Toward Sustainable Mental Health: Development and Validation of the Brief Anxiety Scale for Climate Change (BACC) in South Korea

Abstract

Climate change disrupts lives globally and poses significant challenges to mental health. Although several scales assess climate anxiety, many either conflate symptoms with coping responses or fail to adequately capture the core symptomatology of anxiety. Hence, this study aimed to develop and validate the Brief Anxiety Scale for Climate Change (BACC), a self-report measure designed to assess symptoms of climate anxiety. A preliminary pool of 21 items was generated based on the diagnostic criteria for generalized anxiety disorder and climate-related stress. Study 1 (n = 300) explored the factor structure via an exploratory factor analysis while Study 2 (n = 400) independently validated the structure via a confirmatory factor analysis (CFA). Analyses of the internal consistency, content validity, and discriminant validity helped refine the scale to a final 13-item version with two factors: cognitive and functional impairment. The CFA results indicated that all the fit indices met the recommended thresholds, and the final version demonstrated excellent internal consistency (Cronbach’s α = 0.92). Additionally, latent correlations revealed that climate anxiety was moderately associated with generalized anxiety and depression. The BACC was developed to identify individuals in the community who experience climate anxiety beyond an adaptive level, thereby promoting sustainable mental health in the context of climate change. These findings suggest that the BACC is a promising tool for assessing climate anxiety. With better identification, mental health professionals, community practitioners, and policymakers can utilize the scale to develop climate-sensitive public health programs and tailored intervention strategies.

1. Introduction

The United Nations (UN) has highlighted the widespread and persistent nature of the climate crisis and recognized its profound impact on global mental health [1]. According to a World Health Organization (WHO) policy brief, climate change poses significant psychological risks, including emotional distress, anxiety, depression, and suicidal behavior. Furthermore, the WHO highlights the importance of integrating mental health and psychological support (MHPSS) into national climate change response strategies to address these challenges [1].

Among the various mental health effects of climate change, anxiety has emerged as the most prominent emotional response [2]. One of the key reasons for this is that climate change represents a real, persistent, and long-term threat that is difficult to predict, control, or resolve [3]. Anxiety, by nature, is a psychological response linked to the activation of the behavioral inhibition system (BIS), which regulates individuals’ sensitivity to threats and their tendency to avoid potential harm [4]. In the context of climate change, this unresolved and existential threat can intensify uncertainty and worry about the future, including concerns for one’s own safety and that of one’s children [5]. As a result, such prolonged emotional responses may develop into clinically significant levels of anxiety [3].

This form of anxiety, when specifically related to climate change, is commonly referred to as “climate anxiety”. It is defined as the anxiety experienced by individuals as a result of their perception of climate change [3,6]. A high proportion of adults reported climate anxiety in 2022 (20 s: 43%, 30 s: 57.6%, 40 s: 50.8%, 50 s: 50.3%, 60+: 41.7%) [7], indicating that, in most age groups, nearly half the population was concerned about climate change. Notably, this trend was most pronounced among youths and adults.

Various instruments have been proposed to assess climate anxiety. The Hogg Eco-Anxiety Scale (HEAS) [8] and Climate Change Anxiety Scale (CCAS) [9] are two of the most prominent tools that have laid foundational groundwork through their multidimensional approaches. While these scales capture the broad characteristics of climate anxiety and have made important contributions to the field, they exhibit certain limitations in fully addressing the clinically and developmentally salient features of the construct.

While the HEAS provides a well-structured approach for assessing climate-related worry, it does not include irritability, a symptom that is particularly salient in developmental contexts. It can manifest as a behavioral expression of climate-related distress [10] and is widely recognized as a transdiagnostic feature across anxiety and mood disorders [11]. In particular, irritability is considered a core symptom of anxiety disorders among adolescents [12,13]. Excluding this symptom may limit the developmental sensitivity of the scale, particularly in capturing adolescent-specific manifestations of anxiety.

In addition, the scale does not include items that capture feedback from others regarding preoccupation with climate change. Relying solely on self-reports may limit the accuracy of symptom assessment among adolescents; hence, it is important to incorporate observational input from others, such as through a multi-informant approach, which is repeatedly emphasized in the assessment of youth psychopathology [14,15,16,17]. Therefore, including items that indirectly reflect others’ perspectives may enhance the scale’s clinical sensitivity in assessing adolescent populations.

The CCAS represents a foundational effort to systematically assess the multidimensional nature of climate anxiety through two key factors: cognitive-emotional and functional impairment. While it has made significant contributions to the field, some items may be interpreted as assessing behavioral coping responses rather than emotional or cognitive distress regarding climate change (e.g., “I go away by myself and think about why I feel this way about climate change” or “I write down my thoughts about climate change and analyze them”).

