The Role of Religion in Suicidal Behavior,Attitudes and Psychological Distress Among University Students: A Multinational Study

Abstract

The purpose of this study was to determine the associations between religion, suicidal behavior, attitudes and psychological distress in 5572 students from 12 countries by means of a self-report questionnaire. Our results showed that an affiliation with Islam was associated with reduced risk for suicide ideation, however affiliating with Orthodox Christianity and no religion was related to increased risk for suicide ideation. While affiliating with Buddhism, Catholic religion and no religion was associated with lowered risk for attempting suicide, affiliation with Islam was related to heightened risk for attempting suicide. Affiliation with Hinduism, Orthodox Christianity, Catholicism, other religions and with no religion was associated with decreased risk for psychological distress but those reported affiliating with Islam evinced greater risk for psychological distress. The associations of the strength of religious belief to suicidal ideation and attempts were in the expected direction for most but had a positive relation in respondents affiliating with Catholicism and other religions. Students reporting affiliation with Islam, the Christian Orthodox religion and Buddhism were the least accepting of suicide but they displayed a more confronting interpersonal style to an imagined peer with a suicidal decision. It was concluded that the protective function of religion in educated segments of populations (university students) and in university students residing in Muslim countries where freedom from religion is restricted or religion is normative and/or compulsory is likely to be limited. Our findings suggest that public policies supporting religious freedom may augment the protective function of religion against suicide and psychological distress.

Keywords

religion suicide psychological distress attitudes cross-cultural

Introduction

A well-established regularity in suicidology is that the rates of suicidal behavior show large intersocietal variation (Värnik, 2012). The main research variable in this study, religion, has also usually been assumed to be related to the intersocietal variation of suicidal behaviors (Bertolote, & Fleischmann, 2015). For instance, low suicide rates in persons affiliating with Islam and in nations with predominantly Muslim populations are attributed either to the explicit condemnation of suicide by Islamic scriptures (Lester, 2006; Shah, & Chandia, 2010) or to underreporting (Pritchard, & Amanullah, 2007). All religions have explicit or implicit sanctions against self-killing with varying degrees of severity (Nelson, Hanna, Houri, & Klimes-Dougan, 2012).

Religion is a complex but unified system of beliefs, practices and institutions about the sacred (Durkheim, 1976). From an evolutionary perspective it has been argued that although religion has some fitness costs (e.g., costly ritual behavior that appears to be maladaptive) it nevertheless confers some indirect fitness benefits by harnessing threat-detection cognitive systems and increasing cohesion at the group level (Boyer, & Bergstrom, 2008; Boyer, & Liénard, 2006). According to Purzycki, Haque, and Sosis (2014), religious systems help increase the adaptive capacities of believers in a constant social, physical, psychological and existential flux by providing explanations for a complex and mysterious universe, to cope with death and other forms of psychological suffering and to sustain social order. As self-reflecting agents, human beings are aware of their own inevitable death and this is a source of existential anxiety. A study by Vail, Arndt, and Abdollah, (2012) provides empirical support to the view that religious beliefs help people cope with this existential anxiety. The authors, in three studies, have demonstrated that making participants' own mortality salient enhanced their level of religiosity, belief in a higher power, and belief in a God congruent with their belief system.

Suicide is an extreme form of desperation. How does religion protect its believers against such desperation? Colucci and Martin (2008) and Stack (2000) have identified three theoretical formulations about the link between religion and suicide. The first one is social integration theory (Durkheim, 1897/1951), and the second one is network theory (Pescosolido, 1990). Both theories argue that religion exercises its protective function against suicide through developing informal relationships and promoting community ties that increase networking and feelings of integration. The third one is religious commitment theory (Stark, Doyle, & Rushing, 1983; Stack, 2000). It assumes that core religious beliefs opposing suicide prevent individuals from considering suicide as an option during difficult times.

The social integration, network and religious commitment theories outlined above suggest that religion is associated with better mental health. The scientific literature indicates that religion may be a protective factor against diminished psychological well-being (Bonelli & Koenig, 2013; Kang & Romo, 2011). Findings from a recent longitudinal study by Zou, Huang, Maldonado, Kasen, Cohen and Chen, (2014), for example, showed that religious service attendance reduces depressive symptoms significantly. Note that reduced mental health is one of the strongest risk factors for suicidal behavior. Religion is also assumed to be a deterrent against suicidal behavior (Dervic, Oquendo, Grunebaum, Ellis, Burke, & Mann, 2004; Wu, Wang, & Jia, 2015). Further, acceptability of suicide is lower in religious than in nonreligious people (Stack & Kposowa, 2011), and higher levels of suicide acceptability was found to be associated with contemplating, planning and attempting suicide (Joe, Romer, & Jamieson, 2007). However, some research including a recent review by Lawrence, Oquendo, and Stanley (2015), present a complex picture of the relation between religion and suicidal behavior. Lawrence et al. (2016) concluded that religious affiliation was protective against suicidal attempts and suicide but not against suicidal ideation.

There are two major methodological flaws of the scientific endeavor investigating the link between religion and psychological distress and/or suicidal behavior. First, most of the previous empirical work on the issue is carried out with samples from industrialized Western nations. Second, except for the work by Sisask et al. (2010) most studies are of ecological nature which includes other confounding variables such as levels of economic development, education, secularization, etc. Bearing these methodological considerations in mind, we tested whether religious affiliation and the strength of one's religious belief were related to nonfatal suicidal behavior, psychological distress, and attitudes towards suicide and suicidal individuals. We aimed to test our hypothesis further by asking whether this relationship differed by religion. We sought answers to these questions in university students from 12 nations. From a methodological point of view, in order to make cross-cultural comparisons the samples should be comparable to each other (Boehnke, Lietz, Schreier, & Wilhelm, 2011). University students are similar in several important aspects such as age, level of education, and level of intellectual capacity. As a result, we chose to study college students to make samples from different countries as comparable as possible.

Methods

Participants

A total of 5572 (55.3% women) university students (mean age = 22.1, SD = 3.5 years) from 12 countries volunteered to participate in the study. Samples were from the countries that were identified by Inglehart and Baker (2000) as belonging to four culture zones (1) Confucian (China, population = 1,355,692,576, sample n = 627 and Japan, population = 127,103,388, sample n = 246), (2) Islamic (Iran, population =80,840,713, sample n = 1000; Jordan, population = 7,930,491, sample n = 436; Palestine (West Bank), population = 2,731,052, sample n = 358; Saudi Arabia, population = 27,345,986, sample n = 413; Turkey, population = 81,619,392, sample n = 497 and Tunisia, population = 10,937,521, sample n = 484), (3) English speaking (UK, population = 63,742,977, sample n = 150 and USA, population =318,892,103, n = 239) and (4) Catholic (Austria, population = 8,223,062, sample n = 627 and Italy, population = 61,680,122, sample n = 471). Participants were recruited from one public university in each country and from two public universities in Jordan and Palestine.

