Abstract
Suicide is a leading cause of death among US adolescents, with firearms being a common means of youth suicide. As crisis services have expanded to include digital modalities (e.g., text; online chat), we still know little about the willingness of youth who carry firearms to contact crisis services. The present study examines whether public middle and high school students in Florida who carried handguns reported greater reluctance to contact crisis services in 2024. Findings indicate that nearly 3 in 4 youth (72.69%) reported reluctance to contact crisis services via one or more modalities and nearly 1 in 3 (31.75%) reported reluctance across all modalities. Even after adjusting for covariates, reluctance was significantly higher among youth who carry handguns—a relationship that emerged across youth with and without recent suicidality. Future research should further unpack the reasons for this association to inform strategies to enhance willingness to contact crisis services.
Introduction
Suicide remains a critical public health issue among U.S. adolescents. Recent data identify suicide as the second leading cause of death among youth aged 10 to 14 and the third leading cause among those aged 15 to 19 (Centers for Disease Control and Prevention, 2023). Furthermore, national evidence indicates that 21.1% of U.S. high school students have seriously considered suicide, 17.6% have made a suicide plan, and 10.1% having attempted suicide (Bommersbach et al., 2025). Among youth aged 15 to 24, firearms have emerged as the prominent means of suicide, accounting for more than half of suicide-related fatalities in 2023 (Centers for Disease Control and Prevention, 2023).
In response to the ongoing mental health crisis, crisis intervention services have historically relied on telephone hotlines offering brief, confidential, cost-free counseling and referrals (Gould et al., 2021, 2022). More recently, these services have expanded beyond traditional telephone support to include digital modalities such as text and online chat. Notably, approximately 66% of individuals who access crisis support via text are between the ages of 14 and 24 (Pisani et al., 2022), and an estimated 40% of chat users are under the age of 18 (Gould et al., 2021). Despite the high rates of suicide among young people in the US and the increasing availability of crisis intervention services via telephone, text, and chat platforms, research investigating the factors that influence youths’ decisions to engage in these services remains limited.
One such factor that may influence youths’ decisions to access crisis services is handgun carrying. In recent decades, youth handgun carrying has ticked upwards, with one Florida-based study estimating a shift in prevalence of 3.7% in 2002 to 6.0% in 2022 (Wright-Kelly et al., 2025). Importantly, firearm access among youth is a well-established risk factor for suicide (Reinbergs et al., 2024). Still, far less is known about how youth firearm-related behaviors—such as handgun carrying—may contribute to reluctance to contact crisis services, even during suicidal crises. Studies have shown that youth who are exposed to community violence, have histories of victimization, or report other adversities are significantly more likely to report anxiety and/or depression, suicidality, and carrying handguns (Christensen et al., 2025; Miliauskas et al., 2022; van Geel et al., 2022). Research also suggests that these histories of childhood exposure to violence and trauma can elevate firearm-related behaviors – including handgun carrying—by elevating “the anticipation of adverse event(s)” and perceptions of environmental uncertainty and/or threat (i.e., viewing the world as a dangerous place) (Altikriti et al., 2025; Brennan, 2023; Bryan et al., 2023). These processes can lead to sustained hypervigilance and arousal, especially when threats and sources of stress are unpredictable or uncertain (Anestis & Bryan, 2021; Bryan et al., 2023). When youth carry handguns under these conditions, they typically do so for self-defense or protection (Chavez et al., 2022; Oliphant et al., 2019; Simon et al., 2022). This may be especially true when youth experience cultural and/or medical mistrust and perceive formal institutions (i.e., police and criminal justice system, schools, social services) as ineffective, biased, or unsafe (Sierra-Arévalo, 2016; Sokol et al., 2022). Taken together, these dynamics suggest that youth who carry handguns may exhibit a heightened reluctance to contact crisis services compared to their non-carrying peers, which may distance some of the most vulnerable youths from the very supports intended to prevent harm.
Utilizing data from a large statewide representative sample of middle school and high school students in the state of Florida, the present study is the first to examine whether adolescents who carry handguns exhibit greater reluctance to contact crisis services via multiple modalities (i.e., 988 lifeline and other crisis services via phone, text, and chat).
Youth Handgun Carrying in the Context of Violent Traumas
A long line of research has indicated that youth handgun carrying is not a random phenomenon but is socially patterned – underpinned by social-structural determinants of health (Beardslee et al., 2021; Lee et al., 2025; Yang et al., 2025). Among the most salient and robust correlates of youth handgun carrying is direct and vicarious exposure to community violence – including firearm violence – and related early life traumas (Baiden et al., 2024; Beardslee et al., 2021; Comer & Connolly, 2023; Lee et al., 2025; Rosenbaum et al., 2024). For instance, a recent national study revealed that US adolescents exposed to violence in their community were 2.34 times as likely to carry a gun compared to their counterparts, and those personally threatened or injured with a weapon were 2.91 times as likely to carry a gun (Baiden et al., 2024). Additionally, multiple recent studies point to the social contagion of gun-related events, revealing that heightened exposures to guns and gun-related violence (including being shot at or witnessing someone being shot at) are among the most robust predictors of gun carrying among young people (Beardslee et al., 2021; Comer & Connolly, 2023). Relatedly, these exposures often occur within a broader landscape of childhood trauma (Taylor et al., 2025), which is a known risk factor for adolescent handgun carrying (Jones et al., 2023; Yang et al., 2025). It is within this context of violent traumas that handgun carrying most often occurs, which may motivate these youth—who are likely to perceive their environments as unpredictable, threatening, and unsafe—to seek out firearms as a source of protection (Chavez et al., 2022; Oliphant et al., 2019; Simon et al., 2022). Under these circumstances, some youth may even exhibit a willingness to use guns defensively when or if they are threatened, given that early life trauma can heighten threat sensitivity, which in turn can elevate defensive gun use among individuals with access to firearms (Altikriti et al., 2025).
