Religiousness and lifestyle among Europeans in SHARE

Religiousness and lifestyle among Europeans in SHARE

p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 7 4 e8 1 Available online at www.sciencedirect.com Public Health journal homepage: www.elsevier.com/puhe ...

775KB Sizes 2 Downloads 67 Views

p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 7 4 e8 1

Available online at www.sciencedirect.com

Public Health journal homepage: www.elsevier.com/puhe

Original Research

Religiousness and lifestyle among Europeans in SHARE €ren Mo €ller b, N.C. Hvidt c,d, R. Lindahl-Jacobsen a L.J. Ahrenfeldt a,*, So a Interdisciplinary Center for Research and Education on Population Change, University of Southern Denmark, Odense, Denmark b OPEN e Odense Patient Data Explorative Network, Odense University Hospital and Department of Clinical Research, University of Southern Denmark, 5000 Odense, Denmark c Research Unit of General Practice, Department of Public Health, University of Southern Denmark, 5000 Odense, Denmark d Academy of Geriatric Cancer Research (AgeCare), Odense University Hospital, 5000 Odense, Denmark

article info

abstract

Article history:

Objectives: Religiousness is associated with longevity and better physical health, which may

Received 1 June 2018

be due to lifestyle choices. Here, we examine associations between religiousness and

Received in revised form

health, explained by lifestyle.

11 September 2018

Study design: This is a longitudinal study.

Accepted 11 September 2018

Methods: Data came from 23,864 people aged 50 and above included in the Survey of Health, Ageing and Retirement in Europe in 2004e2005 and followed up during 11 years. Results: Praying and taking part in a religious organization were associated with lower odds

Keywords:

of smoking [odds ratio (OR) ¼ 0.82, 95% confidence interval (CI): 0.73, 0.92 and 0.61, 95% CI:

Religiousness

0.53, 0.70], alcohol consumption (OR ¼ 0.71, 95% CI: 0.64, 0.78 and OR ¼ 0.76, 95% CI: 0.67,

Smoking

0.85), physical inactivity (OR ¼ 0.88, 95% CI: 0.79, 0.98 and OR ¼ 0.54, 95% CI: 0.48, 0.61), and

Alcohol consumption

doing no vigorous physical activity (OR ¼ 0.92, 95% CI: 0.85, 0.98 and OR ¼ 0.63, 95% CI: 0.58,

Physical activity

0.68). Furthermore, religious organizational involvement lowered the odds of sleep prob-

Sleep problems

lems (OR ¼ 0.83, 95% CI: 0.76, 0.91), whereas being religiously educated lowered the odds of

Body weight

high body weight (OR ¼ 0.87, 95% CI: 0.79, 0.96). The more religious (people who prayed, took part in a religious organization and were religiously educated) had lower odds of smoking, alcohol consumption, physical inactivity, and sleep problems than other respondents, and compared with people who only prayed, they had lower odds of smoking, physical inactivity, and sleep problems. People who only prayed had lower odds of alcohol consumption but higher odds of sleep problems than the non-religious. Conclusions: This study confirms that the positive relations between religiousness and health to an important degree can be explained by lifestyle. © 2018 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.

* Corresponding author. Department of Public Health, University of Southern Denmark, J. B. Winsloews Vej 9B, 5000 Odense C, Denmark. Tel.: þ45 6550 3844. E-mail address: [email protected] (L.J. Ahrenfeldt). https://doi.org/10.1016/j.puhe.2018.09.009 0033-3506/© 2018 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.

p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 7 4 e8 1

Introduction

Methods

The majority of studies on religiousness and health have demonstrated that religious involvement predicts greater longevity and is positively related to a variety of physical, behavioral, and mental health outcomes.1 The reasons for the positive associations are multiple and complex but include, for instance, strong community and psychosocial factors, more positive emotions, and a healthier lifestyle.2e5 Individuals who participate in religious activities and services are more likely to limit unfavorable lifestyle behaviors such as smoking6,7 and alcohol consumption.8e10 Furthermore, people with regular church attendance are more often physically active or engage in regular exercise.11,12 There is less research on health behaviors such as sleep, and results are inconsistent;3 however, there is some indication that religious people sleep better, which may influence both physical and mental health.3 In contrast, research on religion and body weight has shown mixed findings with religiousness being associated with higher body weight in most studies.13 The majority of research on associations between religious participation and health comes from the United States, whereas studies in other regions are sparser.2 For instance, less evidence is available as to whether religiousnessehealth associations exist in more secular countries or whether associations vary across religious groups.2 In our recent study,14 we examined some of these associations in various European regions. Using a longitudinal study design based on the Survey of Health, Ageing and Retirement in Europe (SHARE), we found that taking part in a religious organization had strong associations with health, in particular activity limitations and depressive symptoms, whereas religious education by parents lowered the odds of poor selfrated health and long-term health problems. However, little evidence was found that prayer was associated with health, which is in agreement with previous research.15,16 Recent cross-sectional evidence from SHARE17 investigated associations between praying and being religiously educated and multiple behavioral risk factors (MBRFs) for chronic diseases including high body weight, smoking, risky alcohol consumption, and physical inactivity.17 They found that both praying and being religiously educated were associated with fewer MBRFs. Here, we aim to verify associations between religiousness and health explained by lifestyle in a large sample of 23,864 middle-aged and older Europeans in SHARE, followed up for 11 years. We hypothesize that religiousness lowers the odd of smoking, alcohol consumption, physical inactivity, sleep problems, and perhaps high body weight but that religiousness would have the greatest benefits in those who are more religious. Thus, in addition to single religiosity measures, we also investigate different combinations of religiousness and their associations with lifestyle and examine whether associations differ by gender and religious affiliations.