The gender and age distributions differed significantly among countries (gender, χ2 = 294.56, df = 11, p < 0.001; age, F(11, 5407) = 105.61, p < .001). There were more female participants than male participants. However, in the Japanese and Saudi Arabian subsamples, there were more men than women. The US sample was the youngest and the UK sample was the oldest. The number of siblings of students differed significantly between samples (F(11, 5560) = 366.45, p < .001). Jordanian and Palestinian samples had the highest number of siblings and the Chinese had the lowest number of siblings.

Out of the larger data set, so far we have published two papers on the prevalence of suicidal behavior and psychological distress (Eskin et al., 2016a) and attitudes towards suicide and suicidal persons (Eskin et al., 2016b) in which we describe sociodemographic characteristics of participants and the administration of research processes in more detail. In the first paper we found that 29% of the whole group reported having considered killing themselves, 7% reported having attempted to kill themselves and more than 1/3 of the sample scored above the conservative GHQ-12 ≥ 5 cut-off point. Significantly more participants from Austria and the UK reported suicidal ideation while significantly fewer students from China, Italy, Saudi Arabia, Tunisia and Turkey reported suicidal ideation. Significantly more students from Jordan, Palestine, Saudi Arabia, and to some extent from Turkey reported attempting suicide, but significantly fewer participants from Austria, China, Italy, Japan and the US did so. In the second paper we found that the Austrian, the UK, the Japanese and the Saudi Arabian samples were the most accepting of suicide and the Tunisian, the Turkish, the Iranian, and the Palestinian samples were the least accepting. The Turkish, the US, the Italian, and the Tunisian samples were the most socially accepting of a suicidal friend and the Japanese, the Saudi Arabian, the Palestinian and the Jordanian samples were the least accepting of such a friend. Acceptance of suicide was inversely related to social acceptance for a suicidal friend. We also found in the second paper that individuals were more likely to report more suicidal behavior and psychological distress, if they had accepting attitudes towards suicide but rejecting attitudes towards suicidal persons. Further, participants with an accepting attitude to suicide but rejecting attitude to suicidal persons were more likely to be from countries with high suicide rates and significantly high suicidal behavior reporting samples.

Instrument

A self-administered questionnaire was utilized to collect data. The questionnaire included questions about nonfatal suicidal behavior, religious affiliation and strength of religious belief, attitudes towards suicide and suicidal individuals, individualistic-collectivistic value orientations and a measure of psychological distress. In this paper, the associations of religious affiliation and strength of religious belief to nonfatal suicidal behavior, attitudes and psychological distress are reported. (In order to keep a clear focus, the associations of individualism-collectivism values to suicidal behavior, psychological distress and suicidal attitudes will be reported in a separate paper).

Demographics

In a short demographics questionnaire that was prepared by the lead researcher, students were asked to indicate their gender, age, and the number of siblings they have.

Religious affiliation

One question (What is your religion?) asked about the participants' religious affiliation based on 15 choices (Buddhism, Hinduism, Roman Catholicism, Protestantism, Islam, Christian Orthodox, Judaism, Shintoism, African Religion, I have no religion, Shamanism, Sikhism, Taoism, Mormonism and Other than these). Since few participants in our sample reported having affiliated with African religion (n = 22), Shintoism (n = 20), Taoism (n = 13), Judaism (n = 10), Shamanism (n = 4), Mormonism (n = 4) and Sikhism (n = 3), these were merged into the “other religion” category. Based on historical claims (For instance Durkheim's) about the relationship between Protestantism and suicide we divided Christianity into three categories (Catholic, Orthodox and Protestant).

Strength of religious belief

Another question (What is the strength of your religious belief?) asked about the strength of participants' religious belief on a 7-point Likert scale ranging from “None whatsoever (0)” to “Very strong (6)”.

Suicidal behavior

There were five questions about past and current suicidal behavior (Yes = 1, No = 0):

(1) Have you ever thought of killing yourself? (2) Have you, during the past 12 months, thought of killing yourself? (3) Do you have thoughts of killing yourself right now? (4) Have you ever made an attempt to kill yourself? (5) Have you, during the past 12 months, made an attempt to kill yourself?

Participants who responded affirmatively to all or at least one of the first three questions were dichotomized as having suicidal ideation and participants responding affirmatively to both or one of the questions 4 and 5 were dichotomized as having attempted suicide.

Psychological Distress

The 12-item General Health Questionnaire (GHQ-12; Goldberg & Williams, 1988) is the most widely used screening tool for assessing psychological distress. The reliability and validity of the GHQ-12 are well established (Goldberg, Gater, Sartorius, Ustun, Piccinelli, Gureje, & Rutter, 1997). The standard method of 0-0-1-1 scoring was used in the study. In this method, a score of 0 was assigned to the first two low stress alternatives and a score of 1 was given to the two high stress alternatives. The method results in individual scores that range from 0 to 12. The internal consistency reliability of the GHQ-12 in this study was 0.87 with item-total correlations ranging from 0.45 to 0.62. Goldberg et al., (1997) suggested varying GHQ-12 cut-off points from a low of 2 to a high of 4 in 15 centers. Therefore, we used three cut-off (GHQ-12 ≥ 3, 4, 5) points in this study.

Attitudes towards suicide

The 24-item Eskin's Attitudes towards Suicide Scale (E-ATSS, Eskin, 1999a, 2004, 2013; Eskin et al., 2016b) with 5-point Likert type response options ranging from “Completely disagree (1)” to “Completely agree (5)” was used to measure students' attitudes towards suicide. A principle component analysis with varimax rotation extracted six factors: 1. Acceptability of suicide (α = 0.91); 2. Punishment after death (α = 0.93); 3. Suicide as a sign of mental illness (α = 0.94); 4. Communicating psychological problems (α = 0.79); 5. Hiding suicidal behavior (α = 0.82); 6. Open reporting and discussion of suicide (α = 0.62) that explained 73.10% of the total variance.

Attitudes towards suicidal persons

Eskin's Social Reactions to Suicidal Persons Scale (E-SRSPS) begins with a short description of “an imagined suicidal close friend” who decides to kill him/herself and share it with the participant. By means of 20 possible reactions to this friend, students were asked how they would react or feel on 5-point Likert scales ranging from “Completely disagree (1)” to “Completely agree (5)” (Eskin, 1999a, 2004, 2013; Eskin et al., 2016b). A principle component analysis with varimax rotation extracted four factors: 1. Social acceptance (α = 0.90); 2. Helping (α = 0.83); 3. Disapproval of suicidal disclosure (α = 0.77); 4. Emotional involvement (α = 0.63) that explained 60.73% of the total variance.

Procedure

First, a battery of surveys was prepared by the principle investigator (M. Eskin). Then, the other researchers were invited via email to join the study. All researchers worked for a university and collected their portion of the data at their institution, but in Jordan and Palestine, data were also collected at a second university. On the first page of the questionnaire packet, students were told that the study was anonymous. Contact information of the investigator collecting data was provided on the first page of the survey for participants to get in touch and ask possible questions they may have had.

