| Article Access Statistics|
| Viewed||300 |
| Printed||8 |
| Emailed||0 |
| PDF Downloaded||36 |
| Comments ||[Add] |
Click on image for details.
|Year : 2021
: 63 | Issue : 5 | Page
|Cluster analysis of risky behaviors among the youth in Western Iran: Determining correlates and comparing clusters based on severity of disability and attitude toward mental health help-seeking
Habibolah Khazaie1, Farid Najafi2, Behrooz Hamzeh3, Azita Chehri4, Afarin Rahimi-Movaghar5, Masoumeh Amin-Esmaeili5, Mehdi Moradi-Nazar3, Sepideh Khazaie1, Ali Zakiei1, Saeed Kamasi6, Yahya Pasdar3
1 Sleep Disorders Research Center, Kermanshah University of Medical Sciences, Kermanshah, Iran
2 Development and Health Promotion Research Center, Kermanshah University of Medical Sciences, Kermanshah, Iran
3 Research Center for Environmental Determinants of Health, School of Public Health, Kermanshah University of Medical Sciences, Kermanshah, Iran
4 Department of Psychology, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran
5 Iranian National Center for Addiction Studies, Tehran University of Medical Sciences, Tehran, Iran
6 Department of Neuroscience and Psychopathology Research, Mind GPS Institute, Kermanshah, Iran
Click here for correspondence address and
|Date of Submission||29-Sep-2020|
|Date of Decision||13-Jun-2021|
|Date of Acceptance||01-Jul-2021|
|Date of Web Publication||12-Oct-2021|
| Abstract|| |
Background and Aims: The objectives of the study were (i) cluster analysis of risky behaviors; (ii) determining correlates; and (iii) comparing clusters with regard to the attitude toward mental health help seeking.
Methods: The current cross-sectional study is a part of the first phase of the Persian Youth Cohort. From October 2014 to January 2017, 2990 individuals from Ravansar City in western Iran completed structured interviews and national and international standard questionnaires. The obtained data were analyzed using two-stage cluster analysis, multinomial logistic regression analysis, and Chi-square test.
Results: This model provided three distinct clusters: (i) low-risk group with mild distress, (ii) high-risk group with high distress, and (iii) violent group with medium distress. Some sociodemographic and lifetime psychiatric disorders were the correlates of unhealthy clusters (P < 0.05). Compared to the reference cluster, a higher number of members in unhealthy clusters were suffering from medium to severe disability. Nevertheless, the participants in these clusters were less inclined to mental health help seeking.
Conclusions: More than half of the youth were suffering from suicidal and violent behaviors. Since high-risk participants are less inclined to mental health help seeking, the health policymakers can successfully utilize the results in planning general health programs.
Keywords: Disability, help seeking, high-risk behaviors, psychiatric problems, substance-related disorders
|How to cite this article:|
Khazaie H, Najafi F, Hamzeh B, Chehri A, Rahimi-Movaghar A, Amin-Esmaeili M, Moradi-Nazar M, Khazaie S, Zakiei A, Kamasi S, Pasdar Y. Cluster analysis of risky behaviors among the youth in Western Iran: Determining correlates and comparing clusters based on severity of disability and attitude toward mental health help-seeking. Indian J Psychiatry 2021;63:424-32
|How to cite this URL:|
Khazaie H, Najafi F, Hamzeh B, Chehri A, Rahimi-Movaghar A, Amin-Esmaeili M, Moradi-Nazar M, Khazaie S, Zakiei A, Kamasi S, Pasdar Y. Cluster analysis of risky behaviors among the youth in Western Iran: Determining correlates and comparing clusters based on severity of disability and attitude toward mental health help-seeking. Indian J Psychiatry [serial online] 2021 [cited 2021 Oct 22];63:424-32. Available from: https://www.indianjpsychiatry.org/text.asp?2021/63/5/424/328091
| Introduction|| |
Risky behaviors such as aggression and violence, suicidal thoughts and attempts, and substance and alcohol abuse are considered as the most important factors threatening public and social health., Violent behaviors are defined as observable presentations of aggressiveness often intended for injuring or damaging others. On the other hand, suicidal tendency involves suicidal thoughts, plans, attempts, and suicide directed toward the individual himself or herself. Depending on the type of violence and injury, violent behaviors are seen in approximately 1%–29% of the adults. However, the prevalence of suicidal thoughts and attempts varies based on the utilized definition, the follow-up period, and the target population. The prevalence of suicidal thoughts, plans, and attempts are reported as 0.8–14.1, 6.2–6.7, and 0.7%–4.2% in the general population, respectively.,,, With regard to the general population in Iran, suicide is a prevalent problem, which has been increasing during the recent years, particularly among the youth.,,
Demographic factors such as age, gender, education level, unemployment, and marital status,,, drug addiction,,, alcohol abuse,,,, mental disorders,, life events and emotional problems, chronic psychological distress, disability, and social capital, are among the most important causes of aggressive and suicidal behaviors and their corresponding mortality. On the other hand, in the United States, deaths from high-risk behaviors such as suicide are more than forty thousand cases per year and this annual rate has increased by 24% over the past 15 years. This problem, as one of the top ten causes of death, exerts a heavy financial load on the health and economic system of any country.
