CanJPsychiatry 2013;58(4):225-232

Originai Research

Religious Attendance, Spirituality, and Major Depression in Canada: A14-Year Follow-up Study
Lloyd Balbuena, MS, PhD^; Marilyn Baetz, MD, FRCPC^; Rudy Bowen, MD, CM, FRCPC^
' Research Assooiate, Department of Psychiatry, University of Saskatchewan, Saskatoon, Saskatchev^ran. 'Professor and Head, Department of Psychiatry, University of Saskatchewan, Saskatoon, Saskatchewan. Correspondence: Department of Psychiatry, University of Saskatchewan, Saskatoon Heaith Region, Room 119 Eliis Hali, 103 Hospitai Drive, Saskatoon, SK S7N 0W8; m.baetz@usask.ca. ^Professor and Director of Clinical Research, Department of Psychiatry, University of Saskatchewan, Saskatoon, Saskatchewan.

Key Words: prospective foilow-up of a nationaily representative sample, risk of depression overtime, religion, spirituality Received May 2012, revised, and accepted September 2012.

Objective: Although there have been numerous studies on the relation of religion or spirituality and major depression, few used a longitudinal, nationally representative sample. Our study sought to examine the effect of religious attendance, self-declared importance of spiritual values, and self-identification as a spiritual person on major depression. Method: Data coming from 8 waves (1994 to 2008) of the longitudinal Canadian National Population Health Survey were used. People (n = 12 583) who were not depressed at baseline (1994) were followed during 14 years. Depression at each cycle was assessed using the Composite International InterviewShort Form for Major Depression. Weibull proportional hazards regression was used to model longitudinal risk of depression, with religious attendance or spirituality as a predictor. Results: At baseline, monthly religious attenders tended to be older, female, and married, compared with occasional and nonattenders. The Weibull regression model revealed a 22% lower risk of depression for monthly attenders (hazard ratio 0.78, 95% Cl 0.63 to 0.95), compared with nonattenders, after controlling for age, household income, family and personal history of depression, marital status, education, and perceived social support. Neither self-reported importance of spiritual values nor identification as a spiritual person was related to major depressive episodes. Conclusion: Attending religious services at least monthly has a protective effect against major depression.

Pratique religieuse, spiritualit, et dpression majeure au Canada : une tude de suivi sur 14 ans
Objectif : Bien que de nombreuses tudes aient trait de la relation de la religion ou de la spiritualit avec la dpression majeure, peu d'entre elles ont t menes l'aide d'un chantillon longitudinal, nationalement reprsentatif. Notre tude cherchait examiner l'effet sur la dpression majeure de la pratique religieuse, de l'importance auto-dclare des valeurs spirituelles, et de l'auto-identification en tant que personne spirituelle. Mthode : Les donnes issues de 8 cycles (1994 2008) de l'Enqute nationale sur la sant de la population, enqute longitudinale canadienne, ont t utilises. Les personnes (n = 12 583) qui n'taient pas dprimes au dpart (1994) ont t suivies durant 14 ans. chaque cycle, la dpression tait value l'aide de la version abrge de l'entrevue internationale composite pour la dpression majeure. La rgression des risques proportionnels de Weibull a servi modeler le risque longitudinal de dpression, avec la pratique religieuse ou la spiritualit comme prdicteur. Rsuitats : Au dpart, les pratiquants religieux mensuels tendaient tre gs, de sexe fminin, et maris, comparativement aux pratiquants occasionnels et aux non-pratiquants. Le modle de rgression de Weibull a rvl un risque 22 % plus faible de dpression pour les pratiquants mensuels (rapport de risques 0,78; IC 95 % 0,63 0,95), compar aux non-pratiquants aprs contrle de l'ge, du revenu du mnage, des antcdents familiaux et personnels de dpression, du statut matrimonial, de l'instruction, et du soutien social peru. Ni l'importance auto-dclare des valeurs spirituelles ni l'identification comme personne spirituelle n'taient lies aux pisodes de dpression majeure. Conclusion : Assister aux offices religieux au moins une fois par mois a un effet protecteur contre la dpression majeure.
www.rheCJP.ca
TheCanadianJournalofPsychiatry, Vol58, No4,April2013 225