Mistrust, Uncertainty, and Perceived Threat as Barriers to Crisis Services
Just as there are stark disparities in handgun carrying among youth, there are also notable inequities in access to and receipt of crisis services. For example, Secaira et al. (2026) recently posited a mental health disparities framework on the 988 lifeline, highlighting how insufficient attention to the structural/systemic context in which implementation occurs may exacerbate existing mental health disparities in receipt of crisis care, despite overall improvement. Thus, identifying precisely how systemic inequities create barriers to crisis services is critical for these services to promote mental health equity in the population. Despite a lack of research focusing on adolescents specifically, recent data suggests that nearly 30% of adults are not willing to use 988 in a crisis. Beyond willingness, 87.1% report at least one concern about using 988, with top concerns being potential law enforcement involvement, forced hospitalization, potential service fees, and perceived stigma (Ueda et al., 2026). Such concerns can dissuade individuals in crisis from seeking help via crisis services like 988 and are likely concentrated among the most marginalized and disenfranchised segments of society, who already report higher levels of institutional mistrust (Boyd et al., 2025; Brayne, 2014; Wana et al., 2026).
To illustrate, evidence suggests that institutional mistrust reduces the use of crisis hotlines among vulnerable groups in crisis (Wana et al., 2026), and that lower rates of help seeking in this context often occur within a backdrop of violent and/or childhood trauma (Boyd et al., 2025; van den Berk-Clark & Patterson Silver Wolf, 2017). In this sense, contact crisis services cannot be understood merely as isolated, individual choices, but rather as “deeply contextual responses to historical and ongoing systemic inequities that have eroded trust in mental health institutions” (Boyd et al., 2025, p. 3). Data also suggests that, under these circumstances, younger samples with more recent trauma exposure may be especially unlikely to seek out formalized crisis services (van den Berk-Clark & Patterson Silver Wolf, 2017). Moreover, gun-involved violent traumas – which are common among youth who carry handguns – may further hamper mental health help-seeking by engendering uncertainty, hypervigilance, and a deep sense of unsafety and sensitivity to threat (Altikriti et al., 2025; Brennan, 2023; Bryan et al., 2023; Liebschutz et al., 2010).
Why Youth Who Carry Handguns May Be Reluctant to Contact Crisis Services
Considering the evidence of (1) the context of violent traumas in which youth handgun carrying is frequently embedded, and (2) the role of mistrust, uncertainty, and perceived threat in creating barriers to contacting crisis services, it is reasonable to expect youth who carry handguns to be reluctant to contact crisis services. First, studies have shown a heightened sense of threat sensitivity, uncertainty, and perceived danger among individuals who carry handguns, all of which may underpin an unwillingness to seek mental health help via contacting formalized crisis services. This work aligns with Buttrick’s (2020) coping model of protective gun ownership, which argues that gun possession for protection—which is typical in the case of handgun carrying (Oliphant et al., 2019)—symbolically serves as “an aid to manage psychological threats” to safety and control, which are rooted in a belief that “the world is a dangerous place and that society will not keep them safe” (p. 835). We contend that Buttrick’s (2020) model is highly relevant to youth carrying handguns considering that (a) similar coping processes are known to occur in youth who carry handguns (Brennan, 2023), (b) violent trauma histories are commonplace among these youth (Baiden et al., 2024; Comer & Connolly, 2023; Rosenbaum et al., 2024), and (c) trauma, especially when recent, can reduce mental health help seeking—particularly among young people—in part due to fear of negative responses and difficulty trusting others (Truss et al., 2023; van den Berk-Clark & Patterson Silver Wolf, 2017).
Second, considering the backdrop of elevated dangerousness perceptions and threat sensitivity for youth carrying handguns, we also know medical mistrust is heightened among groups exposed to high levels of community violence. For instance, young men with violent injuries from stabbings and shootings often report challenges engaging with healthcare services, as they describe “blurred lines” between surveilling institutions that administer punishment (e.g., policing, corrections) and institutions providing medical and mental health services (Liebschutz et al., 2010, p. 1372). Given the erosion of trust in mental health professionals among young people reporting compounding traumas and violence exposure (Boyd et al., 2025; Truss et al., 2023), it is reasonable to expect a similar pattern for youth who carry handguns, who report higher rates of gun violence exposure (Comer & Connolly, 2023).