Researchers have argued in favor of incorporating coping responses into climate anxiety scales to capture a broader range of psychological reactions; however, the specific behavioral items included in the CCAS do not align with the three well-established coping categories commonly discussed in climate anxiety literature: problem-focused, emotion-focused, and meaning-focused coping. This discrepancy not only complicates their interpretation as valid indicators of symptomatology [18,19,20]; it also blurs the boundaries between symptoms and coping responses, which may hinder the scale’s ability to capture the severity of climate anxiety symptoms as a clinical construct.

A validation study noted these concerns, where some items that reflected coping reactions demonstrated particularly low factor loadings or failed to demonstrate meaningful associations with established measures of anxiety and depression [21]. Such findings suggest that, although symptom expression and coping responses may interact, they represent conceptually distinct psychological constructs, similar to how general anxiety and depression are evaluated. Notably, the Generalized Anxiety Disorder 7-item Scale (GAD-7) and Patient Health Questionnaire-9 (PHQ-9), both of which are widely used diagnostic scales, do not include coping strategies as part of symptom measurement [22,23]. This reflects a broader consensus to separate symptom assessments from coping evaluations.

Supporting this conceptual distinction, McWilliams et al. [24] demonstrated that the association between coping strategies and psychological distress can vary in direction based on the specific coping strategy employed. This variability suggests that coping responses should not be considered equivalent to psychological symptoms. These findings highlight the value of designing measurement tools that isolate core symptomatology from coping mechanisms to enhance the construct validity of climate anxiety assessments.

Building on these considerations, this study aimed to develop the Brief Anxiety Scale for Climate Change (BACC) to reliably assess the core symptoms of climate anxiety, focusing exclusively on symptoms rather than coping behaviors. The scale was developed with active reference to the diagnostic criteria for generalized anxiety disorder (GAD) in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) [25]. Although existing scales have provided valuable foundations for symptom-based assessments, this scale incorporated additional clinically relevant items that are particularly important for capturing the developmental and social-contextual features of climate anxiety. This study assessed the following hypotheses:

H1. 

The BACC would demonstrate high reliability and validity in adolescents and adults.

H2. 

The BACC would demonstrate significant discriminant validity with the generalized anxiety scale (Mental Health Screening Tool for Anxiety Disorders, MHS:A) [26].

This study contributes toward a further comprehensive understanding of the high levels of climate anxiety experienced by South Korean adolescents and adults. This provides a crucial foundation for developing intervention programs designed to assist individuals in coping with climatic anxiety.

2. Study 1

2.1. Methods

2.1.1. BACC Development Procedure

The research team comprehensively reviewed and synthesized previous studies on the development of climate anxiety assessment tools, climate change status surveys, and sustainable development education to inform the selection of measurement domains for the BACC. To conduct this review, the relevant literature was searched using Google Scholar, PsycINFO, and PubMed. Climate anxiety was defined as a state characterized by the initial recognition of climate change as a crisis, followed by worry and anxiety regarding its impact on individuals, their immediate environment, and society. By acknowledging that, similar to anxiety in general, climate anxiety can manifest along a continuum from adaptive to pathological levels, we aimed to develop an assessment tool that encompassed this range.

Hence, the measurement domains of BACC were constructed with active reference to the diagnostic criteria for GAD in the DSM-5 [25]. Specifically, the first domain measured excessive worry regarding the climate crisis and difficulty in regulating such worry; the second domain captured the emotional and physical changes associated with climate anxiety; and the third domain assessed functional impairments in daily life, academic or occupational performance, and social activities resulting from climate anxiety. Based on these domains, a preliminary item pool was constructed for BACC.

In addition to these diagnostic foundations, the item development process also drew upon theoretical frameworks of climate anxiety [3,8,9]. The reviewed studies underscore key features of climate anxiety, including excessive worry, emotional distress, uncontrollable rumination, and functional impairments. These features were incorporated into the scale’s item construction to ensure its capacity to adequately capture core psychological symptoms associated with climate anxiety.

The research team, comprising two integrated master’s/doctoral students, one doctoral student, and one professor who specialized in clinical and counseling psychology, reviewed the preliminary item pool for BACC. The process involved advice and feedback from researchers from the Korean National Arboretum’s Exhibition, Education, and Research Division. Consequently, items with appropriate content validity were selected, which yielded 21 survey items for the final preliminary questionnaire. Specifically, items 1–10 measured “worry and preoccupation with the climate crisis”, items 11–17 assessed “physical, emotional, and cognitive symptoms related to climate anxiety”, and items 18–21 measured “functional impairments in daily life due to climate anxiety”.