Setting and study population

75

SHARE is a cross-national multidisciplinary survey of community-dwelling Europeans aged 50 years and older covering a broad variety of areas including health, lifestyle, and socio-economic status. Data were collected using computer-assisted personal interviews, supplemented by a self-administered paper and pencil questionnaire given to participants after the interview. The latter was used to ask questions on more sensitive topics such as psychological wellbeing and religion.18 This study included respondents aged 50 and older from wave 1 of SHARE, who were followed up in waves 2 (2006e2007), 4 (2011), 5 (2013), and 6 (2015). Wave 3 does not contain health data and is, therefore, not included in this study. Ten European countries were eligible; however, Greece was not a part of waves 4 and 5, and the Netherlands did not participate in wave 6. France was not included because the French questionnaire did not contain questions on religiousness. The response rates varied between countries, being, for example, 40.3% in Belgium and 67.1% in Denmark in wave 1.19

Measures of religiousness Information about religiousness was available from three questions in wave 1. Taking part in a religious organization was examined as part of the personal interview (‘Have you done any of these activities in the last month?’) including seven possible answers, one of them being ‘taken part in a religious organization (church, synagogue, mosque, etc.)’, with the possible answers ‘yes’ or ‘no’. The two other religiosity measures were assessed by the drop-off questionnaire including ‘praying’ (‘Thinking about the present, about how often do you pray?’) dichotomized into praying and not praying and religious education (‘Have you been educated religiously by your parents?’) answered by ‘yes’ or ‘no’. Furthermore, the question on religious affiliation (‘What religion do you belong or feel attached to mostly?’) came from the drop-off questionnaire, categorized into Protestant, Catholic, other (including Orthodox, Jewish, Muslim, and other), and none. We classified religiousness into three categories: the more religiousdpeople who prayed, had taken part in a religious organization during the previous month, and were religiously educated by their parents; the less religiousdpeople who prayed, without taking part in a religious organization, or being religiously educated; and the non-religiousdpeople who did not pray, did not take part in a religious organization, and were not religiously educated. Moreover, three comparisons were constructed: comparison 1dthe more religious vs all other respondents; comparison 2dthe more religious vs the less religious; and comparison 3dthe less religious vs the non-religious.

76

p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 7 4 e8 1

Lifestyle variables We assessed self-reported information about current smoking, daily alcohol consumption, physical activity, body weight, and sleep. Smokers are people who reported that they were currently smoking, which include cigarettes, cigars, and pipes; all others were categorized as non-smokers. Alcohol consumption was defined by intake of any alcoholic beverages, such as beer, cider, wine, spirits, or cocktails daily or almost every day during the previous 3 months. In wave 1, alcohol consumption referred to the 6 months preceding the survey. Information about smoking and alcohol was available in waves 1e5. Physical inactivity was based on two questions: ‘How often do you engage in activities that require a moderate level of energy such as gardening, cleaning the car, or doing a walk?’ and ‘How often do you engage in vigorous physical activity such as sports, heavy housework, or a job that involves physical labor?’ People who answered ‘hardly ever or never’ were categorized as being physically inactive or doing no vigorous physical activity. High body weight was based on the body mass index (BMI) calculated as weight (kg)/height (m2). Individuals were considered to have high body weight if they had a BMI of 25.0 kg/m2 or above.20 Sleep problems were assessed by respondents reporting ‘yes’ to having trouble sleeping recently.

