Mental preparedness in future responders: analyzing anxiety and stress profiles

Abstract

Background

Anxiety sensitivity refers to the fear of anxiety-related sensations arising from beliefs that these sensations have harmful physical, cognitive, or social consequences. In disaster response training, such sensitivity may critically impair performance. This study examined anxiety sensitivity profiles among associate degree students enrolled in disaster- and emergency-related programs, focusing on gender- and grade-based differences across physical, cognitive, and social dimensions.

Methods

Using a cross-sectional survey design, data were collected from 301 students (178 females, 123 males; Mage = 20) via the Anxiety Sensitivity Index-3 (ASI-3). The study was conducted following institutional ethical approval (GO2025/1645) with informed consent. The ASI-3 demonstrated high internal consistency in the present sample (Cronbach's α = 0.88). Analyses included Welch's t-test, ANOVA, and item-level χ² tests.

Results

Total anxiety sensitivity scores did not differ significantly by gender (t = − 1.85, p < 0.066, d = 0.21) or grade level (F = 3.76, p < 0.054). However, female students exhibited significantly higher anxiety on physical and social items, including fear of increased heartbeat (χ² = 28.4, p < .001, φ = 0.31), chest pain, sweating, and fear of fainting in public.

Conclusions

These findings suggest the presence of gender-linked vulnerability patterns in specific anxiety domains, supporting the integration of targeted cognitive-behavioral and stress-regulation modules within disaster training curricula. However, given the cross-sectional design and convenience sampling, such recommendations warrant longitudinal validation to confirm modifiability and educational impact.

Introduction

Sudden events such as disasters, accidents and fires have not only physical effects but also serious psychological effects. Research demonstrates that disaster workers face substantial mental health risks, with post-traumatic stress disorder (PTSD) prevalence rates reaching 20–30% among international cohorts [1] and approximately 25% among Turkish search and rescue personnel following major earthquakes [2]. These elevated rates underscore the imperative for preemptive resilience training to mitigate chronic stress cascades before field deployment. In the event of such incidents, it is imperative that personnel possess not only technical expertise and proficiency, but also exhibit an elevated degree of psychological resilience. Psychological resilience can be achieved by quickly recovering from stressful situations [3]. Uncertainty, intense stress, time pressure and making decisions that have a direct impact on human life, especially at the moment of first contact, are among the main factors that increase the psychological burden on responders. In this context, understanding the level of mental preparedness of individuals who will respond to disasters and emergencies is critical not only at the individual level, but also at the systemic level.

Understanding psychological resilience requires examining anxiety sensitivity, as disaster responders operating across different phases from initial shock to recovery face varying physiological and cognitive demands that can trigger anxiety responses. People trained in disaster and emergency response need to develop their mental resilience as well as their technical skills. Their weaknesses in psychological competence will lead to situations that cause difficulties not only for themselves but also for the team members they work with. It is very important for the smooth running of these professions that they are not overwhelmed by the responsibilities they take on when responding to emergencies and that they have mental and physical stamina.

During the reaction phase, when responders experience panic and must make rapid decisions, those with high anxiety sensitivity may misinterpret their physiological arousal (e.g., increased heart rate) as dangerous, compromising their performance. In order to improve the mental resilience, disaster psychology and decision-making skills of students who aspire to professions such as search and rescue, firefighting, first aid and emergency assistance, it is considered necessary that these courses be delivered by academics who are experts in their field in the schools where they receive technical training and that the mental resilience of students be measured through real case examples.

This study addresses these gaps by providing the first comprehensive assessment of anxiety sensitivity profiles among Turkish associate degree students in disaster and emergency management programs. By examining both overall patterns and item-level gender differences in physical, cognitive, and social anxiety concerns, this research establishes baseline data necessary for curriculum development and identifies specific psychological vulnerabilities that warrant educational intervention. Unlike previous studies focusing on post-deployment outcomes, this pre-deployment assessment enables proactive rather than reactive approaches to psychological preparedness, potentially preventing the high rates of occupational mental health problems documented in active disaster workers. Moreover, by focusing on associate degree students who constitute a significant portion of Turkey's disaster response workforce yet receive less research attention than university level programs, this study contributes to a more inclusive understanding of psychological preparedness across educational levels.

Despite growing recognition of mental health challenges among disaster workers, research examining anxiety sensitivity specifically among students training for disaster response professions remains limited, particularly in the Turkish context. A systematic review of Turkish literature reveals only three published studies examining anxiety or stress among active disaster responders (combined N < 500), with none investigating pre-deployment student populations [2, 4]. This contrasts sharply with international research, which includes over 50 studies on military veterans and emergency personnel. Furthermore, while university-level emergency management programs have received some research attention, associate degree programs—which comprise approximately 40% of Turkey's disaster response workforce—remain virtually unexamined. While previous studies have documented PTSD prevalence among active responders [2, 5], few have investigated pre-deployment psychological profiles of future responders during their educational preparation. Understanding anxiety sensitivity patterns among students is crucial, as this construct represents a modifiable risk factor that can be addressed through targeted educational interventions before field deployment. Therefore, this study aimed to: (1 assess the overall anxiety sensitivity levels of associate degree students enrolled in disaster and emergency management programs, (2 examine potential differences in anxiety sensitivity based on gender and academic level; and (3 identify specific anxiety patterns related to physical, cognitive, and social symptoms that may require targeted educational support. By establishing baseline anxiety sensitivity profiles among pre-deployment students, this research provides foundational data for developing evidence-based psychological preparedness programs within disaster management curricula.

Psychosocıal challenges of disasters and emergencies

Disasters and emergencies are situations that adversely affect people and the environment and cause material and moral loss in the area where they occur. In such instances, both victims and those providing first aid are at greater risk of harm. A comprehensive understanding of the psychological distress experienced in this process is imperative for the successful realisation of future improvement efforts. The process can be considered in four distinct phases, contingent on the magnitude of the disaster and the damage it engenders [6]: 250).

The initial phase following the event is characterised by the psychological shock process. This phase, which can extend beyond 24 h, is accompanied by a range of physiological symptoms and cognitive reactions. The subsequent phase is the reaction process, during which disaster victims and responders experience feelings of panic and aggression in response to the emergency. In the aftermath of an emergency, the initial phase is characterised by a period of acute distress for those affected by the disaster. This stage is often characterised by a profound sense of loss and bereavement. During this period, emergency responders are able to undertake a comprehensive assessment of the situation to ascertain the extent of the damage and the needs of the affected population. In conclusion, the final observed process is the recovery phase. This is the phase in which disaster victims experience adaptation and recovery, and it is also the period in which efforts are made to improve the situation compared to the previous one [7]:42).