Since the items in the originally constructed preliminary item pool were biased toward measuring the pathological dimension of climate anxiety, we adjusted the overall difficulty of the items downward to ensure that the questionnaire could distinguish between pathological and adaptive levels of climate anxiety. Furthermore, the response measurement employed frequency anchoring, which is commonly used in the measurement of anxiety-related symptoms (i.e., never, rarely, sometimes, often, always), via a 5-point Likert scale that ranged from 0 to 4.

2.1.2. Participants and Data Collection

A total of 300 participants, comprising 150 adolescents and 150 adults who resided in South Korea, completed the BACC online in 2023. This study was approved by the Institutional Review Board of Korea University (KUIRB-2024-0314-01), and informed consent was obtained from all participants. The participants’ mean age was 28.93 years (SD = 15.18). They were recruited through a research company specializing in online panel surveys, which provided an overview of the study to eligible adolescents (aged 14–18 years) and adults (aged 18 years and older). Those who voluntarily agreed to participate completed the survey after they provided informed consent. Since Corney and Lee [27] suggested a sample size of 300 as a reasonable standard for a multivariate factor analysis, our sample size was considered adequate. Participants received KRW 5000 (approximately USD 5) as compensation.

Table 1. Sample demographics.

Characteristic	Total (N = 300)	Adolescents (N = 150)	Adults (N = 150)
Age
Mean (SD)	28.93 (15.18)	16.61 (1.13)	41.24 (12.48)
Gender
Male	117 (39.0%)	56 (37.3%)	61 (40.7%)
Female	183 (61.0%)	94 (62.7%)	89 (59.3%)
Final Education Level
Elementary School	35 (11.7%)	35 (23.3%)	-
Middle School	117 (39.0%)	115 (76.7%)	2 (1.3%)
High School	27 (9.0%)	-	27 (18.0%)
2–4 Year College	108 (36.0%)	-	108 (72.0%)
Master’s or Higher	13 (4.3%)	-	13 (8.7%)

summarizes the demographic characteristics of Study 1’s sample.

2.1.3. Measures

Brief Anxiety Scale for Climate Change (BACC)

The BACC was developed to assess symptoms of climate anxiety experienced during the past week. The initial version comprised 21 items across three domains: excessive worry and preoccupation with the climate crisis, physical, cognitive, and emotional symptoms associated with climate anxiety, and functional impairments in daily life. Each item was rated on a 5-point Likert scale that ranged from 0 (never) to 4 (always), based on the frequency of symptom experiences during the past week. Internal consistency was first evaluated via the preliminary 21-item version. Subsequently, 13 items were selected for an exploratory factor analysis (EFA) based on content validation and preliminary psychometric evaluations.

Mental Health Screening Tool for Anxiety Disorders (MHS: A)

Anxiety symptom severity was assessed using the Mental Health Screening Tool for Anxiety Disorders (MHS: A) [26], a validated measurement tool designed for the early detection of GAD in primary care environments. The MHS: A is an 11-item self-report questionnaire that uses a 5-point Likert scale (0–4) to measure the frequency of anxiety symptoms experienced over the past two weeks. Total scores are classified as minimal (0–10), mild (10–20), moderate (20–30), or severe (>30). It demonstrated excellent internal consistency with Cronbach’s alpha values of 0.957 and 0.956 for the offline and online versions, respectively. This study utilized an online format to enhance participants’ convenience [25].

Mental Health Screening Tool for Depressive Disorders (MHS: D)

Depressive symptoms were assessed via the Mental Health Screening Tool for Depressive Disorders (MHS: D) [28], a 12-item instrument developed to facilitate the early detection of major depressive disorder (MDD) in primary care settings. Participants respond on a 5-point Likert scale, and higher scores reflect a greater severity of depressive symptoms experienced over the past two weeks according to the diagnostic criteria. Based on the total scores, depressive symptom severity is classified into minimal (0–8), mild (8–12), moderate (12–20), and very severe (>20) categories. Both offline and online versions have demonstrated excellent internal consistency (Cronbach’s α = 0.943 and 0.945, respectively). In this study, the online version was utilized in order to enhance the convenience to the participants [28].

The MHS: A and MHS: D were selected due to their cultural specificity and methodological suitability for the current study. Unlike many translated scales, both were developed based on Korean participants’ response patterns and validated specifically for detecting anxiety and depressive disorders in Korean populations. Notably, the validation studies included both offline and online versions, and we confirmed their reliability across formats [26,28]. Given that our study employed an online survey method, the use of these tools ensured cultural and technological consistency with their intended application.

2.1.4. Data Analysis

All statistical analyses were conducted via R 4.4.2 [29]. Analyses were conducted using the initial 21 items from BACC, a test developed for this study. Its internal consistency was first evaluated by calculating the Cronbach’s alpha coefficients and item–total correlations.