Statistical analyses Associations between religiousness and lifestyle were analyzed by logistic regression models with robust standard errors, estimating odds ratios (ORs) and 95% confidence intervals (CIs). In the overall model, we included all people interviewed in wave 1, who were followed up in at least one of waves 2, 4, 5, and 6. Repeated observations from the same individual were taken into account by clustering,21 and the models were adjusted for region, gender, age at interview (continuous), educational level, marital status, and employment. Furthermore, we fitted an interaction model, including terms for interaction between the wave and religiosity measures, showing estimates for the associations between religiousness and lifestyle in the individual follow-up waves. The overall model was repeated examining possible interactions between religiosity measures and gender and between religiosity measures and religious affiliations. In addition, we investigated the associations between religiousness and lifestyle in a cross-sectional design using data from wave 1. This analysis included the cross-sectional individual probability weights supplied by SHARE. We corrected the main analysis for multiple testing using the HolmeBonferroni method via R, version 3.3.1. For all other analyses, we used Stata, version 14.2.

Results In Table 1, we report descriptive statistics of participants in wave 1 and of people from wave 1 who were followed up in waves 2, 4, 5, and 6, respectively. In wave 1, 68.9% reported praying, 12.0% took part in a religious organization, and 76.1% were religiously educated.

Praying was associated with lower odds of smoking (OR ¼ 0.82, 95% CI: 0.73, 0.92) and alcohol consumption (OR ¼ 0.71, 95% CI: 0.64, 0.78) in the overall model (Fig. 1A) and in most waves (Supplementary Table 1). The associations were modified by gender, with lower odds of smoking for women (OR ¼ 0.60, 95% CI: 0.51, 0.71) but not for men (OR ¼ 1.05, 95% CI: 0.91, 1.21). Praying was associated with lower odds of alcohol consumption for both sexes, but the odds were slightly lower for women (OR ¼ 0.60, 95% CI: 0.51, 0.69) than for men (OR ¼ 0.78, 95% CI: 0.70, 0.88; results not shown in table). Moreover, we found associations between praying and physical inactivity, both moderate (OR ¼ 0.88, 95% CI: 0.79, 0.98) and vigorous (OR ¼ 0.92, 95% CI: 0.85, 0.98) (Fig. 1A). These associations remained significant in wave 2 (Supplementary Table 1). Taking part in a religious organization was associated with lower odds of smoking (OR ¼ 0.61, 95% CI: 0.53, 0.70), alcohol consumption (OR ¼ 0.76, 95% CI: 0.67, 0.85), physical inactivity (OR ¼ 0.54, 95% CI: 0.48, 0.61), doing no vigorous physical activity (OR ¼ 0.63, 95% CI: 0.58, 0.68), and sleep problems (OR ¼ 0.83, 95% CI: 0.76, 0.91) in the overall model (Fig. 1B) and in most waves (Supplementary Table 1). We found no associations between religious organization and high body weight. Being religiously educated was associated with lower odds of high body weight (OR ¼ 0.87, 95% CI: 0.79, 0.96) overall (Fig. 1C) and in waves 4e6 (Supplementary Table 1). An overall association between religious education and smoking was indicated (OR ¼ 0.89, 95% CI: 0.78, 1.01), but an interaction with gender showed opposite effects for men and women with higher odds of smoking (OR ¼ 1.19, 95% CI: 1.01, 1.41) for men but lower odds for women (OR ¼ 0.65, 95% CI: 0.54, 0.77). A similar pattern was found for alcohol consumption with higher odds for men (OR ¼ 1.17, 95% CI: 1.02, 1.34) but no association for women (OR ¼ 0.94, 95% CI: 0.79, 1.11; not shown in table). For the more religious, we found lower odds of smoking (OR ¼ 0.60, 95% CI: 0.50, 0.71), alcohol consumption (OR ¼ 0.83, 95% CI: 0.73, 0.95), physical inactivity (OR ¼ 0.50, 95% CI: 0.43, 0.58), no vigorous physical activity (OR ¼ 0.59, 95% CI: 0.54, 0.65), and sleep problems (OR ¼ 0.78, 95% CI: 0.70, 0.87) compared with all other respondents (comparison 1) (Fig. 1D, Supplementary Table 1). When compared with the less religious (comparison 2), we found lower odds of smoking (OR ¼ 0.48, 95% CI: 0.36, 0.63), physical inactivity (OR ¼ 0.71, 95% CI: 0.54, 0.93), no vigorous physical activity (OR ¼ 0.73, 95% CI: 0.61, 0.86), and sleep problems (OR ¼ 0.75, 95% CI: 0.62, 0.90) for the more religious (Fig. 1E, Supplementary Table 1). For comparison 3, we found lower odds of alcohol consumption (OR ¼ 0.66, 95% CI: 0.54, 0.81) but higher odds of sleep problems (OR ¼ 1.24, 95% CI: 1.07, 1.44) for the less religious compared with the non-religious (Fig. 1F, Supplementary Table 1). Most associations remained significant after adjusting for multiple testing (Supplementary Table 1). When investigating interactions between religiosity measures and religious affiliations, we only found a few interactions (Table 2), while no differences between affiliations were detected for most associations. No associations were found between religiousness and lifestyle for those who did not have a religious affiliation. The strongest associations were found for people with ‘other’ religious affiliations. In this