Understanding how anxiety sensitivity manifests across these disaster phases is crucial for preparing future responders. In the reaction phase, elevated social anxiety sensitivity may exacerbate panic via fear of observable symptoms such as trembling (odds ratio = 2.1 in emergency contexts; [8]), potentially prolonging distress into the recovery phase. Similarly, physical anxiety sensitivity—particularly fear of cardiovascular symptoms—may lead responders to misinterpret normal physiological arousal during high-intensity rescue operations as cardiac emergencies, resulting in premature task withdrawal. Cognitive anxiety sensitivity, characterized by fear of concentration loss or mental confusion, may impair critical decision-making during the assessment phase when rapid triage is essential. This phase-specific manifestation of anxiety sensitivity dimensions underscores the need for tailored psychological preparation that addresses distinct vulnerabilities across the disaster response continuum.

Theoretical background

Anxiety sensitivity: definition and dimensions

Anxiety sensitivity is defined as the tendency to misinterpret anxiety symptoms as negative or dangerous. Reiss and McNally [9] conceptualized it as 'the belief that bodily sensations associated with anxiety have harmful consequences'—a cognitive vulnerability that extends beyond trait anxiety. While trait anxiety reflects a general propensity to experience anxiety, anxiety sensitivity specifically concerns the fear of anxiety itself and its perceived physical, psychological, and social ramifications. People with high anxiety sensitivity tend to perceive threats more intensely and catastrophize physical and cognitive symptoms of anxiety. This situation can result in chronic stress, especially among those in high-risk occupations. For those working in disaster response, controlling this sensitivity is crucial to performing tasks and maintaining psychological resilience.

Anxiety sensitivity functions as a cognitive vulnerability that amplifies acute stress responses into chronic psychopathology via catastrophic misinterpretation mechanisms. Longitudinal research demonstrates that anxiety sensitivity prospectively predicts the development of PTSD following trauma exposure, with path coefficients ranging from β = 0.32 to 0.45 in structural equation models [10, 11]. Specifically, individuals with elevated baseline anxiety sensitivity who experience traumatic events are 2.8 times more likely to develop PTSD compared to those with low anxiety sensitivity, even when controlling for trauma severity and prior psychiatric history [12]. The mechanism operates through a fear-of-fear cycle: anxious individuals misinterpret normal physiological arousal during stressful situations as evidence of impending danger or loss of control, leading to heightened distress, avoidance behaviors, and ultimately functional impairment.

Anxiety sensitivity has been identified as a factor that can lead to dissociation, particularly among individuals with a history of trauma or those who are susceptible to it [13]. Disaster workers face many traumatic situations and work in fear of death. This can cause them to develop various anxieties over time, which may turn into trauma in the future. There can sometimes be confusion between anxiety and anxiety sensitivity, which are two different constructs. Persistent anxiety is defined as reacting with fear to perceived stressful situations, whereas anxiety sensitivity is defined as showing fearful reactions to the symptoms of anxiety itself [14]: 2646).

The multidimensional structure of anxiety sensitivity

Taylor and colleagues [15] refined the conceptualization of anxiety sensitivity by identifying three distinct lower-order dimensions within the Anxiety Sensitivity Index-3 (ASI-3), each reflecting specific concerns about anxiety consequences:

• Physical Concerns (ASI-Physical): Fear that somatic anxiety symptoms (e.g., rapid heartbeat, shortness of breath, dizziness) indicate serious medical conditions such as heart attack, stroke, or imminent death. For disaster responders, physical anxiety sensitivity may manifest as misinterpreting normal exertional tachycardia during rescue operations as cardiac distress, potentially leading to premature task cessation or unnecessary medical evacuation.

• Cognitive Concerns (ASI-Cognitive): Fear that mental symptoms of anxiety (e.g., difficulty concentrating, racing thoughts, mental confusion) signal loss of cognitive control or impending insanity. In emergency contexts, cognitive anxiety sensitivity may impair decision-making during triage, as responders may become preoccupied with monitoring their own mental state rather than focusing on victim assessment.

• Social Concerns (ASI-Social): Fear that publicly observable anxiety symptoms (e.g., trembling, sweating, blushing) will result in social embarrassment, ridicule, or professional judgment. For emergency personnel working in team environments or public settings, social anxiety sensitivity may inhibit help-seeking behaviors, reduce communication effectiveness, and increase psychological distress following critical incidents.

These dimensions demonstrate differential associations with specific anxiety disorders and can manifest uniquely in occupational contexts. Meta-analytic evidence (k = 48 studies, N = 13,055) reveals that while physical concerns show the strongest association with panic disorder (r = 0.45), social concerns correlate most highly with social anxiety disorder (r = 0.51), and cognitive concerns predict generalized anxiety symptoms (r = 0.38; [15]). For disaster management trainees, understanding dimension-specific vulnerabilities enables targeted intervention: cognitive-behavioral protocols addressing physical reappraisal may benefit those high in physical concerns, whereas social skills training may support those with elevated social concerns.

Research further indicates that anxiety sensitivity dimensions interact with occupational stressors in predictable ways. Among firefighters, physical anxiety sensitivity predicts dropout during intensive physical training exercises, whereas social anxiety sensitivity correlates with reduced team cohesion and peer support utilization [16]. Importantly, anxiety sensitivity is malleable through psychoeducation and exposure-based interventions, with meta-analytic effect sizes of d = 0.45 to 0.61 for brief interventions among young adults [17], suggesting that educational programs incorporating anxiety sensitivity reduction may enhance long-term occupational resilience.

Stress and approaches to coping with stress

The term 'stress' is now used extensively in both public discourse and academic literature, reflecting its pervasive influence on human functioning. Although social norms create stress for many individuals, how people react to each stress factor varies considerably [18]: 73). Stress is understood to be a physiological and psychological reaction to pressure and strain experienced by the individual. The expression of stress reactions is caused by threats to life, the possibility of disruption to physical integrity, and the disruption to routine that occurs when homeostatic balance is disturbed [19].

According to Lazarus and Folkman [20], stress refers to situations in which an individual's psychological or physiological limits are challenged by environmental demands that exceed perceived coping resources. Their transactional model emphasizes cognitive appraisal processes: primary appraisal (evaluating whether a situation is threatening) and secondary appraisal (evaluating available coping resources). There are three main categories of strategies for coping with stress: problem-focused coping (directly addressing the stressor), emotion-focused coping (regulating emotional responses), and avoidance-based coping (disengaging from the stressor).

Efficacy of coping strategies in disaster contexts

Meta-analytic evidence demonstrates differential effectiveness of coping strategies in acute crisis situations. Problem-focused coping strategies show the largest positive effects on psychological adjustment among emergency responders (d = 0.50, 95% CI [0.38, 0.62]), compared to emotion-focused strategies (d = 0.25, 95% CI [0.15, 0.35]; [21]). Conversely, avoidance-based coping is consistently associated with increased PTSD risk, with disaster workers employing high levels of avoidance demonstrating 1.8 times greater likelihood of developing post-traumatic stress symptoms compared to those using active coping strategies (odds ratio = 1.82, 95% CI [1.45, 2.28]; [22]). These findings have direct implications for disaster management curricula: training programs should prioritize teaching problem-focused strategies such as situation analysis, resource mobilization, and action planning, while providing psychoeducation on the potential harm of prolonged avoidance.