All researchers were requested to only undertake data collection after having received approval from relevant institutional review boards for research ethics. With the exception of Austria where such an approval was formally not necessary according to the relevant legal requirements and regulations, ethical approval was obtained in all study sites. In the United Kingdom, the IRB approved the study first but later the data collection was stopped prematurely by the IRB due to one member's concerns over the possible distress the suicide-related questions could cause. Almost all of the researchers in this study reported having no legal sanctions against suicidal behavior in their respective country. However, this was not the case for Jordan. According to the Jordanian Penal Code, “The person who attempts suicide will be punished by imprisonment from three months to two years.”

Statistical Analyses

First, the eight religion categories were coded as dummy variables and then, their associations with suicidal behavior and psychological distress were tested by calculating the odds ratios (ORs) by using the binary logistic regression analysis procedure. In calculating the ORs, age, sex and the number of siblings were taken as covariates and hence their effects were controlled for. In the first series of ORs, the control or reference group included participants who reported not affiliating with a religion. In the second series of ORs reported in Table 1, the control or reference group included participants who reported affiliating with any religion and those not affiliating with a religion. Point-biserial correlation coefficients were calculated between strength of religious belief and suicidal behavior, and psychological distress. Two one-way Multivariate Analyses of Covariances (MANCOVAs) were performed to explore the relation between religious affiliation (IV) and suicidal attitudes (DVs) by entering the age and number of siblings as covariates. Scheffé procedure was employed to do multiple comparisons. Finally, Pearson product-moment correlation coefficients were calculated between the strength of religious belief and suicidal attitudes.

Table 1.

Odds Ratios (OR) between religious affiliation and suicidality and scores above the three cut-off points of GHQ-12.

Religious Affiliation	Suicide Ideation		Suicide Attempt		GHQ-12 ≥ 3		GHQ-12 ≥ 4		GHQ-12 ≥ 5
Religious Affiliation	OR	95% CI	OR	95%CI	OR	95%CI	OR	95% CI	OR	95%CI
Buddhism	1.16	0.84–1.62	0.35^a	0.13–0.96	0.1.13	0.84–1.51	1.16	0.86–1.56	1.19	0.87–1.62
Hinduism	1.50	0.92–2.43	0.87	0.32–2.42	0.46^b	0.28–0.76	0.61	0.36–1.02	0.64	0.37–1.12
Islam	0.73^b	0.63–0.84	2.11^b	1.60–2.78	1.83^b	1.60–2.08	1.82^b	1.60–2.09	1.79^b	1.55–2.06
Christian Orthodox	1.65^a	1.11–2.43	0.57	0.21–1.57	0.74	0.49–1.12	0.57^a	0.36–0.90	0.49^b	0.29–0.83
Christian Protestant	1.26	0.82–1.92	0.32	0.08–1.30	0.70	0.45–1.07	0.69	0.44–1.09	0.73	0.45–1.19
Christian Catholic	1.04	0.85–1.28	0.44^b	0.26–0.75	0.74^b	0.61–0.89	0.64^b	0.52–0.78	0.64^b	0.51–0.80
Other Religion	1.18	0.87–1.59	1.27	0.74–2.19	0.70^a	0.52–0.95	0.77	0.57–1.08	0.80	0.58–1.12
Have no religion	1.22^b	1.04–1.42	0.69^a	0.49–0.96	0.68^b	0.59–0.79	0.71^b	0.61–0.83	0.70^b	0.59–0.82

Associated Wald Statistics are significant at: ^aP < 0.05; ^bP < 0.01

Results

Religious affiliation and belief

The distribution of religious affiliation by country showed that the majority of participants in each country reported affiliating with the dominant religious tradition, with the exception of China, The UK and The USA where the majority reported affiliating with no religion. The majority affiliations by country were as follows: Austria (n = 254, affiliation with Roman Catholicism = 41,5%), China (n = 471, affiliation with no religion = 80.2%), Iran (n = 1000, affiliation with Islam = 100%), Italy (n = 226, affiliation with Roman Catholicism = 49.1%), Japan (n = 105, affiliation with Buddhism = 46.7%), Jordan (n = 414, affiliation with Islam = 98.8%), Palestine (n = 347, affiliation with Islam = 97.2%), Saudi Arabia (n = 403, affiliation with Islam = 97.6%), Tunisia (n = 470, affiliation with Islam = 97.3%), Turkey (n = 450, affiliation with Islam = 93.0%), UK (n = 69, affiliation with no religion = 46.0%), USA (n = 69, affiliation with no religion = 28.9%).

The mean (SD) strength of religious belief ratings on a 7-point scale by country were as follows: Austria 1.7 (1.7), China 1.2 (1.3), Iran 4.2 (1.5), Italy 2.4 (2.1), Japan 1.7 (1.5), Jordan 4.7 (1.1), Palestine 4.7 (1.3), Saudi Arabia 4.7 (1.4), Tunisia 4.6 (1.3), Turkey 4.1 (1.6), UK 2.4 (2.3), and USA 3.1(1.8).

The mean (SD) strength of religious belief ratings by affiliated religion were as follows: Buddhism 2.51 (1.51), Hinduism 3.21 (1.82), Islam 4.48 (1.40), Orthodox 3.02 (1.82), Protestant 2.71 (1.95), Catholic 2.73 (1.85), and other religion 2.88 (2.13).

Religious affiliation, suicidal behavior, and psychological distress

The odds ratio (95% CI) of participants affiliating with any religion having thought of suicide relative to those with no religious affiliation was 0.82 (0.70 – 0.96), Wald = 6.30, df = 1, p < .05. The odds ratio (95% CI) of participants affiliating with a religion having attempted suicide relative to those with no religious affiliation was 1.45 (1.04 – 2.03), Wald = 4.91, df = 1, p < .05. The odds ratios (95% CI) of participants affiliating with a religion having scored above the three cut-off points of GHQ-12 relative to those with no religious affiliation were at GHQ-12 ≥ 3, 1.46 (1.26 – 1.70), Wald = 25.23, df = 1, p < .01, at GHQ-12 ≥ 4, 1.41 (1.21 – 1.65), Wald = 19.27, df = 1, p < .01, and at GHQ-12 ≥ 5, 1.43 (1.21 – 1.69),Wald = 18.16, df = 1, p < .01.

The associations of religious affiliation to suicidal behavior and psychological distress are shown in Table 1. The table shows that while participants who affiliated with Orthodox religion and those unaffiliated with a religion were significantly more likely to report suicidal ideation; participants who affiliated with Islam were less likely to report suicidal ideation. Participants who affiliated with no religion, Buddhism and Catholicism were significantly less likely to report attempting suicide but those affiliating with Islam were significantly more likely to report attempting suicide. Participants affiliating with Hinduism, Catholicism, other religions and those having been unaffiliated with a religion were significantly less likely to score above the cut-off point of GHQ-12 ≥ 3 but participants affiliating with Islam were significantly more likely to score above the three cut-off points of GHQ-12.