77

p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 7 4 e8 1

Table 1 e Baseline characteristics of respondents in wave 1 (2004e2005) of SHARE, who were followed up in waves 2 (2006e2007), 4 (2011), 5 (2013), and 6 (2015). Numbers are N (%) unless otherwise stated. Variables Background variables Number Age in years, mean (SD) Male sex Educational level Lower Medium High Marital status Married/registered partnership Divorced or unmarried Widowed Employment status Employed Unemployed Retired Religiosity measures Belong to a religion Protestant Catholic Orthodox, Muslim, or other None Praying Taking part in a religious organization Religiously educated by parents Comparisons More religiousc Less religiousd Not religiouse Outcomes Current smokingf Alcohol consumption (every day alcohol consumption for the last 3 months)f,g Physical inactivity No vigorous physical activity High body weight (BMI  25) Sleep problems

Wave 1 (2004e2005)a

Wave 2 (2006e2007)b

23,864 64.6 (10.0) 10,946 (45.9) 23,648 12,434 (52.6) 6920 (29.3) 4294 (18.2) 23,833 17,497 (73.4) 2732 (11.5) 3604 (15.1) 23,426 6449 (27.5) 1510 (6.5) 15,467 (66.0)

16,509 64.1 7505 16,391 8576 4717 3098 16,501 12,145 1882 2474 16,246 4597 1010 10,639

16,263 4442 (27.3) 7362 (45.3) 2181 (13.4) 2278 (14.0) 11,137 (68.9) 2823 (12.0) 12,378 (76.1)

11,738 3084 5199 1832 1623 8132 2135 8999

Wave 4 (2011)b

(28.3) (6.2) (65.5)

10,518 63.3 4706 10,432 5226 3043 2163 10,512 8011 1224 1277 10,338 3129 682 6527

(26.3) (44.3) (15.6) (13.8) (69.6) (13.0) (76.8)

7596 2247 3976 204 1169 5006 1085 5688

(9.7) (45.5) (52.3) (28.8) (18.9) (73.6) (11.4) (15.0)

Wave 5 (2013)b

Wave 6 (2015)b

(30.3) (6.6) (63.1)

9715 62.6 4301 9631 4806 2770 2055 9709 7455 1149 1105 9549 3088 642 5819

(32.3) (6.7) (60.9)

9246 62.1 (8.2) 4107 (44.4) 9185 4656 (50.7) 2654 (28.9) 1875 (20.4) 9243 7097 (76.8) 1077 (11.7) 1069 (11.6) 9138 3009 (32.9) 539 (5.9) 5590 (61.2)

(29.6) (52.3) (2.7) (15.4) (66.2) (10.4) (75.0)

7019 2036 3722 185 1076 4632 1048 5238

(29.0) (53.0) (2.6) (15.3) (66.3) (10.8) (74.5)

6739 1668 3107 1181 783 4756 1327 5163

(8.9) (44.7) (50.1) (29.2) (20.7) (76.2) (11.6) (12.2)

(8.5) (44.3) (49.9) (28.8) (21.3) (76.8) (11.8) (11.4)

(24.8) (46.1) (17.5) (11.6) (70.8) (14.4) (76.8)

1887 (8.3) 1222 (19.5) 2435 (14.3)

1478 (9.3) 869 (19.7) 1699 (13.9)

723 (7.1) 578 (18.4) 1224 (15.5)

703 (7.5) 551 (19.0) 1144 (15.7)

913 (10.3) 514 (20.8) 947 (13.4)

4699 (40.9) 5208 (21.8)

2932 (18.2) 3370 (20.4)

1603 (15.5) 2419 (23.0)

1289 (13.3) 2112 (21.7)

e e

1903 6795 9991 4852

1397 4838 6249 3283

1486 4684 5730 3138

2680 9727 14,104 7155

(11.3) (40.9) (60.3) (30.4)

(11.6) (41.6) (61.7) (29.7)

(13.5) (46.8) (61.1) (31.7)

(15.3) (48.3) (61.2) (33.0)

1428 4387 5549 2810

(15.5) (47.5) (62.2) (31.7)