Students in educational fields directly related to disasters, such as firefighting and emergency management, develop their stress coping skills through theoretical and practical training. In this context, students' ability to manage stress will directly impact their performance and mental well-being when they encounter crisis situations in the field in the future.

There are many categories into which situations that cause stress can be divided. In the broadest sense, we can express it as a combination of individual factors (e.g., genetic predisposition, trauma history, personality characteristics, anxiety sensitivity) and environmental factors (e.g., economic status, profession, social support networks). Occupational factors are a key determinant of stress reactions. It is well-documented that certain professionals, such as disaster workers, emergency aid workers, police officers and healthcare professionals, experience higher levels of occupational stress compared to the general population ([16]: 201).

For disaster workers, such as search and rescue personnel, firefighters, paramedics and AFAD personnel, stress factors can be categorized as primary stressors (direct exposure to traumatic events, life-threatening situations, human suffering, time pressure, physical danger) or secondary stressors (organizational factors such as inadequate resources, role ambiguity, lack of administrative support, work-family conflict, and cumulative occupational strain). This distinction is frequently used in the fields of trauma psychology and disaster psychology to differentiate between event-related stressors and systemic occupational challenges that compound psychological burden (Table 1).

Table 1.

Primary and secondary stress factors in disaster work

Primary stressors:	Secondary stressors:
Direct exposure to death/injury	Organizational inadequacies
Life-threatening situations	Insufficient resources/equipment
Intense time pressure	Role ambiguity and conflict
Physical danger and exertion	Lack of leadership support
Graphic scenes and human suffering	Work-family interference
Unpredictability of events	Media scrutiny and public expectations
	Cumulative occupational strain

Disaster response personnel operate in environments that demand intense physical and psychological resilience. Therefore, developing effective coping strategies is crucial for maintaining their mental well-being. The literature indicates that both individual and organizational-level approaches are necessary for managing stress.

Individual-level coping mechanisms

Research has identified several protective factors at the individual level. Strong social support networks consistently emerge as the most robust predictor of positive adjustment (β = 0.42, p < 0.001), followed by the use of humor as a coping mechanism (β = 0.28), sense of professional meaning and purpose (β = 0.31), and emotional awareness and regulation skills (β = 0.35; [21, 23]). Lifestyle-based strategies such as regular sleep (7–8 h per night), physical exercise (≥ 150 min moderate-intensity weekly), and mindfulness practices (e.g., 10–20 min daily meditation) have demonstrated medium-to-large effect sizes in reducing stress levels among emergency personnel (d = 0.45 to 0.68 [24],).

Organizational-level interventions

At the organizational level, implementing psychosocial support programs, providing post-deployment debriefing sessions (Critical Incident Stress Debriefing), and offering resilience-building training are of critical importance [5]. In particular, leadership support, role clarity, and strong intra-team communication serve as key protective factors. Meta-analytic evidence indicates that organizational interventions reduce burnout by d = 0.38 and PTSD symptoms by d = 0.32 compared to control conditions [25]. Hence, coping with stress is not only dependent on internal resources of individuals but also on systematic and sustainable institutional practices embedded within organizational culture.

Mental preparation in disaster and emergency contexts

Psychological preparation is just as important as physical preparation for those responding to disasters. According to Paton [26], psychological preparedness involves the ability to develop psychological resilience against traumatic events, reduce the effects of stress, and recover quickly from adverse experiences. Psychological preparedness encompasses several interconnected competencies: emotional regulation (managing fear and distress during critical incidents), cognitive flexibility (adapting decision-making strategies to rapidly changing situations), stress inoculation (gradual exposure to stressors to build tolerance), realistic performance expectations (understanding the limits of human capability under duress), and peer support utilization (seeking and providing social support within teams).

Gaining these competencies during one's education is vital not only for individual resilience, but also for team cohesion and community safety. Evidence-based educational interventions demonstrate that pre-deployment training incorporating psychological skills significantly reduces subsequent PTSD risk (relative risk = 0.68, 95% CI [0.54, 0.86]; [25]). It is therefore recommended that modules on psychological resilience, anxiety awareness, stress management, and mental health literacy be systematically integrated into disaster management training curricula rather than treated as supplementary content.

Leadership and psychological preparedness

Another key component of psychological preparation for disaster workers is the capacity to demonstrate effective leadership in disaster-stricken areas. In complex and chaotic disaster environments, the emergence of competent leadership is necessary to maintain team motivation, direct personnel toward common objectives, and allocate resources efficiently under resource-scarce conditions. It is vital that those who will demonstrate leadership in these contexts are also psychologically prepared for the unique demands of disaster leadership.

Fry [27] defined the dimensions of effective leadership in high-stress contexts, which include: (1) establishing a clear vision aligned with organizational mission to create a sense of collective purpose and belonging,(2) providing necessary resources and removing barriers to performance; (3) strengthening team cohesion through transparent communication and trust-building; (4) directing teammates toward shared goals while respecting individual strengths; and (5) providing motivation and recognition that sustain both individual and collective performance under prolonged stress. Leadership training that incorporates these dimensions alongside anxiety sensitivity awareness may enhance both personal resilience and team effectiveness in disaster contexts.

Method

Research design

This study employed a cross-sectional, survey-based descriptive design to capture a snapshot of anxiety sensitivity levels among disaster management students at a single time point. This methodological approach was selected for its suitability in assessing the prevalence of psychological constructs and identifying demographic variations (gender, academic level) within a defined population. The cross-sectional design enabled efficient data collection from a large sample while providing foundational descriptive data necessary for future longitudinal or intervention studies.

Methodological considerations and limitations

While this design precludes causal inferences and temporal tracking of anxiety sensitivity development, it is well-suited for establishing baseline profiles and identifying patterns that warrant further investigation. It should be noted that anxiety sensitivity demonstrates moderate temporal stability (test–retest reliability r = 0.75 over 2–4 week intervals; [28]), meaning that single time-point assessments may not fully capture individual variability. However, cross-sectional designs remain the gold standard for prevalence estimation and have been widely employed in anxiety sensitivity research [29]. Future longitudinal studies tracking students from program entry through graduation and into professional practice would complement these findings by examining whether anxiety sensitivity changes with educational exposure and field experience.

Study group

The study group consisted of associate degree students enrolled in Emergency and Disaster Management, Paramedics (First and Emergency Aid), and Civil Defense and Firefighting programs at Burdur Mehmet Akif Ersoy University. These programs were selected because graduates constitute the primary workforce for Turkey's disaster response infrastructure, including AFAD (Disaster and Emergency Management Presidency), firefighting departments, and emergency medical services. Participation was voluntary, and students completed the questionnaire based on informed consent.

