Research

Although there is much empirical research to support the view that religion is protective against suicide, and some contrary evidence, there have been few prospective studies. According to Koenig et al, ^{Reference Koenig, King and Carson2} out of 141 studies on religion and suicide conducted to date, only two ^{Reference Lubin, Glasser, Boyko and Barell3,Reference Thompson, Ho and Kingree4} were prospective or longitudinal. One of these studies comprised a retrospective trawl of a national mortality database, and the other (which was not originally designed for the purpose of studying the influence of religion on suicide) did not study completed suicides. Indeed, as Koenig et al note, the design of prospective studies of completed suicide is challenging, requiring large sample sizes and prolonged periods of follow-up to ensure sufficient numbers of observed suicides.

The publication in this issue of the Journal of the first prospective study of religious service attendance and completed suicide is therefore significant. Based on the Third National Health and Nutrition Examination Survey (NHANES III) in the USA, Kleiman & Liu have shown that those who frequently attended religious services (i.e. ⩾24 times per year) were less than half as likely to die by suicide than those who attended services less frequently. ^{Reference Kleiman and Liu5} Following more than 20 000 participants over 12-18 years, 25 individuals were identified as having died by suicide, slightly fewer than might have been expected on the basis of national statistics. Given this small number of actual suicides, sample power in this study is somewhat more modest than one might have expected. It must also be noted that this data-set was not constructed for the purpose at hand and we know nothing about the specific religious affiliations, experiences, beliefs or other religious behaviours of participants - only the frequency with which they attended religious services. Notwithstanding these limitations, this is a significant piece of research, and the authors conclude that frequent religious service attendance is a long-term protective factor against suicide.

How might religion help?

There is evidence to suggest that religion may help people to cope better with life stresses, reduce the incidence of depression and substance misuse, facilitate recovery from depression, enhance social support and provide sources of hope and meaning (pp. 179-180). ^{Reference Koenig, King and Carson2} Any or all of these considerations provide possible explanations for the putative protective benefit identified in the study by Kleiman & Liu. Although interpretations of such studies are debated, ^{Reference Cook6} and negative findings are also reported (e.g. King et al ^{Reference King, Marston, McManus, Brugha, Meltzer and Bebbington7} ), the overall balance of findings is thought by many to reveal a benefit for religious beliefs and practices in relation to mental health and well-being.

In an ideal research design, we would know about more than simply religious service attendance. Although the data available to Kleiman & Liu are an improvement on the crude measures of religion employed in the only two previous prospective studies of completed suicide, and although religious service attendance is an objective measure widely employed in research of this kind, it is frustrating not to know more about the religious beliefs, behaviours, experiences and traditions of these participants. However, what we do know is not unimportant. Whatever kinds of religious services these individuals attended, we know that the teachings and traditions to which they were exposed would have had particular implications for anyone contemplating suicide.

Religious teachings on suicide

In Hebrew scripture, suicide is implicitly condemned in the commandment not to commit murder (Exodus 20:13) and both Jewish and Christian commentators have generally seen suicide to be forbidden by this commandment. Augustine of Hippo, for example, argued that the commandment is not qualified (it does not say ‘You shall not murder your neighbour’ but simply ‘You shall not murder’), that no scriptural permission to die by suicide is granted, and that it is better for the Christian to suffer than to commit the sin of taking a life (City of God I.16-27). In conclusion, he states that ‘suicide is monstrous’. ^{Reference Bettenson and O'Meara8} More recently, in contrast, it has been noted that suicide is not explicitly condemned in Judeo-Christian scripture, ^{Reference Koch9} and attitudes have been more sympathetic towards the suffering of those who consider taking their own lives, tending to emphasise more the importance of divine and human compassion (e.g. From Despair to Hope, p. 207 ¹⁰ ).

The Qur’an is more explicit in its injunction against suicide:

As a result, suicide is illegal in many Islamic countries, and it has been suggested that suicide is under-reported in the Islamic world.

In Eastern faith traditions, attitudes towards suicide may be understood as more ambiguous, ^{Reference Coghlan, Ali, Cook, Powell and Sims12} and Durkheim took Hinduism as a religion within which examples of altruistic suicide might be identified, but the general sense of respect for life, and concerns about the impact on karma, still provide a prevailing understanding that suicide is wrong.

Although the study by Kleiman & Liu does not address the situation of those who would describe themselves as ‘spiritual but not religious’, it is possible that this group do not experience protection against either depressive disorder ^{Reference King, Marston, McManus, Brugha, Meltzer and Bebbington7} or suicidal behaviour. ^{Reference Rasic, Belik, Elias, Katz, Enns and Sareen13} It might be expected that they attend religious services less frequently, but this group almost certainly includes some frequent service attenders (e.g. among those who espouse more modern styles of worship and who eschew traditional understandings of ‘religion’).

Implications for clinical practice

The findings of Kleiman & Liu add to the already large body of evidence obtained from cross-sectional studies, the great majority of which show lower levels of suicidal ideation and behaviour among those who are religious. ^{Reference Koenig, King and Carson2} It would therefore seem wise to take religion into account when assessing suicidal risk, albeit remembering that religion is no guarantee against suicide (as the study by Klieman & Liu clearly demonstrates). Although this is something that merits discussion with patients at risk of suicide, Koenig et al rightly note (p. 190) ^{Reference Koenig, King and Carson2} that caution should be displayed if considering spiritual/religious interventions with such a group, as no studies to date have examined effectiveness (or harm) of such interventions. This therefore again raises the question of whether and how boundaries can be drawn in psychiatry in such a way as to define what may properly be discussed, and when, and how, in relation to spirituality and religion. ^{Reference Cook6} The College Position Statement, Recommendations for Psychiatrists on Spirituality and Religion, ^{Reference Cook14} suggests some of the general ways in which we might seek to ensure good practice, but it is also clear that much is properly left to the judgement of the psychiatrist and that patients contemplating suicide are especially vulnerable.