Finally, the blurred lines across surveilling institutions administering health services v. those administering punishment noted by Liebschutz et al. (2010) may have unique and notable implications for system avoidance among youth who carry handguns. In her theory on criminal justice contact and institutional attachment, Brayne (2014) posits and finds evidence that, compared to their counterparts, individuals with criminal justice contact (e.g., via police contact, arrest) are less likely to interact with a variety of surveilling (i.e., record-keeping) institutions—including healthcare—in an effort to evade further surveillance. Importantly, youth who carry handguns are more likely to have histories of criminal justice contact (e.g., via police contact, arrest) compared to their counterparts (Booty et al., 2026; Vaughn et al., 2017) and thus may exhibit greater system avoidance. Compounding the risk of generalized system avoidance is (a) the documented fear of possible law enforcement involvement reported by many 988 callers (Ueda et al., 2026) and (b) potential for gun possession and/or carrying to arise in crisis conversations, as counselors are trained in lethal means safety counseling and will ask callers about firearm access and devise a plan to remove access to lethal means (Draper & McKeon, 2024). These dynamics could spark fears of firearm confiscation—and possibly punishment for illicit carrying—and thereby dissuade youth who carry handguns from contacting crisis services at all.
Ultimately, since violent traumas—including gun violence exposures—are common among youth who carry handguns, these youth are at risk of clinically meaningful mental health symptoms—including possible suicidality (Boccio et al., 2025; Romero et al., 2017). Also, to the extent they face barriers to accessing, seeking, and/or receiving appropriate crisis services, youth who carry handguns will likely continue to incur a high risk of suicidality. Thus, it is vital that we seek to better understand and remedy potential reluctance to contact crisis services among youth who carry handguns, as these youth are not only at high risk of experiencing mental health crises, but also have ready access to a highly lethal means of suicide.
The Current Study
This study is the first to examine youth handgun carrying as a risk factor for reluctance to contact crisis services via several modalities among a recent statewide sample of youth. We propose the following three research questions:
Methods
Data
This study employed data drawn from the 2024 Florida Youth Substance Abuse Survey (FYSAS) (Florida Department of Children and Families, 2024). The FYSAS is an annual survey of public middle school and high school students in the state of Florida. Data collection for the FYSAS is sponsored by the Florida Departments of Children and Families, Education, and Health. In recent years, the FYSAS has been administered using either a traditional pencil and paper format or using an internet-based system, depending on county preference. Items included in the FYSAS questionnaire are primarily concerned with adolescent substance use. However, each year the survey also includes items assessing a variety of daily activities, peer relationships, school engagement, and acts of misconduct (including handgun carrying). Of note, several items concerning suicidal ideation, suicidal behavior, and contacting crises services were added to the questionnaire beginning in 2023.
Each year, a representative sample of adolescents is selected using a two-stage cluster sampling strategy. First, groups of public middle schools and high schools are selected excluding adult education, correctional, and special educational schools. Probability of selection of individual schools was tied to enrollment size—wherein larger schools have a higher probability of selection. Then, survey administrators randomly selected classrooms within the selected schools. For the 2024 survey, 356 high schools and 405 middle schools were selected to participate in the survey. 346 of the 405 selected middle schools participated in the survey along with 306 of the 356 selected high schools. Overall, 48,352 students completed the survey in 2024. After removing surveys where students (a) appeared to exaggerate substance use and delinquent behavior, (b) reported using a fictitious drug, (c) demonstrated inconsistent patterns of responses, (d) completed less than 25% of the survey, and/or (e) completed the survey for the wrong grade group, 44,755 respondents remained. For the current analyses, all respondents who were missing data on measures related to contacting crisis services, suicidality, and handgun carrying were dropped, leading to a final analytic sample of 38,316 respondents.
Missing data on the covariates was handled using multiple imputation with chained equations to produce and merge 20 datasets. Examination of cases missing data on the key independent and dependent variables revealed missingness is associated with observed covariates (e.g., age, sex, race, maternal education, living in a rural area, having a parent in the military, and living in a household that speaks a primary language other than English) indicating the data are not missing completely at random (MCAR). Analysis of missing data on the covariates after restricting the sample to respondents with complete data on the key independent and dependent variables with Little’s test further indicates the data are not MCAR (p < .05). As such, analyses for this study were conducted on the assumption that the data are missing at random (MAR). Prior to imputation, missing data for most covariates were below 1%, and only one variable (i.e., maternal education) was above 10%. Descriptive statistics are presented prior to imputation. To test the robustness of our findings, all models were estimated prior to and after imputation. The findings from the analyses, including the non-imputed data, yielded the same general pattern of findings as the analyses with the imputed data. 1 This analysis utilized de-identified secondary data from the 2024 FYSAS and was therefore exempt from Institutional Review Board review.
Outcome Measures
Reluctance to contact crisis services was measured using four separate indicators of self-reported likelihood of contacting crisis services. All items are asked using a hypothetical scenario in which the respondent was feeling overwhelmed, upset, or suicidal.
Reluctance to Call 988
Reluctance to Call 988 was measured using a single item. Prior to the item, respondents were given the prompt “988 is a 24/7, confidential crisis line that connects individuals who need to talk about suicidal thoughts and other difficulties, with a crisis counselor.” Then they were asked “If you were feeling overwhelmed, upset, or suicidal, would you call 988?” Response options included “not likely,” “somewhat likely,” and “very likely.” This item has been recoded as a dichotomous indicator of Reluctance to Call 988 where respondents who answered “somewhat likely” and “very likely” were coded as “0” and respondents who answered “not likely” were coded as “1.” Descriptive statistics for this measure and all other measures included in this study are displayed in Table 1.
