Suicidal ideation, plan, and attempt among men who have sex with men in Nepal: Findings from a cross-sectional study

Men who have sex with men (MSM) are at increased risk for suicide, with a much higher prevalence of suicidality than the general population. While there is a growing interest in the identification of risk factors for suicidal behaviors globally, the understanding of the prevalence and risk factors for suicidal behaviors among MSM in the context of low- and middle-income countries is almost non-existent. Therefore, this study aimed to investigate suicidal ideation, plan, and attempts, and related factors among MSM in Nepal. A cross-sectional respondent driven survey was conducted on 250 MSM between October and December 2022. Bivariate and multivariable logistic regression was used to evaluate independent correlates of suicidal behaviors of MSM. Overall, the lifetime prevalence of suicidal ideation, plans, and attempts among MSM in this study were 42.4%, 31.2%, and 21.6%, respectively. MSM with depressive symptoms (aOR = 5.7, 95% CI = 2.4–14.1), advanced education (higher secondary and above; aOR = 2.9, 95% CI = 1.4–6.1), and smoking habit (aOR = 2.5, 95% CI = 1.2–5.3) were at increased risk for suicidal ideation. Similarly, those with depressive symptoms (aOR = 2.2, 95% CI = 1.1–4.8) and advanced education (aOR = 2.7, 95% CI = 1.2–5.7) were more likely to plan suicide, whereas young MSM were significantly more prone to attempting suicide (aOR = 2.7, 95% CI = 1.3–5.8). Interestingly, MSM with moderate to severe food insecurity were 2–3 times more likely to think about, plan, or attempt suicide (ideation: aOR = 3.5, 95% CI = 1.6–7.7; plan: aOR = 3.7, 95% CI = 1.6–8.3; attempt: aOR = 2.2, 95% CI = 1.1–4.6). The results suggest the importance of early assessment of suicidal behaviors among MSM and the need for tailored interventions to simultaneously address mental health problems and food insecurity to reduce suicide-related problems among Nepalese MSM.


Introduction
The World Health Organization (WHO) estimates approximately 703,000 people die by suicide annually globally [1].For every suicide, there are 20 other people likely to make a suicide attempt, and many more having serious thoughts of suicide.And of all global suicides, 77% occur in low-and middle-income countries (LMIC) [1], characterized by limited resources to prevent suicidal behaviors, such as suicidal ideation, suicide attempt, and suicides.In such settings, suicidal behaviors remain a low public health priority, so reliable and comprehensive data are unavailable [2].This makes it difficult to accurately assess the magnitude of suicidal behavior, particularly among sexual and gender minority groups (e.g., men who have sex with men; MSM), its associated risk factors, and the effectiveness of preventive strategies.
Previous studies have shown that members of the sexual and gender minority (SGM) group are at greater risk of suicidal behaviors than their heterosexual counterparts [3,4].In general, there is a wide range of risk factors that contribute to increased risk of suicidal behaviors, including life stress, stress perception, substance use, sociodemographic background, food insecurity, chronic conditions, and various forms of distress (e.g., depression, anxiety, hopelessness, pain), that apply to the general population [5][6][7][8][9].One possible explanation for the elevated risk of suicidal behaviors among MSM may be their experiences of stressors unique to SGM groups, such as the burden of disclosing their sexual identity, discrimination, stigma, institutional prejudice, social exclusion, and violence mainly based on their sexual orientation [10].The combined impact of these risk factors in SGM individuals increases their susceptibility to suicide compared to the risk factors experienced by the general population [11].The overall suicide rate in Nepal is 7.2 per 100,000 population.Among males, the suicide rate is higher at 8.2 per 100,000, while among females, it is relatively lower at 6.3 per 100,00 [12].However, suicidal ideation among gay and bisexual men in Nepal was 46.9% [13].
Despite the growing concerns about suicidal behaviors among MSM globally [14], much of this research has been conducted in high-income country settings [3,5,15,16].The extent to which these findings apply to SGM populations in LMIC settings, like Nepal, remains unknown.Particularly in Nepali culture, traditional gender roles significantly shape societal and familial expectations towards opposite-sex marriage and reproduction.As a result, MSMs who do not conform to these norms may be stigmatized and excluded by their families [17], contributing to an increased risk of suicide.Despite the growing number of suicides among MSM in Nepal [14], there is limited information regarding suicidal thoughts and conduct.So, there is a substantial need to scale research on mental health and suicidality among this population.This paper, therefore, aimed to examine the prevalence of lifetime suicide ideation, plan, and attempts, and associated factors among MSM in Nepal, a low-income country located in the WHO region of South-East Asia where 40% of all global suicides [2].The results of this study will provide an important insight into the prevalence of suicidal behaviors and may help to inform the development of targeted suicide prevention intervention strategies for MSM in Nepal and other low-income countries with similar social and cultural settings.