Original Research

ith mystical rituals and systems of beliefs, neither organized religion nor individual spirituality is an ideal field for scientific study. However, as William James' argued, the origins and claims of religion and spirituality do not diminish their pragmatic value. It is by its outcomes, such as social integration,^ making sense of suffering,^ and helping to take control over (or gaining acceptance of) external forces,"* that religion should be assessed by science. Specifically, in terms of mental health, there is intriguing, albeit inconsistent, evidence that religion, understood as public worship attendance, protects against depression or alleviates its symptoms in at least some populations.'* A few studies have reported a higher incidence or level of depression with higher religiosity.'" In each case, a specific aspect of religion was at play. For instance, moral injunctions against unmarried cohabitation" and a reward-seeking or extrinsic religious orientation' were both associated with higher depression. Counterintuitive results have also been reported: while religious belief promotes hopefulness, it also predicts higher depression.'- These results have not been consistent or cornparable because of the use of different measures and generally cross-sectional study designs. Since 2000, several longitudinal studies on religion and depression have been published. The findings reported in these studies are also inconsistent. In the Baltimore Epidemiologie Catchment Area Study,'-* religious worship was unrelated to episodes of major depression over a 10year period. This result is consistent with a study conducted among African Americans that reported no protective effect of religious attendance from depression, although those who reported receiving religious guidance were half as likely to be depressed 4 years later.'''-" In a recent study," the offspring of biologically predisposed parents were followed up at years 10 and 20. While religious salience at year 10 predicted lower risk of depression, religious attendance was unrelated to risk. In contrast, Braam et al,'* following an elderly population, reported an inverse relation between religious attendance and level of depression symptoms. Likewise, among elderly African Americans living with cancer, religious attendance was inversely related to levels of depressive symptoms." A recent study'** of communitydwelling elderly adults in Alabama found that higher intrinsic religiosity (that is, self-rated importance of religion) predicted a decline in depressive symptoms in 4 years of follow-up. Neither prayer nor religious attendance predicted the course of depression symptoms. Spirituality, an overlapping concept with religion, has also been well-studied regarding depression. In contrast to religion, which is yoked to particular traditions, spirituality is about finding transcendent meaning in life"a more

Clinical Implications
Religious attendance could have the same protective effect on depression occurrence as a university education or high perceived social support. The lack of association between self-identification as a spiritual person or self-rated spirituality and depression may result from the low reliability of the measures used. Alternatively, something extra in religious attendance over and above religious self-identity and spirituality accounts for the protective effect. Limitations The timing of a depressive episode is based on participant recall. The data collection cycle was done at 2-year intervals thereby allowing for the possibility of unobserved or unrecalled episodes.

individually centred activity. Many spirituality studies have been done in the context of end-of-life or palliative carea timely occasion for pondering existential purpose. People with spiritual beliefs have a more active coping style, enjoy life more, and are more hopeful." Interestingly, spirituality and religiosity have been shown to have discordant relations with depression. With acquired immune deficiency syndrome (AIDS) and cancer patients,^" more spiritual people were less depressed, while more religious ones were the opposite. Baetz et al* found that religious attendance was associated with less depression, while spirituality was associated with higher levels of depression. They discussed that certain types of depression may be associated with search for spiritual meaning or vice versa, or that the person that is depressed may be less likely to attend services. Similarly, Hayward et aF reported that private religious activity was associated with a higher occutTence of depression. The inconsistency of results in the literature can be partly explained by research design limitations. Two limitations of primary concern are the predominance of crosssectional studies and the focus on elderly or chronically ill populations. Cross-sectional studies cannot disentangle whether depression or religiosity came first. Conversely, the focus on elderly or ill populations sets a very high threshold before religion shows an effect. Additionally, medication data are generally not taken into account. These limitations can only be addressed by prospective studies in cornmunity-dwelling people who are representative of the general population. These too are beset by problems, such as attrition, as it is common to include only the subset of people interviewed or measured at multiple time points. Analytically, protective (or harmflil) effects of religiosity or spirituality can also be missed when timing is not taken into account. Without a tirne variable, all events (that is, depression) are treated equally, whether they occurred early or late in a given time span. This approach also neglects to relate the count of adverse events to the number of people exposed to risk at a given time.
wmi.LaRCP.ca

Abbreviations
HR MDE NPHS hazard ratio major depressive episode National Population Healtti Survey