Religious belief, suicidal behavior, and psychological distress

Table 2 displays the point-biserial correlation coefficients between suicidal behavior and psychological distress and strength of religious belief according to religious affiliation. The strength of religious belief correlates inversely significantly with suicidal ideation in participants affiliating with Hinduism, Islam, Orthodox Christianity, and Catholic religions. A negative correlation was observed between strength of religious belief and suicide attempts in students affiliating with Islam but there were positive correlations between religious belief and suicide attempts in participants affiliating with Catholicism and those affiliating with other religions. The correlation coefficients between religious belief and scoring above the three cut-off points of GHQ-12 were negative and reached significance only in participants affiliating with Islam.

Table 2.

Point-biserial correlation coefficients between strength of religious belief and suicidality and scores above the three cut-off points of GHQ-12 by religious affiliation.

Religious Affiliation	Suicide Ideation		Suicide Attempt		GHQ-12 ≥ 3		GHQ-12 ≥ 4		GHQ-12 ≥ 5
Religious Affiliation	r	df	r	df	r	df	r	df	r	df
Total sample	− 0.14^b	4042	− 0.01	4041	− 0.04^a	4249	0.02	4249	0.02	4249
Buddhism	0.03	189	– 0.07	189	– 0.05	187	– 0.09	189	– 0.06	187
Hinduism	– 0.26^a	71	– 0.03	71	– 0.07	71	– 0.09	71	– 0.01	71
Islam	– 0.17^b	2863	– 0.10^b	2862	– 0.04^a	3128	– 0.06^b	3128	– 0.06^b	3128
Christian Orthodox	– 0.21^a	109	– 0.06	109	0.01	96	0.09	96	0.05	96
Christian Protestant	– 0.13	97	0.02	89	– 0.02	89	0.01	89	0.01	89
Christian Catholic	– 0.10^a	511	0.10^a	511	0.01	490	– 0.02	490	– 0.03	490
Other Religion	– 0.01	202	0.17^a	202	– 0.02	188	– 0.04	188	– 0.01	188

P values are significant at: ^aP < 0.05; ^bP < 0.01, one-tailed.

Religious affiliation and suicidal attitudes

The means and standard deviations for E-ATSS and E-ATSPS factor scores are presented in Table 3. The one-way MANCOVA on attitudes towards suicide factors produced a significant main effect for religious affiliation, F_(7,5419) = 74.30, p < .0001. Univariate F tests showed religious affiliation differences on all factors: acceptability of suicide (F = 95.29, p < .0001), punishment after death (F = 553.61, p < .0001), suicide as a sign of mental illness (F = 60.65, p < .0001), communicating psychological problems (F = 12.28, p < 0.0001), hiding suicidal behavior (F = 59.59, p < .0001), and open reporting and discussion of suicide (F = 22.87, p < .0001).

Table 3.

Means and standard deviations of the attitudes towards suicidality factors by participants' religious affiliation

Attitude factors	Buddhism		Hinduism		Islam		No religion		Orthodox		Protestantism		R. Catholic		Other religion
Attitude factors	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD	M	SD
Attitudes to suicide
Acceptability of suicide	2.07	0.80	2.11	0.98	1.62	0.81	2.35	0.91	1.78	0.76	2.37	0.90	2.14	0.85	2.12	0.92
Punishment after death	2.31	0.77	2.33	0.96	3.86	0.97	1.84	0.79	2.51	1.00	2.19	0.90	2.04	0.86	2.31	1.00
Suicide as sign of mental illness	2.67	0.97	2.82	1.11	3.43	1.25	2.45	1.01	2.96	1.13	2.76	1.00	2.78	1.10	2.77	1.06
Communicating problems	3.73	0.66	4.01	0.70	3.73	0.90	3.66	0.83	4.12	0.74	3.85	0.65	3.99	0.70	3.82	0.85
Hiding suicidal behavior	2.98	0.96	2.22	0.90	2.90	1.10	2.37	0.98	2.11	0.99	2.06	0.91	1.95	0.98	2.52	1.14
Open reporting & discussion	2.70	0.85	3.18	0.94	3.06	1.06	2.83	1.07	3.18	1.07	3.22	0.90	3.47	0.84	2.85	1.05
Attitudes to suicidal persons
Social acceptance	3.64	0.75	3.96	0.86	3.89	0.83	3.88	0.66	4.13	0.66	3.88	0.60	3.97	0.57	3.86	0.76
Helping	3.94	0.57	4.12	0.64	4.09	0.78	4.05	0.60	4.21	0.57	4.18	0.51	4.29	0.55	3.96	0.68
Disclosure disapproval	2.91	0.72	2.59	0.95	2.79	0.83	2.58	0.76	2.36	0.74	2.46	0.64	2.37	0.71	2.68	0.73
Emotional involvement	3.05	0.89	3.00	0.97	3.46	0.89	3.10	0.90	3.16	0.95	2.63	0.83	2.99	0.94	3.03	0.97

E-ATSS factor scores from the lowest to the highest:

Acceptability of suicide: Islam, Orthodox, Buddhism, Hinduism, Other religion, Catholicism, No religion, Protestantism

Punishment after death: No religion, Catholicism, Protestantism, Buddhism, Other religion, Hinduism, Orthodox, Islam

Suicide as sign of mental illness: No religion, Buddhism, Protestantism, Other religion, Catholicism, Hinduism, Orthodox, Islam

Communicating problems: No religion, Buddhism, Islam, Other religion, Protestantism, Catholicism, Hinduism, Orthodox

Hiding suicidal behavior: Catholicism, Protestantism, Orthodox, Hinduism, No religion, Other religion, Islam, Buddhism

Open reporting & discussion: Buddhism, No religion, Other religion, Islam, Hinduism, Orthodox, Protestantism, Catholicism

E-SRSPS factor scores from the lowest to the highest:

Social acceptance: Buddhism, Other religion, No religion, Protestantism, Islam, Hinduism, Catholicism, Orthodox

Helping: Buddhism, Other religion, No religion, Islam, Hinduism, Protestantism, Orthodox, Catholicism

Disclosure disapproval: Orthodox, Catholicism, Protestantism, No religion, Hinduism, Other religion, Islam, Buddhism

Emotional involvement: Protestantism, Catholicism, Hinduism, Other religion, Buddhism, No religion, Orthodox, Islam

The one-way MANCOVA on attitudes towards an imagined suicidal friend factors yielded a significant main effect for religious affiliation, F_(7,5419) = 17.92, p < 0.0001. Univariate F tests showed significant religious differences on all factors: social acceptance (F = 7.61, p < .0001), helping (F = 12.04, p < .0001), disapproval of suicidal disclosure (F = 13.39, p < .0001) and emotional involvement (F = 39.27, p < .0001) factors.

In order to ease the comprehension of our findings, mean E-ATSS and E-ATSPS factor scores were ordered from the lowest to the highest under the respective tables for affiliated religion. The results of post-hoc comparisons with Scheffe tests are given in Table 4.

Table 4.