It is clear that better understanding and identification of social, psychological, and therapeutic factors related to risky behaviors such as suicidal thoughts and behaviors can contribute to developing and expanding the measures that can be taken to prevent such dangerous behaviors. However, one of the barriers for realizing such an important task is that the distribution of risky behaviors in different age and gender groups with their unique sociocultural context does not follow a single pattern., To mitigate this problem, classifying the general population in consistent groups and distinguishing the profile of risky behaviors based on the population may possibly be effective in identifying predictors and consequences. Cluster analysis is a highly useful statistical method for partitioning a population based on common characteristics and a wide range of disorders. Clustering mental health will facilitate the identification of target groups for providing on-time intervention and will also help health policymakers provide services in line with the preferences and needs of the population. Based on these considerations, the current study was carried out to realize three objectives: (i) cluster analysis of risky behaviors and partitioning the participants; (ii) determining sociodemographic correlates, substance and alcohol abuse disorders, and social capital for each one of the clusters; and (iii) comparing clusters with regard to the severity of disability and attitude toward mental health help-seeking.
| Methods|| |
Design and context
The Youth Cohort is one of the subsets of the Comprehensive Prospective Epidemiological Research Studies in (PERSIAN) Cohort. The PERSIAN is a homogenous national cohort study started in 2014 covering 170 thousand adults in the age range of 35–70 years. This study surveys the population-based information in the fields of medicine, epidemiology, health, and nutrition of adults. The PERSIAN birth, youth, and elderly cohort are carried out along with the PERSIAN cohort (http://www.persiancohort.com/aboutus/). The PERSIAN youth cohort focuses on gathering information with regard to lifetime mental disorders, substance and alcohol abuse, suicidal and violent behaviors, traffic and nontraffic injuries, outpatient services and hospitalization due to psychological problems, and mortality. In this prospective study, the participants are followed up on for at least 3–5 years using phones and face-to-face meetings.
Participants in the current cross-sectional study
The first phase of the youth cohort includes 9 thousand individuals in three cities, i.e., Ravansar in the west, Fasa in the south, and Rafsanjan in the center of the country (3 thousand individuals in each city). This sample size was determined using the 80% power if alpha equals to 0.05 using the assumptions (incidence in nonexposed = 0.03; exposed = 0.02) method. The information in the current cross-sectional study is in relation to the population of Ravansar City as a part of the youth cohort. The geographical and population information for Ravansar City has been presented before elsewhere. Since the target population of the youth cohort includes urban and rural youth aged 15–34 years in this region, 3 thousand individuals were investigated. The sample which was randomly selected represents the majority of the young population in this region. However, 9 individuals were eliminated because of incomplete questionnaires. Furthermore, 1 participant was eliminated by the statistical software application. The entrance criteria for the study included the above-mentioned age range plus at least 5 years of residence in this city and having at least 5 years of completed education.
Gathering the required data
Signing up and gathering information was performed during the first phase from May 2014 to May 2016. The demographic and socioeconomic information, family history of psychiatric problems, using 12-month, and lifetime outpatient services for psychiatric diseases were obtained using standard questionnaires. To evaluate lifetime psychiatric disorders and the prevalence rate of drug and alcohol abuse, the Composite International Diagnostic Interview, version 2.1, was used based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR). The sociodemographic information form, suicidal and violent behaviors questionnaire, Kessler Psychological Distress Scale (K10), Sheehan Disability Scale, social capital questionnaire, and a questionnaire on attitude toward mental health help seeking were other instruments used in the current study.