226 '4' LaRevuecanadiennedepsychiatrie, vol58, no4, avril2013

Religious Attendance, Spirituality, and Major Depression in Canada: A14-Year Follow-up Study

Our study airned to analyze the effect of religious attendance, importance of spiritual values, and self-identification as a spiritual person on occurrence of major depression. Our objective is to contribute to the few studies with a prospective longitudinal design in a nationally representative sample. In our study, we used a survival model, in which the dependent variable is the triple (t^, ?,, d) where t^ and t^ define the span of time and c/is the observed outcome of either failure (that is, MDE) or censorship. In this way, the main effects of religious attendance and spirituality are assessed by how they modity risk for depression.

while the spirituality variables were measured in wave 2 (1996).

Covariates
We analyzed possible confounding variables, including age, sex, marital status, educational attainment, incotne adequacy, and family and personal history of depression. Income adequacy was categorized into low, middle, or high and was calculated by dividing household income by the number of people living in the household. Family and personal history of depression were entered as covariates because these are known risk factors for major depression. Perceived social support was included in the models because previous results'"" showed its association with religious attendance. Perceived social support was the summed score of 4 items (1 = yes; 0 = no) that reflected whether the respondent felt that they had somebody to confide in, they could count on, that could give them advice, and who made them feel loved. Although the psychometric properties of this scale have not been validated, it has been used in several published articles.'"^^

Method
Sample
The sample used in our study consists of longitudinal respondents from the Canadian NPHS. The NPHS is a biennial survey conducted by Statistics Canada starting in 1994 to collect information on health and socioeconomic characteristics of the national population. Among 17 276 rnernbers of the cohort, we excluded those who reported a MDE ( = 849), those who were younger than 16 years of age ( = 2747), and those with missing data ( = 3252) at baseline (1994), giving us a sample of 12 583. According to the practices of the NPHS, the longitudinal respondent gave verbal consent to be interviewed and was free to refuse in subsequent cycles. Interviews were conducted by trained interviewers using a computer-assisted interview system. In cycle 1, 75% of interviews were done in person, but starting at cycle 2, 95% were done by telephone.-^' Dependent Variable Our dichotomous depression status variable is derived from the Composite International Diagnostic Interview Short Form for Major Depression,-^ which defines MDE with a cut-off score of 0.9. This scale has been shown to distinguish 93% of major depression cases in the US National Comorbidity Survey.-^ In addition to depression status assessed at each of the 8 cycles, respondents were asked to report the specific month of the previous year that they got depressed, when applicable. Timing of depression was combined with depression status to serve as the response variable in survival analysis, as described previously.

Statistical A nalysis
These demographic and health variables were compared across the categories of religious attendance to establish baseline similarity or differences. As the NPHS is a complex survey, we weighted our estimates (means and proportions) with Statistics Canada-provided probability weights, which take into account selection probability and loss to follow-up. With the SVY prefix in Stata,^' 500 bootstrap replications were used to calculate 95% confidence intervals of estimates. As is typical of longitudinal studies, there is significant participant attrition over the follow-up period (that is, right censorship) so survival analysis was used to model risk of depression across religious attendance categories. With a slow, upward trend of depression over time, the Weibull proportional hazards regression technique was used with religious attendance as a predictor variable. Compared with Cox regression, which disregards the relation of risk with time, Weibull regression has an additional parameter (a), that allows for risk to increase, decrease, or stay constant with time. Our survival model allowed for repeated MDEs for the same person. Similar regression models were developed with the spirituality variables entered separately as predictors to avoid collinearity. Stata^''version 12.1 was used for all statistical procedures.

Independent Variables
We divided frequency of religious attendance into 3 categories: monthly or more frequently, less than rnonthly to yearly, and not at all. Two spirituality variables importance and self-identitywere responses to single questions: "Do spiritual values or your faith play an important role in your life?" and "How religious or spiritual are you?" The former is answered by yes or no, while the latter allows for a range of responses from very to not at all. These variables were both dichotomized for analysis purposes. We ran a regression model with a 4-category version of the spirituality variable as a predictor (that is, instead of a dichotomized variable) but this did not alter the result. Religious attendance was measured at baseline.
wmi.TheCJP.ca