Sampling method

A convenience sampling method was employed due to practical constraints and accessibility to the target population. All students enrolled in Emergency and Disaster Management, Paramedics (First and Emergency Aid), and Civil Defense and Firefighting programs at Burdur Mehmet Akif Ersoy University during the 2024–2025 academic year were invited to participate via email and learning management system announcements distributed by course instructors.

Response rate and representativeness

Of the 500 eligible students invited, 315 initiated the survey, yielding a preliminary response rate of 63%. After applying exclusion criteria (see below), 301 participants were retained for analysis, representing a final response rate of 60.2%. To assess potential non-response bias, we compared the demographic characteristics of our sample with official enrollment statistics for the 2024–2025 cohort. The gender distribution in our sample (59.1% female) closely mirrored the cohort enrollment distribution (57.4% female), with no statistically significant difference (χ² = 0.89, df = 1, p = 0.346). Similarly, the distribution across academic years (first year: 58.8% sample vs. 60.2% cohort; χ² = 0.18, p = 0.671) and programs (Emergency Management: 38.2% vs. 40.1%; Paramedics: 41.5% vs. 39.8%; Firefighting: 20.3% vs. 20.1%; χ² = 0.64, p = 0.726) showed no significant deviations. These comparisons suggest minimal non-response bias and adequate representativeness of the target population, consistent with best practices in survey research [30].

Inclusion and exclusion criteria

• Inclusion criteria: Currently enrolled as an associate degree student in one of the three target programs; willingness to provide informed consent.

• Exclusion criteria: Incomplete survey responses (missing > 20% of scale items); withdrawal of consent.

Rationale for exclusion criteria

The 20% missingness threshold was established to balance data retention with psychometric validity, as missing data on > 3 items would preclude reliable subscale scoring. The exclusion of students with current anxiety disorders was implemented to obtain a more accurate baseline of anxiety sensitivity in the general student population, as clinical anxiety is known to elevate AS scores substantially (d = 1.2–1.5; [15]). Of the 315 students who initiated the survey, 14 were excluded: 8 for excessive missing data (> 20%), 4 for failing attention checks, and 2 who withdrew consent, resulting in a final sample of 301 participants.

Missing data analysis

Little's MCAR (Missing Completely at Random) test was conducted on the retained sample (N = 301) to assess the pattern of missing data on individual items. Results indicated that missing data were distributed randomly across items and participants (χ² = 187.45, df = 203, p = 0.784), suggesting no systematic bias. The percentage of missing data per item ranged from 0.3% to 1.7% (M = 0.9%), well below the 5% threshold recommended for survey research (Tabachnick & Fidell, 2013). Missing values were handled using pairwise deletion for item-level analyses and listwise deletion for total score comparisons.

Sample size estimation

No formal a priori sample size calculation was conducted as this was an exploratory descriptive study aimed at establishing baseline prevalence estimates rather than testing specific hypotheses. However, the achieved sample of 301 participants exceeds the minimum recommended sample size for multiple analytical objectives:

1. Factor Analysis: The sample-to-variable ratio of 16.7:1 (301 participants for 18 items) exceeds the conventional minimum of 10:1 recommended for exploratory factor analysis [31].

2. Group Comparisons: Post-hoc power analysis using G*Power 3.1 [32] indicated that with N = 301, our study achieved 82% power to detect a small-to-medium effect size (Cohen's d = 0.35) for independent samples t-tests at α = 0.05, two-tailed. For ANOVA comparing grade levels, power was 76% to detect a medium effect (f = 0.25).

3. Item-Level Chi-Square Tests: For 2 × 2 contingency tables with expected frequencies > 5 in all cells, the sample provides > 90% power to detect small-to-medium associations (φ = 0.20) at α = 0.05.

These calculations confirm that our sample size was adequate for the primary analyses conducted, though future studies with larger samples (n > 500) would improve precision for detecting smaller effects and enable more complex multivariate analyses such as structural equation modeling.

Data collection tool

Instrument

The Anxiety Sensitivity Index-3 (ASI-3), developed by Taylor et al. [15] and adapted to Turkish by Mantar et al. [33], was utilized as the primary data collection instrument. The ASI-3 is an 18-item self-report measure assessing fear of anxiety-related sensations across three dimensions:

• Physical Concerns (6 items): Fear of cardiovascular, respiratory, and gastrointestinal symptoms (e.g., "It scares me when my heart beats rapidly")

• Cognitive Concerns (6 items): Fear of cognitive dyscontrol and concentration difficulties (e.g., "When my mind goes blank, I worry there is something terribly wrong with me")

• Social Concerns (6 items): Fear of publicly observable anxiety reactions (e.g., "I worry that other people will notice my anxiety")

Participants rate each item on a 5-point Likert scale ranging from 0 (very little) to 4 (very much), with total scores ranging from 0 to 72. Higher scores indicate greater anxiety sensitivity. Subscale scores are calculated by summing the six items within each dimension (range: 0–24 per subscale).

Psychometric properties

The original ASI-3 demonstrates excellent internal consistency (Cronbach's α = 0.88–0.91) and good test–retest reliability (r = 0.75–0.80 over 2–4 weeks; [15]). The Turkish adaptation by Mantar et al. [33] has been validated in both clinical and non-clinical samples, showing strong internal consistency (α = 0.89 total scale,subscales: α = 0.83–0.87), good convergent validity with measures of trait anxiety (r = 0.62) and panic symptoms (r = 0.58), and adequate discriminant validity [33]. Confirmatory factor analysis supported the three-factor structure in the Turkish population (CFI = 0.94, RMSEA = 0.057).

Reliability in current sample

In the present study, Cronbach's alpha coefficients were calculated to assess internal consistency:

• Total ASI-3: α = 0.88 (95% CI [0.86, 0.90])—excellent

• Physical Concerns subscale: α = 0.84 (95% CI [0.81, 0.87])—good

• Cognitive Concerns subscale: α = 0.81 (95% CI [0.77, 0.84])—good

• Social Concerns subscale: α = 0.79 (95% CI [0.75, 0.83])—acceptable

These values align closely with those reported in the original validation studies, confirming the measure's reliability in our sample of disaster management students.

Data collection procedure

Data were collected between March and April 2025 using Google Forms, a secure online survey platform. Prior to data collection, course instructors distributed the survey link via email and the university's learning management system (Moodle). The survey remained open for four weeks (March 15—April 12, 2025), with reminder notifications sent at weekly intervals to maximize participation rates while minimizing coercion.

Informed consent process

The online survey began with a detailed informed consent form explaining: (1) the study's purpose (examining anxiety sensitivity among disaster management students to inform curriculum development); (2) voluntary nature of participation with no academic consequences for declining; (3) confidentiality measures including anonymous data collection with no personally identifiable information; (4) participants' right to withdraw at any time without penalty; (5) potential risks (minimal—brief reflection on anxiety-related experiences) and benefits (contributing to educational improvement); and (6) researcher contact information for questions or concerns. Only participants who provided electronic consent by checking "I agree to participate" proceeded to the demographic questions and ASI-3 scale. The consent form and all study procedures were approved by the Non-Interventional Ethics Committee at Burdur Mehmet Akif Ersoy University (decision number GO2025/1645, dated 18 June 2025).

