Determination of the social related factors of suicide in Iran: a systematic review and meta-analysis

Suicide is an important cause of death around the world [1] and is considered to be a major health problem for most countries [2, 3] imposing a substantial financial burden [4]. Current literature estimates the rate of attempted suicide in Iran to be 41.8 and 64.5 per 100,000 people for males and females respectively [5]. The World Health Organization reported that globally, approximately 815,000 people had committed suicide in the year 2000, with attempted suicide believed to be ten to twenty times higher for the same period [6].

The French sociologist, Émile Durkheim, was the first to consider that social determinants could be an integral factor of disease and to show that suicide is a social phenomenon [7, 8]. Durkheim hypothesized that suicide had two important aspects, social integration and moral integration [9]. Feeling lonely, experiencing family conflicts and low levels of community integration are social factors revealed by western psychological theorists to be involved in the etiology of suicide [9]. However, these factors may vary significantly for different countries due to distinctions within social and cultural constructs.

Studies from developing countries reveal that stressful social events can be an important trigger for suicide attempts [10]. Moreover, there are reports that the family environment in particular [11] and quality of the marital relationship can be predictors of suicidal behavior [12, 13]. The effect of pre-existing psychiatric disorders and past history of suicidal ideation are well recognized as predictors for future incidences [11]. However, there is a paucity of data examining the role of social risk factors in suicidal behavior for developing countries, resulting in a gap in the existing evidence.

Suicidal behavior is often the result of a complex matrix of clinical [14, 15], familial [16], economic [17], political [18] and even geographical [19] variables. Indeed, recent studies are now beginning to observe a link between climate fluctuations and suicide [19‐21]. However, none of these studies have established a definitive association between the climate and social related factors resulting in suicide, such as familial conflict. Iran in particular, is an arid to semi-arid region with significant climactic variations across each province [22]. With new evidence reporting even greater degrees of fluctuations in rainfall [23] and ground water levels [24, 25] concerns are being raised over the psychological effect this may yield in the Iranian population.

This lack of precise and reliable evidence paring the social aspect of suicide with considerable climatic variations found in each Iranian province highlights a need for a comprehensive study to evaluate the available literature. Subsequently, the purpose of this study was to estimate the prevalence of social factors resulting in suicide in Iran through meta-analyses and to explore the causes of between-study variability.

A total of 2,526 citations were retrieved through electronic database screening, producing a total of 20 studies from 16 provinces of Iran that met all of the eligibility criteria [27‐46]. The characteristics of each included study are reported in Table 1 and a Quorum flow chart outlining the details related to the selection process are presented in Figure 1. A total of 12,005 subjects were available for assessment from the 20 included studies. Overall the STROBE methodological quality assessment produced a maximum score of 31 and mean of 13.26.

This review of 20 studies found that social factors contributing to attempted suicide include family conflicts (30%), marital problems (26%), economic constrains (12%) and educational failures (5%), with family conflicts found to be the most prevalent for this Iranian study population. Results of the meta-regression also found that climate types and sample size produced significant levels of heterogeneity (Table 2). Subgroup analysis based on the type of climate showed an interaction with family conflict and the central plateau foothills area, more so than with any other climate in the country (Figure 2). A possible explanation might be that with the exception of Tehran (Capital of Iran), provinces in this subgroup are considered disadvantaged, particularly over the past twenty years. In addition, three studies were conducted over twelve years ago (between 1992 and 1999). With the rapid growth of public health in Iran in recent years, it seems likely that this higher prevalence of family conflicts from the central plateau foothills can potentially be attributed to the accumulation of older studies in this subgroup rather than climate type. Following a sensitivity analysis and the removal of the Mazandaran 2008 study (due to a small sample size; n = 100), subgroup analysis revealed a significant difference between family conflicts and climate for the region of the Capsian sea coast. Capsian sea coast located to the north of Iran have a largely temperate climate, with the Alborz mountain range in the surrounding area causing the terrain to be dominated by cloud-cover and consistent rainfall. Subsequently, our findings support the growing evidence of a relationship between human mood and the seasons [47‐49]. A biological explanation may relate to the photoperiodic mechanisms of the sun’s rays [50], with new evidence indicating that serotonin transporter availability is altered in response to diminished light exposure with lower levels of serotonin found in winter months [51]. The amount of daily sunshine and global radiation can influence serotonin-1A receptor binding in the limbic brain regions of health subjects, highlighting a relationship between seasonal factors and the regulation of serotonergic transmissions [51]. Further research is required to determine the degree to which this potential relationship between exposure to sunlight and suicidal behavior exists. Using the results of suicide attempts, subgroup analysis suggests that family conflicts were less prevalent among people who survived from the suicide attempt (nonfatal), indicating that family may have an important role in the outcome. However, only two studies included fatal suicide with the meta-analysis producing high levels of heterogeneity. Consequently, a comparison between these subgroup was difficult highlight a paucity of data and as such the results should be interpreted with caution.