Descriptive Statistics.
Reluctance to Contact a Crisis Counselor
Reluctance to Contact a Crisis Counselor was measured using a question where respondents were asked, “If you were feeling overwhelmed, upset, or suicidal how likely would you be to use phone, text, or chat to talk with a crisis counselor.” Underneath this question, respondents were asked about each contact method separately: Phone, Text, or Chat. For Phone, Text, and Chat options, respondents were provided the following response options (separately for each item): “not likely,” “somewhat likely,” “and very likely.” Each item was recoded as a dichotomous indicator of reluctance to contact a crisis counselor (whether by phone, text, or chat) where “somewhat likely” and “very likely” are coded as ‘0’ and “not likely” is coded as “1.”
Reluctance to Contact Crisis Services via One or More Modalities
A variable for reluctance to contact crisis services via one or more modalities was constructed using all previous outcome measures. Specifically, respondents who answered “not likely” to one or more of the previous items (i.e., call 988, phone a counselor, text a counselor, or chat a counselor) were coded as “1” and all remaining respondents were coded as “0.”
Reluctance to Contact Crisis Services Across All Modalities
To assess reluctance to contact crisis services across all modalities, we constructed a variety index for reluctance to contact crisis services using the 988, phone, text, and chat crisis contact modality items (0–4). Then, respondents who indicated reluctance to contact crisis services through all four modalities were coded as “1” while those who did not report reluctance across all four items were coded as “0.”
Predictor Measures
Handgun Carrying in Last 12 Months
Handgun carrying in the 12 months prior to the survey was measured in two separate ways using a single item from the questionnaire. Specifically, respondents were asked to indicate “How many times in the past year (12 months) have you: carried a handgun?” Response options for this item included “never” (0), “1 or 2 times” (1), “3 to 5 times” (2), “6 to 9 times” (3), “10 to 19 times” (4), “20 to 29 times” (5), “30 to 39 times” (6), to “40+ times” (7). For the present analyses, this variable was employed in two ways. First, it was recoded as a dichotomous indicator of carrying a handgun in the past 12 months, where 0 = did not carry a handgun and 1 = did carry a handgun. Second, the variable was employed in its original coding schema as a measure of the frequency of handgun carrying in the past 12 months.
Controls
Suicidality
Suicidality was measured using responses to three items assessing suicidal ideation and behaviors. To illustrate, respondents were asked “During the past 12 months, did you ever seriously consider attempting suicide?” Respondents were also asked “During the past 12 months, did you ever make a plan about how you would attempt suicide?” Response options for these two items included 0 = no and 1 = yes. Then, respondents were asked “During the past 12 months, how many times did you actually attempt suicide?” Response options spanned from “0 times” to “6 or more times.” Using these three items we created a measure of suicidality where respondents who did not report any suicidal thoughts, plans, or behaviors were coded as ‘0’ and respondents who reported any suicidal thoughts, plans, or attempts in the past 12 months were coded as “1.”
Additional Controls
Models for this study were estimated controlling for age, male sex, racial identity (non-Hispanic-Black, Hispanic, non-Hispanic White, non-Hispanic Other, Multiracial), living in a rural area, maternal education, having a parent in the military, and living in a home where the primary language is a language other than English. Age was measured in years. Male sex was measured using a single item where respondents were asked to indicate their sex where 0 = female and 1 = male. Race was coded using five dichotomous indicators of racial identity. Hispanic is coded so that 0 = non-Hispanic, Black, Other, or Multiracial and 1 = Hispanic. Black is coded so that 0 = Hispanic, White, Other, or Multiracial and 1 = non-Hispanic Black. White is coded so that 0 = Hispanic, Black, Other, or Multiracial and 1 = Non-Hispanic White. Other is coded so that 0 = Hispanic, Black, White, or Multiracial and 1 = Non-Hispanic Asian, Non-Hispanic Native American or Alaska Native, Non-Hispanic Pacific Islander or Native Hawaiian, or Non-Hispanic “other” racial identity. Multiracial is coded so that 0 = only reported one racial identity (or Hispanic-White) and 1 = reported more than one racial identity (excluding Hispanic-White). For ease of interpretation, non-Hispanic White is used as the reference category for the models in this study.
Living in a rural area was measured using a single item where respondents were asked to indicate where they are currently living. Response options included “on a farm,” “in the country, not on a farm,” or “in a city, town, or suburb.” This item has been recoded so that 0 = “in a city, town, or suburb” and 1 = “on a farm” or “in the country, not on a farm.” Maternal education was measured using an item where respondents were asked to indicate the highest level of education completed by their mother. Response options ranged from “completed grade school or less” to “graduate or professional school after college.” Having a parent in the military was measured using a dichotomous indicator of whether one of their parents, stepparents, or guardians is in the military where 0 = no and 1 = yes. Living in a home where the primary spoken language is a language other than English was measured using an item where respondents were asked to indicate “what is the language you use most often as home?” Response options included “English,” “Spanish,” or “Another Language.” This item is coded so that 0 = “English” and 1 = “Spanish” or “Another Language.”
