Abstract

Introduction

Suicidal ideations (SI), also known as suicidal thoughts, refer to impulses, desires, and obsessions related to death. Prevalence of suicidal ideation was 14 percent. The current study assumed that identifying the true predictors of SI would allow for a greater understanding of suicide risk.

Methods

In this cross-sectional population-based study, 17,644 residents aged from 15 to 90 years were selected through a multi-stage sampling method from 22 districts of Tehran. Using hypothesized causal models, the pathways through which various variables influenced the components of SI were identified. Also, the applicability of the compensatory and risk-protective models of resiliency for the prediction of SI was tested by using the interaction multiple regression analyses.

Results

SI was experienced by 13.44% of the study population. SI was more prevalent听with individuals between the ages of 15 to 24. There are no differences between men and women when covariate analysis is used. The significant predictions by anxiety and physical activity (moderate) indicate support for the compensatory model for male and female, while the significant prediction by those indicates support for the risk-protective model for females.

Conclusion

PA would both lessen the detrimental impact of anxiety on suicidal thoughts and potentially reduce the probability of SI. It's important to develop and evaluate PA-enhancing treatments, especially for woman who are experiencing anxiety.

Introduction

Suicidal ideations (SI), also known as suicidal thoughts or ideas, refer to a range of impulses, desires, and obsessions regarding death [1]. Suicide is the fourth leading cause of death among 15鈥�29-year-olds [2] and each year more than 700,000 people (represents a global suicide rate of 9.0 (13.7 men, 5.4 women) per 100,000 population take their own lives globally [3]. In addition to that prevalence, six more persons are negatively impacted by suicide in significant psychological, social, and economic ways. According to the World Health Organization (WHO), someone attempts suicide every three seconds and commits suicide every forty seconds around the globe [4]. Suicide is a major issue affecting public health that accounts for 1% of the global disease burden [4] and has increased by 60% over the past 45 years [5]. Middle east and Africa regions are among the most attempted suicide rates. Iran is at the top of the suicide and suicide attempt rates among all the Middle Eastern countries with rates of 5.2 and 193.49 per 100,000 people respectively. Also, a history of chronic diseases such as mental health issues, low socioeconomic status, low grades of education, and drug addiction are the most important risk factors for suicidal attempts [6,听7].

A recently conducted systematic review in Iran showed the prevalence of suicide death was 8.14 and the prevalence of suicide attempts in the whole population was 131 (152 per 100,000 women and 128 per 100,000 men) per 100,000 people [8].

Since suicide can be prevented, WHO stated that conducting empirical research on suicide risk factors in order to identify particularly vulnerable individuals is of the utmost importance. Over the past few decades, numerous studies have investigated the associated risk factors and showed that SI is predictive of attempted and completed suicides [9,10,11]. Female gender, parental psychopathology, childhood family adversities, unemployment and prior history of suicidality can be considered as a predictor or risk factors for suicidal behaviors (i.e. suicidal ideation, plans and attempts) in both developed and developing countries [12].

Regarding adolescents, gender difference and adverse childhood experiences (e.g. physical and sexual abuse, bullying, parental neglect) can be considered as potential risk factors for SI in low- and middle-income countries [13].

In other hand despite 50 years of research, prediction accuracy for outcomes has only been slightly better than chance. No broad or subcategory reliably predicted outcomes, and studies rarely considered multiple risk factors [14].

Efforts have been made to identify interventions that effectively reduce SI risks. Although most of the interventional studies have been done in high-income countries with the USA and South Korea at the top, nearly 80% of the suicides in the world occurred in low/middle-income countries [15]. Also, especially walking and jogging are the two significant interventions that decrease the SI index in both genders. Yoga and strength training are the most effective for older and younger people respectively [16]. In a study conducted in the southwest of Iran among older adults, it was found that the prevalence of suicidal ideation is higher among women. However, this association does not remain significant when considering independent factors (it was significant in the crude model but not in the multivariate or adjusted model [17]. Nevertheless, the majority of studies have consistently noted that gender differences are associated with a female preponderance in suicidal ideation [18].