Subsequently, an EFA was conducted to examine the underlying factor structure. Prior to the EFA, preliminary analyses were performed to assess the normality of item distributions following Chou and Bentler’s guidelines [30], with skewness values < |2| and kurtosis values < |7| considered acceptable. The suitability of the data for factor analysis was further evaluated via the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy and Bartlett’s test of sphericity. EFA was conducted using maximum-likelihood extraction with Oblimin rotation, which allows the factors to be correlated, using the Psych package (version 2.4.6.26). A parallel analysis was performed to determine the appropriate number of factors to retain [31].

To examine the BACC’s discriminant validity, a confirmatory factor analysis (CFA) model was constructed, which included the BACC, generalized anxiety (MHS:A), and depression (MHS:D) factors; furthermore, the latent correlations among these factors were estimated.

To improve clarity and reproducibility, the primary analysis steps conducted in R are summarized in Figure 1 using pseudocode format.

2.2. Results

2.2.1. Internal Consistency

Cronbach’s alpha was calculated to assess the internal consistency of the symptom scale using R 4.4.2. Cronbach’s alpha for the symptom scale was 0.94, which indicates a very high level of internal consistency. The 95% confidence interval for Cronbach’s alpha was between 0.93 and 0.95, which confirmed the stability of the reliability estimates.

Table 2. Mean, standard deviations, Cronbach’s alpha, and item–total correlations of the Brief Anxiety Scale for Climate Change.

Item	Mean (SD)	Cronbach’s Alpha If Item Deleted	Standardized Item–Total Correlation
1 (Perceived threat to safety)	2.48 (0.92)	0.94	0.54 ***
2 (Perceived threat to earth)	2.70 (1.06)	0.94	0.47 ***
3 (Perceived destruction)	2.40 (1.06)	0.94	0.63 ***
4 (Future concerns)	2.70 (1.06)	0.94	0.43 ***
5 (Uncontrollable worry)	1.71 (1.22)	0.94	0.74 ***
6 (Search for Climate Change Information in Media)	1.84 (1.06)	0.94	0.69 ***
7 (Perceived helplessness)	2.27 (1.08)	0.94	0.68 ***
8 (Perception of severity)	2.77 (0.99)	0.94	0.46 ***
9 (Thinking about climate change)	2.34 (1.03)	0.94	0.68 ***
10 (Worry about impact)	2.54 (0.99)	0.94	0.66 ***
11 (Emotional distress)	1.10 (1.19)	0.94	0.80 ***
12 (Nightmares)	0.81 (1.15)	0.94	0.74 ***
13 (Distraction)	0.88 (1.11)	0.94	0.80 ***
14 (Insomnia)	0.86 (1.15)	0.94	0.77 ***
15 (Restlessness)	1.20 (1.24)	0.94	0.80 ***
16 (Irritability)	0.86 (1.16)	0.94	0.74 ***
17 (Fear of an impending threat)	1.26 (1.19)	0.94	0.78 ***
18 (Social disruption)	0.80 (1.11)	0.94	0.72 ***
19 (Impaired performance)	0.78 (1.18)	0.94	0.71 ***
20 (Others’ feedback on excessive worry)	0.86 (1.16)	0.94	0.76 ***
21 (Academic/work disruption)	0.79 (1.18)	0.94	0.74 ***

Note. *** p < 0.001, Cronbach’s alpha values are standardized.

presents the means and standard deviations for individual items, Cronbach’s alpha coefficients when removing individual items, and standardized item–total correlations. Item-by-item calculations of Cronbach’s alpha revealed that no items needed to be removed to improve internal consistency. Nevertheless, a few items showed relatively low correlations with the overall score, as indicated by their standardized item–total correlations, which were below 0.5. Item–total correlation is an important indicator of how consistently each item measures the overall scale. Therefore, the researchers conducted an internal review and concluded that these items measured a normal range of anxiety for climate change compared with the other items. As a result, three items were removed: item 2 (perceived threat to Earth), item 4 (future concerns), and item 8 (perception of severity).

To refine the scale, an item-level analysis was conducted to assess the extent to which each item contributed meaningfully to the overall construct being measured. Following a thorough evaluation of the content validity, five items were eliminated. Although the original scale aimed to assess both adaptive and pathological levels of climate anxiety, certain items introduced conceptual inconsistencies. Specifically, items 6 (“I search for information about climate change in the media (e.g., YouTube, TV, internet searches)”) and 9 (“I think about issues related to climate change”) were removed as they primarily reflected general interest in climate change or coping behaviors rather than symptoms of climate anxiety.

Additionally, items 17, 19, and 21 were eliminated due to excessive redundancy, as they closely overlapped with other items. These decisions were made based on a rigorous review conducted by two clinical psychologists and two doctoral researchers who specialized in clinical psychology. This ensured that each item was evaluated based on its theoretical and empirical validity. Through this refinement process, the final 13-item scale retained strong content validity and provided a further concise and conceptually robust measurement tool.