Analytic Strategy
The analytic strategy for this study involves several linked steps. First, we address research question 1a and 1b by first calculating the descriptive statistics pertaining to all study variables (including reluctance to contact crisis services and handgun carrying) for the full sample, and subsequently for the sample stratified by reluctance to contact crisis services–with ancillary analyses examining reluctance (a) via one or more modalities and (b) across all modalities. Next, we address research question 2a by employing logistic regression to examine youth who report recent handgun carrying (i.e., past 12 months) exhibit significantly greater reluctance to contact crisis services (a) via specific modalities, (b) via one or more modalities, and (c) across all modalities. Similarly, we also address research questions 2b by employing logistic regression to examine youth who report more frequent handgun carrying exhibit significantly greater reluctance to contact crisis services (a) via specific modalities, (b) via one or more modalities, and (c) across all modalities. Finally, to address research question 3, we examined associations between handgun carrying in the past 12 months and all indicators of reluctance to contact crisis services by youth suicidality.
Results
In the first step of the analysis (research question 1a), we examined descriptive statistics among the full analytic sample (n = 38,316) for each form of reluctance to contact crisis services, suicidality, and handgun carrying in the past 12 months. As can be seen in Table 1, approximately 18% (17.90%) report suicidal thoughts, plans, or attempts in the past 12 months. Nearly 55% (54.89%) report reluctance to call 988, 51.75% report reluctance to phone a crisis counselor, 43.42% report reluctance to text a crisis counselor, and 50.65% report reluctance to contact a crisis counselor via chat. In total, 72.69% of respondents in the full sample report any reluctance to contact crisis services (via one or more of the modalities examined). Moreover, 27.31% report no reluctance to contact crisis services, 18.65% report reluctance to contact crisis services through one modality, 11.80% report reluctance to contact crisis services through two modalities, 10.49% report reluctance to contact crisis services through three modalities, and 31.75% report reluctance to contact crisis services through all available modalities. In addition, approximately 7% (6.98%) of the sample reported carrying a handgun in the past 12 months. For correlations between reluctance modalities, see Table A1 of the appendix.
After examining descriptive statistics in the full sample, we then stratified the sample by reluctance to contact crisis services (research question 1b). As can be seen, approximately 22% (21.81%) of respondents who report reluctance to contact crisis services through one or more modalities also report suicidality. In contrast, only 7.50% of respondents who report having no reluctance to contact crisis services also report suicidality. In addition, 7.80% of respondents who report reluctance to contact crisis services through one or more modalities also report carrying a handgun in the past 12 months compared to 4.82% of respondents who did not report any reluctance. Chi-square tests reveal statistical differences for both suicidality and handgun carrying between respondents who report any reluctance to contact crisis services and those who did not report any reluctance. In terms of key demographics, findings from Table 1 also reveal that females and Non-Hispanic Black youth were overrepresented among youth reporting any reluctance to contact crisis services. For instance, while female youth comprised 53.25% of youth reporting any reluctance, they only comprised 44.80% of youth reporting no reluctance (p < .01). Similarly, while Non-Hispanic Black youth comprised 15.54% of youth reporting any reluctance, they only comprised 11.39% of youth reporting no reluctance (p < .01). In ancillary analyses, we compared descriptive statistics between youth reporting reluctance via some modalities v. across all modalities. Overall, findings suggest significant differences between these groups—in their handgun carrying, suicidality, and a number of demographics (including age, sex, race/ethnicity, and rurality). For more details, see Table A2 of the Appendix.
In the next step of the analysis (research questions 2a and 2b), we estimated multivariable logistic regression models to examine associations between handgun carrying in the past 12 months and reluctance to contact crisis services. As can be seen in Table 2, carrying a handgun in the last 12 months is positively and significantly associated with reluctance to call 988 (OR = 1.665; p < .01), reluctance to phone a crisis counselor (OR = 1.499; p < .01), reluctance to text a crisis counselor (OR = 1.481; p < .01), reluctance to chat a crisis counselor (OR = 1.498; p < .01), reluctance contact crisis services via one or more modalities (OR = 1.822; p <.01), and reluctance to contact crisis services across all modalities (OR = 1.477, p < .01). Additionally, frequency of handgun carrying in the past 12 months was also positively and significantly associated with reluctance to contact crisis services for each of the contact modalities (988: OR = 1.213, p < .01; Phone: OR = 1.144, p < .01; Chat: OR = 1.160, p < .01; Text: OR = 1.165, p < .01), reluctance to contact via one or more modalities (OR = 1.234; p < .01), and reluctance to contact across all modalities (OR = 1.165, p < .01). To unpack these associations further, we used predictive margins to plot the predicted probability of any reluctance according to frequency of carrying a handgun in the past 12 months in Figure 1. Examination of Figure 1A reveals that the predicted probability of reporting reluctance to contact via one or more modalities appears to increase as the frequency of handgun carrying increases. In addition, we plotted the probability of reluctance to contact crisis services across all modalities as a function of frequency of handgun carrying. As can be seen in Figure 1B, the predicted probability of reluctance to contact crisis services through all modalities increases as the frequency of handgun carrying increases.