Study design and participants
Data were drawn from a cross-sectional study of 250 MSM conducted between October and December 2022 in Kathmandu Valley.The Kathmandu Valley is comprised of three districts: Kathmandu, Bhaktapur, and Lalitpur.Kathmandu district is the national capital, densely populated, and the largest metropolitan city, whereas Bhaktapur and Lalitpur are neighboring districts inside the valley.

Inclusion criteria
Individuals were eligible if they identified as MSM; were aged 18 years or older; understood Nepali or English; and were willing to undergo screening for HIV and Syphilis.

Exclusion criteria
The study excluded participants who did not come through referrals from previous study participants and those who did not identify as MSM, transgender males, or were below 18 years of age.

Study procedures
Respondent-driven sampling or a non-random sampling technique based on social networks and typically employed for populations that are difficult to access, was used to recruit study participants [18].We initiated the recruitment chains with five MSM "seeds", purposively selected based on recommendations from community-based organizations working with MSM, with attention given to socio-demographic and geographic representation.Each seed who completed the interviewer-administered questionnaire was given five recruitment coupons to recruit potential peers.Each successive participant received five coupons to enlist more peers in the study as shown in Fig 1.
Trained research assistants administered questionnaires to participants face-to-face using Qualitrics XM in a private room, which took approximately 40 minutes to complete.Each participant received an incentive of 1000 Nepalese Rupee (~USD 8) for study participation and an additional 500 Nepalese Rupee (~USD 4) for each of up to five eligible peers they successfully recruited into the study.

Study variables
The dependent variables in the study included suicidal behaviors.Suicidal behaviors were assessed using the suicidality module of the World Mental Health Composite International Diagnostic Interview (WHM-CIDI), a commonly used tool for measuring suicidal ideation, plan, and thoughts in different settings in many countries, including Nepal.This module includes an assessment of lifetime occurrences of suicide ideation ("Have you ever thought about committing suicide?"), plans ("Have you ever made a plan for committing suicide?"), and attempts ("Have you ever attempted suicide?") [19].The independent variable included information about socio-demographic, behavioral, and health-related characteristics (Table 1).The tools used in this study have been previously used in similar settings and similar study populations in Nepal [13,20].

Socio-demographic characteristics
Socio-demographic data included age, province, education, monthly income, religion, employment status, and sexual orientation.Age was dichotomized to <25 years and �25 years.Participants from all provinces other than Bagmati province were merged into a single category: outside Bagmati, and those from Bagmati province were included in the Bagmati province category.Participants' educational levels were divided into two categories: up to SLC and Intermediate and above, where those with education level of intermediate level or higher were merged into a single category.In addition, participants were asked to report their monthly income in Nepalese rupees (NPR), and this was reported with the categories: <Nrs.20,000(~USD 150) and �20,000 and above.The sexual orientation variable was assessed in 2 groups: Gay and Bisexual.Categories for collecting information on participants' religion-Hindu, Buddhist, and others-were adopted from Nepal's Health Management Information System.The participant's employment status was measured by yes/no responses to whether they were currently employed.

Behavioral characteristics
The behavioral characteristics included questions about whether the participants used substance within six months (yes/no), ever smoked (yes/no), used alcohol within 12 months (yes/ no), had condomless sex in the past six months (yes/no), ever engaged in sex work, i.e., involving in sexual activity by receiving money, materials or gifts (yes/no), disclosed sexual orientation (yes/no) and ever detained by the police (yes/no).