2.2.2. Exploratory Factor Analysis (EFA)

An EFA was conducted using the maximum-likelihood method with direct Oblimin rotation based on a 13-item scale. Prior to factorization, KMO and Bartlett’s tests were performed to ensure that the data were suitable for factorization. The overall measure of the sampling adequacy value for the KMO test was 0.96, and Bartlett’s test of sphericity yielded a value of χ² (136) = 4543.82, with a p-value of <0.001, which confirmed that the data were suitable for factor analysis. Moreover, the skewness and kurtosis values for all items met the acceptable criteria, with skewness values below |2| and kurtosis values below |7|, indicating no severe violations of normality.

The results of the EFA indicated that the two-factor model provided an adequate fit, explaining 68% of the variance between the items. Standardized loadings of each factor were high for most items, and the inter-factor correlation was 0.45. The first and second factors explained 46% and 22% of the variance, respectively. Parallel analysis also supported the two-factor solution as the eigenvalues of the first two factors exceeded those of the simulated data. Furthermore, the scree plot (Figure 2) confirmed the appropriateness of the two-factor model. The EFA model demonstrated an excellent fit, with the root mean square of residuals (RMSR) = 0.02, the Tucker–Lewis Index (TLI) = 0.986, and the root mean square error of approximation (RMSEA) = 0.042 (90% CI: 0.022–0.06).

Factor 1, designated as “functional impairment”, primarily encompassed items that pertained to challenges with daily activities and sleep attributable to climate anxiety. Factor 2, designated as “cognitive impairment”, encompassed items that reflected excessive worry, preoccupation with climate change, and emotional dysregulation.

2.2.3. Correlations with Generalized Anxiety and Depression

To evaluate whether the BACC captured a distinct construct separate from existing diagnostic tools for GAD and MDD, a CFA model was used, which included items from the BACC and anxiety (MHS: A) and depression scales (MHS: D). Correlations between the three latent factors were assessed. BACC’s correlations with the anxiety and depression scales were 0.641 and 0.595, respectively.

3. Study 2

3.1. Method

Study 2 was conducted to independently validate the two-factor structure identified in Study 1 and to reassess the reliability of the final 13-item version of the BACC. By employing a new sample, this follow-up study aimed to further examine the psychometric stability of the scale and its applicability to adolescent and adult populations.

3.1.1. Participants and Data Collection

A total of 400 participants, comprising 151 adolescents and 249 adults who resided in South Korea, completed the BACC online in 2024 (age M [SD] = 32.14 [18.77]). The procedures for Study 2 were also approved by the Institutional Review Board of Korea University under the same protocol (KUIRB-2024-0314-01), and informed consent was obtained from all participants. A research company specializing in online panel surveys provided an explanation of the study to a panel of adolescents (aged 14–17 years) and adults (aged 18 years and older). The survey was conducted with those who volunteered to participate and provided informed consent. Participants were compensated KRW 5000 (approximately USD 5) for completing the survey. Data were collected to conduct a CFA and reassess the internal consistency of the BACC.

Table 3. Study 2 sample demographics.

Characteristic	Total (N = 400)	Adolescents (N = 151)	Adults (N = 249)
Age
Mean (SD)	32.14 (18.77)	15.72 (1.11)	42.10 (17.39)
Gender
Male	201 (50.3%)	80 (53.0%)	121 (48.6%)
Female	199 (49.7%)	71 (47.0%)	128 (51.4%)
Final Education Level
Elementary School	66 (16.5%)	66 (43.7%)	-
Middle School	124 (31.0%)	85 (56.3%)	39 (15.7%)
High School	54 (13.5%)	-	54 (21.7%)
2–4 Year College	135 (33.8%)	-	135 (54.2%)
Master’s or Higher	21 (5.3%)	-	21 (8.4%)

provides the detailed demographic information of Study 2’s sample.

3.1.2. Measures

Study 2 employed the same measures as those used in Study 1. The BACC comprised 13 items for assessing cognitive and functional impairments related to climate anxiety rated on a 5-point Likert scale that ranged from 0 (not at all) to 4 (extremely). General anxiety and depressive symptoms were also assessed using the MHS:A and MHS:D subscales, respectively. Higher scores on each scale indicated greater levels of the respective symptoms.

3.1.3. Data Analysis

All statistical analyses were conducted via R 4.4.2 [29]. Analyses were performed using the 13 items of the BACC finalized in Study 1. Internal consistency was assessed via Cronbach’s alpha coefficients to evaluate its reliability.