Regression Models of the Association Between Handgun Carrying and Reluctance to Contact Crisis Services (n = 38,316).
p <.05, **p < .01.
(a) Probability of reluctance via one or more modalities as a function of frequency of handgun carrying with 95% confidence intervals. (b) Probability of reluctance across all modalities as a function of frequency of handgun carrying with 95% confidence intervals.
In the final step of the analysis (research question 3), we examined the relationship between handgun carrying and reluctance to contact crisis services stratified by suicidality. As displayed in the top half of Table 3, handgun carrying in the past 12 months is positively and significantly associated with reluctance to contact crisis services for three of the four individual contact modalities (988: OR = 1.313, p < .01; Chat: OR = 1.267, p < .01; Text: OR = 1.200, p < .05), reluctance to contact via one or more modalities (OR = 1.596, p < .01), and reluctance to contact across all modalities (OR = 1.207, p < .05) among respondents with suicidality (n = 6,858). Similarly, frequency of handgun carrying is significantly associated with reluctance to contact crisis services for three of the four individual contact modalities (988: OR = 1.112, p < .01; Chat: OR = 1.104, p < .01; Text: OR = 1.107, p < .01), reluctance to contact via one or more modalities (OR = 1.196, p < .01), and reluctance to contact across all modalities (OR = 1.103, p < .01) in the sample of respondents who report suicidality. In contrast, ever carrying a handgun in the past 12 months and frequency of handgun carrying in the past 12 months are not significantly associated with reluctance to phone a crisis counselor. Figure 2A presents the predicted probability of reporting reluctance to contact crisis services via one or more modalities across frequency of handgun carrying in respondents who reporting suicidality. As can be seen, this relationship follows a similar pattern to the pattern seen in the full the sample where the probability of reporting reluctance to one or more modalities appears to increase as the frequency of handgun carrying the past 12 months increases. Similarly, Figure 2B presents the predicted probability of reluctance to contact crisis services across all modalities as a function of the frequency of handgun carrying. As can be seen, the predicted probability of reluctance to contact crisis services across all modalities increases as frequency of handgun carrying increases among respondents who report suicidality.
Regression Models of the Association between Handgun Carrying and Reluctance to Contact Crisis Services Stratified by Suicidality.
Note. “Suicidality” is defined as reporting suicidal thoughts, plans, or attempts. *p < .05, **p < .01. While covariates are suppressed to conserve space, all models are estimated controlling for age, sex (male = 1), non-Hispanic Black, Hispanic, Non-Hispanic Other, Multiracial, living in a rural area, having a parent in the military, and living in a household that speaks a primary language other than English.
(a) Probability of reluctance via one or more modalities as a function of frequency of handgun carrying for respondents who reported suicidal thoughts, plans, or attempts with 95% confidence intervals. (b) Probability of reluctance across all modalities as a function of frequency of handgun carrying for respondents who reported suicidal thoughts, plans, or attempts with 95% confidence intervals. (c) Probability of reluctance via one or more modalities as a function of frequency of handgun carrying for respondents who did not report suicidal thoughts, plans, or attempts with 95% confidence intervals. (d) Probability of reluctance across all modalities as a function of frequency of handgun carrying for respondents who did not report suicidal thoughts, plans, or attempts with 95% confidence intervals.
Then, we examined the relationship between handgun carrying and reluctance to contact crisis services among respondents who did not report suicidality. The bottom half of Table 3 reveals that handgun carrying is positively and significantly associated with reluctance to contact crisis services via each of the individual modalities (988: OR = 1.533, p < .01; Phone: OR = 1.373, p < .01; Chat: OR = 1.395, p < .01; Text: OR = 1.452, p < .01), along with reluctance via one or more modalities (OR = 1.581, p < .01), and across all modalities (OR = 1.428, p < .01) among respondents who did not report suicidality. The same pattern of results emerged for frequency of handgun carrying in the past year (988: OR = 1.192, p < .01; Phone: OR = 1.121, p < .01; Chat: OR = 1.140, p < .01; Text: OR = 1.156, p < .01; Reluctance via one or more modalities: OR = 1.179, p < .01; Reluctance across all modalities: OR = 1.158, p < .01) for respondents who did not report suicidality. 2 The probability of reluctance to contact crisis services via one or more modalities is plotted as a function of the frequency of handgun carrying in Figure 2C. Similar to the previous figures, the probability of reluctance to contact via one or more modalities increases as the frequency of handgun carrying increases. A graph of the predicted probability of reluctance to contact across all modalities as a function of the frequency of handgun carrying in Figure 2D shows a similar pattern to the previous graphs, wherein the predicted probability of reluctance to contact across all modalities increases as the frequency of handgun carrying increases among respondents who did not report suicidality.