Regarding culturally unique suicidal strategies, one qualitative study attempted to identify themes based on cultural contexts in Iran. This study revealed five categories that explain suicide attempts, including personal factors and life experiences, family dynamics, social and educational influences, psychological-emotional problems, and stress coping strategies [19].

Prior research was inspired by medical understanding and frameworks, and thus focused primarily on medical diagnosis and treatment of depression [20]. Indeed, cognitive behavioral therapy and some antidepressants have been identified as effective interventions for depression. Protective factors include certain coping strategies (e.g., problem-focused coping and emotion-focused coping), [21] social support, and self-esteem [22]. Additionally, exercise or physical activity can act as a protective measure against suicidal thoughts and attempts. Kim et al. found that moderate physical activity, as per official guidelines [23], is negatively associated with SI [24]. The impact of exercise on reducing mental disorders such as depression and anxiety may also indirectly contribute to lower rates of suicidal behavior [25].

Previously risk factors such as social anxiety have been identified for suicide attempts in emerging adulthood [26]. Studies have shown significant negative correlations between all levels of PA and suicidal behavior among older people [27, 28]. Resiliency models proposed for assessing the relationships between risk and protective factors and adjustment can provide valuable insights into the protective role of PA in the relationship between mental health issues and suicidal ideation.

At least four generic resiliency models currently exist, including the compensatory model, the risk-protective model, the challenge model, and the protective-protective model [29,30,31]. According to the compensatory model, risk and protective factors have additive effects on maladjustment. According to the risk-protective paradigm, maladjustment is related to the interaction between risk and protective factors. Specifically, higher levels of the protective factor could mitigate the effect of risk factor on maladjustment [32].

Since suicidal ideation logically precedes suicidal acts, the current study assumed that identifying the true predictors (after adjusting by the pertinent variables) of suicidal ideation would allow for a greater understanding of suicide risk. The purpose of this study was to understand how anxiety and physical activity interact in predicting suicidal ideation in a large population that are likely applicable to the entire population of men and women across various age groups.

Methods

The studied sample was drawn from the Urban HEART project's second round. It developed by the WHO Centre for health development located in Kobe, Japan, may provide an example of efforts to reduce urban inequality. Urban HEART helps countries and districts to systemically producing evidence to evaluate and address to unfair health conditions and inequity in urban settings [33]. The Urban Health Equity Assessment and Response Tool (Urban HEART) project was a large population-based cross-sectional survey conducted in Tehran, Iran. In-home interviews were conducted with the respondents. Before collecting data, all interviewers attended a two-day training workshop [34].

Participants

Men and women over the age of 15 from the population of Tehran were included in the current study, with the final sample consisting of 17,644 individuals (54% being female) who correctly completed all questionnaire sections. Respondents were stratified and randomly selected (regardless of their professions) to test the questionnaire. The questionnaires were asked in an interview which took around 25 min within the household. If the respondent was illiterate or unable to self-administer the questionnaires, they were completed through an interview. The Urban HEART-2 cross-sectional survey was conducted in 22 districts of Tehran, Iran's capital city and includes 3,4700 households. Our study was approved by the Ethics Committee of Tehran University of Medical Sciences.

Urban HEART-2 (UH-2) based on response-oriented and assess the equity of health in Tehran city, which can demonstrate improved visions to decision-makers in health policy and lead to an improvement in community health. Six action areas were defined which included 鈥減hysical environment and infrastructure鈥�, 鈥渟ocial and human development鈥�, 鈥渆conomics鈥�, 鈥済overnance鈥�, 鈥渃ultural鈥� and 鈥渉ealth and nutrition鈥�.

Data collection

Each of Tehran's 22 districts and 380 localities was classified as a stratum (with the exception of six neighborhoods that were inaccessible for security reasons and therefore excluded from the study). Because some of the respondents were illiterate, the interviewers completed the GHQ by reading the items audibly to all interviewees. For some subjects, repeated questioning was required to obtain an answer. Totally 34,700 families from 22 districts in Tehran were covered by this population-based survey [35] and more than 22,500 participants responded to our health questionnaire (rate鈥�=鈥�65%). After removing records with missing and invalid data from 22,500 initial health-related datasets, 17,644 participants remained for data analyses (rate鈥�=鈥�78%) Fig.听1. (Study sample).