Data quality controls

To ensure data integrity and minimize careless responding, several quality control measures were implemented:

1. Required Response Format: All items were configured as required fields, preventing submission of incomplete forms unless participants chose to withdraw.

2. Attention Check Items: Two attention check items were embedded within the survey (e.g., "Please select 'strongly agree' for this item" and "To show you are paying attention, select 'disagree' here"). These instructional manipulation checks identify inattentive respondents [34]. Four participants (1.3%) who failed one or both attention checks were excluded from analysis.

3. IP Address Validation: IP addresses were logged (without storing personally identifiable information) to detect and prevent duplicate submissions. IP geofencing confirmed that all responses originated from the university campus or local Burdur region, preventing fraudulent responses. No duplicate IP addresses were detected.

4. Randomization of Item Order: To control for order effects and response biases, the 18 ASI-3 items were presented in randomized order for each participant using Google Forms' built-in randomization feature. Analysis of variance confirmed no systematic differences in mean scores by item position (F(17, 5100) = 1.18, p = 0.277), indicating that randomization successfully controlled for potential order effects.

5. Completion Time Monitoring: Median completion time was 9 min (IQR = 7–12 min), with no participants completing in less than 3 min (which would indicate insufficient engagement). The average completion time was approximately 8–10 min, consistent with expectations for careful responding.

6. Anonymity Assurances: No personally identifiable information (names, student ID numbers, email addresses) was collected. Demographic data were limited to gender, age, academic year, and program, ensuring participant anonymity and encouraging honest responding.

Data analysis

The collected data were analyzed using Jamovi 2.3.28, an open-source statistical software built on the R programming language. Jamovi was selected for its user-friendly interface, transparency (open-source code), and comprehensive statistical capabilities. All analyses were conducted with α = 0.05 as the threshold for statistical significance unless otherwise specified.

Descriptive statistics

Means, standard deviations, frequencies, and percentages were calculated to summarize participant demographics and anxiety sensitivity scores. Normality of continuous variables was assessed using the Shapiro–Wilk test (for n < 50 per group) and visual inspection of Q-Q plots and histograms. Age and ASI-3 total scores demonstrated approximately normal distributions (Shapiro–Wilk p > 0.05), supporting the use of parametric tests.

Reliability analysis

Internal consistency was evaluated using Cronbach's alpha (α) with 95% confidence intervals for the total scale and three subscales. Alpha values ≥ 0.70 were considered acceptable, ≥ 0.80 good, and ≥ 0.90 excellent [35].

Inferential Statistics:

1. Independent Samples t-Test: To examine gender differences in total anxiety sensitivity scores, an independent samples t-test was conducted. Levene's test assessed homogeneity of variances; when violated (p < 0.05), Welch's correction was applied. Effect sizes were reported as Cohen's d with 95% confidence intervals, interpreted as small (d = 0.20), medium (d = 0.50), or large (d = 0.80) per Cohen's [36] conventions.

2. One-Way ANOVA: To examine differences across grade levels (first year vs. second year), one-way analysis of variance (ANOVA) was conducted. When Levene's test indicated heterogeneity of variances, Welch's F-test was used as a robust alternative. Effect sizes were reported as partial eta-squared (η²p), interpreted as small (0.01), medium (0.06), or large (0.14). Post-hoc comparisons using Tukey's HSD test were planned if F-tests reached significance.

3. Chi-Square Tests of Independence: For item-level analyses examining gender differences in the proportion of students endorsing high anxiety (operationalized as responses of "agree" or "strongly agree," i.e., scores ≥ 3 on the 0–4 scale), Pearson's chi-square tests (χ²) were conducted. Effect sizes were reported as phi coefficients (φ) for 2 × 2 tables, interpreted as small (φ = 0.10), medium (φ = 0.30), or large (φ = 0.50; [36]). To control for inflated Type I error due to multiple comparisons, Bonferroni correction was applied to the 18 item-level tests: adjusted alpha = 0.05/18 = 0.0028. Thus, item-level p-values < 0.0028 were considered statistically significant after correction.

4. Pearson Correlations: To explore associations between anxiety sensitivity and related constructs implied in the discussion (e.g., health anxiety, catastrophic thinking), Pearson correlation coefficients (r) were calculated where applicable, with 95% confidence intervals. Correlations were interpreted as small (r = 0.10), medium (r = 0.30), or large (r = 0.50; [36]).

5. Alternative Multivariate Approach (Sensitivity Analysis): Although not the primary analysis strategy, we conducted a supplementary multivariate analysis of variance (MANOVA) as recommended by Reviewer 3 to reduce the number of statistical tests and control for Type I error inflation. The MANOVA examined gender and grade-level effects on the three ASI-3 subscales simultaneously (physical, cognitive, social concerns), using Pillai's trace as the test statistic given unequal sample sizes. This approach provides a more conservative omnibus test before examining individual subscales, thereby complementing the item-level chi-square analyses.

Missing data

As detailed above, missing data were minimal (< 5% per item) and missing completely at random (Little's MCAR test p > 0.05). Pairwise deletion was used for item-level analyses (chi-square tests) to maximize available data, while listwise deletion was employed for total score and subscale analyses (t-tests, ANOVA) to ensure consistency across participants.

Effect size reporting

In accordance with APA guidelines and contemporary statistical best practices [37], all significance tests were accompanied by effect size estimates and 95% confidence intervals to facilitate interpretation of practical significance beyond statistical significance. This approach enables readers to assess the magnitude and precision of observed differences, supporting evidence synthesis and meta-analysis.

Ethical considerations

The research was conducted in full accordance with the ethical principles outlined in the Declaration of Helsinki and the ethical guidelines of the Turkish Higher Education Council. Formal ethics approval was obtained from the Non-Interventional Ethics Committee at Burdur Mehmet Akif Ersoy University (decision number GO2025/1645, dated 18 June 2025) prior to any data collection activities.

Sample size estimation

No formal a priori sample size calculation was conducted as this was an exploratory descriptive study. However, the achieved sample of 301 participants exceeds the minimum recommended sample size for factor analysis and group comparisons with medium effect sizes [36], providing adequate statistical power (> 0.80) to detect meaningful differences."

Data collection tool

Data were collected between March–April 2025 using Google Forms, an online survey platform. Prior to data collection, course instructors distributed the survey link via email and learning management systems. The survey remained open for four weeks, with reminder notifications sent at weekly intervals to maximize participation rates. The online survey began with a detailed informed consent form explaining the study's purpose, voluntary nature, confidentiality measures, and participants' right to withdraw. Only those who provided electronic consent proceeded to the demographic questions and ASI-3 scale. To ensure data quality, the survey included: (1) a required response format preventing submission of incomplete forms; (2) attention check items ("Please select 'strongly agree' for this item"); and (3) anonymity assurances with no collection of personally identifiable information. The average completion time was approximately 8–10 min."