Discussion
The present study examined associations between youth handgun carrying and reluctance to contact crisis services among a representative sample of public middle and high school students in the state of Florida in 2024. Our study yielded three key findings. First, we found that nearly 3 in 4 youth (72.69%) reported reluctance to contact crisis services via one or more of the modalities examined (e.g., call 988; contact a crisis counselor via phone, chat, or text), and nearly 1 in 3 (31.75%) reported reluctance to contact crisis services across all modalities. Reluctance to call 988 (54.89%) or phone a crisis counselor (51.75%) were the most commonly endorsed items, whereas reluctance to text a crisis counselor was somewhat less commonly endorsed (43.42%). Given that participants in our sample are at a life stage where vulnerability to suicidality is high (Bommersbach et al., 2025; Centers for Disease Control and Prevention, 2024)—with nearly 1 in 5 reporting suicidality in this sample—this level of reluctance to seek help from crisis services is particularly concerning and may have implications for youth receipt of formal crisis services. Still, future research should seek to move beyond the concept of reluctance to contact crisis services and explore whether these factors shape actual youth receipt of crisis services. Furthermore, unpacking the reasons why youth view certain modalities more favorably (e.g., text a crisis counselor v. calling 988) is also an important direction for future research. Extant research shows about two-thirds of individuals accessing crisis support via text are between the ages of 14 and 24 (Pisani et al., 2022), which points to the promise of using texting as a modality to reach young people specifically. We also found some significant differences in reluctance by youth demographics, with female, rural, and non-Hispanic Black youth exhibiting somewhat higher odds of reluctance via one or more modalities. Future research should center on the experiences of these groups to understand the reasons for their heightened reluctance.
Second, chi-square tests revealed a significantly greater likelihood of suicidality and handgun carrying among respondents who report reluctance to contact crisis services. This suggests that the very young people with the greatest need for crisis services (i.e., those reporting suicidality) and at the greatest risk of potentially lethal suicide attempts (given readily available access to handguns) are often the same young people who are most reluctant to contact crisis services (Swanson et al., 2021). These findings persisted in multivariable models. Specifically, youth who carried a handgun during the 12 months prior to the survey (v. those who did not) reported significantly greater reluctance to call 988, phone a crisis counselor, text a crisis counselor, and chat with a crisis counselor, independent of a host of potential confounders (age, sex, race/ethnicity, rurality. maternal education, etc.). The highest point estimates emerged in the case of calling 988, which builds upon research showing that 87.1% of individuals express one or more concerns specifically with using 988 during a mental health crisis—which includes law enforcement involvement and the possibility of being sent to jail (Ueda et al., 2026). While our data cannot speak to the precise reasons for the association between handgun carrying and reluctance to contact crisis services, prior research indicates that youth who carry handguns (v. those who do not) have higher rates of violent traumas—including gun violence exposures (Comer & Connolly, 2023)—which have also been linked to reduced mental health help seeking (Truss et al., 2023; van den Berk-Clark & Patterson Silver Wolf, 2017). Additionally, it is possible that youth who carry handguns may exhibit greater reluctance to contact crisis services out of fear of being punished for possessing (and likely unlawfully carrying) a handgun, and potentially having their handgun taken away. Such fears may be rooted in the fact that 988 counselors are trained in lethal means safety counseling and will therefore ask callers about firearm access and devise a plan to remove access to lethal means (Draper & McKeon, 2024). It is also possible that some of the underlying characteristics of youth who carry handguns are also the characteristics that would underlie an unwillingness to seek help from formal sources. For instance, individuals who carry handguns report significantly higher threat perceptions (Altikriti et al., 2025) and broadly more negative cognitions about and distrust of the world and institutions (Bryan et al., 2023; Bryan et al., 2020; Simon et al., 2022), which may reduce their willingness to proactively seek help from a crisis line/crisis counselor. Future research should unpack the constellation of pre-existing risk factors and potential mechanisms that underlie this association.
Finally, the association between youth handgun carrying and a reluctance to contact crisis services held across groups with and without suicidality (i.e., suicidal thoughts, plans, and/or attempts). This finding suggests that, even for young people who did not report suicidality in the 12 months prior to the survey, those who carried a handgun during that time period (v. those who did not) still reported a greater reluctance to contact crisis services if they personally experience a crisis in the future. This suggests that any connection between handgun carrying and suicidality is not fully responsible for the patterns in the data around handgun carrying and reluctance to contact crisis services. While we do not posit a causal relationship between these factors, we nevertheless contend that additional research into why this reluctance to contact crisis services exists among youth who carry handguns is critical. In the long run, such research can help inform strategies to address reluctance to contact crisis services across modalities among youth who carry handguns.
Limitations
Despite its strengths and contribution to the literature, our study is not without its limitations. First, our data are cross-sectional, and we are therefore unable to make causal claims. Even so, the goal of the study was to understand whether youth who carry handguns are more reluctant to contact crisis services than their peers—not whether carrying handguns causes youth to become more reluctant to contact these services. As outlined previously, we acknowledge that the reasons for these differences are likely complex and merit additional research. Even so, we posit that understanding differences in youth willingness to contact crisis services—including as it relates to prior and ongoing risk behaviors like handgun carrying – is a critical step in devising formal crisis resources that all youth—including those most at risk of suicide – are open to utilizing. Second, while our sample is representative of middle and high school students in the state of Florida in 2024, our findings cannot be generalized to youth who are disconnected from school or outside the state of Florida. Future research should seek to replicate these findings in other contexts and with youth who are disconnected from school.
