Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Reliability, validation and norms of the Chinese version of Anxiety Sensitivity Index 3 in a sample of military personnel

  • Wenpeng Cai ,

    Contributed equally to this work with: Wenpeng Cai, Wei Dong, Yu Pan

    Roles Funding acquisition, Investigation, Writing – original draft, Writing – review & editing

    Affiliation Department of Psychology and Mental Health, Second Military Medical University, Shanghai, China

  • Wei Dong ,

    Contributed equally to this work with: Wenpeng Cai, Wei Dong, Yu Pan

    Roles Funding acquisition, Investigation, Writing – review & editing

    Affiliation Department of Psychology and Mental Health, Second Military Medical University, Shanghai, China

  • Yu Pan ,

    Contributed equally to this work with: Wenpeng Cai, Wei Dong, Yu Pan

    Roles Formal analysis, Investigation

    Affiliation Department of Medical Psychology, General Hospital of PLA, Beijing, China

  • Cun Wei,

    Roles Funding acquisition

    Affiliation Department of Psychology and Mental Health, Second Military Medical University, Shanghai, China

  • Shuimiao Zhang,

    Roles Investigation

    Affiliation Department of Neurology, Jinan Military General Hospital of PLA, Jinan, China

  • Bin Tian,

    Roles Investigation

    Affiliation Department of Psychology and Mental Health, Second Military Medical University, Shanghai, China

  • Jin Yan,

    Roles Conceptualization, Funding acquisition

    Affiliation Department of Psychology and Mental Health, Second Military Medical University, Shanghai, China

  • Guanghui Deng

    Roles Conceptualization, Funding acquisition, Supervision, Writing – review & editing

    Affiliation Department of Psychology and Mental Health, Second Military Medical University, Shanghai, China


This study aimed to explore the properties of the Chinese version of the Anxiety Sensitivity Index– 3 (ASI-3) in a sample of military personnel. Using non-probabilistic sampling, the Chinese version of the ASI-3 was administered to 3,077 valid participants aged 16 to 36 years old (M = 22.35, SD = 3.57) from nine military units. The Depression Anxiety Stress Scales-21 (DASS-21) and The State-Trait Anxiety Inventory (STAI) were used to assess the construct validity. A one-way analysis of variance was conducted to compare the differences in the services and positions of the participants. It was found that ①The indices of confirmatory factor analysis met the standard values, which supported the hypothesis of the three-factor model of the original ASI-3; ②ASI-3 was significantly associated with DASS-21 in positive ways, which indicated the high convergent validity; on the other hand, the correlation between ASI-3 and TAI was relatively low, which indicated there was an empirical discrimination between anxiety sensitivity and trait anxiety. ③The Cronbach’s α coefficients were 0.926 for the total scale and 0.828–0.841 for the three subscales; ④At a cut-off score of 16, the sensitivity and specificity levels were 71.1% and 76.7%, respectively, where the sum of sensitivity and specificity becomes the maximum, accompanied with improvement of PPV and NPV; ⑤There were significant findings in the ASI and subscales among the five services and four positions. This study provides new evidence that the Chinese version of the Anxiety Sensitivity Index-3 has good validity and reliability and could be applied as an effective tool to assess anxiety sensitivity in military personnel. Our recommendations to researchers and practitioners are that the three factor model should be replicated across some different special forces and the items and constructs could be modified on Chinese culture.


Within the cognitive-behavioral framework, dysfunctional beliefs play an important role in emotional disorders [1]. For example, social phobia is considered to be affected by exaggerated beliefs that are viewed as threatening (e.g., when I tremble in the presence of others, I fear what people might think of me)[2]. Panic disorder is considered to be influenced by catastrophic beliefs of the danger of internal sensations (e.g., When I notice my heart skipping a beat, I worry that there is something seriously wrong with me) [3]. These dysfunctional cognitions cause catastrophic misinterpretations of danger cues, thereby triggering situational anxiety and fear [4]. Specifically, anxiety sensitivity (AS) is thought to be a cognitive diathesis including a set of trait-like dysfunctional beliefs about harmful consequences of anxious arousal.