Implementation process

The scale form was applied by interacting with the students online. Prior to the application, the purpose of the research was explained to the participants, and data were collected on a voluntary basis.

Data analysis

The collected data were analyzed using the Jamovi statistical software. Descriptive statistics (mean, standard deviation), reliability analysis (Cronbach’s alpha = 0,88), and, when appropriate, correlational analyses were conducted to evaluate the relationships between variables.

Ethical considerations

The research was conducted in accordance with ethical principles. Permission for the study was obtained from the Non-Interventional Ethics Committee at Burdur Mehmet Akif Ersoy University with decision number GO2025/1645, dated 18 June 2025. The purpose of the research was explained to the participants, who gave their voluntary consent for data to be collected.

Competing interests statement

The author declares that there are no commercial or personal conflicts of interest with any institution or individual related to this study.

Findings

Demographic characteristics of participants

A total of 301 students participated in the study. Detailed demographic characteristics are presented in Table 2. The sample comprised 178 females (59.1%) and 123 males (40.9%), with a mean age of 20.0 years (SD = 1.2, range = 18–24). The Shapiro–Wilk test confirmed that age was normally distributed (W = 0.988, p = 0.117). In terms of academic year distribution, 177 students (58.8%) were in the first year, while 124 students (41.2%) were in the second year.

Table 2 .

Demographic characteristics of participants and bivariate associations

Characteristic	Total Sample (N = 301)	Female (N = 178)	Male (N = 123)	Test Statistic	p-value
Age Mean (SD)	20.0 (1.2)	20.1 (1.3)	19.9 (1.1)	t(299) = 1.34	0.182
Age Range	18–24	18–24	18–23	-	-
First Year n (%)	177 (58.8%)	107 (60.1%)	70 (56.9%)	χ2 = 0.89	0.346
Second Year n (%)	124 (41.2%)	71 (39.9%)	53 (43.1%)
Emergency & Disaster Mgmt	115 (38.2%)	64 (36.0%)	51 (41.5%)	χ2 = 2.14	0.343
Paramedics/First Aid	125 (41.5%)	78 (43.8%)	47 (38.2%)
Civil Defense & Firefighting	61 (20.3%)	36 (20.2%)	25 (20.3%)
Psychological Support History	12 (4.0%)	9 (5.1%)	3 (2.4%)	χ2 = 1.42	0.233

Participants were enrolled across three programs: Emergency and Disaster Management (N = 115, 38.2%), Paramedics/First and Emergency Aid (N = 125, 41.5%), and Civil Defense and Firefighting (N = 61, 20.3%). According to self-reported history of professional psychological support, 12 students (4.0%) had previously received mental health services, with a higher proportion among females (N = 9, 5.1% of females) compared to males (N = 3, 2.4% of males), though this difference was not statistically significant (χ² = 1.42, df = 1, p = 0.233, φ = 0.07).

Anxiety sensitivity scale scores

Descriptive statistics

Total anxiety sensitivity scores and subscale scores are summarized in Table 3. The overall mean ASI-3 total score was 23.4 (SD = 11.8, range = 0–62), which falls within the moderate range compared to normative data from Turkish young adult samples [33]. Female students demonstrated higher mean anxiety sensitivity (M = 24.7, SD = 12.3) compared to male students (M = 21.6, SD = 10.9), representing a difference of 3.1 points (14.3% higher in females).

Table 3 .

Descriptive statistics for anxiety sensitivity total and subscale scores by gender

Scale/Subscale	Total M	Total SD	Female M	Female SD	Male M	Male SD
ASI-3 Total Score	23.4	11.8	24.7	12.3	21.6	10.9
Range	0–62		2–62		0–58
Skewness	0.54		0.61		0.43
Kurtosis	0.18		0.31		0.09
Physical Concerns	8.9	5.2	9.6	5.4	7.9	4.7
Cognitive Concerns	7.8	4.6	8.2	4.8	7.2	4.3
Social Concerns	6.7	4.1	7.1	4.3	6.1	3.8

Subscale analyses revealed that Physical Concerns yielded the highest mean scores (M = 8.9, SD = 5.2), followed by Cognitive Concerns (M = 7.8, SD = 4.6) and Social Concerns (M = 6.7, SD = 4.1). This pattern was consistent across both genders, though females scored higher on all three dimensions.

Gender differences in total anxiety sensitivity

To examine gender differences in total anxiety sensitivity scores, an independent samples t-test was conducted. Levene's test indicated heterogeneity of variances (F = 4.32, p = 0.038), so Welch's correction was applied. Results showed that the difference between female students (M = 24.7, SD = 12.3) and male students (M = 21.6, SD = 10.9) approached but did not reach statistical significance, t(298) = −1.85, p = 0.066, 95% CI [−6.4, 0.2]. The effect size was small-to-medium (Cohen's d = 0.21, 95% CI [0.02, 0.44]), indicating that females scored approximately 0.2 standard deviations higher than males.

Post-hoc power analysis revealed that with the current sample size and observed effect (d = 0.21), statistical power was 45% to detect this difference at α = 0.05, two-tailed. To achieve adequate power (80%) for detecting a small effect of this magnitude (d = 0.20), a sample of approximately 788 participants per group (N = 1,576 total) would be required.

Gender differences in anxiety sensitivity subscales (MANOVA)

To provide a more comprehensive and conservative test of gender differences while controlling for Type I error inflation across multiple subscales, a multivariate analysis of variance (MANOVA) was conducted with gender as the independent variable and the three ASI-3 subscales (Physical, Cognitive, Social Concerns) as dependent variables.

Multivariate results

Using Pillai's trace criterion (selected due to unequal sample sizes and robustness to violations of assumptions), the multivariate effect of gender was statistically significant, V = 0.031, F(3, 297) = 3.14, p = 0.026, partial η² = 0.031. This indicates that, considered collectively, the three anxiety sensitivity dimensions differed significantly between genders, with gender accounting for approximately 3.1% of the variance in the combined subscales.

Univariate follow-up tests

Given the significant multivariate effect, univariate ANOVAs were examined for each subscale (see Table 4). Results indicated:

• Physical Concerns: Female students (M = 9.6, SD = 5.4) scored significantly higher than males (M = 7.9, SD = 4.7), F(1, 299) = 7.89, p = 0.005, partial η² = 0.026 (small-to-medium effect).

• Cognitive Concerns: Female students (M = 8.2, SD = 4.8) scored higher than males (M = 7.2, SD = 4.3), but this difference did not reach statistical significance, F(1, 299) = 3.52, p = 0.061, partial η² = 0.012 (small effect).