Gathering the data and the interviews were performed by two trained psychology majors. The training for interviewers was provided by the PERSIAN cohort team based on the standard protocols and materials prepared by the World Health Organization. Due to the high number of questionnaires and interviews, ten participants would be invited to one of the health centers for the interviews each day. After obtaining the informed consent form for participating in the study and providing the necessary guarantees for the participants on the confidentiality of their information, the data of each participant would be obtained in one of the health centers in the city or the village in 40–60 min. The answers of the participants would be recorded immediately on the electronic database connected to the central server of the Youth Cohort. The daily procedure was supervised by a general physician and the central research team in Tehran would supervise the study and its stages.
The sociodemographic information form
The sociodemographic data include gender and age, the number of completed years of education, marital status (single, married, divorced/widowed), and the occupational group (employed, unemployed, student, housekeeper). These data are obtained in a standard way from the IranMHS protocol.
Composite International Diagnostic Interview (Version 2.1)
The lifetime version of the Composite International Diagnostic Interview (Version 2.1) was used for determining psychological disorders based on the fourth edition of the DSM-IV-TR and the tenth revision of the International Statistical Classification of Diseases and Related Health Problems-10. This tool has been developed by the World Health Organization, and it is designed to be used by experienced interviewers. This structured interview has a good inter-rater and retest reliability in various cultures and languages. The Persian version of this tool also has a good reliability, except for psychoses. In the current study, lifetime spread of disorders related to abuse of any form of substance and alcohol abuse, and psychiatric and mental disorders including major depressive disorder, persistent depressive disorder (dysthymia), generalized anxiety disorder, and obsessive-compulsive disorder were screened using this tool.
Kessler Psychological Distress Scale
This 10-item scale was developed in 2003 by Kessler et al. for screening psychological problems in the general population. The questions are scored based on the Likert spectrum from never (score of zero) to always (score of 4), and the total score ranges from 0 to 40. This scale which is mainly used for measuring the level of anxiety and depressive symptoms has good validity and reliability. The Iranian form of this scale also has good validity.
The inventory of suicidal thoughts and behaviors
The lifetime Persian version of this questionnaire, which has been developed for IranMHS, is derived from the amended format utilized in World Mental Health Survey. This questionnaire includes 9 items related to suicidal thoughts, serious thoughts, plans, attempts, number of tries, attempt in the past 12 months, the method of suicide attempt, the serious intent for dying, and referral to health centers. This scale has a good inter-rater reliability in Iranian samples. In the current study, only the data related to serious suicidal thoughts, suicidal plans, and history of suicidal attempts (all in the yes/no format) were entered into the analysis.
This two-part questionnaire includes six items, and it evaluates the occurrence of any violent behaviors in the past 12 months. The first part of the questionnaire, which includes 4 items, evaluates destructing objects, self-injury, violence against family, and violence against people outside the family. In the second part, two items evaluate the legal ramifications of the violence and the need for health-care services by the victim. This tool has a good inter-rater reliability among the general population of Iran (kappa coefficient = 0.6). In the current study, the participants who reported any of the violent behaviors in the first part of the questionnaire were registered as people with violent behaviors, and those reporting any of the items of the second part of the questionnaire were recorded as people with serious violent behaviors.
Social Capital Questionnaire
This questionnaire includes four parts. The first three parts include nine items, evaluating voluntary participation, a sense of belonging and friendship, and sense of trust. The questions are scored based on a Likert spectrum from never (a score of 1) to completely (a score of 5). The total score for each part ranges from 9 to 45. The fourth part, which includes only one item, is related to family/social support. This item is coded as “yes/no/not sure.”
Questionnaire on attitude to mental health help-seeking
This questionnaire includes four items. These items measure the attitude of the individual toward visiting psychotherapists, ease of communicating one's problems, sense of embarrassment because of stigma, and visiting a health professional in case of being addicted, respectively. The responses are classified into five categories, i.e., definitely yes, probably yes, probably no, definitely no, and not sure.
Sheehan Disability Scale
This scale measures the severity of the disability related to any mental disease. The visual analog measure of this scale ranks the severity of disability as no disability (=0), mild disability (=1–3), medium disability (=4–6), and severe disability (=7–10). This scale has been successfully used in population-based studies in Iran,, and its Persian version has acceptable validity and reliability.