Subsequently, a CFA was conducted via the maximum likelihood method with the lavaan package (version 0.6.19) to validate the two-factor structure identified in Study 1. Model fit was evaluated based on the chi-square statistic (χ²), comparative fit index (CFI), TLI, RMSEA, and standardized root mean square residual (SRMR).

To examine its discriminant validity, we evaluated the latent correlations among the BACC, generalized anxiety (MHS:A), and depression (MHS:D) factors estimated via the CFA.

For clarity and reproducibility, the key analysis steps conducted in R for Study 2 are presented below as pseudocode (See Figure 3).

3.2. Results

3.2.1. Internal Consistency

The BACC demonstrated excellent internal consistency with the validation sample. Cronbach’s alpha coefficient was 0.92, which indicated high reliability. The 95% confidence interval for Cronbach’s alpha was 0.92–0.94 based on both Feldt’s and Duhachek’s methods. Item–total correlations were satisfactory.

Table 4. Item means, standard deviations, and item–total correlations for the Brief Anxiety Scale for Climate Change (BACC).

Item	Mean (SD)	Cronbach’s Alpha If Item Deleted	Standardized Item–Total Correlation
1 (Perceived threat to safety)	2.6 (0.90)	0.92	0.49 ***
3 (Perceived destruction)	2.5 (1.03)	0.92	0.56 ***
5 (Uncontrollable worry)	2.1 (1.23)	0.92	0.69 ***
7 (Perceived helplessness)	2.4 (1.07)	0.92	0.54 ***
10 (Worry about impact)	2.5 (0.98)	0.92	0.47 ***
11 (Emotional distress)	1.6 (1.31)	0.91	0.80 ***
12 (Nightmares)	1.4 (1.34)	0.91	0.83 ***
13 (Distraction)	1.5 (1.30)	0.91	0.82 ***
14 (Insomnia)	1.5 (1.36)	0.91	0.84 ***
15 (Restlessness)	1.6 (1.26)	0.91	0.84 ***
16 (Irritability)	1.5 (1.30)	0.91	0.82 ***
18 (Social disruption)	1.4 (1.31)	0.91	0.80 ***
20 (Others’ feedback on excessive worry)	1.4 (1.33)	0.91	0.80 ***

Note. *** p < 0.001; Cronbach’s alpha values are standardized.

presents the standardized item–total correlations for each item.

These results suggest that the BACC items consistently measured the intended climate anxiety construct across both cognitive and functional impairment domains. In Study 1, the Cronbach’s alpha for the initial 21-item version was 0.94, indicating that the shortened 13-item version retained a comparable level of internal consistency.

3.2.2. Confirmatory Factor Analysis (CFA)

To confirm the two-factor structure identified in Study 1, we conducted a CFA using the 13 final items.

Table 5. Summary of goodness-of-fit indices for CFA.

Model Tested	χ2 (df)	CFI	TLI	RMSEA	SRMR
Two-factor model	216.266 *** (64)	0.960	0.951	0.077	0.071

Note. *** p < 0.001.

and

Table 6. Standardized loadings (pattern matrix) for the Brief Anxiety Scale for Climate Change.

Item	Standardized Coefficient	p-Value
Factor 1 (Functional Impairment)
11	0.790	<0.001
12	0.899	<0.001
13	0.853	<0.001
14	0.898	<0.001
15	0.838	<0.001
16	0.857	<0.001
18	0.852	<0.001
20	0.873	<0.001
Factor 2 (Cognitive Impairment)
1	0.683	<0.001
3	0.630	<0.001
5	0.702	<0.001
7	0.653	<0.001
10	0.647	<0.001

Note. p < 0.001; Extraction method: maximum likelihood. Rotation method: Oblimin with Kaiser Normalization. Factor loadings less than |0.1| are suppressed for clarity.

summarize the goodness-of-fit indices for the CFA and standardized factor loadings, respectively. Both the CFI (0.960) and TLI (0.951) exceeded the recommended threshold of 0.90, which indicated a good model fit. The RMSEA value was 0.077, which was below the criterion of 0.08, and the SRMR value was 0.071, which met the recommended cut-off value of 0.08. Collectively, these results supported the adequacy of the two-factor model.

Standardized factor loadings ranged from 0.63–0.90 across the two factors. All items loaded significantly on their respective factors, supporting the theoretical structure of the BACC. Factor 1 (Functional Impairment) included eight items (items 11, 12, 13, 14, 15, 16, 18, and 20), whereas Factor 2 (Cognitive Impairment) included five items (items 1, 3, 5, 7, and 10). In contrast, a confirmatory factor analysis using the original 21 items yielded lower model fit indices (CFI = 0.899, TLI = 0.887, RMSEA = 0.099, SRMR = 0.119), suggesting that the shortened version improved model fit without sacrificing reliability.