Fig.听1

Study sample

Variables

In the demographic section of the questionnaire, participants reported their age, relationship status, highest education level achieved, occupation, and number of family members.

Socio-demographic variables

Age of participant were categorized in three group; 15鈥�29 years 鈥測oung adult鈥�, 30鈥�60 years 鈥渁dults鈥� and over 60 years as elderly people. Marital status was defined in three categories namely; never married, married and previously married (divorced/widow). The education of responders was categorized in three levels; illiterate/primary education, secondary education and diploma and higher education. Occupation status in this study was defined in three subgroups including; employed, unemployed and retired. The economy level was defined as sufficiency of family鈥檚 income to support their life costs and categorized in low, middle and high levels. Chronic illnesses in the family include diabetes, cancer, myocardial infarction, stroke, asthma, hypertension, and osteoporosis.

The participants were requested to provide information regarding the number of family members living in their household or a total number of family members, in order to determine the size of their family. The coding scheme assigns a value of 1 to indicate a small family with 1 to 3 members, a value of 2 to represent a medium family consisting of 4 or 5 individuals, a value of 3 for a large family comprising6 or 7 individuals, and a value of 4 indicating a very large family size consisting of 8 or more individuals [36].

GHQ-28

In community and non-psychiatric clinical settings, the General Health Questionnaire (GHQ) is the most commonly used screening instrument for detecting psychiatric disorders [37]. In the present study, the available Farsi version of the GHQ-28 was used [38]. A review of studies on the validity of the GHQ鈥�28 in various countries, including Iran, demonstrates its high validity and reliability as a community screening instrument for mental disorders [39]. Four 7-item subscales are: somatic symptoms, anxiety and insomnia, social dysfunction, and severe depression. It permits the evaluation of mental health on four dimensions corresponding to these four scales. Regardless of version type, the structure of every query is always identical. To derive scale scores with more variability, we used Likert scoring (0123) instead of the original scoring procedure (0011) [40].

Suicidal ideation

Suicidal ideation was assessed on the basis of responses to 4 questions contained in the 28-item GHQ [37]. The subjects were instructed to respond to queries about how they had felt "over the past few weeks" as per standard GHQ protocol. Responses of 鈥渘ot at all鈥�, 鈥渘o more than usual鈥�, 鈥渞ather more than usual鈥� or 鈥渕uch more than usual鈥� were recorded for 2 questions: 鈥淗ave you recently felt that life is not worth living?鈥�, and 鈥淗ave you recently found yourself wishing you were dead and away from it all?鈥�. Similarly, responses of 鈥渄efinitely not鈥�, 鈥淚 don鈥檛 think so鈥�, 鈥渉as crossed my mind鈥� or 鈥渄efinitely has鈥� were recorded for the other 2 questions: 鈥淗ave you recently had thoughts of the possibility that you might do away with yourself?鈥�, and 鈥淗ave you recently found the idea of taking your own life kept coming into your mind?鈥� A binary scoring method was used that negative response; 鈥渘ot at all鈥�, 鈥渘o more than usual鈥�, 鈥渄efinitely not鈥� and 鈥淚 don鈥檛 think so鈥� given 0 score and 1 score was assigned to other options (positive response) [41]. By adding the binary scores for each of the 4 questions, giving a score from 0 to 4, a scale of suicidal ideation can be devised, and this can be used to compare subjects鈥� responses with other instruments employed. Thus the degree of suicidal ideation can be related to measures of hopelessness [42] self-esteem [43] anomie [44] depressive affect [43] and locus of control [45] as well as the subscales of the GHQ [46].

Anxiety

The Likert scoring method, ranging from 0 to 7, was employed to assess levels of anxiety within the subdomain of the GHQ-28. This approach allowed for the determination of anxiety scores, ranging from minimum to severe. Total sum scores were created for anxiety (7 items 8鈥�14). Responses of 鈥渘ot at all鈥�, 鈥渘o more than usual鈥�, 鈥渞ather more than usual鈥� or 鈥渕uch more than usual鈥� were recorded for 7 questions: 鈥渓ost much sleep over worry?鈥�, 鈥渉ad difficulty in staying asleep more once you are off?鈥�, 鈥渇elt constantly under strain?鈥�, 鈥渂een getting edgy and bad-tempered?鈥�, 鈥渂een getting scared or panicky for no good reason?鈥�, 鈥渇ound everything getting on top of you?鈥�, and 鈥渂een feeling nervous and strung-up all the time?鈥�.