Post-hoc comparisons of country E-ATSS factor scores* (above the diagonal) and E-SRSPS factor scores** (below the diagonal) with Scheffe tests

Religions	Buddhism	Hinduism	Islam	No religion	Orthodox	Protestantism	R. Catholic	Other religion
1. Buddhism		5	1, 2, 3, 6	1, 2, 5	4, 5, 6	5, 6	5, 6	5
2. Hinduism			1, 2, 3, 5	2
3. Islam	1, 4	4		1, 2, 3, 5, 6	2, 3, 4, 5	1, 2, 3, 5	1, 2, 3, 4, 5, 6	1, 2, 3, 5
4. No religion	1, 3		3, 4		1, 2, 3, 4	4	1, 2, 3, 4, 5, 6	2
5. Orthodox	1, 3		3			1	1, 2
6. Protestantism	3, 4		3, 4	4	4
7. R. Catholic	1, 2, 3		2, 3, 4	2, 3				5, 6
8. Other religion			4				2, 3

*E-ATSS factors: 1. Acceptability of suicide, 2. Punishment after death, 3. Suicide as a sign of mental illness, 4. Communicating psychological problems, 5. Hiding suicidal behavior, 6. Open reporting and discussion of suicide.

**E-SRSPS factors: 1. Social acceptance, 2. Helping, 3. Disclosure disapproval, 4. Emotional involvement

Strength of religious belief and suicidal attitudes

Table 5 presents the correlation coefficients between the strength of religious belief and attitudes towards suicide and suicidal persons. The highest positive correlation was between punishment after death and participants' strength of religious beliefs, and the highest negative one was between participants' strength of religious beliefs and acceptability of suicide in total sample. For these factors similar patterns were observed for the participants affiliated with Hinduism, Orthodox Christianity, Protestantism, and Catholicism.

Table 5.

Correlation coefficients between attitudes towards suicidality and participants' strength of religious beliefs.

Attitude factors	Total sample		Buddhism		Hinduism		Islam		Orthodox		Protestantism		R. Catholic		Other religion
Attitude factors	r	df	r	df	r	df	r	df	r	df	r	df	r	df	r	df
Attitudes to suicide
Acceptability of suicide	−0.38^c	5572	−0.09	190	−0.50^c	75	−0.23^c	3145	−0.37^c	109	−0.40^c	97	−0.39^c	512	−0.17^a	202
Punishment after death	0.61^c	5572	−0.03	190	0.43^c	75	0.22^c	3145	0.57^c	109	0.63^c	97	0.47^c	512	0.34^c	202
Suicide as sign of mental illness	0.26^c	5572	−0.06	190	−0.12	75	0.07^c	3145	0.10	109	−0.00	97	−0.05	512	0.02	202
Communicating problems	0.10^c	5572	0.14^a	190	0.23	75	0.15^c	3145	0.21^a	109	0.15	97	0.12^b	512	0.09	202
Hiding suicidal behavior	0.17^c	5572	−0.10	190	−0.05	75	0.03	3145	−0.10	109	0.03	97	0.00	512	0.08	202
Open reporting & discussion	0.05^c	5572	−0.06	190	0.15	75	0.00	3145	−0.06	109	−0.14	97	−0.01	512	−0.04	202
Attitudes to suicidal persons
Social acceptance	0.03^a	5572	0.24^b	190	0.06	75	0.01	3145	0.22^a	109	0.00	97	0.14^b	512	0.14	202
Helping	0.05^b	5572	0.22^b	190	−0.17	75	0.06^b	3145	0.09	109	−0.14	97	0.09^a	512	0.01	202
Disclosure disapproval	0.13^c	5572	−0.16^a	190	0.17	75	0.09^c	3145	0.14	109	0.14	97	0.07	512	−0.09	202
Emotional involvement	0.19^c	5572	0.06	190	0.17	75	0.09^c	3145	0.25^b	109	0.12	97	0.22^c	512	−0.05	202

P values are significant at: ^aP < 0.05; ^bP < 0.01; ^cP < 0.001.

Discussion

The present study aimed at testing the associations of religious affiliation and the strength of religious belief to suicidal behavior and attitudes, and psychological distress in student samples in 12 countries. Scientific investigations have often showed that being affiliated with a religion is associated with better mental health (Zou et al., 2014) and a lower propensity for suicidal behavior (Dervic et al., 2004; Sisask et al., 2010; Wu et al., 2015). Our results are in line with these findings for suicidal ideation, but contrast them for suicide attempts and psychological distress. In this study, participants reporting an affiliation with a religion were found to be at a greater risk for attempting suicide and scoring above the cut-off points of GHQ-12 than those reporting no religious affiliation. These results are in stark contrast to findings from a recent review of 89 studies by Lawrence et al. (2016) who found religion to be a protecting factor against suicide attempts, but not against suicidal ideation. The studies covered in this review were conducted mainly with samples from Western nations. The overwhelming majority of participants from Muslim countries in the present study reported having affiliated with religion. Therefore, the difference between our findings and the findings from Lawrence et al. might be a reflection of social, economic and political conditions in Muslim countries. This is an overview, but does it apply to affiliation with specific religious traditions?

An examination of the associations of affiliation with specific religious traditions to suicidal behavior and psychological distress presents a more nuanced picture (Table 1). While being affiliated with Islam was associated with reduced odds for suicide ideation, affiliating with Orthodox religion and not affiliating with a religion were associated with heightened odds of considering suicide. Except for the case of Orthodox Christianity, our findings are in line with theoretical expectations. Since no country adhering to Orthodox religion was included in our study, the association between affiliation with Orthodox religion and suicide ideation might be due to minority stress. However, an examination of the associations of religious affiliation to suicide attempts and psychological distress reveals contrasting patterns. Affiliations with Buddhism and Catholicism were associated with reduced odds of attempting suicide. To our surprise, affiliating with no religion was associated with reduced odds for attempting suicide, which initially seemed to be a theoretically inconsistent relation. Another theoretically contradictory finding is the relation between affiliation with Islam and reports of suicide attempts. In line with findings from previous work (Eskin, 1999b; Eskin, Palova, & Krokavcova, 2014; Eskin, Voracek, Stieger, & Altinyazar, 2011; Lester, 2006) our study demonstrates that participants who reported an affiliation with Islam in our sample have heightened odds of having attempted suicide despite the prohibiting attitude of their religion to suicide, and their scoring above the three cut-off points of GHQ-12.

Why is affiliation with Islam associated with reduced odds for suicide ideation but heightened odds of attempting suicide and higher levels of psychological distress? Two plausible explanations may be offered for these contradictory findings. First, according to Schacter, Guerin, and St Jacques (2011), individuals have a tendency to distort and reconstruct their past memories into meaningful personal narratives. As a result, students from Muslim countries may be reconstructing their own past memories to fit them in with the doctrines of their religious faith. As a concrete act, however, suicide attempt may not be amenable to distortion. Second, the observed associations might reflect the effects of contextual factors. An overwhelming majority of participants affiliated with Islam are from Islamic countries, while only 61 were from other countries. According to the Freedom in the World 2015 report by Freedom House (2015) only Tunisia is classified as “free” in terms of political rights and civil liberties, while Turkey is classified as “partly free”, and the other Muslim countries included in our study are classified as “not free”. Religious affiliation has a normative character in Islamic countries. In other words, there is freedom to religion but freedom from religion is limited and involves risks for life and/or social exclusion. Our findings suggest that freedom of choice is essential for religion to exercise its protective function. One can argue that when a belief system is coerced either by law or by social norms, it loses its protective function and becomes a risk factor.