Results
At baseline (1994), monthly attenders were more likely to be older, female, married, and members of organizations, as compared with occasional or nonattenders. Low-income earners were overrepresented in nonattenders. There are suggestive differences across religious attendance groupings related to rnental health. Monthly attenders had lower levels of distress, were less likely to have had previous episodes of depression, and a family history of depression.
TheCanadianJournaiofPsychiatry, Voi58, No4,Aprii2013 -^ 227

Original Research

Table 1 D e m o g r a p h i c characteristics

of 12 583 NPHS respondents at baseline (1994) by religious attendance

Mean or % (95% CI)

Characteristic Age, years

Monthly or more often (a) 48.1 (47.5 to 48. 7) 41.8 (41.2 to 42.3)

Occasionally (b) 40.6 (39.9 to 41.3) 48 (47.3 to 48.7)

Not at all (c) 41.2 (40.7 to 41.6) 53.3 (52.8 to 53.7)

Pairwise significant a>{b = c)

Sex, male

a< b<c

Marital status Married Single 70.4 (69.8 to 70.9) 15.8 (15.3 to 16.3) Previously married 13.8 (13.5 to 14.2) 16.9 (16.5 to 17.3) 61.9 (61.3 to 62.5) 26.2 (25.7 to 26.8) 11.9 (11.5 to 12.3) 17.1 (16.5 to 17.7) 60.6 (60.1 to 61.1) 26.2 (25.7 to 26.7) 13.2 (12.9 to 13.5) 18.5 (18.1 to 18.9) b<c<a a> b>c

a<(b = c)

Low income

{a = b)<c

Highest education attained <Secondary 28.4 (27.9 to 28.9) 15.2 (14.8 to 15.6) 24.4 (23.9 to 25.0) 31.9 (31.5 to 32.5) 3.76 (3.74 to 3.79) 47.8 (47.1 to 48.4%) 19.16 (18.5 to 19.8) 9.27 (8.9 to 9.7) 27.55 (26.9 to 28.2) 26.8 (26.2 to 27.4) 15.9 (15.4 to 16.4) 24.3 (23.6 to 24.9) 33.1 (32.5 to 33.7%) 3.79 (3.77 to 3.81) 26.2 (25.6 to 26.8%) 21.05 (20.3 to 21.8) 10.42 (9.9 to 11.0) 29.29 (28.6 to 30.0) 27.4 (26.9 to 27.9) 17 (16.6 to 17.4) 25.4 (24.9 to 25.9) 30.2 (29.7 to 30.7) 3.7 (3.68 to 3.73) 20.8 (20.4 to 21.3%) 21.96 (21.4 to 22.5) 12.76 (12.3 to 13.2) 28.89 (28.3 to 29.5) (a = b)>c a> {b = c) {a = b)<c

Secondary graduate

Some post-secondary

b <c a = b; a = c c <a< b

College or university graduate

Perceived social support index

Member of organizations

a> b > c

Ever had episodes of depression

a <{b = c)

Ever been diagnosed of depression

a <b <c

Biological parent or sibling diagnosed with depression

a<{b = c)

Estimates were calculated using sampling and bootstrap weights from Statistics Canada All figures are proportions except for age and perceived social support which are means.

compared with the other 2 groups. Perceived social support did not differ between monthly and occasional attenders but was lower in nonattenders. There were stark differences between groups in organizational metnbership. About onehalf of monthly attenders joined organizations, while the proportion was one-quarter and one-fifth for occasional and nonattenders, respectively (Table 1). Over the 14-year follow-up period, there were 2355 cases of depression, of which 1645 were new and 710 were recurrences (data not shown). Table 2 displays the yearby-year tally of people by depression or loss to followup. In the survival model, religious attendance remained
228 La Revue canadienne de psychiatrie, vol 58, no 4, avril 2013

significant after controlling for income adequacy, age, family and personal depression history, marital status, sex, education, and perceived social support. There was a 22% lower risk of depression for rnonthly attenders, compared with nonattenders (HR 0.78; 95% Cl 0.63 to 0.95). We did not find a protective effect for occasional attendance (HR 0.91; 95% CI 0.76 to 1.09). Monthly religious attendance had about the same magnitude of effect as college education and perceived social support (Table 3). There was a doseresponse relation in the protection afforded by religious attendance. Neither the importance of spiritual values (HR 0.88; 95% CI 0.72 to 1.07) nor identifying as a spiritual www.LaRCP.ca