Physical activity

GPAQ

To measure physical activity, we utilized consumption of energy Metabolic Equivalents (MET)-min/week and physical activity level (low, moderate, and high) [47]. The following categories of physical activity are defined in terms of MET-mins per week:

High level: vigorous-intensity activity in at least 3 days with a consumption of at least 1500 MET-min/week or a combination of walking and moderate or vigorous-intensity activities in at least 7 days with a consumption of at least 3000 MET-min/week. Moderate level: vigorous-intensity activity for at least 20 min in 3 or more days moderate-intensity activity or walking for at least 30 min in 5 or more days or a combination of walking and moderate or vigorous-intensity activity in at least 5 days with a consumption of at least 600 MET-min/week. Low level: not meeting the criteria for either high or moderate level [48].

GPAQ

Physical activity variables were derived from a valid and reliable Persian version [47, 49] of the Global Physical Activity Questionnaire (GPAQ- version 2) [48] developed by WHO, which has demonstrated high validity and reliability for assessing physical activity in a national surveillance in middle-income developing countries [50].

Statistical analysis

Data preprocessing and analysis for this study were conducted using Stata version 17. Following the rectification of unusual and missing data, categorical variables were characterized using frequency and percentage measures. The examination of SI patterns entailed tabulation, employing a Chi-squared test to explore associations at the socio-demographic level in men and women separately. To ascertain the linkage between socio-demographic factors, physical activity level, anxiety, domestic violence, tabaco use (as covariates) and SI, a two-step logistic regression analysis was executed. Initially, a univariable logistic regression was carried out over gender subgroups. Subsequently, employing the forward method, a multivariable logistic regression analysis was performed to evaluate the adjusted impact of each factor on SI within the study cohort. The independent variables considered encompassed age, gender, marital status, family size, occupation, housing situation, economic status, family tobacco use, family addiction, family violence, anxiety, and physical activity. SI was considered as a Primary outcome in this study. A confidence interval of 95% was applied, and statistical significance was determined at a threshold of p鈥�&濒迟;鈥�0.05.

The compensatory and risk-protective models were evaluated using the previously proposed multiple regression techniques [30, 32]. Each model was evaluated separately for men and women. In the same analysis, the compensatory and risk-protective models were evaluated by regressing the outcome measure (suicidal ideation) on the risk factor (anxiety), the relevant protective factor (moderate PA), risk factor鈥壝椻�塸rotective factor. Significant prediction by both the risk and protective factors indicates support for the compensatory model.

PA was determined as a quantitative variable with low was 1 and moderate was 2 and excluded high levels. Anxiety was determined as an ordinal variable. Interaction multiple regressions was tested. The compensatory model would suggest that anxiety will contribute positively, while PA will contribute negatively to the prediction of suicidal ideation [51, 52].

Results

A total of 17,644 individuals participated in the study, with 46% (8133) being male and 9515 being females. The age spectrum of participants spanned from 15 to 90 years, with the majority falling within the 25 to 34-year bracket. Among the male participants, 73% were married, 20% were unemployed, and 44% reported a middle-income level. In the female participants most of them being married, and reported low economy level (46%) with unemployment status (85%). The educational distribution revealed that a significant proportion possessed secondary education (53% in males and 58% in females). When considering aspects related to mental health and its determinants, suicidal ideation (SI) was reported by 12.53% of the men and 14.23% of women. Additionally, anxiety symptoms were disclosed by 31.95% and 38.43% of the men and women, respectively. Descriptive socio-demographic and other pertinent characteristics of the study participants are outlined in Table听1.