The data related to continuous variables are reported as the mean and standard deviation, and the discrete data are reported in values and percentages. Before performing the main analysis, the components of suicide, including suicidal thoughts, plans, and attempts, along with violent behaviors were coded as No (=0)/Yes (=1). The score of psychological distress was entered into the analysis in the form of mean and standard deviation. Since there are both stratified and continuous variables in the study, two-step cluster analysis (TSCA) was performed for identifying clusters. This analysis method was used due to the large size of the sample and the presence of continuous and discrete variables. TSCA determines the importance ranking of the classification variables, which play a role in predicting the model, and determines the number of clusters automatically. The fitting of the model was determined based on Schwarz's Bayesian information criterion using the average silhouette coefficient. The silhouette coefficient is a measure of internal validity, which ranges from 0 to 1. The scores closer to 1 indicate that the model is better. At the next stage, suicidal and violent behaviors were compared among the clusters using the Chi-square test.
At the next stage, the multinomial logistic regression was performed for identifying the correlates of the derived clusters. All the sociodemographic variables (gender, age, marital status, education level, occupation), lifetime history of psychiatric and mental disorders, lifetime spread of abuse of any substance and alcohol abuse, and social capital (voluntary participation, sense of belonging and friendship, social/family support) were entered into the model simultaneously. Since there are three clusters, cluster 1 (healthy participants) was considered as the reference cluster, and adjustment was applied for sex and age. The results of the analysis are presented as adjusted odds ratios with 95% confidence intervals.
As the final step, the severity of disability and attitude toward mental health help-seeking in the entire population and for individual clusters were evaluated. The severity of disability was rated as no disability, mild disability, medium disability, and severe disability. Each one of the items in the questionnaire on attitude toward seeking help is defined in the form of classes (definitely yes, probably yes, probably no, definitely no, I don't know). To evaluate the significance of the difference between individual clusters and the reference cluster, single-variable Chi-square test was used. To do so, the frequency proportion of each cluster compared to the reference cluster was calculated and after assigning weights for the frequencies, the significance of the difference between the two clusters was calculated. All the statistical analyses were performed using SPSS20 (IBM Corp., Armonk, NY, USA) software application. All the tests had two ranges and statistical significance was defined as P < 0.05.
| Results|| |
[Table 1] presents the profile of the risky behaviors obtained from the TSCA, as well as the summary of the model, as can be seen from the table, the silhouette measure of cohesion and separation is completely acceptable. Suicidal thoughts and plans along with violent behaviors played the most prominent role in determining the clusters. Based on the results depicted in this table, there is a significant difference between the clusters with regard to all the components of risky behaviors and psychological distress (P < 0.001). This model proposes three clusters characterized as (i) low-risk group with mild distress, (ii) high-risk group with high distress, and (iii) violent group with medium distress. The members of cluster 1 (46.5%) do not show any suicidal or violent behaviors, and they have a mild level of psychological distress. On the other hand, suicidal and violent behaviors are seen in 26.2%–81.5% and 11.8%–72.6% of the participants in cluster 2 (19.9%), respectively. Moreover, psychological distress in this group was at an abnormal level. Finally, all the participants in cluster 3 (33.6%) only showed nonserious violent behaviors with a medium level of psychological distress.
|Table 1: High-risk behaviors profile derived from two-stage cluster analysis (n=2990)|
Click here to view
Sociodemographic correlates and family history related to the clusters
[Table 2] depicts the characteristics of participants based on individual clusters at the baseline. Moreover, this table shows the results of the multinomial logistic regression after adjustment for all sex and age. The information in this table indicates that among the demographic variables, there is a significant relationship between education level and marital status and the unhealthy cluster (cluster 2) (P < 0.05). In addition, there is a significant relationship between sex and cluster 3 (P < 0.05). Compared to the reference cluster (cluster 1), the participants in these clusters are male and they have lower education levels and a significantly higher number of divorces. Furthermore, the members of cluster 2 have a weaker social capital, and the lifetime prevalence of psychiatric disorders for these participants is 5.9 times higher (P < 0.05). On the other hand, the lifetime prevalence rate of psychiatric disorders in participants of cluster 3, compared to the reference cluster, is 1.6 times higher. Moreover, the score of sense of trust, as one of the components of social capital, in this group is lower than that of the reference cluster (P < 0.05).
|Table 2: The results of multinomial regression logistic for identifying correlates|
Click here to view
Comparing clusters with regard to severity of disability and attitude to mental health help seeking
[Table 3] depicts the results of comparing the clusters with regard to the severity of disability and attitude to mental health help seeking. As can be seen, compared to the reference cluster, the frequency of medium to severe disability in other groups, particularly cluster 2, is significantly higher. Compared to cluster 1, the participants in clusters 2 and 3 are less embarrassed and can more easily communicate their problems for a health professional; however, they are less inclined to mental health help seeking (P < 0.05).