However, when religious belief is taken into account, participants' strength of religious belief was inversely associated with suicidal ideation in the total sample, and in the students reporting an affiliation with Hinduism, Islam, Orthodox Christianity, and Catholicism. Of note, the strength of belief was unrelated to suicide attempts in the total sample but inversely related to attempting suicide in participants affiliating with Islam, and positively in those affiliating with the Roman Catholic religion and other religions. Strength of religious belief thus seems to be protective against psychological distress in Muslim participants, but the magnitudes of associations are negligible.

Suicide acceptability is related to broader cultural values and is higher in nonreligious than in religious people (Stack & Kposowa, 2016). The findings from our study support previous findings. For instance, acceptability of suicide was the highest in students who reported affiliations with Protestantism, no religion, and the Catholic religion while it was the lowest in participants affiliating with Islam, Orthodox Christianity and Buddhism. Participants affiliated with Islam expressed the lowest acceptance for suicide, which is in line with the findings reported by Stack and Kposowa (2011). The highest mean of punishment after death factor scores was seen for the participants affiliating with Islam, Orthodox Christianity, and Hinduism, while the lowest scores were in those affiliating with no religion, the Catholic religion and Protestantism. Strength of religious belief was inversely related to acceptance of suicide and positively related to a belief in punishment after death.

An examination of participants' attitudes towards an imagined suicidal friend revealed that the lowest social acceptance of suicide and helping for a suicidal friend were observed in students reporting an affiliation with Buddhism, other religion and no religion while the highest scores were seen in those affiliating with the Orthodox religion, the Catholic religion, Hinduism and Protestantism. Participants affiliating with the Orthodox, Catholic and Protestant religions were the least disapproving of a suicidal disclosure in contrast to those affiliating with Buddhism, Islam and no religion. Respondents affiliating with Protestantism, Catholicism and Hinduism displayed the least emotional involvement in a fictitious suicidal peer, while those affiliating with Islam, Orthodox religion and with no religion were emotionally most involved in a peer with a suicidal decision. The strength of a participant's religious belief was positively related to these attitudinal factors.

Taken together, our findings indicate that the association between religiosity, suicidal behavior, suicidal attitudes, and psychological distress involves complex issues reflecting the effects of both personal and contextual factors. Religious affiliation and the strength of religious belief may be protective against suicidal behavior and psychological distress for some people in some sociocultural contexts but not so for others in other sociocultural contexts. Recently, Lester and Walker (2017) have showed religiosity to be protective for suicidal ideation in European American students but not in African American students. An unexpected finding from our research is the higher percentages of suicidal attempts and psychological distress in participants who reported affiliating with Islam despite the lowest suicide acceptance in this group. It may be that religiosity is protective for most participants, but its protective function seems to be limited for people affiliating with Islam for attempting suicide and to some extent for those affiliating with Orthodox Christianity for suicide ideation. The associations of the strength of religious belief to suicidal behavior are in large part in the expected direction but in participants reported affiliating Catholic religion and other religions, stronger belief was related to more suicide attempts. The association between religious belief and psychological distress was negligible.

Limitations

Although results from this multinational study provide some valuable insights into the relation between religion, suicidal behavior, attitudes and psychological distress, a number of considerations should be taken into account when generalizing our results. First, university students may not be representative of the general populations of the countries under consideration. Second, the preventative function of religion for adverse mental health may not be as strong in educated segments of populations as it is in the general populations. Third, religious affiliation and strength of religious belief may be too general concepts to examine the effects of religion. Fourth, our findings from this cross-sectional study should not be interpreted in a causal way. Finally, one should bear in mind a possible volunteer bias in our study. Since the data were collected from consenting students, it is possible that those who chose to participate in the study were somewhat different than those who did not consent to be part of the study. For instance, the students who refused to participate may have greater propensity to suicide or psychological distress.

Implications

Our findings have a number of research and policy implications. First, future research may benefit from examining the association between religious orientation and coping in relation to suicidal phenomena and psychological distress instead of using global variables such as affiliation and strength of belief. Second, more research is needed on possible links between suicidal phenomena and psychological well-being in Muslim countries. Third, the influence of minority stress on suicide and psychological distress should be addressed in future research. Next, future research should address social policies or interventions for minimizing adverse effects of minority stress. Finally, sociopolitical measures enhancing not only freedom to religion but also freedom from religion may augment the protective function of religion against suicide and psychological distress in Muslim countries.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

Mehmet Eskin, PhD, is a professor of clinical psychology at the Department of Psychology, College of Social Sciences and Humanities, Koc University, Istanbul, Turkey. Dr Eskin earned his bachelor's degree in psychology from the Department of Psychology, Middle East Technical University in Ankara. He did his graduate training in clinical psychology in the Departments of Psychology in both Middle East Technical University and University of Oslo, Norway. He received his PhD in psychology from the Department of Psychology, Stockholm University, Sweden. Dr. Eskin authored six books and nearly 100 scholarly articles on suicide, problem-solving therapy, sexual orientation, personality disorders. He is an elected member of The Science Academy of Turkey.

Senel Poyrazli, PhD, works as Professor of Counseling Psychology at the Pennsylvania State University – Harrisburg and identifies as an international psychologist. Her research primarily involves college students' psychosocial adjustment processes; with cross-cultural competency being an additional research interest. Dr. Poyrazli is a fellow of the American Psychological Association (APA) and is actively involved within the organization. Dr. Poyrazli also serves as an editor of the Eurasian Journal of Educational Research and was a consulting editor for Perspectives in International Psychology: Research, Consultation, Practice.

Mohsen Janghorbani, PhD, is an emeritus professor of epidemiology in the Department of Epidemiology and Biostatistics at Isfahan University of Medical Sciences, Iran. Prof. Janghorbani researches chronic disease epidemiology. He is currently the Principal Investigator of the Isfahan Endocrine and Metabolism Research Center, Iran. His published works focus on chronic disease epidemiology, particularly diabetes research.

Seifollah Bakhshi, MSc in Epidemiology, is the Technical Director of Incubator Center of Health Technology in Shahrekord University of Medical Sciences. He is currently expert of research in Vice Chancellery for research & technology in Shahrekord University of Medical Sciences. His Master's Thesis is about “The prevalence of suicide ideation and factors associated among students of Isfahan University of Medical Sciences, 2013–2014”. His published works focus on mental health & epidemiology of suicide.

Mauro Giovanni Carta is a full professor of translational medicine University of Cagliari, Head of Liaison Psychiatric and Psychosomatic Unit University Hospital of Cagliari, Italy. Co-Chairman WHO-UNICA agreement on Quality Rights in the Mediterranean Area.