Table 1 The related characteristic of study population by SI in gender subgroup

Each of the variables detailed in Table听1 exhibited a substantial association with the prevalence of suicidal ideation (SI) both in men and women. For instance, individuals aged 15 to 29 demonstrated the highest SI prevalence in contrast to other age cohorts (p鈥�<鈥�0.001). Similarly, individuals who were previously married or never married, those with lower familial income, limited engagement in PA, those exposed to familial violence and tobacco use, as well as individuals experiencing anxiety, exhibited a direct positive relationship with SI prevalence within the population of men and women. Domestic violence and tabaco use were more related with SI prevalence in women compared to men (27.6% and 19.7% vs. 22.4 and 16.4).

Table 2 Univariable Logistic regression of SI prevalence by related factors in men and women

These associations are visually depicted in Fig.听2, which illustrates the intricate patterns of SI prevalence across subgroups defined by age, education, family-size, and economy level.

Fig.听2

Prevalence (%) of SI by related factors (age_cat, education, economy and family_size) in males and females

The outcomes of univariable and multivariable sex-specific logistic regression analyses are provided in Table听2 and Table听3. Within age categories, men aged 15 to 29 years exhibited a1.6-fold increased odds of experiencing SI in comparison to those older than 60 years (OR鈥�=鈥�1.67, p鈥�<鈥�0.001). Also, for women this association were declarable (OR鈥�=鈥�1.44, p鈥�<鈥�0.001). This association remained statistically significant only in women after accounting for other covariates within the model (OR鈥�=鈥�1.68, p鈥�<鈥�0.001). Conversely, no significant probability of SI was discerned among the remaining age groups relative to the reference group of individuals older than 60 years in the adjusted model both in men and women.

Table 3 Multivariable logistic regression of SI prevalence by related factors in men and women

Individuals classified as previously married or never married exhibited an elevated Odds of SI in both gender in crude model but in the multivariable model only women shown this relationship in terms of marital status and SI (OR鈥�=鈥�1.50, p鈥�<鈥�0.001). Notably, a small and very large family size demonstrated a positive association with SI prevalence when compared to medium family sizes in univariable analysis. But in the adjusted model only women in very large family had more Odds of SI compared to those who in the medium family size (OR鈥�=鈥�1.37, p鈥�=鈥�0.02). Furthermore, unemployment emerged as a factor associated with an increased odds of SI in men but not in women, with an average elevation of 1.37 times compared to retired individuals, even after adjusting for other pertinent variables (OR鈥�=鈥�1.37, p鈥�=鈥�0.006). Among the observed associations, one of the most robust was between SI and anxiety (OR鈥�=鈥�10.43, p鈥�<鈥�0.001 in men and OR鈥�=鈥�11.57, p鈥�<鈥�0.001 in women). Despite the positive relationship between factors such as; family economy, tobacco use, domestic violence, and SI within the adjusted model in both gender groups, a contrasting effect was apparent for moderate levels of physical activity (PA). Compared to the category of low activity, moderate PA exhibited a protective influence on SI (OR鈥�=鈥�0.80, p鈥�=鈥�0.003 in men and OR鈥�=鈥�0.85, 辫鈥�=鈥�0.02 in women).

The results of the analysis for testing the compensatory and risk- protective models in which Physical activity served as the protective factor (Table听4). The nonsignificant predictions of PA failed to support the compensatory model for males (OR鈥�=鈥�1.65; CI 95% [1.63鈥�1.75]). The significant prediction by anxiety鈥壝椻�塒A (moderate) indicates a support for the risk-protective model for males. This risk protective model shows that the odds of suicidal ideation decreased by an average of 27% for each unit increase in PA.

Table 4 Results of testing the compensatory model and risk-protective model to predict SI prevalence for males

Table 5 showed that the results of the analysis for testing the compensatory and risk-protective models in females, which Physical activity served as the protective factor for females. The significant predictions by both anxiety and PA (moderate) indicate support for the compensatory model, while the significant prediction by anxiety鈥壝椻�塒A (moderate) indicates support for the risk-protective model for females. Furthermore, Figs.听3 and 4 show the anxiety鈥壝椻�塒A interaction effect on SI in the males and females respectively.