Religious Attendance, Spirituality, and Major Depression in Canada: A 14-Year Follow-up Study

Table 2 Proportion of people by depression status Year 1994 1996 1998 2000 2002 2004 2006 2008 Not depressed, n 12 583 10 896 9965 8965 8140 7550 6873 6156 Depressed, n 0 366 378 400 335 350 282 244 Lost to follow-up, n 0 1321 2240 3218 4108 4683 5428 6183 Depressed, % 0.0 3.4 3.0 4.5 4.1 4.6 4.1 4.0

Figure 1 Survival curves by religious attendance

o-

d
_ >
(/}

o
00

o
in

50

100 Analysis time

Monthly or more Does not attend

150
- Occasionally

200

person (HR 1.14; 95% CI 0.85 to 1.51) was protective against depression. The rate of depression stratified by religious attendance is shown in Figure 1. The proportional hazards assumption was tested by stratifying the regression model according to each of the categorical covariates. We found no evidence that the proportional hazards assumption was violated.

Discussion
In our study, we examined the protective role of religious attendance and spirituality on major depression using a longitudinal survey representative of the Canadian population. Using a cohort of people without depression at baseline, we found cross-sectional and long-term effects of religious attendance. Significantly fewer monthly attenders reported ever having episodes or a diagnosis of depression (Table 1). This cross-sectional association suggests a protective effect of religions attendance. By selecting a cohort of people not depressed initially and following them over time, our longitudinal analysis
www.TheCJP.ca

establishes the precedence of religious or spiritual practice over depression, if it occurs at all. One previously published study" reported that disability does not lead to lower religious attendance, while another reported that women, but not men, are likely to stop attending services after onset of depression.^* While we cannot rule out the possibility that more frequent religious attendance indexes more favourable antecedent conditions (such as family upbringing or social status), the magnitude of differences at baseline does not seem to be clinically meaningful or tip the scales overwhelmingly in favour of the monthly attenders. For example, although fewer monthly attenders had a family history of depression, more of them were female, a known risk factor. In our multivariate survival model, we found that a previous MDE is the biggest risk factor for a subsequent depression, and this is consistent with published studies. The shape parameter of our Weibull model (1.68) confirms conventional thinking that risk of depression has an increasing trajectory with age.'*' In addition, our model clarified the inverse
The Canadian Journal of Psychiatry, Vol 58, No 4, April 2013 229

Original Research

Table 3 Weibull proportional hazards model of risk for MDE

Variable Age in 1994 younger than 16 years Sex, female Not married Income adequacy Highest education <Secondary Secondary graduate Some post-secondary College or university degree No history of depression in biological family Not previously depressed Perceived social support Frequency of religious attendance ^Monthly Occasionally Does not attend Alpha (shape parameter) 0.78 0.90 1.00 1,68
0.04 0.08 0.08 -2.45 -1.03 0.01 0.30 0.63 to 0.95 0.76 to 1.09

HR 0.98 1.46 0.96 0.92

SE=' 0.00 0.14 0.06 0.05

z
-5.78

P
<0.001

95% Cl" 0.97 to 0.99

4.13 -0.71 -1.66

<0.001 0.48 0.10

1.22 to 1.75 0.85 to 1.08 0.83 to 1.02

1.00 1.04 0.86 0.78 0.69 0.17 0.84

0.13 0.10 0.09 0.07

Reference category 0.28 -1.24 -2.27 -3.78 0.78 0.22 0.02 <0.001 0.81 to 1.33 0.68 to 1.09 0.63 to 0.97 0.57 to 0.84