Maria Francesca Moro, MD, is a psychiatrist and a post-doctoral fellow in human rights and health of minorities at the University of Cagliari. She is currently completing a PhD in Epidemiology at Columbia University. Her research focuses on the impact of human rights violations on mental health.

Ulrich S. Tran, PD, DSc, is a senior lecturer of methods in psychology at the Department of Basic Psychological Research and Research Methods, School of Psychology, at the University of Vienna, Austria. He is an executive board member of the Wiener Werkstaette for Suicide Research. His published works cover topics in clinical research (suicide, psychotraumatology, psychotherapy, mindfulness), laterality research, and individual differences research.

Martin Voracek, DSc, DMSc, PhD, is Full Professor of Psychological Research Methods – Research Synthesis, and Head of the Research Methods Unit, and Deputy Head of the Department of Basic Psychological Research, School of Psychology, at the University of Vienna, Austria. He is a founding member and executive board member (Deputy Chair) of the Wiener Werkstaette for Suicide Research. His research interests include methodology for and applications of research synthesis (meta-analysis and systematic reviews) and open science, individual differences research, and suicide research.

Anwar Mechri, MD, is a professor of psychiatry in the Department of Psychiatry at the University Hospital of Monastir (Tunisia). Dr. Anwar Mechri is the local primary researcher of the suicidal behavior and attitudes multinational project. Its published works focus on schizophrenia and bipolar disorders and suicidal behaviors.

Khouala Aidoudi, MD, is a public health doctor in the emergency department of the Regional Hospital of Ben Guerdane (Medenine, Tunisia). She is the local secondary researcher of the suicidal behavior and attitudes multinational project.

Motasem Hamdan, PhD is the Dean of Public Health, Al-Quds University in East Jerusalem and an associate professor of health policy and management. Dr. Hamdan published research focus on health policy, human resources for health, quality and patient safety, and mental health services research.

Hani Nawafleh, PhD, PhD, is the Dean, Princess Aisha Bint Al-Hussein College of Nursing & Health Sciences and Professor of Community Health Nursing at Al-Hussein Bin Talal University, Jordan. Dr. Nawafleh is a former board member of the Jordanian Nursing Council and former member of the Jordanian Parliament. Dr. Nawafleh studies community mental health and nursing practices. He is currently the Principal or CoPrincipal Investigator on many research projects. His published works focus on community mental health services research, nursing practices, infection control issues, and health education.

Jian-Min Sun, PhD in psychology, is a professor of management in the Department of Management and International Business at The University of Auckland, New Zealand. Dr. Sun researches applied psychology in the workplace, particularly antecedents and mechanisms of employee well-being as well as the relationship with performance. He is currently the principle investigator of Chinese business in New Zealand and cultural engagement and mental health of employees. His publications focus on human resource management practices and employee performance and well-being, leadership and employees' attitude/performance, cross-cultural issues in leadership and management.

Chris Flood, RGN, RN (MH), BSc, MSc, PhD, is a mental health nurse, researcher and senior lecturer with over 27 years' experience of working in and around the NHS either in clinical practice, research or the training of staff. Chris currently works at City, University of London as a senior lecturer and has research interests in patient safety, health economics and suicide and has participated in a number of clinical trials and economic evaluations and is currently a co-applicant on studies exploring peer support in stroke patients and psychosocial support for families of children with long-term gastrostomy tubes in situ. His published work has included work on the cost-effectiveness of occupational therapy versus social work, an economic evaluation in the use of advance directives for service users with psychosis, an economic evaluation of laparoscopic surgery versus medication and studies into safety issues such as the safe use of Midazolam medication and evaluating mental health staff knowledge of resuscitation procedures.

Dr Louise Phillips is a Senior Lecturer in Mental Health at City University and a mental health nurse. She currently leads the pre-registration Mental Health nursing programme. Before starting her academic career, she practiced in a variety of settings including the voluntary sector and the NHS and worked for a number of years as a Community Psychiatric Nurse in the King's Cross area of London. In addition to her nursing qualification she has a BA in European Cultural History and completed her PhD on approaches to the body and mental illness at the University of Kent at Canterbury in 2003, which later became a book entitled: Mental Illness and the Body: Beyond Diagnosis. Her current research interests include perinatal mental health and motherhood experiences of nursing students.

Kouichi Yoshimasu, MD, PhD, is a child psychiatrist, a clinical psychologist, and an associate professor at Wakayama Medical University (Japan) as well as research collaborator at the Mayo Clinic (USA). His research interests are epidemiology and etiology of mental disorders, especially neurodevelopmental disorders.

Kanami Tsuno, MPH, PhD, is the junior associate professor of Department of Hygiene at School of Medicine, Wakayama Medical University. Dr. Tsuno researches psychosocial risk factors and mental health. She is currently the principal investigator of the workplace incivility and mental health study in Japan. Her published works focus on occupational mental health research, especially antecedents and consequences of workplace bullying and incivility among workers.

Dr Omar Kujan, DDS, DipOPath, MSc, PhD, currently holds an academic appointment of Senior Lecturer in Oral Pathology at UWA Dental School, The University of Western Australia. He has received his PhD degree from The University of Manchester, School of Dentistry in 2006. He was then appointed by the University of Manchester as a Clinical Research Fellow. He subsequently took a training post at Oral Pathology Department, Guy's Hospital, London, UK. He held formerly a position of an assistant professor of oral pathology and medicine at the Department of Oral Medicine and Diagnostic Sciences, and director of Skills Development and Continuing Education Unit, Al-Farabi Colleges, Riyadh, Saudi Arabia. He has over 60 publications in peer review international journals, in addition to several book chapters. His primary research interests are: oral cancer and potentially malignant disorders; screening, prognostic and predictive markers, oral diseases, dental education and students' perceptions.

Hacer Harlak, PhD, is a professor of social psychology and head of the Psychology Department at Adnan Menderes University in Aydin-Turkey. Dr. Harlak teaches social psychology and joint issues of social and clinical psychology in both undergraduate and graduate level. Her published works focus on health behavior and attitudes, communication skills training, and intergroup relations.

Yousef S. Khader, BDS, MSc, MSPH, MHPE, FFPH, ScD, is a professor of Epidemiology, Medical Education and Biostatistics in the Department of Community Medicine, Public Health and Family Medicine at Jordan University of Science & Technology. Dr Khader researches non-communicable diseases including mental health. He is currently the principal investigator of the National Diabetes study, and the Women Empowerment study. His published works focus on public health services and the burden of non-communicable diseases, and mental health problems in the Eastern Mediterranean Region.

References

Bertolote

J. M.

Fleischmann

(2015) A global perspective in the epidemiology of suicide. Suicidologi 7: 6–8.

Boehnke

Lietz

Schreier

Wilhelm

(2011) Sampling: the selection of cases for culturally comparative psychological research. In: Matsumoto

van de Vijver

F. J. R.

(eds) Cross-Cultural Research Methods in Psychology, New York: Cambridge University Press, pp. 101–129.

Bonelli

R. M.

Koenig

H. G.

(2013) Mental disorders, religion and spirituality 1990 to 2010: a systematic evidence-based review. Journal of Religion and Health 52: 657–673.