Table 5 Results of testing the compensatory model and risk-protective model to predict SI prevalence for females

Fig.听3

Suicidal ideation as a function of anxiety鈥壝椻�塒A for males

Discussion

To our knowledge, there is little information on the relationship between SI and its protective or risk variables (risk factors: being unemployed, having a history of family addiction, experiencing familial violence, and belonging to a larger family size and protective factors decrease the likelihood of experiencing SI or mitigate the impact of risk factors (like moderate PA and having a moderate family size)) in a large sample of individuals with suicidal ideation, as well as the application of two resilience models to SI (how risk and protective factors interact to influence SI). As we mentioned before resilience models such as compensatory and risk-protective. Compensatory model suggests the presence of protective factors can directly reduce the impact of risk factors, leading to better outcomes. For example, engaging in moderate PA can independently lower the likelihood of SI. Risk-Protective model proposes higher levels of protective factors can decrease the negative effects of risk factors.

We are reporting a 13.44% prevalence of SI in the study population. This result underscores the multifaceted nature of SI, indicating that its prevalence is influenced by various significant demographic variables such as age, gender, and family size. A population-based study of 82 countries found a similar overall pooled prevalence of SI at 14.0%. These findings emphasize the need to consider diverse demographic and lifestyle factors to better understand the prevalence of SI. There was significant regional variation in the prevalence of mental health problems. The highest pooled prevalence of suicidal ideation was observed in the Africa Region (21.0%) and the lowest was in the Asia region with 8.0% [53].

Being unemployment emerged as a significant factor associated with increased odds of SI, even after adjusting for other variables. Individuals classified as previously married or never married exhibited an elevated risk of SI in both unadjusted and adjusted models. Interestingly, moderate family size appeared to have a protective effect on SI prevalence when compared to larger family sizes. Furthermore, familial tobacco use, family addiction history, and familial violence showed positive relationships with SI in the adjusted model. Addressing emotional needs relates to the fulfillment that comes from connecting or reconnecting with the individual鈥檚 family, friends, and the relationships they have with others and with themselves [54, 55].

Individuals aged 15 to 24 demonstrated an increasing likelihood of experiencing SI for both genders compared to those older than 65 years, even after adjusting for potential confounders. Despite young women potentially facing different challenges than young men, such as societal pressures, gender-based discrimination, and varied coping mechanisms that may influence the prevalence and intensity of suicidal ideation differently, [56,57,58].

Our results indicate that the likelihood of SI in the 15鈥�29 age group is higher for men than for women (1.67 vs. 1.44). However, both genders show significantly elevated likelihood compared to the reference group which was aligned with the study of Kerr et. Al that indicated boys and men (Ages 9 to 29 Years) were at high risk of SI [59]. Conversely, no significant probability of SI was found among other age groups relative to the reference group of older individuals. Adolescence is a developmental stage associated with an increased risk for the onset and growth of suicidal thoughts and behaviors [60]. In fact, it has been documented that the lifetime prevalence of SI among adolescents is 12.1% [61]. In light of the fact that suicidal ideation is a strong predictor of suicide attempts and completed suicide, it is crucial to develop models of vulnerability and protection against suicidal ideation among young adolescents in order to implement early prevention and intervention programs [62].

These risk factors are multifactorial, intricate, and interrelated. Most are connected to their families and schools. The significance of adequate parental regulation including supervision, monitoring, rule-setting, and other forms of behavioral control is one of the most well-documented findings from research on adolescents and their families [63, 64]. The risk factors associated with adolescent suicide behavior are multifactorial. Most are related to the family and the school. The importance of adequate parental regulation involving supervision, monitoring, rule-setting, and other forms of behavioral control is documented in [63, 64]. Parental bonding and a sense of family connection are also associated with lower levels of depression, and SI [65]. As we presented having a family with moderate size appeared to be a protector and family addiction history, and familial violence showed positive relationships with SI.

While initially, women displayed a higher likelihood of SI than men, this difference disappeared when accounting for covariates. Previous research has indicated that the effects of social risk and protective factors are modified by gender [66,67,68]. For women and men, it was evident that the risk-protective model was supported when PA (moderate) was the protective factor. These results suggest that a moderate level of PA reduces the effects of anxiety on suicidal ideation among men and women. For women only, results regarding PA also supported the compensatory model. This result suggests that for women, the strength of the association between suicidal ideation and anxiety is reduced when PA is considered. Previous data led to inconsistent findings with regards to the impact of gender on suicidal ideation. The potential explanation for these findings could be attributed to the difference in ways to cope performed by individuals of different genders [69]. In a study conducted among 15鈥�25 year olds in Malaysia, it was shown that suicidal thoughts are more common in men than women and also age can be a predictive factor in men [70]. Suicide rates were higher among men compared to women [71]. However, suicide attempts were most prevalent among young women, closely followed by young men [72].