Health-related and psychological characteristics
Participants were asked whether they had ever tested HIV (yes/no) and STIs (yes/no) to know their status.Information on their health coverage was collected via the dichotomized response of yes/no.Three categories within the past 6 months, 6 months to 2 years, and more than 2 years were used to assess the participant's last visit to the doctor.Participants were also asked whether they worried about being negatively judged by the healthcare workers (yes/no).

Depressive symptoms
The Patient Health Questionnaire tool was used to assess the individuals' depressive symptoms.Each of the nine DSM-IV criteria received a score from "0" (not at all) to "3 (nearly every day) using the PHQ-9 instrument.A composite score of 0-27 was calculated, and a score of >10 was regarded as having depressive symptoms [21].The Cronbach's alpha value for the PHQ-9 was 0.85.

Daytime sleepiness
An 8-item Epworth Sleepiness Scale was used to assess daytime sleepiness.On a 4-point scale (0-3), participants were asked to rate their usual chances of dozing off or falling asleep while

Food insecurity
Participants' food insecurity was measured using an 8-item Food Insecurity Experience Scale (FIES), which included questions regarding participant's food-related behaviors and experiences associated with increasing difficulties in accessing food due to scarce resources.
Responses to the 8 questions were simple binary choices of "Yes" or "No".Total scores for these responses, which ranged from 0 to 8, were added together.And the final scores were classified into 2 categories: food secure (0-3); and severe/moderately food insecure (4-8) [23].The Cronbach's alpha value for the FIES was 0.94.

Experience of violence
Experience of past-year intimate partner violence among participants was assessed based on the 4-item Hurt, Insult, Threaten and Scream (HITS) screening tool.Responses to each question were recorded with a 5-point frequency format: never, rarely, sometimes, often, and frequently, and the score values ranged from 4 to 20.The final scores were categorized as: Normal (0-10) and Violence (11-25) [25].The Cronbach alpha value for the HITS was 0.86.

Data analysis
The statistical software IBM SPSS Version 26.0 (IBM Corp, New York, USA) was used for data analysis.Descriptive statistics were used to summarize the data, including frequency and percentages for categorical variables and mean and standard deviation for continuous variables.The Chi-square test was used to assess the relationship between categorical independent and dependent variables.Multivariable logistic regression analysis was conducted to identify potential factors associated with the outcome variable.The adjusted odds ratio (AOR) was calculated with a 95% confidence interval (CI), and a p-value below 0.05 was statistically significant.Bivariate analysis was used to identify significant variables for inclusion in the adjusted regression analysis, with a 10% significance level used as the criterion for inclusion [26].
We tested for multi-collinearity by calculating the variance inflation factor (VIF scores) for each variable in the predictor's models.We set the cutoff VIF score of 5, and no variable was found to have a higher VIF score greater than 5.The model with the lowest Akaike Information Criterion (AIC) value was selected as the best-fitted model.The likelihood ratio and Hosmer Lem show tests were for the goodness of fit of the model.The ROC curve with AUC (area under the curve) of suicidal ideation, plan and attempt with adjusted independent variables was reported on S1-S3 Tables and figures on S1-S3 Figs respectively.The regression model was explained by the equation: Where Y is the expected probability for the outcome variable to occur, b0 is the constant/ PLOS GLOBAL PUBLIC HEALTH intercept, b 1 through bn are the regression coefficients and X 1 through X n are distinct independent variables, and e is the error term.

Result
Socio-demographic, behavioral, and health-related outcomes Participant characteristics are described in Table 2.The mean age of participants was 27.6 (SD = 8.9) years, and over half of the participants were single (64.4%), unemployed (56.6%), and had completed a high school degree and above (58%).The majority of participants reported having used illegal substances within the past 6 months (84.8%), ever smoked tobacco (72.8%), and used alcohol in the past 12 months (71.6%).Almost half of the participants reported engaging in condomless sex in the past 6 months (48.4%).Of the total participants, 19.6% were found to have depressive symptoms, 49.6% had poor social support, and 29.2% had either moderate or severe food insecurity.