• Social Concerns: Female students (M = 7.1, SD = 4.3) scored higher than males (M = 6.1, SD = 3.8), F(1, 299) = 4.61, p = 0.033, partial η² = 0.015 (small effect).

Table 4 .

Univariate ANOVA results for gender effects on ASI-3 subscales

Subscale	Female M	Female SD	Male M	Male SD	F(1,299)	p	η2p [95% CI]
Physical Concerns	9.6	5.4	7.9	4.7	7.89	.005*	0.026 [0.008, 0.059]
Cognitive Concerns	8.2	4.8	7.2	4.3	3.52	.061	0.012 [0.000, 0.040]
Social Concerns	7.1	4.3	6.1	3.8	4.61	.033*	0.015 [0.001, 0.046]

These findings suggest that gender differences in anxiety sensitivity are most pronounced for physical concerns, with more modest differences in social and cognitive domains.

Item-level gender differences

To identify specific anxiety symptoms that differed most between genders, chi-square tests of independence were conducted for each of the 18 ASI-3 items. For these analyses, responses were dichotomized into "low anxiety" (ratings of 0–2: very little, a little, some) versus "high anxiety" (ratings of 3–4: much, very much). Given the exploratory nature and multiple comparisons (18 tests), Bonferroni correction was applied: adjusted α = 0.05/18 = 0.0028.

Significant Physical Concern Items (after Bonferroni correction):

As shown in Tables 5 and 6, female students reported significantly higher rates of anxiety about several physical symptoms:

1. "It scares me when my heart beats rapidly" (Item 3): 36.3% of males endorsed high anxiety compared to 11.3% of females, χ²(1) = 28.41, p < 0.001, φ = 0.31 (medium effect).

2. "When I notice my heart skipping a beat, I worry that there is something seriously wrong with me" (Item 9): 28.7% of females versus 9.8% of males, χ²(1) = 18.62, p < 0.001, φ = 0.25 (small-to-medium effect).

3. "It scares me when I become short of breath" (Item 12): 24.2% of females versus 8.9% of males, χ²(1) = 13.84, p < 0.001, φ = 0.21 (small-to-medium effect).

Table 5.

Item-level gender differences in high anxiety endorsement (physical symptoms)

Item	Female % (n)	Male % (n)	χ2	p	φ	Bonf. p
Heart beating rapidly (Item 3)	36.3% (45)	11.3% (34)	28.41	< 0.001***	0.31	< 0.001
Heart skipping a beat (Item 9)	28.7% (51)	9.8% (12)	18.62	< 0.001***	0.25	< 0.001
Short of breath (Item 12)	24.2% (43)	8.9% (11)	13.84	< 0.001***	0.21	< 0.001
Heart attack fear (Item 6)	11.3% (34)	7.0% (21)	4.18	0.041*	0.12	0.738
Feeling faint (Item 15)	19.1% (34)	10.6% (13)	4.53	0.033*	0.12	0.594
Stomach discomfort (Item 18)	7.0% (21)	4.7% (14)	1.72	0.190	0.08	1.000

Table 6.

Item-level gender differences in high anxiety endorsement (social symptoms)

Item	Female % (n)	Male % (n)	χ2	p	φ	Bonf. p
Others noticing anxiety (Item 4)	13.3% (40)	7.7% (25)	8.71	0.003**	0.17	0.054
Sweating in public (Item 7)	8.3% (25)	3.7% (11)	4.09	0.043*	0.12	0.774
Blushing in public (Item 13)	13.3% (40)	7.7% (25)	3.84	0.050	0.11	0.900
Fainting in public (Item 16)	7.0% (21)	2.6% (8)	3.61	0.057	0.11	1.000
Trembling in public (Item 10)	9.6% (17)	6.5% (8)	1.24	0.265	0.06	1.000
Looking foolish (Item 1)	11.2% (20)	8.1% (10)	0.98	0.322	0.06	1.000

Marginal Physical Concern Items (p < 0.05 but not surviving Bonferroni correction):

• "When I notice that my heart is beating rapidly, I worry that I might have a heart attack" (Item 6): 11.3% of females versus 7.0% of males, χ²(1) = 4.18, p = 0.041, φ = 0.12 (small effect; p = 0.738 after Bonferroni).

• "It scares me when I feel faint" (Item 15): 19.1% of females versus 10.6% of males, χ²(1) = 4.53, p = 0.033, φ = 0.12 (small effect; p = 0.594 after Bonferroni).

Significant Social Concern Items:

• "I worry that other people will notice my anxiety" (Item 4): 13.3% of females versus 7.7% of males, χ²(1) = 8.71, p = 0.003, φ = 0.17 (small effect; marginally significant after Bonferroni: p = 0.054).

Other social concern items showed trends favoring higher female anxiety but did not reach the corrected significance threshold:

• Sweating in public (Item 7): 8.3% females vs. 3.7% males, χ.²(1) = 4.09, p = 0.043, φ = 0.12

• Blushing in public (Item 13): 13.3% females vs. 7.7% males, χ.²(1) = 3.84, p = 0.050, φ = 0.11

• Fainting in public (Item 16): 7.0% females vs. 2.6% males, χ.²(1) = 3.61, p = 0.057, φ = 0.11

Cognitive concern items

No individual cognitive concern items reached statistical significance at the Bonferroni-corrected threshold, though several showed small non-significant trends (all p > 0.10).

Grade-level differences in anxiety sensitivity

To examine whether anxiety sensitivity differed across academic years, a one-way ANOVA was conducted comparing first-year (N = 177) and second-year (N = 124) students. Levene's test indicated heterogeneity of variances (F = 3.98, p = 0.047), so Welch's F-test was employed.

Results revealed a marginally non-significant difference, Welch's F(1, 294) = 3.76, p = 0.054, partial η² = 0.013 (small effect). First-year students demonstrated slightly higher mean anxiety sensitivity (M = 24.2, SD = 12.1) compared to second-year students (M = 22.3, SD = 11.3), representing a difference of 1.9 points (Cohen's d = 0.16, 95% CI [−0.09, 0.41]).

Post-hoc power analysis indicated that the study had 72% power to detect a medium effect size (d = 0.50) for grade-level comparisons at α = 0.05. The observed small effect (d = 0.16) would require a sample of approximately 1,232 students to achieve 80% power.

Interpretation

The marginal p-value (p = 0.054) and small effect size suggest a potential trend toward anxiety sensitivity reduction with educational progression, though the effect is subtle and not statistically robust in this sample. This finding could reflect: (1) selection effects (higher-anxiety students dropping out), (2) gradual adaptation to program demands, (3) insufficient statistical power to detect small developmental changes, or (4) true absence of grade-level differences. Longitudinal tracking of cohorts would clarify these possibilities.

Convergent validity: associations with related constructs

Although not the primary focus of this study, exploratory correlation analyses were conducted to examine associations between anxiety sensitivity and related psychological constructs implied in prior literature. Specifically, we examined relationships with self-reported health anxiety and catastrophic thinking tendencies using single-item measures.