SD, Standard deviation; BMI, body mass index; SHARE, Survey of Health, Ageing and Retirement in Europe. Regions: Northern Europe (Denmark and Sweden), Western Europe (Austria, Belgium, Germany, Switzerland, the Netherlands), and Southern Europe (Italy, Spain, and Greece). However, Greece was not part of waves 4 and 5, and the Netherlands was not part of wave 6. a Participants in wave 1. b Participants in wave 1 who were followed up in the specific waves. c Praying, taking part in a religious organization, and religiously educated. d Praying, not taking part in a religious organization, and not religiously educated. e Not praying, not taking part in a religious organization, and not religiously educated. f The question was not asked in wave 6. In wave 1, there is a large number (12,369) of missing answers to the question about the current smoking. If the missing values were included in percentage calculations, the prevalence of smoking in wave 1 was 19.7%. g In wave 1, alcohol consumption was measured within the last 6 months.

group, the Orthodox comprised 82.6%, while 1.9% were Jewish and 2.4% were Muslims, and 13.1% reported that they belong to other affiliations. Associations between religiousness and lifestyle were overall similar between Catholics and Protestants with one exception: praying was associated with lower odds of alcohol consumption in Catholics, whereas no association was found among Protestants. Results from the cross-sectional analysis were quite similar to those from the overall longitudinal model; however, being more religious was associated with high body weight in the cross-sectional model (OR ¼ 1.26, 95% CI: 1.08, 1.48),

whereas no difference was found in the longitudinal model (OR ¼ 1.02, 95% CI: 0.91, 1.14) (Fig. 1).

Discussion Based on a large longitudinal sample of middle-aged and older Europeans, we found strong associations between religiousness and lifestyle, suggesting that positive associations of religiousness and health coulddat least largelydbe explained by lifestyle.

78

p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 7 4 e8 1

Fig. 1 e Associations between religiosity measures and lifestyle in a cross-sectional and a longitudinal setting based on wave 1 participants (2004e2005) from SHARE, who were followed up in waves 2e6 (2006e2015), adjusted for European region, gender, age at interview, education, the marital status, and employment. OR, odds ratio; CI, confidence interval.

The positive findings among religiousness, smoking, and alcohol are no surprise as many religions have recommendations ordin some cases even prohibitionsdagainst alcohol intake and smoking.22 Although few religions have prescriptions and recommendations for exercise or physical activity, we found evidence of a positive effect of religious service attendance on physical activitydboth moderate and vigorous in line with the majority of previous research.1,3 What is at play, then, might be what one could call a religious ecology of the body.23 Thus, for instance, when religious people believe that the body is the temple of God or the divine

spirit, such notions transfer to more subtle ways of taking care of the body.24 However, despite inverse associations between religiousness and physical activity, research on religion and body weight has shown mixed findings with religiousness being associated with higher body weight in the majority of studies;13 however, four25e28 of the five25e29 multivariate analyses in longitudinal studies did not fully support this. In line, we found no associations between praying or taking part in a religious organization and body weight, but our research indicated that being religiously educated by parents lowered the odds of high body weight. Another interesting finding in

79

p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 7 4 e8 1

Table 2 e Significant interactions between religiosity measures and religions in a longitudinal mixed-effects setting based on wave 1 participants (2004e2005) from SHARE, who were followed up in at least one of waves 2 (2006e2007), 4 (2011), 5 (2013), and 6 (2015). Protestant

Catholic

Orthodox, Muslim, Jewish, or other affiliation

No religious affiliation

OR (95% CI)a

OR (95% CI)a

OR (95% CI)a

OR (95% CI)a

1.03 (0.10, 1.24) 0.97 (0.06, 1.09)

0.67 (0.58, 0.77) 1.00 (0.89, 1.13)

0.51 (0.31, 0.84) 0.59 (0.42, 0.82)

0.72 (0.51, 1.02) 0.96 (0.75, 1.22)

0.76 (0.08, 0.92) 0.94 (0.11, 1.19) 0.95 (0.10, 1.18)

0.77 (0.67, 0.88) 0.92 (0.78, 1.07) 0.96 (0.83, 1.10)

0.61 (0.50, 0.74) 1.37 (1.12, 1.68) 0.63 (0.52, 0.77)

2.19 (0.96, 5.03) 1.35 (0.57, 3.16) 0.57 (0.21, 1.51)

0.89 (0.09, 1.08) 0.90 (0.05, 1.01) 0.99 (0.06, 1.12)

0.87 (0.64, 1.18) 0.94 (0.76, 1.16) 0.89 (0.72, 1.09)

1.49 (1.00, 2.21) 0.55 (0.41, 0.72) 0.58 (0.43, 0.76)

1.12 (0.88, 1.43) 0.98 (0.83, 1.15) 1.11 (0.92, 1.32)

b

Praying Alcohol consumption No vigorous physical activity Religious organizationb No vigorous physical activity High body weight (BMI  25 kg/m2) Sleep problems Religiously educatedb Current smoking No vigorous physical activity Sleep problems