|Table 3: The disability and help-seeking attitude separated by the clusters*|
Click here to view
| Discussion|| |
- Three clusters were identified based on suicidal and violent behaviors and psychological distress. A healthy cluster without any high-risk behaviors with mild psychological distress; a cluster with various high-risk behaviors and high psychological distress; and a cluster with only violent behaviors and a medium level of psychological distress
- Only 46.5% of the samples (cluster 1) do not show any high-risk behaviors. In other words, almost 20% of the total samples suffer from suicidal behaviors and 53.5% of the total sample (clusters 2 and 3) suffer from violent behaviors
- Sex, education level, marital status, lifetime psychiatric disorders, and components of social capital are the most important predictors of unhealthy clusters
- Compared to the reference cluster, a higher number of participants in unhealthy clusters suffer from medium to severe disability. While the participants in these clusters show that they are less embarrassed of the stigma and are capable of easily communicating their problems with a health professional, they are less inclined to mental health help seeking.
The maximum prevalence rates of violent and suicidal behaviors in the general population have been reported as 29% and 14%, respectively., In line with these studies, the prevalence of problems in the current study was high. In our study, the prevalence rates of violent and suicidal behaviors were close to 50% and 20%, respectively. One of the reasons behind the high rate of prevalence for risky behaviors in the sample is the age range of the participants. The target population of the previous studies was often adults from various age groups. This is while in our study, only adolescence and young adults aged 15–34 years were evaluated. Previous studies also confirm the high prevalence rate of risky behaviors in younger age groups.,, On the other hand, risky behaviors are increasing, particularly among Iranian youngsters.,
Our results show that there is a significant relationship between lower levels of education and the cluster with risky behaviors. Education is an important factor in the capability of directing health-care and effective utilization of social and economic resources related to health. Higher education levels also have a positive impact on personal health behaviors. On the other hand, previous research shows that the risk of suicide and high-risk behaviors is directly related to higher education levels., Based on the results of a number of other studies, there is a relationship between education level and risky behaviors.,
In the participants of cluster 2, with a number of high-risk behaviors and high levels of psychological distress, the rate of marriage was lower and the divorce rate was higher than the healthy cluster. Being married is a protecting factor against risky behaviors, particularly suicide, and divorce can increase the likelihood of such behaviors threefold., Divorce and the mental pressures due to loneliness and reduced social support depending on the cultural contexts and differences may risk the mental health of the individual. In particular, in societies such as Iran where there is a significant emphasis on marriage, divorce may pave the way for the manifestation of psychiatric problems. These individuals usually have a higher level of psychological distress, confirmed by the results of the current study. Therefore, divorce can be considered an important risk factor for the occurrence of risky behaviors.,
Another finding of the study indicates that there is a highly significant relationship between lifetime psychiatric disorders and unhealthy clusters. Chronic psychological distress and psychiatric disorders are the most important causes of risky behaviors, violence, and suicide., The results of recent review studies indicate that the level of violent and suicidal behaviors is significantly higher among people suffering from affective problems and psychotic disorders, particularly schizophrenia, compared to ordinary people.,
With regard to the current sample, the prevalence rate of lifetime abusing various drugs and alcohol abuse in unhealthy clusters (6.2%–9.4%) is higher than the reference cluster (2.4%). This finding is in line with the results of previous studies.,,,, Alcohol abuse, on its own, is considered a risk factor for violence and suicide, and if it is accompanied by other risk factors such as young age and being single, the risk will increase. Alcohol consumers drink alcoholic beverages to facilitate relations with others and increase compatibility with their environment and its challenges. In fact, alcohol is one of the options for escaping social-mental challenges, which after a while may be replaced with more dangerous options such as violent and suicidal behaviors.,, Besides alcohol, drug abuse will facilitate the occurrence of violent behaviors among the youth. It is worth mentioning that in the current study, there was no significant relationship between lifetime drug and alcohol abuse and unhealthy clusters. This is due to the very low prevalence rate of drug abuse among the participants (only 5.1% of the entire sample).
In line with previous reports,, it is found that there is a relationship between poor social capital and increasing violent and suicidal behaviors. According to Noguchi, social capital at the personal and social levels is an important protective factor against suicidal thoughts. Social capital is defined as participation, trust, relational networks, and cooperation, and it enables individuals to participate in social activities in a positive manner. It is obvious that in the case of poor social participation and trust, followed by isolation, the likelihood of the occurrence of violent behaviors will increase.