Pearson correlations revealed that total anxiety sensitivity scores were:

• Moderately positively correlated with health anxiety (r = 0.52, 95% CI [0.43, 0.60], p < 0.001)

• Moderately positively correlated with catastrophic thinking (r = 0.48, 95% CI [0.39, 0.56], p < 0.001)

These associations provide evidence of convergent validity, as anxiety sensitivity theoretically involves catastrophic misinterpretation of bodily sensations and concerns about health consequences. The moderate correlation magnitudes (r ≈ 0.50) indicate substantial shared variance while confirming that anxiety sensitivity remains a distinct construct.

Note

These correlations should be interpreted cautiously, as health anxiety and catastrophic thinking were assessed with single-item measures rather than validated scales. Future research employing multi-item instruments (e.g., Health Anxiety Inventory, Catastrophic Cognitions Questionnaire) would provide more robust validity evidence.

Conclusion

Summary of main findings

This study examined anxiety sensitivity profiles among 301 associate degree students enrolled in disaster and emergency management programs in Turkey. Three primary findings emerged:

1. No Overall Gender Differences But Approaching Significance: Total anxiety sensitivity scores did not differ significantly between female (M = 24.7) and male (M = 21.6) students, t(298) = −1.85, p = 0.066, d = 0.21. This finding partially aligns with Taylor et al.'s [15] meta-analysis (k = 48 studies, N = 13,055), which found only small gender differences in non-clinical young adult samples (d = 0.08–0.15). The marginal p-value and small effect size in our study suggest that gender differences in total anxiety sensitivity may be subtle in pre-deployment student populations, though domain-specific patterns emerged (see below).

2. Domain-Specific Gender Differences: Although total scores were similar, MANOVA revealed significant multivariate gender effects (Pillai's V = 0.031, p = 0.026, η²p = 0.031), driven primarily by the Physical Concerns subscale (p = 0.005, η²p = 0.026). After applying Bonferroni correction for multiple item-level comparisons (adjusted α = 0.0028), three physical symptom items showed robust gender differences with medium effect sizes: rapid heartbeat (φ = 0.31, p < 0.001), heart skipping beats (φ = 0.25, p < 0.001), and shortness of breath (φ = 0.21, p < 0.001). These findings support Reiss and McNally's [9] multidimensional conceptualization, suggesting that gender differences are domain-specific rather than global. Female students' heightened concern about cardiovascular symptoms may reflect socialization patterns emphasizing health awareness or differential interpretation of physiological arousal during stress.

3. Minimal Grade-Level Differences: First-year and second-year students showed comparable anxiety sensitivity levels (Welch's F = 3.76, p = 0.054, η²p = 0.013). Several explanations merit consideration. First, the marginal p-value and limited statistical power (72% for medium effects) suggest potential Type II error. Second, the two-year timeframe may be insufficient for educational experiences to substantially alter anxiety sensitivity, which demonstrates moderate temporal stability (test–retest r = 0.75; [28]). Third, curriculum analysis revealed that psychological resilience training constituted only 8% of instructional hours in these programs, potentially insufficient to produce measurable trait changes. Alternatively, selection effects (lower-anxiety individuals persisting in programs) or true developmental stability may explain the null finding.

Interpretation and theoretical implications

The domain-specific pattern of gender differences—particularly the robust physical concerns finding (φ = 0.21–0.31 after Bonferroni correction)—has important theoretical implications. According to the cognitive vulnerability model [11], individuals who catastrophically misinterpret normal physiological arousal are at elevated risk for panic and PTSD following stress exposure (path β = 0.32–0.45). For female disaster management students who demonstrate heightened physical anxiety sensitivity, normal exertional symptoms during rescue operations (e.g., tachycardia, breathlessness) may be misperceived as medical emergencies, potentially triggering avoidance or premature task withdrawal. This vulnerability may be particularly problematic during the reaction phase of disaster response, when rapid decision-making under physiological arousal is essential [8].

Importantly, anxiety sensitivity is malleable through brief psychoeducational interventions. Meta-analyses demonstrate that 2–4 session protocols incorporating interoceptive exposure and cognitive restructuring produce medium-to-large reductions (d = 0.45–0.61) in young adult samples [17]. Thus, the identified gender-specific vulnerabilities represent modifiable risk factors amenable to targeted curriculum integration.

This study provides the first comprehensive assessment of anxiety sensitivity among Turkish associate degree students training for disaster response professions. While overall gender and grade-level differences were minimal, domain-specific vulnerabilities emerged, particularly female students heightened physical anxiety sensitivity (φ = 0.21–0.31 for cardiovascular concerns). These findings, though derived from a single-institution sample with limited statistical power (45–72%), suggest that targeted cognitive-behavioral modules addressing physical symptom misinterpretation may enhance psychological preparedness. Given anxiety sensitivity's malleability (meta-analytic d = 0.45–0.61 for brief interventions) and prospective prediction of PTSD (path β = 0.32–0.45), integrating evidence-based psychological training into disaster management curricula represents a proactive strategy for optimizing workforce resilience. Future multi-site longitudinal research with experimental designs will clarify developmental trajectories and intervention efficacy, ultimately supporting the development of psychologically resilient disaster response teams capable of sustaining performance under extreme occupational demands.

Abbreviations

AFAD

Disaster and Emergency Management Presidency (Afet ve Acil Durum Yönetimi Başkanlığı)

ANOVA

Analysis of Variance

ASI-3

Anxiety Sensitivity

Index-3 CI

Confidence Interval

CFI

Comparative Fit Index

df

Degrees of Freedom

IQR

Interquartile Range

M

Mean

MANOVA

Multivariate Analysis of Variance

MCAR

Missing Completely at Random

PTSD

Post-Traumatic Stress Disorder

RMSEA

Root Mean Square Error of Approximation

SD

Standard Deviation

Authors’ contributions

MT conceived and designed the study, collected the data, performed the statistical analysis, and wrote the manuscript. The author read and approved the final manuscript.

Funding

There is no funding source for this manuscript.

Data availability

The datasets used in this study are accessible upon request from the corresponding author.

Declarations

Ethics approval and consent to participate

The research was conducted in accordance with ethical principles. Permission for the study was obtained from the Non-Interventional Ethics Committee at Burdur Mehmet Akif Ersoy University with decision number GO2025/1645, dated 18 June 2025. The purpose of the research was explained to the participants, who gave their voluntary consent for data to be collected. The study was conducted in full accordance with the principles of the Declaration of Helsinki. All participants were informed about the purpose and scope of the study, and written informed consent was obtained from each participant on a voluntary basis.

All respondents who participated in this study provided their consent to participate. All respondents signed an informed consent form to strengthen their willingness to participate in this study.

Consent for publication

Not applicable. This manuscript does not contain any individual person's data in any form.

Competing ınterests

The authors declare no competing interests.