OR, odds ratio; CI, confidence interval; SHARE, Survey of Health, Ageing and Retirement in Europe; BMI, body mass index. a Estimates adjusted for gender, age at interview, education, marital status, and employment. b No significant interactions with religious affiliation were found for the remaining associations.

our study was that taking part in a religious organization and being more religious lowered the odds of sleep problems, possibly because of the effects of religious affiliation and practice on mental health.3,30 Another finding in our study was that the more religious people had increased odds of high body weight in the cross-sectional study, whereas no association was found in the longitudinal setting. This might indicate that being more religious could help reduce high body weight over time. In this study, we also investigated associations between different combinations of religiousness and lifestyle. Recent evidence14,31 lends support to the suggestion of two epidemiological forces with opposite directions in religiousness and health with well-integrated and long-standing reliance on religion (‘restful religiousness’) being associated with positive health outcomes, whereas the opposite holds for religiousness that has been sparked by crisis and disease (‘crisis religiousness’).31 Crisis religiousness often features negative religious coping elements that may by themselves lead to negative mental and physical health outcomes14,32,33 including higher odds of depressive symptoms.14 We found that religiousness had the largest benefits in those who were more religious, providing evidence for positive associations between restful religiousness and lifestyle. However, in this study, only sleep problems were associated with crisis religiousness, showing higher odds of sleep problems for people who only reported praying compared with the non-religious respondents. Although women are often more religious than men34,35 and that they overall have a healthier lifestyle,36 few associations between religiousness and lifestyle were modified by gender; however, some associations were modified by religious affiliations. The fact that people with ‘other’ religious affiliations have the most pronounced associations with lifestyle was a surprise finding for which there may be several explanations. First it may be because they are a minority; minorities are known to be more adamant in practicing their religious beliefs.37,38 Second, there could be a spillover effect on other healthy lifestyle habits from lifestyle-regulating religious practices such as fasting (Orthodox and Muslim) and

halal or Kosher regulations.39 Thus, religious communities seem to be a major pathway to human well-being,40 and our study provides evidence that this is the case also in Europe. Based on the findings from this and current longitudinal studies,2 it seems that religiousnessdparticularly participating in a religious organizationdplays a beneficial role in avoiding negative lifestyle habits, which might be a cause of the lower mortality.41 Thus, owing to the strong association between lifestyle and chronic diseases,20,42 religiousness could be assessed as a potential determinants of disease prevention. Thus, our research suggests that for people who are already religious, taking part in a religious organization might be encouraged as a kind of meaningful social participation influencing the lifestyle and sleep quality. Furthermore, it clearly points to the importance of countering loneliness (which is considered a major obstacle to the quality of life, health, and longevity in the modern society43 and has even been found to impact bad sleep44) and thus advancing meaningful communities in our societies, even for individuals who are not religious. This article has several strengths. The study was conducted with data from Europe, where data on religion-health analyses are sparser,2 and included multiple waves of longitudinal data, which are preferable to cross-sectional data, to avoid reverse causation.45 We used several measures of religiousness, including having received a religious education, an exposure that has been relatively unexamined in prior research,2 and used a number of outcomes on lifestyle and weight. The large sample from 10 European countries allowed us to examine associations by religious affiliation and gender and to adjust for major potential confounders. A limitation in SHARE is the relatively low response rate and attrition from the sample over the six waves. Moreover, misclassification is inevitable because of self-reporting, and measures of religiousness from the self-administrated questionnaire had approximately onethird of missing values. In contrast, the question ‘taken part in a religious organization’ had only 2% of missing values. In conclusion, our study showed strong associations between religiousness and lifestyle and aimed to verify that the positive relations between religiousness and health to an

80

p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 7 4 e8 1

important degree can be explained by lifestyle, which might be of particular importance for public health.

Author statements Acknowledgments This article uses data from SHARE waves 1, 2, 4, 5, and 6; see € rsch-Supan et al. (2013) for methodological details.18 The Bo SHARE data collection has been primarily funded by the European Commission through FP5 (QLK6-CT-2001-00360), FP6 (SHARE-I3: RII-CT-2006-062193, COMPARE: CIT5-CT-2005028857, SHARELIFE: CIT4-CT-2006-028812), and FP7 (SHAREPREP: N 211909, SHARE-LEAP: N 227822, SHARE M4: N 261982). Additional funding from the German Ministry of Education and Research, the Max Planck Society for the Advancement of Science, the U.S. National Institute on Aging (U01_AG09740-13S2, P01_AG005842, P01_AG08291, P30_AG12815, R21_AG025169, Y1-AG-4553-01, IAG_BSR06-11, OGHA_04-064, HHSN271201300071C), and from various national funding sources is gratefully acknowledged (see www. share-project.org).