3.2.3. Correlations with Generalized Anxiety and Depression

To further examine the construct validity of the BACC, we analyzed its latent correlations with generalized anxiety (MHS: A) and depression (MHS: D) via CFA. The three-factor CFA model demonstrated an acceptable-to-good model fit (CFI = 0.929, TLI = 0.924, RMSEA = 0.063, and SRMR = 0.054).

The latent variable that represented climate anxiety demonstrated a moderate correlation with generalized anxiety (r = 0.526) and depression (r = 0.505), indicating that climate anxiety was related to, yet distinct from, broader emotional symptoms. These findings provide evidence for the discriminant validity of the BACC and suggest that, while climate anxiety overlaps with general anxiety and depression to an extent, it also captures a distinct psychological construct.

3.2.4. Final Items of the BACC

The final version of the BACC comprised 13 items that assessed cognitive and functional impairments related to climate anxiety.

Table 7. Final items of the Brief Anxiety Scale for Climate Change (BACC).

No	Items
1	I worry that climate change will threaten people’s safety (e.g., economic, social, physical safety).
2	I think climate change will destroy what I value most.
3	Once I start worrying about climate change, I find it hard to stop.
4	I worry that I might not be able to cope with climate change.
5	I worry about the impacts of climate change.
6	I cry when I think about climate change.
7	I have nightmares about climate change.
8	Thinking about climate change makes it hard for me to concentrate on what I’m doing.
9	Thinking about climate change makes it difficult for me to fall asleep.
10	Thinking about climate change keeps me from feeling comfortable.
11	Thinking about climate change makes me become easily annoyed or irritated.
12	My worries about climate change make it hard for me to enjoy time with my family or friends.
13	People around me say I think about climate change too much.

Note. Original item numbers were reordered after item selection. Items were translated via forward and backward translation procedures.

presents the English versions of the finalized items. Original item numbers were reordered after item selection and refinement. Items were translated through a forward and backward translation procedure. This translation was conducted by graduate students in a combined master’s and doctoral program specializing in clinical and counseling psychology, who possessed native-level English proficiency. This process was conducted to ensure conceptual and linguistic equivalence.

4. Discussion

This study aimed to develop and validate the BACC through two independent samples that comprised 700 adolescents and adults who resided in South Korea. Study 1 (n = 300) focused on the initial validation of the BACC and analyzed its internal consistency, content validity, and factor structure via an EFA, while Study 2 (n = 400) independently confirmed the two-factor structure via a CFA and reassessed internal consistency. Starting from an initial pool of 21 items, statistical and theoretical refinements led to a final 13-item version of the BACC. The final scale demonstrated strong internal consistency, a robust factor structure, and good discriminant validity with good model fit indices (CFI = 0.96, TLI = 0.95, RMSEA = 0.077, SRMR = 0.071), thereby supporting its construct validity.

The content validity of the BACC was established through a multi-step development process. Items were initially constructed based on DSM-5 diagnostic criteria for GAD and reviewed in relation to existing climate anxiety measures. Based on expert consensus and external feedback, appropriate items were selected or revised. In addition, item difficulty was adjusted to capture a continuum from adaptive to pathological levels of climate anxiety. To address the limitations of previously used tools, new items were also developed to reflect clinically important symptoms that have not previously been assessed—such as irritability (item 7), which is particularly relevant in adolescent populations. Originally, the items were developed based on the DSM-5 diagnostic criteria for GAD and conceptualized into three domains: (1) excessive worry about the climate crisis, (2) physical, emotional, and cognitive symptoms, and (3) functional impairments in daily life. However, the EFA results from Study 1 supported a two-factor structure that encompassed cognitive and functional impairment. This two-factor structure was subsequently confirmed by the CFA in Study 2, which demonstrated excellent model fit indices.

The first factor represented functional impairments caused by climate anxiety, which encompassed difficulties in daily life, academic or occupational performance, and social interaction. The second factor reflected cognitive impairment related to climate anxiety, specifically, excessive worry and preoccupation with climate change. This revised factor structure suggested that, rather than forming a distinct category, the physical and emotional symptoms of climate anxiety were more closely aligned with either functional impairments or cognitive manifestations of climate anxiety.

Although previous studies on climate anxiety conceptualize climate anxiety as comprising cognitive and functional impairments, these previous factor analyses were based on the full 22-item version rather than a refined 13-item structure [9]. This study contributes to the existing literature by providing empirical support for a two-factor model using a newly developed 13-item scale. These findings suggest that the core dimensions of climate anxiety and cognitive and functional impairment are stable across different measurement approaches. This indicates that climate anxiety can be reliably assessed.

Analyses of the associations between climate anxiety, generalized anxiety, and depression revealed consistent findings across the two independent samples.