Boyer

Bergstrom

(2008) Evolutionary perspectives on religion. Annual Review of Anthropology 37: 111–130.

Boyer

Liénard

(2006) Precaution systems and ritualized behavior. Behavioral and Brain Sciences 29(6): 635–641.

Colucci

Martin

(2008) Religion and spirituality along the suicidal path. Suicide and Life-Threatening Behavior 38(2): 229–244.

Dervic

Oquendo

M. A.

Grunebaum

M. F.

Ellis

Burke

A. K.

Mann

J. J.

(2004) Religious affiliation and suicide attempt. American Journal of Psychiatry 161(12): 2303–2308.

Durkheim

(1976) The elementary forms of the religious life, New York, NY: George Allen & Unwin.

Durkheim, E. (1951). Suicide: A Study in Sociology [1897]. Illinois: The Free Press.

10.

Eskin

(1999a) Social reactions of Swedish and Turkish adolescents to a close friend's suicidal disclosure. Social Psychiatry and Psychiatric Epidemiology 34: 492–497.

11.

Eskin

(1999b) Gender and cultural differences in the 12-month prevalence of suicidal thoughts and attempts in Swedish and Turkish adolescents. Journal of Gender, Culture and Health 4: 187–200.

12.

Eskin

(2004) The effects of religious versus secular education on suicide ideation and suicidal attitudes in adolescents in Turkey. Social Psychiatry and Psychiatric Epidemiology 39(7): 536–542.

13.

Eskin

(2013) The effects of individualistic-collectivistic value orientations on non-fatal suicidal behavior and attitudes in Turkish adolescents and young adults. Scandinavian Journal of Psychology 54: 493–501.

14.

Eskin

Palova

Krokavcova

(2014) Suicidal behavior and attitudes in Slovak and Turkish high school students: a cross-cultural investigation. Archives of Suicide Research 18(1): 58–73.

15.

Eskin

Voracek

Stieger

Altinyazar

(2011) A cross-cultural investigation of suicidal behavior and attitudes in Austrian and Turkish medical students. Social Psychiatry and Psychiatric Epidemiology 46(9): 813–823.

16.

Eskin

Sun

J. M.

Abuidhail

Yoshimasu

Kujan

Janghorbani

Voracek

(2016a) Suicidal behavior and psychological distress in university students: a 12 nation study. Archives of Suicide Research 20: 369–388.

17.

Eskin

Kujan

Voracek

Shaheen

Carta

M. G.

Sun

J. M.

Tsuno

(2016b) Cross-national comparisons of attitudes towards suicide and suicidal persons in university students from 12 countries. Scandinavian Journal of Psychology 57: 554–563.

18.

Freedom House (2015). https://freedomhouse.org/sites/default/files/01152015_FIW_2015_final.pdf. Accessed: 15 April 2015.

19.

Goldberg, D. P., & Williams, P. (1988). A user’s guide to the GHQ. Windsor, NFER Nelson.

20.

Goldberg

D. P.

Gater

Sartorius

Ustun

T. B.

Piccinelli

Gureje

Rutter

(1997) The validity of two versions of the GHQ in the WHO study of mental illness in general health care. Psychological Medicine 27(01): 191–197.

21.

Inglehart

Baker

W. E.

(2000) Modernization, cultural change, and the persistence of traditional values. American Sociological Review 65: 19–51.

22.

Joe

Romer

Jamieson

P. E.

(2007) Suicide acceptability is related to suicide planning in US adolescents and young adults. Suicide and Life-Threatening Behavior 37(2): 165–178.

23.

Kang

P. P.

Romo

L. F.

(2011) The role of religious involvement on depression, risky behavior, and academic performance among Korean American adolescents. Journal of Adolescence 34(4): 767–778.

24.

Lawrence

R. E.

Oquendo

M. A.

Stanley

(2016) Religion and suicide risk: a systematic review. Archives of Suicide Research 20(1): 1–21.

25.

Lester

(2006) Suicide and Islam. Archives of Suicide Research 10(1): 77–97.

26.

Lester

Walker

R. L.

(2017) Religiosity is a protective factor for suicidal ideation in European American students but not in African American students. OMEGA-Journal of Death and Dying 74(3): 295–303.

27.

Nelson

Hanna

Houri

Klimes-Dougan

(2012) Protective functions of religious traditions for suicide risk. Suicidology Online 3: 59–71.

28.

Pescosolido

(1990) The social context of religious integration and suicide: pursuing the network explanation. Sociological Quarterly 31: 337–357.

29.

Pritchard

Amanullah

(2007) An analysis of suicide and undetermined deaths in 17 predominantly Islamic countries contrasted with the UK. Psychological Medicine 37: 421–430.

30.

Purzycki

B. G.

Haque

Sosis

(2014) Extending evolutionary accounts of religion beyond the mind: religions as adaptive systems. In: Watts

Turner

L. P.

(eds) Evolution, Religion, and Cognitive Science: Critical and Constructive Essays, Oxford University Press,, 2014.

31.

Schacter

D. L.

Guerin

S. A.

St Jacques

P. L.

(2011) Memory distortion: an adaptive perspective. Trends in Cognitive Sciences 15(10): 467–474.

32.

Sisask

Värnik

Kolves

Bertolote

J. M.

Bolhari

Botega

N. J.

Wasserman

(2010) Is religiosity a protective factor against attempted suicide: a cross-cultural case-control study. Archives of Suicide Research 14(1): 44–55.

33.

Shah

Chandia

(2010) The relationship between suicide and Islam: a cross-national study. Journal of Injury and Violence Research 2: 93–97.

34.

Stack

(2000) Suicide: A 15-year review of the sociological literature, Part II: modernization and social integration perspectives. Suicide and Life-Threatening Behavior 30: 163–176.

35.

Stack

Kposowa

A. J.

(2011) Religion and suicide acceptability: a cross-national analysis. Journal for the Scientific Study of Religion 50(2): 289–306.

36.

Stack

Kposowa

A. J.

(2016) Culture and suicide acceptability: a cross-national, multilevel analysis. The Sociological Quarterly 57(2): 282–303.

37.

Stark

Doyle

D. P.

Rushing

(1983) Beyond Durkheim: Religion and suicide. Journal for the Scientific Study of Religion 22: 120–131.

38.

Vail

K. E.

Arndt

Abdollahi

(2012) Exploring the existential function of religion and supernatural agent beliefs among Christians, Muslims, Atheists, and Agnostics. Personality and Social Psychology Bulletin 38(10): 1288–1300.

39.

Värnik

(2012) Suicide in the world. International Journal of Environmental Research and Public Health 9(3): 760–771.

40.

Wang

J. Y.

Jia

C. X.

(2015) Religion and completed suicide: a meta-analysis. PloS One 10(6): e0131715.

41.

Zou

Huang

Maldonado

Kasen

Cohen

Chen

(2014) The efficacy of religious service attendance in reducing depressive symptoms. Social Psychiatry and Psychiatric Epidemiology 49(6): 911–918.