Ethical approval The ethical approval was obtained from the University of Mannheim International Review Board.

Funding Funding was obtained from the German Ministry of Education and Research, the Max Planck Society for the Advancement of Science, the U.S. National Institute on Aging (U01_AG09740-13S2, P01_AG005842, P01_AG08291, P30_AG12815, R21_AG025169, Y1-AG-4553-01, IAG_BSR06-11, OGHA_04e064, HHSN271201300071C), and various national funding sources.

Competing interests None declared.

Author contributions L.J.A. and N.C.H. conceptualized and designed the study. L.J.A. performed data analyses and drafted all versions of the manuscript. S.M. assisted with data analyses. All authors helped interpreting the data and critically revised the entire manuscript. All authors approved the final version of the manuscript.

references

1. Koenig HG. Religion, spirituality, and health: a review and update. Adv Mind Body Med 2015;29:19e26. 2. VanderWeele TJ. Religion and health in Europe: cultures, countries, context. Eur J Epidemiol 2017;32:857e61. 3. Koenig HG, King DE, Carson VB. Handbook of religion and health. Oxford/New York: Oxford University Press; 2012. p. 317.

4. Lindahl-Jacobsen R, Hanson HA, Oksuzyan A, Mineau GP, Christensen K, Smith KR. The male-female health-survival paradox and sex differences in cohort life expectancy in Utah, Denmark, and Sweden 1850e1910. Ann Epidemiol 2013;23:161e6. 5. Morton KR, Lee JW, Martin LR. Pathways from religion to health: mediation by psychosocial and lifestyle mechanisms. Psycholog Relig Spiritual 2017;9:106e17. 6. Brown QL, Linton SL, Harrell PT, Mancha BE, Alexandre PK, Chen KF, et al. The influence of religious attendance on smoking. Subst Use Misuse 2014;49:1392e9. 7. Gillum RF. Frequency of attendance at religious services and cigarette smoking in American women and men: the third national health and nutrition examination survey. Prev Med 2005;41:607e13. 8. Brown TL, Parks GS, Zimmerman RS, Phillips CM. The role of religion in predicting adolescent alcohol use and problem drinking. J Stud Alcohol 2001;62:696e705. 9. Edlund MJ, Harris KM, Koenig HG, Han X, Sullivan G, Mattox R, et al. Religiosity and decreased risk of substance use disorders: is the effect mediated by social support or mental health status? Soc Psychiatry Psychiatr Epidemiol 2010;45:827e36. 10. Michalak L, Trocki K, Bond J. Religion and alcohol in the U.S. National Alcohol Survey: how important is religion for abstention and drinking? Drug Alcohol Depend 2007;87:268e80. 11. Hill TD, Ellison CG, Burdette AM, Musick MA. Religious involvement and healthy lifestyles: evidence from the survey of Texas adults. Ann Behav Med 2007;34:217e22. 12. Strawbridge WJ, Shema SJ, Cohen RD, Kaplan GA. Religious attendance increases survival by improving and maintaining good health behaviors, mental health, and social relationships. Ann Behav Med 2001;23:68e74. 13. Yeary KHK, Sobal J, Wethington E. Religion and body weight: a review of quantitative studies. Obes Rev 2017;18:1210e22. € ller S, Andersen-Ranberg K, Vitved AR, 14. Ahrenfeldt LJ, Mo Lindahl-Jacobsen R, Hvidt NC. Religiousness and health in Europe. Eur J Epidemiol 2017:1e9. 15. Musick MA, House JS, Williams DR. Attendance at religious services and mortality in a national sample. J Health Soc Behav 2004;45:198e213. 16. VanderWeele TJ, Yu J, Cozier YC, Wise L, Argentieri MA, Rosenberg L, et al. Attendance at religious services, prayer, religious coping, and religious/spiritual identity as predictors of all-cause mortality in the black women's health study. Am J Epidemiol 2017;185:515e22. 17. Linardakis M, Papadaki A, Smpokos E, Sarri K, Vozikaki M, Philalithis A. Are religiosity and prayer use related with multiple behavioural risk factors for chronic diseases in European adults aged 50þ years? Publ Health 2015;129:436e43. 18. Borsch-Supan A, Brandt M, Hunkler C, Kneip T, Korbmacher J, Malter F, et al. Data resource profile: the Survey of Health, Ageing and Retirement in Europe (SHARE). Int J Epidemiol 2013;42:992e1001. 19. Bergmann M, Kneip T, Luca D, Giuseppe, Scherpenzeel Annette. Survey participation in the Survey of Health, Ageing and Retirement (SHARE), wave 1e6. Based on release 6.0.0 (March 2017). In: SHARE working paper series 312017. Munich: SHARE-ERIC; 2017. 20. WHO. Global status report on noncommunicable diseases, Geneva 2010. 21. The behavior of maximum likelihood estimates under nonstandard conditions. In: Huber PJ, editor. Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, Volume vol. 1: Statistics; 1967. Berkeley, Calif: University of California Press; 1967.