Correlates of suicide ideation, plans, and attempts
Fig 2 presents the lifetime prevalence of suicidal behaviors among study participants.Specifically, 42.4% of the participants reported suicidal ideation, 31.2% reported plans, and 21.6% reported attempts of suicide.Among the participants who had thoughts about ending their life in the past 12 months (n = 106), 22.6% thought about it multiple times.

Discussion
Although the rate of death by suicide among MSM is increasing, little is known about suicidal behaviors among members of SGM populations (e.g., MSM) in LMIC settings, including Nepal.To this end, we aimed to determine the prevalence and factors associated with suicidal ideation thoughts, plans, and attempts among MSM in Nepal.The lifetime prevalence of suicidal ideation (42.4%), plans (31.2%), and attempts (21.6%) among Nepalese MSM were alarmingly high compared to those from previous studies conducted in Nepal [14,27] and other geographic settings [5,15,28].Not surprisingly, these rates among MSM were higher than among Nepal's general population, i.e., 13.3% suicidal ideation and 9.4% attempt [2,12].It is likely that discrimination, prejudice, stigma, and mental health issues brought about by their sexual orientation may have been attributed to the increased prevalence [16,29,30].Prior studies have indicated that individuals who reveal or self-identify as homosexual or bisexual, the predominant case for participants in this investigation, may experience increased susceptibility to suicide due to feelings of harm, shame, guilt, social exclusion, and loss of support [31].In addition, Nepal faced a significant impact of COVID-19 [32], which increased the risk of infection, isolation, and a stressful environment, contributing to a higher prevalence of suicidal behavior than before [33].Moreover, factors such as financial stress, domestic violence, food insecurity, and limited or variable access to healthcare services might have significantly increased suicidal behavior [33].Furthermore, discrimination against the MSM community in Nepal is prevalent in various sectors, including employment and education.The state institutions' recruitment criteria do not accept the "O" category as a gender choice, which deprives LGBTQ+ individuals of job opportunities in civil service, the army, and the police.This discrimination exacerbates the challenges faced by LGBTQ+ individuals, leading to feelings of hopelessness and despair [34].In addition, the legal system in Nepal does not recognize same-sex marriage, which means that MSM couples cannot legally marry.This lack of recognition of their fundamental rights contributes to the marginalization of LGBTQ+ individuals and further exacerbates their mental health issues.Heterosexual married couples in Nepal can enjoy certain benefits, such as jointly buying property and adopting children, that are unavailable to homosexual couples.This inequality in legal rights reinforces the stigmatization of the MSM community and can lead to an increased feeling of isolation, depression, and suicidal behavior [35].
Notably, the present study found a significant association between food insecurity and suicidal behaviors among MSM, consistent with previous studies' findings [6,36].The mechanism that links food insecurity and suicidal behaviors has been explained in some prior studies [7,37].For example, food insecurity can make it harder for individuals to access social support and other resources, leading to social isolation [38].In addition, food insecurity could be a source of embarrassment, anxiety, and stress, which may be responsible for the exacerbation of mental disorders and, eventually, increase the risk for suicidal behaviors [14,39,40].Alternatively, due to stigmatization and discrimination, MSMs are less likely to find employment opportunities which might result in food insecurity and suicidal behavior [37,41].The government of Nepal and concerned stakeholders could work to develop and implement targeted food assistance programs, like food banks, meal delivery services, or other forms of assistance, for MSMs who are struggling with food insecurity.
Our findings unveiled those depressive symptoms significantly affected suicidal ideation and plans.These results align with that of previous studies [1,42,43].According to the American Foundation for Suicide Prevention (AFSP) and WHO, depression is a significant risk factor and a warning sign for suicide [1,44].The link between depression and suicide may be partially explained by the effects of depression on the central nervous system, specifically the disruption of 5-hydroxytryptamine (5-HT) metabolism and resultant 5-HT deficiency and serotonergic hypoactivity [45].This disrupted 5-HT metabolism is closely associated with an increased risk of suicide [46].Also, barriers to treating depression worsen psychiatric outcomes, including treatment resistance and increased risk for suicide [40].Research has consistently demonstrated a strong association between depression and suicidal behavior, including suicidal ideation and suicide plans [39].Mental health services are often not adequately  depression, psychosis) in its Basic Health Service Package of 2018 [48].There is a need for mental health evaluation and promotion specific to SGM populations [14].
As in previous studies [49,50], younger MSMs were significantly associated with suicidal ideation and plan.One reason might be that younger MSM are more likely to be in the process of discovering and coming to terms with their sexual orientation, which can be a difficult and challenging process [42].They may be more likely to experience feelings of confusion, isolation, and rejection, which can contribute to adverse mental health outcomes, such as suicidality [51].Additionally, younger MSM may be less likely to have access to social support networks, including other members of the LGBTQ+ community, which can leave them feeling isolated and alone [51].Another reason could be that young MSMs are more likely to engage in risky behaviors, such as substance use or unprotected sex, which could contribute to poor judgment and increase the risk of suicidal thoughts and plans.
Notably, MSM with higher educational status had higher odds of suicidal ideation and plan, contrasting with previous studies' findings [52].While our findings do not elucidate the pathways between education attainment and suicidal behaviors, it is possible that other variables may mediate this relationship (e.g., psychiatric disorders, feeling anxious and depressed, tobacco smoking, and alcohol consumption).For example, alcohol consumption and/or tobacco use may be some of the possible health-behavioral pathways linking educational status and suicidal behaviors [53,54].Not surprisingly, smoking was identified as a risk factor for suicidal ideation, as reported in previous studies [54][55][56].Epidemiological investigations conducted in the past have suggested that smoking is associated with a cluster of maladaptive behaviors linked with different psychological disorders and risky activities, including substance and alcohol use, as well as sexual and physical abuse.These behaviors are regarded as significant risk factors for suicide [55,[57][58][59][60][61][62].It is possible that a portion of the observed association between smoking and suicide risk can be attributed to the direct impact of smoking on biological pathways.Specifically, smoking has been shown to reduce the activity of the serotonergic system within the human hippocampus, which may decrease brain serotonin function.This decrease in serotonin function has been linked to increased suicide risk [63][64][65].The results of this study indicate that suicide prevention efforts should include smoking prevention and cessation initiatives as a key component.