It has been found that elderly males are at higher risk for suicide than any other group [73]. Although elderly women make more attempts, elderly men complete suicides more often [11]. A systematic review of suicidal behavior in Europe and America has concluded that suicidal ideation and behavior are more common among women [74]. There is an inconsistency between the pieces of literature. In China, the prevalence of SI was substantially higher among men, whereas the converse was in India for older individuals [27]. Consequently, the principal finding highlights the need for context-based interventions. Considering that individuals with moderate PA had a lower association with SI than those with low PA, strategies such as encouraging regular PA and improving walkability should be considered.

The results highlighted that moderate PA level was associated with a protective effect on SI in comparison to low levels of physical activity. The significance of low physical health as a risk factor appears to be well-established [75]. Although an investigation examined the risk and protective factors for SI from childhood to adulthood on mental and physical health PA had no significant effect [76]. With reference to suicidality, the factors that modulate or serve as buffers against taking one鈥檚 own life are referred to as protective factors [77]. Despite increasing support for protective factors in suicide prevention the positive message of recovery from suicidal ideation and the hope this may engender for suicidal clients has garnered little attention [78]. It has been shown that being physically active, even in moderate form, can enhance life satisfaction and elevate people鈥檚 level of happiness [79].

The study among Korean adolescents demonstrate that physical activity does not have the protective impact that is generally believed to shield adolescents from suicidal behaviors. Adolescents in Korea are more prone to attempt suicide if they engage in frequent, intense physical activity [80]. Addressing physical needs includes promoting a healthy lifestyle and generating interest in previously enjoyable hobbies or pastimes as a means of consolidating internal coping strategies [81, 82]. Perhaps physical activity can be considered as a coping skill and people can benefit from its effect in stressful situations throughout the day [83]. Therefore, it can be said that physical activity is effective in improving mental health by improving the level of mood and as a coping skill, and reduces anxiety symptoms, which are one of the risk factors of suicidal thoughts.

The most robust associations were observed between SI and anxiety in this study. Several national organizations have identified anxiety and its associated disorders as significant risk factors for suicide [84, 85]. Once an individual's suicidal thoughts get activated, feelings of anxiety and agitation may manifest as a way to focus their attention on suicide. This interaction between attentional fixation on suicide and a sense of hopelessness might contribute to an increased risk of suicide [86, 87]. Despite some early research suggesting that there is no association between anxiety disorders and suicidality [88, 89]. Adolescents who attempt suicide, compared to those who experience suicidal ideation, are more likely to have been diagnosed with anxiety disorders [90]. In general, there is a notable correlation between anxiety and its associated issues, and the possibility of experiencing suicidal thoughts and engaging in suicide attempts. These connections have been found to be statistically significant in prospective studies [91].

Limitations.

It is important to acknowledge the study's limitations. Due to the study's use of correlation analysis on concurrently collected data, it is not possible to infer a causal connection between variables. All data in this investigation were collected via self-report from the same source. Second, those who declined to take part in the study may have been more likely to have suicidal ideas and depressive symptoms. Third, the current study examined suicidal ideation as an adjustment factor (used it to control and understand its impact on other variables in our study, but this focus limits the applicability of our results to just those with suicidal thoughts, not those who have taken more severe actions), rather than suicide attempts or completions. Although suicidal ideation is a predictor of suicide attempts and completions [11], it is essential to recognize that suicidal ideation does not lead to suicide in the majority of people. Therefore, we need to take caution when projecting the results of this study to individuals who have attempted or completed suicide.

Conclusion

Overall, our research revealed a SI prevalence of 13.44%, which is in line with worldwide data. We demonstrated that PA would both lessen the detrimental impact of anxiety on suicidal thoughts and potentially reduce the probability of SI. It's important to develop and evaluate PA-enhancing treatments, particularly for those who are experiencing anxiety especially for woman.