Third, our outcomes are based on hypothetical scenarios, which ask young people to estimate their likelihood of contacting crisis services during a potential future mental health crisis. For our purposes, a “not likely” response (i.e., not likely to contact crisis services) is conceptualized as reluctance or a disinclination to use a given crisis service modality. While we anticipate that this measure is related to actual non-use of formal crisis services, usage of these services is not observed in these data, and we cannot speak to youth refusal of mental health help more broadly via other support mechanisms (e.g., non-crisis informal mental health supports). Additionally, the current data do not measure youth assumptions that might underlie their reluctance (e.g., such as beliefs about anonymity and/or confidentiality of crisis conversations). Even so, our findings on the reduced willingness of youth who carry handguns to contact crisis services held among youth who reported suicidality in the past 12 months—a group for whom contacting crisis services in the future is likely highly relevant given their recent mental health history. Still, future research on youth who carry handguns should include questions about their past experiences calling 988 and other crisis services—as well as what transpired during and following the correspondence—to help unpack the approaches that work best to meet the needs of these youth. Fourth, while we posit potential explanations for the observed associations, we were unable to empirically test them. It is critical for future longitudinal research to explore the reasons why handgun carrying is linked to reluctance to contact crisis services, including some of the possibilities discussed previously (e.g., threat sensitivity, institutional mistrust, fears of handgun confiscation/punishment, and/or law enforcement involvement).
Conclusions
Youth who carry handguns are at greater risk of suicide (Swanson et al., 2021), intimating they are likely to benefit from utilizing well-equipped crises services in their hour of need. Even so, our findings point to elevated levels of reluctance to contact crisis services among youth who carry handguns—a population of youth who, given this reluctance, are at risk of being underserved by crisis services. Future research should further unpack the reasons for this association to inform strategies to enhance willingness to contact crisis services and promote adolescent mental health equity.
Footnotes
Appendix
Descriptive Statistics Stratified by Reluctance to Contact Crisis Services via Some Modalities v. Across All Modalities.
| Reluctance to contact via some modalities |
Reluctance to contact across all modalities |
p | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Variable | Mean (%) | SD | Range | n | Mean (%) | SD | Range | n | |
| Ever Carried a Handgun in the Last 12 Months | 0–1 | 0–1 | <.01 | ||||||
| Yes | (6.96) | — | — | 1,092 | (8.87) | — | — | 1,079 | |
| No | (93.04) | — | — | 14,593 | (91.13) | — | — | 11,087 | |
| Times Carried Handgun Last 12 Months | .137 | .654 | 0–7 | 15,685 | .212 | .903 | 0–7 | 12,166 | <.01 |
| Suicidality | 0–1 | 0–1 | <.01 | ||||||
| Yes | (20.52) | — | — | 3,218 | (23.47) | — | — | 2,855 | |
| No | (79.48) | — | — | 12,467 | (76.53) | — | — | 9,311 | |
| Age | 14.223 | 1.957 | 10–19 | 15,661 | 14.581 | 1.906 | 10–19 | 12,147 | <.01 |
| Sex | 0–1 | 0–1 | <.01 | ||||||
| Female | (56.46) | — | — | 8,796 | (49.12) | — | — | 5,936 | |
| Male | (43.54) | — | — | 6,784 | (50.88) | — | — | 6,149 | |
| Race | |||||||||
| Non-Hispanic Black | (14.70) | — | 0–1 | 2,306 | (16.61) | — | 0–1 | 2,021 | <.01 |
| Hispanic | (24.34) | — | 0–1 | 3,817 | (24.89) | — | 0–1 | 3,028 | .287 |
| Non-Hispanic Other | (9.28) | — | 0–1 | 1,455 | (8.87) | — | 0–1 | 1,079 | .241 |
| Non-Hispanic White | (38.44) | — | 0–1 | 6,030 | (35.94) | — | 0–1 | 4,373 | <.01 |
| Multiracial | (13.24) | — | 0–1 | 2,077 | (13.69) | — | 0–1 | 1,665 | .281 |
| Rural | 0–1 | 0–1 | <.01 | ||||||
| Yes | (20.80) | — | — | 3,236 | (23.55) | — | 2,836 | ||
| No | (79.20) | — | — | 12,318 | (76.45) | — | 9,206 | ||
| Maternal Education | 3.283 | 1.343 | 0–5 | 11,587 | 3.169 | 1.386 | 0–5 | 8,858 | <.01 |
| Parent in the Military | 0–1 | 0–1 | .036 | ||||||
| Yes | (14.42) | — | — | 2,249 | (15.32) | — | — | 1,852 | |
| No | (85.58) | — | — | 13,347 | (84.68) | — | — | 10,236 | |
| Language other than English in the Home | 0–1 | 0–1 | .071 | ||||||
| Yes | (16.13) | — | — | 2,515 | (16.94) | — | — | 2,047 | |
| No | (83.87) | — | — | 13,080 | (83.06) | — | — | 10,038 | |
Acknowledgements
The analyses, conclusions, views, and opinions presented here are those of authors alone and should not be attributed to any of the organizations that sponsor the Florida Youth Substance Abuse Survey.
Ethical Considerations
This analysis utilized de-identified secondary data from the 2024 FYSAS and was therefore exempt from Institutional Review Board review.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.