0.01 0.06

-20.33 -2.61

<0.001 0.01

0.15 to 0.21 0.74 to 0.96

Reference category 19.41 <0.001 1.60 to 1.77

' SEs and 95% CIs were calculated using bootstrap weights provided by Statistics Canada

association of religious attendance and depression seen at baseline by isolating the effect of religious attendance over and above other factors. The protection afforded by religious attendance is a modest one, compared with other variables, and it is present only when attendance is at least monthly. We did not find a protective effect for the declared importance of spiritual values or identifying as a spiritual person. This is not entirely unexpected because spirituality has had an inconsistent relation with mental health.'-^ This is in accord with another study-' that found that religious attendance, but not subjective religiosity, retained its significant inverse relation with lifetime mental disorders after controlling for covariates. What mechanisms may account for the effect of religious attendance? Social support is a well-known pathway to improved mental healthparticipation in collective worship gives the person a sense of belonging or rootedness in a community.^" For uprooted people or disadvantaged groups, belonging to a community is a valuable social resource. Koenig et aP' reported that 19 of 20 studies found positive associations between religious involvement and social support. People who have few social ties or are in the periphery of the network have significantly worse depression." Hence it is possible that religion protects against depression by its social integration function. As Durkheim^ theorized, the beneficent influence of religion is its support for a collective life, such as by sustaining one another's beliefs. Consistent with the social support hypothesis of religion, the monthly attenders in our study
230 ^ La Revue canadienne de psychiatrie, vol 58, no 4, avril 2013

were more than twice as likely to be organization members than nonattenders. In contrast to the social support pathway, where the doctrine and content of religious practice are incidental, other theorists emphasize religion's ability to provide meaning and purpose in life.^" Thus far, the evidence for the relation between religion and depression is not conclusive.^*~^' Some ingredient of the religious experience other than behaviours, networks, or attitudes alone" probably contributes to the benefit. From the believers' perspective, they have recourse to divine assistance (even a personal relationship in Christian traditions) and thus are less likely to feel alone with the vicissitudes of life. Indeed, the inverse correlation between religious involvement and depression has been found to be stronger in stressed populations.'' Aside from a benevolent world view, church members are probably more altruistic to those in need,^** implying that the source of social support matters. Two studies^'*'"' reported health benefits from religious social support but not secular social support. Similarly, volunteering in a religious context was associated with better mental health than secular volunteering."' In our study, we controlled for perceived social support as a covariate and the protective effect of monthly attendance remained significant. Hence our result is more consistent with the view that religious attendance has an effect over and above social support. Other pathways by which religion could exert benefit is the cultivation of healthy habits and the reduction of risky behaviour.''- A recent meta-analytic review concluded that religious involvement among adolescents contributes to www.LaRCP.ca

Religious Attendance, Spirituality, and Major Depression in Canada: A 14-Year Follovi/-up Study

better self-appraisal, higher self-esteem, and less syrnptoms of depression.''^ This result is consistent with those conducted in elderly populations.''^''" We are not sure how to explain the different effects of religious attendance and spirituality variables. One possibility is that not all spiritual people avail themselves of the social network resources of religious people. Our result is consistent with a previously reported weak relation between private religious activity and depression."** Alternatively, without the benefit of a specific instrument to measure spirituality, the spiritual people were probably more heterogeneous than attenders. In contrast to religious attendance, which is a straightforward measure, spirituality is more ambiguous. Measures of spirituality tend to include indicators of psychological well-being, thus when spirituality is used to predict mental health outcomes, the reasoning becomes circular.*' Our study has several important limitations. Unlike in clinical trials, the NPHS, as a survey, does not match participant characteristics at the outset or assign them randomly to intervention groups. At baseline, there were already differences in age, sex, and other demographic variables, and statistically controlling for these differences is the best, if imperfect, remedy we have. Secondly, our depression timing variable depends on participant recall of episodes over a previous year and is not as-it-happens measured. Third, NPHS waves are collected every 2 years and therefore, there is interval censoring for depression status in the intervening years. The main strengths of our study are its longitudinal design, large nationally representative sample, and the use of a psychometrically validated depression measure that allowed us to calculate the specific month of its occurrence.

Conclusion
After statistically controlling for socioeconomic and biological covariates of depression, the effect of religious attendance on depression remained significant and lowered the risk of depression in a dose-response fashion.

Acknowledgements
The authors thank Father Ron Griffin, PhD, for his help in conceptualizing the project, and 2 anonymous reviewers. The NPHS dataset is from Statistics Canada. However, the analysis and views herein are entirely those ofthe authors. The authors did not receive funding for this study.