The meta-analysis examining the prevalence of marital problems in cases of attempted suicide produced findings similar to those reported by Janghorbani et al., who found that completed suicide and attempted suicide were more frequent among married couples than single people [5]. This is in contrast to the popular belief that marriage has a protective effect for against psychological or behavioral disorders [5]. Studies are required to investigate the particular effects of marriage itself, or other variables such as culture, economy and psychological factors on suicide attempts. The notable finding in this study was that all married subjects in these nine studies had nonfatal suicide attempts. A probable justification for this might be that married couples use less dangerous methods in attempting suicide, as their primary aim is to invoke a response from their spouse rather than actually complete suicide. These results may be influenced by the predominantly female representation in our study, which is however consistent with more women attempting suicide than men [5, 52, 53].

Unexpectedly, 5% of suicide attempts were found to be because of educational failures (95% CI: 3 to 8%). Previous studies report that people who attempt suicide following educational disappointments are usually aged between 14 and 17 years [54]. Student deaths impose a considerable burden on the society, and as such special attention to this issue is warranted. Due to inadequate reporting of results we were unable to distinguish high school students from university students, thus the 5% prevalence could be attributed to failure in high school exams or university entrance exams. This has the potential to be an important risk factor for many college entrance students taking exams (Konkor) as parents, communities and the individuals themselves place a significant amount of pressure on success for Iranian students in particular. Mental disorders such as depression should be considered as another risk factor for suicide attempts among Iranian students [55‐59]. Nowadays, mental disorders are considered to be one of the most important causative factors related to suicide attempts [60‐63]. There is a possibility that an interaction between mental disorders and educational failures may have a considerable impact on suicide attempts and further research such as case–control studies are needed to address this interaction. In addition, screening for mental disorder among students should be considered as part of standard care in schools and universities to allow early intervention to occur. Educational failures were highest in the Razavi Khorasan province with 18% and Alborz province with 15%, as such future research should target schools in these areas.

The relationship between socio-economic status and suicide has been confirmed through the existing literature [64] as has the subsequent impact on health. The results of this meta-analysis indicate a high prevalence of suicide due to economic constrains with a prevalence of 12% found across the twelve studies. Recent evidence from Iran has found that suicide attempts among young people more frequent than among elders with a negative relationship observed between suicide attempts and age [5]. The most important economic constraint among young people in Iran is likely to be unemployment. Therefore, job creation and encouragement to enter into the workforce particularly for among young people may result in a decrease in attempted suicides.

This study does have several limitations with all pooled analyses containing significant heterogeneity and subsequently should be interpreted with caution. The results should however be considered generalisable as they include a broad geographical cross-section from Iran. Potential factors contributing to the variability include location (setting), time of the study and characteristics of the population. Such heterogeneity is to be expected though considering the diverse cultures and ethnic groups found in Iran. Although many would argue that in the presence of such significant heterogeneity a meta-analysis should not be presented, we believe that providing the reader with the pooled prevalence estimates and a caution relating to the presence of heterogeneity will allow them to obtain a broad perspective examining the social factors related to suicide. Another limitation relates to the possibility that data evaluating risk factors for attempted suicide could be missed due to inadequate reporting of results or a lack of publication. Indeed, publication bias was detected for the majority of outcomes assessed in this study. The search strategy was also limited due to the use of standard search keywords in domestic databases and for this reason all synonyms were retrieved in both Persian and English languages. Moreover, a lack of comprehensive coverage for university databases containing research projects and theses further limit the search strategy. However, we believe that this review still provides the reader with an overview of the current available evidence and highlights that there is a potential gap due to reporting biases that need to be considered in future investigations and research.

Suicide is a multi-factorial problem with socio-related factors interacting reciprocally amongst each other with a great deal of overlap between these variables. Therefore, quantification of the association between these factors is now needed and should be the focus of future studies using case–control designs and meta-analyses reporting odds ratio with pooled interactions.

In conclusion, the results of this meta-analysis indicate that family conflicts are a significant factor associated with suicide in Iran. The area most affected were provinces bordering the Caspian sea coast with subgroup analysis suggesting that climate may be an integral factor contributing to this observation. Returning to the question posed at the beginning of this study, this review has found that social factors have a perceptible role in Iranian suicidology.