2=0.020; F(4,242)=17.215, p=0.000, R2=0.222, ΔR2=0.024]. ORA positively and significantly predicted word list memory and constructional praxis (B=0.250, p=0.014; B=0.182, p=0.006, respectively). The addition of NORA to explain the MMSE-KC significantly increased the explanatory power [F(4,242)=21.764, p=0.000, R2=0.265, ΔR2=0.040]. NORA also significantly predicted MMSE-KC (B=0.397, p=0.000). The addition of the IR to explain word list recall significantly increased the explanatory power[F(4,242)=4.007, p=0.004, R2=0.062, ΔR2=0.030]. IR positively explained wordlist recall (B=0.047, p=0.006) (Table 4).
DISCUSSION
Our study investigated the effect of religiosity on the cognitive function of patients with AD by measuring not only global cognitive function but also its specific sub-domains. To the best of our knowledge, this is the meaningful study to investigate specific cognitive functions according to religious affiliation. The results indicate that the Christian group had better global cognitive functions, verbal memory and recall, and visuospatial ability than the non-religiously affiliated group. And, there was no significant difference in cognitive function between the Buddhist and religiously non-affiliated groups. This finding may reflect that the effect of religious affiliation on cognitive function may differ depending on religion. Post-hoc analyses showed no significant differences in groups; however, the relatively small sample size of the Buddhist group might have limited statistical power, and the findings should therefore be interpreted cautiously.
Although previous studies have suggested that religious affiliation could benefit cognitive function, this study highlights that the beneficial effects could differ depending on religion. Given the cross-sectional design, causal direction cannot be determined. The observed association may reflect reverse causation or selection effects, whereby individuals with better preserved cognitive function are more likely to maintain engagement in religious activities. Therefore, future studies should consider not only whether participants have a religious affiliation but also their level of religious activities. Longitudinal studies are also needed to clarify the temporal and causal relationships between religiosity and cognitive function.
Unlike the Christian group, which showed results consistent with those of a previous study, the Buddhist group showed no significantly better cognitive functions than the religiously non-affiliated group. This could be because Christianity has more religious activities, such as weekly worship, than Buddhism. The Christian group showed higher ORA, NORA, and IR values than did the other groups. Extant literature suggests that Christian participants report more frequent participation in ORA than Buddhist participants [7]. However, this finding may be related to the Christian-centric structure of DUREL, which may underestimate religiosity in Buddhist participants due to cultural mismatches in the IR and ORA subscales.
According to regression analysis, global cognitive and memory functions were predicted by NORA and IR, respectively. Additionally, visuospatial ability and verbal memory function are associated with ORA. Although we could not directly verify the causal relationship between religious activity and cognitive function in this cross-sectional study, we believe that the level of religious activity is an important factor determining the beneficial effects of religious activity on cognitive function. Therefore, we suggest that the level of religious activity may be as important as religious affiliation.
Religious activities may be related to cognitive function. They usually include prayer/meditation, singing, scriptural study, sermons, and general socialization, which can stimulate cognitive faculties [20]. These activities require people to plan, concentrate, and remember several aspects related to them. Repeating prayers, which can be measured using the NORA, indirectly facilitates mental concentration and acts as a word-repeating exercise that may reduce the progression of dementia [21]. One review proposed that religiosity and spirituality may be involved in abstract thinking, which serves to exercise the brain areas necessary for memory retention [2]. Furthermore, a recent cohort study suggested that religious beliefs and experiences are associated with thicker cortices in the left and right parietal and occipital areas [22]. Considering the association between visuospatial ability and the parietal-occipital lobes [23,24], a higher level of constructional praxis might be because of religiosity and spirituality on the brain region responsible for visuospatial ability. However, these neurobiological interpretations could be regarded as hypothetical, as the present study did not include direct neuroimaging or neurophysiological measurements.
Religiosity can affect cognitive function in different ways. Some studies have proposed that religious beliefs corresponding to IR have the potential to influence the cognitive appraisal of negative life events in a way that makes them less distressing [25]. Furthermore, religious prosocial behaviors that can be measured by ORA (e.g., meetings during religious social events) could buffer stress and lead to the human support that is required during difficult times [26-28]. Other studies have suggested that stress gives rise to physiological changes (particularly elevated levels of glucocorticoids) that adversely affect the brain areas responsible for memory, especially the hippocampus region [29-32]. Considering that many studies have suggested that stress and depression negatively affect cognitive function, religiosity seems to have a positive effect on cognitive function by reducing negative emotions.
In this context, we assume that there is no difference between Buddhist and religiously non-affiliated groups in terms of cognitive functions, which could be caused by insufficient levels of religious activities. This result suggests that we need to consider not only whether participants have religious affiliation but also the level of religious variables when studying the effect of religion on cognitive function.
This study had some limitations. First, as a cross-sectional study, our ability to make causal inferences was limited; therefore, a longitudinal approach should be considered in future studies. Second, the sample size of the Buddhist group was relatively small; therefore, the real effects of religiosity and spirituality on cognitive functions may have been underestimated. Third, participants had relatively low educational levels. This might be due to the poor socio-political circumstances induced by the Second World War and the Korean War in South Korea when the participants were of school-going age [33]. This limits the generalizability of the results to more highly educated or Western populations. Fourth, the observed associations may be subject to residual confounding from unmeasured psychosocial and socioeconomic factors, and therefore could not be interpreted as evidence of an independent causal effect. Fifth, the DUREL subscale may be not suitable for assessing IR in non-Christian people. The ORA and NORA were assessed by the frequency of religious activity, but the IR subscale could be affected by the culture and doctrines of each religion. In addition to this, the Christian-centric wording of the IR and ORA subscales may have limited their ability to capture Buddhist forms of intrinsic belief and religious practice, potentially leading to an underestimation of religiosity in Buddhist participants. DUREL may have another limitation because DUREL cannot distinguish “spiritual but not religious” individuals as some studies suggest the subscales of DUREL may not directly measure subjective experiences of spirituality. Therefore, results related to the DUREL may be interpreted with caution. Future studies incorporating validated measures of spirituality would be valuable in further clarifying the distinct and shared contributions of religiosity and spirituality to cognitive function. Furthermore, there is a need to develop scales that can be used by all religious affiliations in future studies.
Compared to the religiously non-affiliated group, the Christian group showed better functioning in terms of global cognitive function, memory, and visuospatial ability, which had a significant association with religious activities and IR; however, the Buddhist group showed no difference from the other group. The Christian group had a higher level of religious activities and IR than the religiously non-affiliated and Buddhist groups. Therefore, these results may suggest that the levels of religious activity and IR may be related to cognitive function. However, the relatively small sample size of the Buddhist group could be an important limitation, underscoring the need for future studies with larger and more balanced samples.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