Our results show that the severity of disability in unhealthy clusters is significantly higher than the healthy cluster. Specifically, in cluster 2, the majority of participants are suffering from medium to severe disability. This disability is probably due to higher rates of divorce and loneliness,, psychological distress, substance and alcohol abuse, and poor social capital. In line with recent reports,, while high-risk participants suffer from a significant level of disability, they do not have a positive attitude toward mental health help seeking and are less inclined to receive any services. The presence of risky behaviors and lack of willingness to seek treatment create a dangerous situation both for the individual himself or herself and for their families and other social groups. It seems that this situation is due to a number of issues: poor insight of mental problems; severe depressive symptoms, and despair and helplessness in solving various crises; personality disorders such as antisocial personality and lack of remorse for violent behaviors; the presence of magical thoughts, megalomania, and thinking that you are capable of doing anything, which will lead to lack of requiring the support of a health professional. Since our results show that these individuals are less embarrassed by the stigma and are capable of easily communication their problems for a health professional, the above-mentioned possibilities are reinforced.
In general, it seems that the prevalence of risky behaviors among participants in western Iran is not that different from the prevalence in other parts of the world. Moreover, the cluster analysis can accurately distinguish and classify low-risk and high-risk participants. However, the point which is more important than other results is that people with risky behaviors do not have a positive attitude and willingness to seek psychological health-care services. This can endanger the health of the individual and the society in a major way and create dangerous and irreversible consequences for the society.
This study was started in 2014, so we utilized structured interviews based on DSM-IV. Therefore, it is recommended that future studies use the DSM5 format. The participants in this study only include about one-third of the sample of the national study, and if this analysis is repeated with the entire sample of 9 thousand participants, the accuracy and generalizability of the results can be improved. The current analysis was carried out based on the lifetime prevalence of suicidal thoughts and behaviors and the 12-month prevalence of violent behaviors. It seems that the lifetime prevalence of violent behaviors can provide more useful results. In the current study, the only attitude toward mental health help seeking was evaluated and investigated. It is recommended that future studies report and compare the real results of the participants' using psychological health-care services.
| Conclusions|| |
The current study was carried out to perform a cluster analysis of risky behaviors, determine their correlates, and compare clusters with regard to the severity of disability and attitude toward mental health help-seeking. Our model provided two high-risk clusters and a healthy cluster. Based on the obtained results, more than half of the youth are suffering from suicidal and violent behaviors, mainly caused by lifetime history of psychiatric disorders and poor social capital. Since compared to healthy participants, high-risk participants are less inclined to seek professional help, the health professionals and policymakers can successfully utilize the results of the current study in planning general health programs and maps.
We thank all those who helped to prepare this work.
Financial support and sponsorship
The project was funded by the Kermanshah University of Medical Sciences (code: 94524).
Conflicts of interest
There are no conflicts of interest.
| References|| |
Maghsoudi A, Jalali M, Neydavoodi M, Rastad H, Hatami I, Dehghan A. Estimating the prevalence of high-risk behaviors using network scale-up method in university students of Larestan in 2014. J Subst Use 2017;22:145-8.
Liu J, Lewis G, Evans L. Understanding aggressive behaviour across the lifespan. J Psychiatr Ment Health Nurs 2013;20:156-68.
Meyer RE, Salzman C, Youngstrom EA, Clayton PJ, Goodwin FK, Mann JJ, et al.
Suicidality and risk of suicide – definition, drug safety concerns, and a necessary target for drug development: A consensus statement. J Clin Psychiatry 2010;71:e1-21.
ten Have M, de Graaf R, van Dorsselaer S, Verdurmen J, van 't Land H, Vollebergh W, et al.
Incidence and course of suicidal ideation and suicide attempts in the general population. Can J Psychiatry 2009;54:824-33.
Gunnell D, Harbord R, Singleton N, Jenkins R, Lewis G. Factors influencing the development and amelioration of suicidal thoughts in the general population. Cohort study. Br J Psychiatry 2004;185:385-93.
Malakouti SK, Davoudi F, Khalid S, Ahmadzad Asl M, Moosa Khan M, Alirezaei N, et al.
The epidemiology of suicide behaviors among the countries of the eastern Mediterranean Region of WHO: A systematic review. Acta Med Iran 2015;53:257-65.
Cao XL, Zhong BL, Xiang YT, Ungvari GS, Lai KY, Chiu HF, et al.
Prevalence of suicidal ideation and suicide attempts in the general population of China: A meta-analysis. Int J Psychiatry Med 2015;49:296-308.
Mirhashemi S, Motamedi MH, Mirhashemi AH, Taghipour H, Danial Z. Suicide in Iran. Lancet 2016;387:29.