p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 7 4 e8 1

22. Hoff A, Johannessen-Henry CT, Ross L, Hvidt NC, Johansen C. Religion and reduced cancer risk: what is the explanation? A review. Eur J Cancer 2008;44:2573e9. 23. Lyons J. Ecology of the body: styles of behavior in human life. Duke University Press; 1987. 24. Braaten CE, Braaten L. The living temple: a practical theology of the body and the foods of the Earth. Wipf and Stock Publishers; 2016. 25. Pasch KE, Nelson MC, Lytle LA, Moe SG, Perry CL. Adoption of risk-related factors through early adolescence: associations with weight status and implications for causal mechanisms. J Adolesc Health 2008;43:387e93. 26. Cline KM, Ferraro KF. Does religion increase the prevalence and incidence of obesity in adulthood? J Sci Study Relig 2006;45:269e81. 27. Kobayashi D, Shimbo T, Takahashi O, Davis RB, Wee CC. The relationship between religiosity and cardiovascular risk factors in Japan: a large-scale cohort study. J Am Soc Hypertens 2015;9:553e62. 28. Nathenson SL, Wen M. Religiousness, physical activity and obesity among older cancer survivors: results from the health and retirement study 2000e2010. Int J Religion Spiritual Soc 2012;2:129e44. 29. Feinstein M, Liu K, Ning H, Fitchett G, Lloyd-Jones DM. Incident obesity and cardiovascular risk factors between young adulthood and middle age by religious involvement: the Coronary Artery Risk Development in Young Adults (CARDIA) study. Prev Med 2012;54:117e21. 30. Koenig HG. Handbook of religion and mental health. San Diego: Academic Press; 1998. 31. Hvidt NC, Hvidtjorn D, Christensen K, Nielsen JB, Sondergaard J. Faith moves mountains-mountains move faith: two opposite epidemiological forces in research on religion and health. J Relig Health 2017 Feb;56(1):294e304. https://doi.org/10.1007/s10943-016-0300-1. 32. Harrison M O, Koenig HG, Hays JC, Eme-Akwari AG, Pargament KI. The epidemiology of religious coping: a review of recent literature. Int Rev Psychiatr 2001;13:86e93. 33. Pargament KI, Smith BW, Koenig HG, Perez L. Patterns of positive and negative religious coping with major life stressors. J Sci Stud Relig 1998;37:710e24.

81

34. Trzebiatowska M, Bruce S. Why are women more religious than men?. Oxford: Oxford University Press; 2012. 35. Deaton A. NBER working paper series: National Bureau of Economic Research, 152712009. 36. Cockerham WC. Health lifestyle theory and the convergence of agency and structure. J Health Soc Behav 2005;46:51e67. 37. Thygesen LC, Hvidt NC, Juel K, Hoff A, Ross L, Johansen C. The Danish religious societies health study. Int J Epidemiol 2012;41:1248e55. 38. Finke R, Stark R. Religious choice and competition. Am Sociol Rev 1998:761e6. 39. Schulz DN, Kremers SPJ, van Osch LADM, Schneider F, van Adrichem MJG, de Vries H. Testing a Dutch web-based tailored lifestyle programme among adults: a study protocol. BMC Publ Health 2011;11:108. 40. VanderWeele TJ. On the promotion of human flourishing. Proc Natl Acad Sci U S A 2017;114:8148e56. 41. Chida Y, Steptoe A, Powell LH. Religiosity/spirituality and mortality. A systematic quantitative review. Psychother Psychosom 2009;78:81e90. 42. WHO. Noncommunicable diseases. World Health Organization; 2017. http://www.euro.who.int/en/health-topics/ noncommunicable-diseases/noncommunicable-diseases. [Accessed 28 August 2017] [cited 2017 August 28]. 43. Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D. Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspect Psychol Sci 2015;10:227e37. 44. Kurina LM, Knutson KL, Hawkley LC, Cacioppo JT, Lauderdale DS, Ober C. Loneliness is associated with sleep fragmentation in a communal society. Sleep 2011;34:1519e26. 45. VanderWeele TJ, Jackson JW, Li S. Causal inference and longitudinal data: a case study of religion and mental health. Soc Psychiatry Psychiatr Epidemiol 2016;51:1457e66.

Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.puhe.2018.09.009.