References
1. James W. The varieties of religious experience: a study in human nature. New York (NY): Mentor; 1958. 2. Durkheim E. The elementary forms ofthe religious life, a study in religious sociology. New York (NY): Collier; 1961. 3. FrankI VE. Man's search for meaning; an introduction to logotherapy. Boston (MA): Beaeon Press; 1963. 4. Koenig HG. Research on religion, spirituality, and mental health: a review. Can J Psychiatry. 2009;54(5):283-291. 5. Baetz M, Bowen R, Jones G, et al. How spiritual values and worship attendance relate to psychiatric disorders in the Canadian population. Can J Psychiatry. 2006;51(10):654-661.
www.rheCJP.ca

6. Baetz M, Griffin R, Bowen R, et al. The association between spiritual and religious involvement and depressive symptoms in a Canadian population. J Nerv Ment Dis. 2004;192(12):818. 7. Koenig HG, Cohen HJ, Blazer DG, et al. Religious coping and depression among elderly, hospitalized medically ill men. Am J Psychiatry. 1992; 149( 12); 1693-1700. 8. Hayward RD, Owen AD, Koenig HG, et al. Religion and the presence and severity of depression in older adults. Am J Geriatr Psychiatry. 2012;20(2):188-192. 9. Maltby J, Day L. Depressive symptoms and religious orientation: examining the relationship between religiosity and depression within the context of other correlates of depression. Pers Indiv Diff. 2000;28(2):383-393. 10. Maseiko J, Gilman SE, Buka S. Religious service attendance and spiritual well-being are differentially associated with risk of major depression. Psychol Med. 2009;39(6):1009-1017. 11. Sorenson AM, Grindstaff CF, Turner RJ. Religious involvement among unmarried adolescent mothersa source of emotional support. Sociol Relig. 1995;56(1):71-81. 12. Murphy PE, Ciarrocehi JW, Piedmont RL, et al. The relation of religious belief and practices, depression, and hopelessness in persons with clinical depression. J Consult Clin Psychol. 2000;68(6):1102-1106. 13. Rasic D, Robinson JA, Bolton J, et al. Longitudinal relationships of religious worship attendance and spirituality with major depression, anxiety disorders, and suicidal ideation and attempts: findings trom the Baltimore Epidemiologie Catchment Area Study. J Psychiatr Res. 2011;45(6):848-854. 14. Ellison CG, Flannelly KJ. Religious involvement and risk of major depression in a prospective nationwide study of African American adults. J Nerv Ment Dis. 2009;197(8):568-573. 15. Miller L, Wickramaratne P, Gameroff MJ, et al. Religiosity and major depression in adults at high risk: a ten-year prospective study. Am J Psychiatry. 20t2;169(l):89-94. 16. Braam AW, Hein E, Deeg DJH, et al. Religious involvement and 6-year course of depressive symptoms in older Dutch citizens: results from the longitudinal aging study Amsterdam. J Aging Health. 2004;16(4):467^89. 17. Musick MA, Koenig HG, Hays JC, et al. Religious activity and depression among community-dwelling elderly persons with cancer: the moderating etect of race. J Gerontol B Psychol Sei Soc Sei. 1998;53(4):S218-S227. 18. SunF, ParkN, Roff L, et al. Predicting the trajectories of depressive symptoms among southern, community-dwelling older adults: the roles of religiousness. Gerontologist. 2010;50:105-106. 19. Weaver AJ, Flannelly KJ. The role of religion/spirituality for cancer patients and their caregivers. South Med J. 2004;97(12): 1210-1214. 20. Nelson CJ, Rosenfeld B, Breitbart W, et al. Spirituality, religion, and depression in the terminally ill. Psychosomatics. 2002;43(3):213-220. 21. Statistics Canada. National Population Health Survey household component cycles 1 to 8 (1994/1995 to 2008/2009) longitudinal documentation [Internet]. Ottawa (ON): Statistics Canada; 2010 [cited 2012 May 30]. Available from: http://www23.statcan.gc.ca/ imdb/p2SV.pl?Function=getSurvey&SDDS=3225&lang=en&db=i mdb&adm=8&dis=2. 22. Kessler RC, Mroezek D. Scoring ofthe UM-CIDI short tbrms. Ann Arbor (MI): University of Michigan Institute for Social Research/ Survey Research Center; 1994. 23. Kessler RC, Andrews G, Mroezek D, et al. The World Health Organization Composite Intemational Diagnostic Interview Short Form (CIDI-SF). Int J Methods Psychiatr Res. 1998;7(4);171-185. 24. Koenig HG, King DE, Carson VB. Handbook of religion and health. 2nd ed. New York (NY): Oxford University Press; 2011. 25. Kaplan MS, Berthelot JM, Feeny D, et al. The predictive validity of health-related quality of life measures: mortality in a longitudinal population-based study. Qual Life Res. 2007; 16(9): 1539-1546. 26. StataCorp LP. Stata version 12.1. College Station (TX): StataCorp LP;2011.
The Canadian Journal of Psychiatry, Vol 58, No 4, April 2013 231