Hassanian-Moghaddam H, Zamani N. Suicide in Iran: The facts and the figures from nationwide reports. Iran J Psychiatry 2017;12:73-7.
Bakhshani N, Bahareh B, Bakhshani S, Lashkaripour K. Suicidal attempts among individuals seeking treatment for substance dependency. Procedia Soc Behav Sci 2010;5:1982-5.
Hallgren KA, Ries RK, Atkins DC, Bumgardner K, Roy-Byrne P. Prediction of suicide ideation and attempt among substance-using patients in primary care. J Am Board Fam Med 2017;30:150-60.
Meltzer H, Brugha T, Dennis MS, Hassiotis A, Jenkins R, McManus S, et al.
The influence of disability on suicidal behaviour. Eur J Disabil Res 2012;6:1-12.
Ahmed S, Stanciu CN. Addiction and suicide: An unmet public health crisis. Am J Psychiatry Resid J 2017;12:3-6.
Swahn MH, Donovan JE. Correlates and predictors of violent behavior among adolescent drinkers. J Adolesc Health 2004;34:480-92.
Pompili M, Serafini G, Innamorati M, Dominici G, Ferracuti S, Kotzalidis GD, et al.
Suicidal behavior and alcohol abuse. Int J Environ Res Public Health 2010;7:1392-431.
Ghanbari B, Malakouti SK, Nojomi M, De Leo D, Saeed K. Alcohol abuse and suicide attempt in Iran: A case-crossover study. Glob J Health Sci 2015;8:58-67.
Kim HS, Salmon M, Wohl MJ, Young M. A dangerous cocktail: Alcohol consumption increases suicidal ideations among problem gamblers in the general population. Addict Behav 2016;55:50-5.
Batty GD, Kivims228;ki M, Bell S, Gale CR, Shipley M, Whitley E, et al.
Psychosocial characteristics as potential predictors of suicide in adults: An overview of the evidence with new results from prospective cohort studies. Transl Psychiatry 2018;8:22.
Wright DR, Fitzpatrick KM. Social capital and adolescent violent behavior: Correlates of fighting and weapon use among secondary school students. Soc Force 2006;84:1435-53.
Kunst AE, van Hooijdonk C, Droomers M, Mackenbach JP. Community social capital and suicide mortality in the Netherlands: A cross-sectional registry-based study. BMC Public Health 2013;13:969.
Orlewska K, Orlewska E. Burden of suicide in Poland in 2012: how could it be measured and how big is it? Eur J Health Econ. 2018;19:409-17.
Cheong KS, Choi MH, Cho BM, Yoon TH, Kim CH, Kim YM, et al.
Suicide rate differences by sex, age, and urbanicity, and related regional factors in Korea. J Prev Med Public Health 2012;45:70-7.
Zhang Y, Ji Y. Clustering of risk behaviors and their social determinants among primary school learners in Beijing, China: A cross-sectional study. Chin Med J 2015;128:1567-73.
] [Full text]
Gellerstedt M, Petersen PH. Partitioning reference values for several subpopulations using cluster analysis. Clin Chem Lab Med 2007;45:1026-32.
Hale DR, Viner RM. Policy responses to multiple risk behaviours in adolescents. J Public Health (Oxf) 2012;34 Suppl 1:i11-9.
Khazaie H, Najafi F, Hamzeh B, Chehri A, Rahimi-Movaghar A, Amin-Esmaeili M, et al
. Cluster analysis of psychiatric profile, its correlates, and using mental health services among the young people aged 15-34: Findings from the first phase of Iranian youth cohort in Ravansar. Soc Psychiatry Psychiatr Epidemiol 2018;53:1339-48.
Rahimi-Movaghar A, Amin-Esmaeili M, Sharifi V, Hajebi A, Radgoodarzi R, Hefazi M, et al.
Iranian mental health survey: Design and field proced. Iran J Psychiatry 2014;9:96-109.
Sharifi V, Amin-Esmaeili M, Hajebi A, Motevalian A, Radgoodarzi R, Hefazi M, et al.
Twelve-month prevalence and correlates of psychiatric disorders in Iran: The Iranian Mental Health Survey, 2011. Arch Iran Med 2015;18:76-84.
Wittchen HU. Reliability and validity studies of the WHO – Composite International Diagnostic Interview (CIDI): A critical review. J Psychiatr Res 1994;28:57-84.
Kessler RC, Andrews G, Colpe LJ, Hiripi E, Mroczek DK, Normand SL, et al.
Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med 2002;32:959-76.
Dadfar M, Lester D, Atef Vahid M, Nasr Esfahani M. The psychometric parameters of the Farsi form of the Kessler psychological distress scale (K10) in psychiatric outpatients. Int J Med Res Health Sci 2016;5:1-7.
Sheehan DV, Harnett-Sheehan K, Raj BA. The measurement of disability. Int Clin Psychopharmacol 1996;11 Suppl 3:89-95.
Amin-Esmaeili M, Motevalian A, Rahimi-Movaghar A, Hajebi A, Hefazi M, Radgoodarzi R, et al.
The translation and psychometric assessment of the Persian version of the sheehan disability scale. Iran J Psychiatry 2014;9:125-32.
Garlow SJ. Age, gender, and ethnicity differences in patterns of cocaine and ethanol use preceding suicide. Am J Psychiatry 2002;159:615-9.
Bozzay ML, Liu RT, Kleiman EM. Gender and age differences in suicide mortality in the context of violent death: Findings from a multi-state population-based surveillance system. Compr Psychiatry 2014;55:1077-84.
Pompili M, Vichi M, Qin P, Innamorati M, De Leo D, Girardi P. Does the level of education influence completed suicide? A nationwide register study. J Affect Disord 2013;147:437-40.
Ansari-Moghaddam A, Ansari H, Mohammadi M, Khosravi S, Sanei-Moghaddam E, Miri-Bonjar M. Predictors of high-risk behaviors in municipal workers and staff in Zahedan, South-East of Iran. Int J High Risk Behav Addict 2016;5:e27555.
Sychareun V, Thomsen S, Faxelid E. Concurrent multiple health risk behaviors among adolescents in Luangnamtha province, Lao PDR. BMC Public Health 2011;11:36.
Yoshioka E. Suicide, socio-economic inequalities, gender, and psychiatric disorders commentary: Educational levels and risk of suicide in Japan: The Japan Public Health Center Study (JPHC) Cohort I. J Epidemiol 2016;26:277-8.
Masocco M, Pompili M, Vichi M, Vanacore N, Lester D, Tatarelli R. Suicide and marital status in Italy. Psychiatr Q 2008;79:275-85.
Fukuchi N, Kakizaki M, Sugawara Y, Tanji F, Watanabe I, Fukao A, et al.
Association of marital status with the incidence of suicide: A population-based Cohort Study in Japan (Miyagi cohort study). J Affect Disord 2013;150:879-85.
Robards J, Evandrou M, Falkingham J, Vlachantoni A. Marital status, health and mortality. Maturitas 2012;73:295-9.
Nawka A, Rukavina TV, Nawková L, Jovanović N, Brborović O, Raboch J. Psychiatric disorders and aggression in the printed media: Is there a link? A central European perspective. BMC Psychiatry 2012;12:19.
Pompili E, Carlone C, Silvestrini C, Nicolò G. Focus on aggressive behaviour in mental illness. Riv Psichiatr 2017;52:175-9.
Stoddard SA, Epstein-Ngo Q, Walton MA, Zimmerman MA, Chermack ST, Blow FC, et al.
Substance use and violence among youth: A daily calendar analysis. Subst Use Misuse 2015;50:328-39.
Noguchi M, Kobayashi T, Iwase T, Suzuki E, Kawachi I, Takao S. Social capital and suicidal ideation in Community-Dwelling older residents: A multilevel analysis of 10,094 subjects in Japan. Am J Geriatr Psychiatry 2017;25:37-47.
Okamoto M, Kawakami N, Kido Y, Sakurai K. Social capital and suicide: An ecological study in Tokyo, Japan. Environ Health Prev Med 2013;18:306-12.
Samokhvalov AV, Popova S, Room R, Ramonas M, Rehm J. Disability associated with alcohol abuse and dependence. Alcohol Clin Exp Res 2010;34:1871-8.
Castelli Dransart DA, Guerry S. Help-seeking in suicidal situations: Paramount and yet challenging. Interactions between significant others of suicidal persons and health care providers. J Clin Med 2017;6:17.
Calear AL, Batterham PJ, Christensen H. Predictors of help-seeking for suicidal ideation in the community: Risks and opportunities for public suicide prevention campaigns. Psychiatry Res 2014;219:525-30.
Research Center for Environmental Determinants of Health, School of Public Health, Kermanshah University of Medical Sciences, Kermanshah
Source of Support: None, Conflict of Interest: None
[Table 1], [Table 2], [Table 3]