Limitations

The main limitation to the study is the relative paucity of reported suicides in most of the Islamic countries. Fourteen countries had an annual suicide rate of less than 25 per million population. Furthermore, we have to consider how reliable and robust are the data. Weaker systems are more likely to underreport suicide. However, we can only use the best information available, which is the consistently collected WHO annual mortality statistics.¹⁹ A minor limitation was that for six of the Islamic countries, WHO reported only the numbers of suicides but no WHO population data were available. This meant that their suicide rates had to be calculated using a different source for population.

Main findings

The null hypothesis can be rejected because each of the four Islamic groups’ suicide and combined undetermined and accidental death rates were markedly different from those of the four Western comparator countries. There was a degree of internal validity in the findings as three of the less-traditional Islamic countries were only moderately significantly different from the Western comparator countries (P < 0.03); the significance of the difference was stronger for the former USSR group (P < 0.001) and markedly higher for the Gulf Arab states and Middle Eastern and North African countries (P < 0.0001). This supports earlier research, which found a likelihood of underreporting of suicide in Islamic countries^{Reference Eskin, AlBuhairan, Rezaeian, Abdel-Khalek, Harlak and El-Nayal14} that appears to have persisted more than a decade later. Furthermore, a very recent Iranian study using a new network method showed that suicide attempts and completed suicides continue to be officially underreported.^{Reference Moradinazar, Najafi, Baneshi and Haghdoost21} Yet probable underreporting still occurs in Western countries, as cultural attitudes persist.^{Reference Papaslanis, Kontaxakis, Christodoulou, Konstantakopoulos, Kontaxaki and Papadimitriou5,Reference Gunnell, Hawton and Kapur15–Reference Pritchard and Hansen18} This was seen in certain groups of doctors and nurses in Germany^{Reference Beine and Schubert22} and in the military in the USA,^{Reference Vannoy, Andrews, Atkins, Dondanville, Young-McCaughan and Peterson23} and there continue to be inconsistencies in the UK and in Israel for younger people.^{Reference Mars, Cornish, Heron, Boyd, Crane and Hawton24,Reference Bakst, Braun, Zucker, Amitai and Shohat25} In a comparative study of 21 Western countries, it was shown that in certain countries some age groups appeared to contain more likely underreporting than others:^{Reference Pritchard and Amanualla6} these included the four Western comparator countries in our study. Indeed, Greece's suicide:UnD+AccD ratio of 3.03 suggests that cultural inhibitions about suicide are still in evidence. In the UK, with its well-established system of coroners’ courts, youth (15–24 years old) suicide rates were at the same level as the rate of undetermined deaths, and it was suggested that bringing in an open verdict avoided confronting the bereaved families with the idea that their child died by their own hand, adding to their grief. In some other Western countries high rates of undetermined deaths occurred in different age groups and more often among females.^{Reference Gunnell, Hawton and Kapur15,Reference Pritchard and Hansen18} So if there is a degree of underreporting of suicides in some Western countries, it is not surprising that this may happen more often in Islamic countries, considering the stronger cultural taboos in Muslim countries.^{Reference Moradinazar, Najafi, Baneshi and Haghdoost21}

The question remains whether, despite the ‘violent’ methods in undetermined and accidental deaths and the significant statistical differences found between the Islamic and Western comparator countries, it can be assumed that they contain underreported suicides. The results can only be indicative but, apart from Iran at 29 per million population, suicides virtually do not occur in the Middle East and Gulf State countries. So, can such low reported rates be believed? If they can, it would imply that either their psychiatric services are far more effective in preventing suicides than those of Western countries, or that psychiatric disturbance with its link to suicidal behaviour is substantially lower than in Western nations. Moreover, it was not just that every Islamic country's undetermined death rates were higher than their suicide rates, but for 19 of the Islamic nations the accidental death rates were higher than their self-inflicted death (suicide) rates.

Practice implications

This broad-brush population study has practice relevance for both Islamic countries and Western countries that have substantial minority Islamic populations. If patients or families do not talk about possibly ‘hurting themselves’, practitioners can harbour a false sense of reassurance that Muslims are exempt from experiencing suicidal ideation. This can lead to less robust enquiry when providing care and treatment, and psychiatric training needs to be aware of this possibility. Sadly, this can be compounded by patients and families who might be reluctant to raise concerns because of the stigma surrounding suicide, which still occurs across cultures, even among those without a religious allegiance.^{Reference Corrigan, Sheehan and Al-Khouja26}

The danger is that if the extent of suicidal behaviour goes unrecognised, or indeed is denied, then the necessary action to support mentally disturbed individuals is undermined. Patients and families need to be able to have trust in services being non-judgemental, respectful of belief systems and confidential.

From a public health and policy point of view, mental health services in all countries, but especially those with strong cultural taboos against suicide, need to have more accurate suicide data to attract resources to reduce the risk of unmet needs.^{Reference Cantor, Leenaars and Lester3–Reference Pritchard and Amanualla6,Reference Khan, Naqvi, Thaver and Prince12–Reference Eskin, AlBuhairan, Rezaeian, Abdel-Khalek, Harlak and El-Nayal14} These results from 21 Islamic countries suggest that the authorities should review their situation and regard suicide as a public health problem as these results appear to be at odds with officially reported levels of suicide.

It is appreciated that the problem of assessing suicide risk for Muslim patients is that, for them and their family, psychiatric distress is often not interpreted as an illness but as sin and failure. This can further diminish the patient's self-esteem and we need to be able to enquire sensitively yet avoid challenging their religious values. It is always necessary to approach the patient from their personal perspective and we draw on the advice of an Islamic author who suggests that the psychiatrist could ask, without offence, ‘have you wished that God would allow you to die somehow?’.^{Reference Rassool27} This could give the practitioner some idea about risk, without adding stress to an already difficult situation.

It is hoped that this study might begin the necessary debate that seeks to prevent psychiatry's ultimate catastrophe, an unnecessary death, with all the reverberations for family and society in both Western but especially Islamic countries.