Original Research

27. Idler EL, Kasl SV. Religion among disabled and nondisabled persons II: attendance at religious services as a predictor of the course of disability. J Gerontol B Psychol Sei Soc Sei. 1997;52(6):S306-S316. 28. Maselko J, Hayward RD, Hanlon A, et al. Religious service attendance and major depression: a case of reverse causality? Am J Epidemiol. 2012;175(6):576-583. 29. Chatters LM, Bullard KM, Taylor RJ, et al. Religious participation and DSM-IV disorders among older African Americans: findings from the National Survey of American Life. Am J Geriatr Psychiatry. 2008;16(12):957-965. 30. Shreve-Neiger AK, Edelstein BA. Religion and anxiety: a critical review of the literature. Clin Psychol Rev. 2004;24(4):379-397. 31. Koenig HG, Larson DB, Larson SS. Religion and coping with serious medical illness. Ann Pharmacother. 2001;35(3):352-359. 32. Rosenquist IN, Fowler JH, Christakis NA. Social network determinants of depression. Mol Psychiatry. 2011;16(3):273-281. 33. Jung CG, Dell WS. Modem man in search of a soul. Dell WS, Baynes CF, translators. New York (NY): Harcourt, Brace; 1950. 34. Koenig HG, Hays JC, George LK, et al. Modeling the crosssectional relationships between religion, physical health, social support, and depressive symptoms. Am J Geriat Psychiatry. 1997;5(2):131-144. 35. George LK, Ellison CG, Larson DB. Explaining the relationships between religious involvement and health. Psychol Inq. 2002;13(3):190-200. 36. Bienenfeld D, Koenig HG, Larson DB, et al. Psychosocial predictors of mental health in a population of elderly womentest of an explanatory model. Am J Geriat Psychiatry. 1997;5(l):43-53.

37. Idler EL, Kasl SV. Religion, disability, depression, and the timing of death. Am J Sociol. 1992;97(4): 1052-1079. 38. Ellison CG, Levin JS. The religion-health connection: evidence, theory, and future directions. Health Educ Behav. 1998;25(6):700-720. 39. Kark JD, Shemi G, Friedlander Y, et al. Does religious observance promote health? Mortality in secular vs religious kibbutzim in Israel. Am J Public Health. 1996;86(3):341-346. 40. Krause N. Exploring the stress-buffering effects of church-based and secular social support on self-rated health in late life. J Gerontol B Psychol Sei Soc Sei. 2006;61(l):S35-S43. 41. Musick MA, Wilson J. Volunteering and depression: the role of psychological and social resources in different age groups. Soc Sei Med. 2003;56{2):259-269. 42. Steinberg L. A social neuroscience perspective on adolescent risktaking. Dev Rev. 2008;8(l):78-106. 43. Yonker JE, Schnabelrauch CA, Dehaan LG. The relationship between spirituality and religiosity on psychological outcomes in adolescents and emerging adults: a meta-analytic review. J Adolesc. 2011;35:299-3I4. 44. Chatters LM. Religion and health: public health research and practice. Annu Rev Public Health. 2000;21:335-367. 45. Krause N. Religiosity and self-esteem among older adults. J Gerontol B Psychol Sei Soc Sei. 1995;50B(5):P236-P246. 46. Koenig HG, George LK, Peterson BL. Religiosity and remission of depression In medically ill older patients. Am J Psychiatry. 1998;155(4):536-542. 47. Koenig HG. Spirituality & health research. West Conshohocken (PA): Templeton Press; 20 U.

232 4* LaRevuecanadiennedepsychiatrie, vol 58, no4, avrii2013

www./.aRCP.ca

Copyright of Canadian Journal of Psychiatry is the property of Canadian Psychiatric Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.