In Study 1, the BACC total score was moderately correlated with generalized anxiety (r = 0.641) and depression (r = 0.595), while, in Study 2, the corresponding correlations were r = 0.526 and r = 0.505, respectively. Replication of similar correlation patterns across different samples strengthened the reliability and robustness of the results, suggesting that, while climate anxiety was related to broader emotional distress, it captured a distinct psychological construct.

These findings were consistent with those of prior research, which found that climate or eco-anxiety, particularly its core symptom dimensions, was moderately associated with generalized anxiety and depression, while remaining conceptually distinct. Clayton and Karazsia [9] reported correlations that ranged from r = 0.47–0.62 between their climate change anxiety subscales (cognitive and functional impairment) and general emotional distress. Meanwhile, Hogg et al. [8] observed moderate associations (r = 0.21–0.46) between the affective symptoms of eco-anxiety and measures of anxiety and depression. These parallel findings suggested that climate-related anxiety, as measured by the BACC, captured a comparable yet distinct dimension of psychological distress. Moderate correlations observed across both the studies highlighted that climate anxiety was intricately tied to individuals’ broader emotional functioning, thus underscoring its relevance as a meaningful and independent construct within the field of mental health, rather than merely a transient or situational concern.

A strength of this study is that the EFA and CFA were conducted on independent samples to enhance the robustness of the identified factor structure. Furthermore, the associations between climate anxiety and broader emotional symptoms, such as generalized anxiety and depression, were examined across two independent samples, which yielded consistent results. This methodological approach strengthened the scale’s reliability and stability.

Despite these strengths, some limitations remain. First, the test–retest reliability was not assessed in this initial validation study. Hence, future studies should investigate and establish the temporal stability of the BACC scores. Second, this study did not empirically compare the BACC with existing climate anxiety measures, such as the CCAS or HEAS. Although the scale’s development process critically reviewed the conceptual and psychometric limitations of the existing tools, future studies should conduct direct comparisons to further establish convergent and incremental validity.

Third, while the BACC demonstrated sound psychometric properties among South Korean adolescents and adults recruited through online surveys, the generalizability of these findings to other populations and cultural contexts may still be limited. Moreover, although the manifestations of climate change—such as rising temperatures or changes in precipitation—may differ across regions [32], the scale was intentionally designed to capture individuals’ psychological responses to climate change as a persistent and abstract threat, rather than reactions to specific events. This approach contributes to enhancing its generalizability; however, cultural and regional factors may nonetheless influence how climate anxiety is experienced and expressed. Therefore, cross-cultural validation studies are needed to evaluate the applicability of the BACC across diverse sociocultural settings.

Fourth, although the BACC was developed based on DSM-5 criteria and established theoretical frameworks, its clinical applicability has not yet been tested in clinical or high-risk populations. The BACC was developed to assess both pathological and adaptive levels of climate anxiety by calibrating item difficulty. Consequently, the present study evaluated the scale’s validity using a community-based sample. Future research should include replication studies with culturally and demographically diverse samples and assess the scale’s diagnostic utility in populations with clinically significant levels of climate-related anxiety.

Although this study did not include these existing scales for direct empirical comparison, the development of the BACC was grounded in a thorough conceptual review of their content, structure, and psychometric limitations. Key concerns were addressed, such as the inclusion of coping items and lack of clinical specificity, particularly regarding adolescent symptom presentation. These distinctions informed the development of the BACC as a symptom-focused measure of climate anxiety. Nevertheless, future studies should be conducted to empirically compare the BACC with existing instruments to further establish its convergent and incremental validity. Such efforts will strengthen construct validity and help confirm the BACC’s generalizability and structural stability across diverse populations and contexts.

Despite the study having some limitations, these findings highlight the BACC’s potential as a promising climate anxiety scale that captures pure symptomatology distinct from generalized anxiety and depression after excluding coping behaviors. Therefore, the BACC may serve as a valuable tool for identifying individuals experiencing clinically significant levels of climate-related distress and guiding appropriate interventions in both research and applied settings.

5. Conclusions

This study developed and validated the BACC, a 13-item self-report measure designed to assess the core symptoms of climate anxiety. This study used two independent samples to identify and confirm a two-factor structure—representing cognitive and functional impairments—that demonstrated consistent associations with broader emotional symptoms, such as generalized anxiety and depression.

By focusing exclusively on symptomatology and excluding coping behaviors, the BACC offers a distinct and clinically meaningful tool for assessing climate anxiety. Future research should examine its temporal stability and directly compare it with existing climate anxiety measures to strengthen its convergent and incremental validity. Ultimately, the BACC fills an important gap in the measurement of climate-related mental-health symptoms by providing a reliable and developmentally sensitive tool.