Limitation
There are several limitations to this study that should be acknowledged.The first limitation is that the study was conducted during the COVID-19 pandemic, so pandemic-related conditions might still influence suicidal behaviors.A longitudinal study design would be necessary to fully understand the pandemic's impact on mental health outcomes.Given the nature of the cross-sectional design, the results should be interpreted solely as associations, and the role of causality may not be inferred.Another limitation is the potential for respondent bias, as the findings were based on subjective reports from MSM participants.Additionally, the study was conducted among MSM living in the Kathmandu Valley, the capital city, so the results cannot be generalizable to MSM in other parts of Nepal [66].Lastly, since our study was conducted within a hard-to-reach population, we encountered challenges related to sample size constraints.Consequently, there may be instances where certain variables exhibit AUC less than 0.5, which could suggest limited discriminatory power.However, it is important to note that our model selection process considered multiple factors, including AIC and VIF, which collectively supported the suitability of our chosen multivariable logistic regression model.This criterion helped us the best fitting model within the constraints of our data set and research context and while the sample sized posed challenges, our approach aimed to maximize the informativeness of the available data.Despite these limitations, the study still provides important evidence of suicidal behavior among MSM, which should be of interest to policymakers and other stakeholders working with MSM communities.

Conclusion
Our findings indicated an increased risk for suicidal behaviors among MSM in Nepal.The results show the importance of early assessment of suicidal behaviors among MSM and the need for tailored interventions to simultaneously address mental health problems and food insecurity to reduce suicide-related problems among Nepalese MSM.

Table 3 . Factors associated with suicidal ideation among study participants (N = 250).
[47] part of care and treatment services in Nepal among MSM.The lack of social welfare net in Nepal presents a significant obstacle to providing mental health services, as payment for most such care must be made out-of-pocket[47].It is only recently that the Department of Health Services (DoHS) included treatment for certain mental health conditions (e.g., available