Reiss and McNally defined anxiety sensitivity as “the fear of arousal-related sensation, arising from beliefs that the sensations have harmful somatic, cognitive, or social consequences (e.g., death, insanity, or social rejection)” [5]. As individuals with high anxiety sensitivity feel anxiety, they must be alarmed by the arousal-related sensations, which give rise to an intensification of anxiety symptoms, fear response, and avoidance behaviors. A large body of research suggests that AS is a strong psychological risk factor for the development and maintenance of anxiety psychopathology [6, 7]. Additionally, existing studies indicate that anxiety sensitivity might be associated with several psychosomatic diseases, such as hypochondriasis, chronic pain, and substance use disorders [8].

There is a similar construct Trait Anxiety, which also reflects individuals’ sensitivity to develop anxious mood. However, trait anxiety is deemed to be the tendency to respond fearfully to various stressors, and anxiety sensitivity refers to a proneness to respond with anxiety to anxiety symptoms[9]. Although fear and anxiety often occur together, this does not mean that these terms are interchangeable as theoretical terms. Anxiety is a rather vague feeling of discomfort while fear is triggered by imminent danger, so that anxiety is driven by uncertainty about anticipated events while fear is the response to certain imminent events. Reiss noted that while trait anxiety predicts future anxiety based on anxiety experiences in the past, AS predicts future anxiety based on the beliefs assessed by the ASI, regardless of the frequency of anxiety experiences in the past [10]. Trait anxiety cannot explain why some people react with anxiety to their own anxiety or anxiety related sensations[11]. One the other hand, Reiss and McNally found that correlations between the ASI and trait anxiety have been consistently lower than the correlations between trait and state anxiety[12, 13]. Taylor and Cox pointed out a modest correlation (r = 0.26) between trait version of STAI and 60 item-Anxiety Sensitivity Profile which is a revised form of the ASI, and concluded that AS and trait anxiety are correlated but distinct factors[14, 15].

The original Anxiety Sensitivity Index (ASI) was composed of 16 items that assess various components of the fear of anxiety-related sensations [16]. A high internal consistency (Cronbach’s α = 0.82–0.88) and acceptable test-retest reliability (14 days, r = 0.75) were verified in two studies [17, 18]. Several factor analytic studies showed 60% of the total variance could be explained by three factors, including fears of negative consequences of arousal-related physical, cognitive, or social sensations [19, 20].

Therefore, Taylor and Cox developed the 36-item Anxiety Sensitivity Index–Revised (ASI-R), which was supposed to assess the following six lower-order factors in greater depth: (a) cardiovascular fears, (b) respiratory fears, (c) gastrointestinal symptoms and fears, (d) fear of cognitive dyscontrol, (e) fear of publicly observable anxiety reactions, and (f) neurological and dissociative symptoms.[8] However, the ASI-R failed to obtain stable factor structures, given the different solutions concluded by different studies [21, 22]. A psychometrically sound, multidimensional measure of AS was supposed to be developed in the following studies.

In response to these concerns, the 18-item Anxiety Sensitivity Index-3 was developed by Taylor et al. to improve the basic psychometric criteria of the original scales [23]. Three subscales of physical, cognitive, and social concerns were included in the ASI-3, which showed strong correlations with (r ≥ 0.5) and higher internal consistency the (Cronbach’s α ≥ 0.73) the 16-item ASI and explained approximately 76% of the total variance.

With the advantage of the structure and a potential relationship with anxiety disorders, the ASI-3 has attracted many researchers’ attention. To date, several cross-cultural studies have contributed to a more universal etiological model of anxiety sensitivity and related anxiety disorders [2427]. Likewise, many Chinese researchers are devoted to demonstrating the role of AS in the development and maintenance of mental disorders [2830]. Li et al. translated the ASI-R and assessed its reliability and validity in a sample of middle school students [31]. Then, Wang et al. revised the Chinese version of the ASI-3 and formulated the ASI-3 norm of Chinese healthy adult women [32]. However, many specific psychometric characteristics remained to be assessed in a larger sample, especially in males, because several studies showed that gender differences existed in anxiety sensitivity and its subscales [20, 33]. Although the majority of soldiers are not engaged in war or conflicts, they have to face many other types of stressors, such as training exercises, heavy workloads, and family separations [34, 35]. These inevitable stressors could give rise to anxiety, depression, and posttraumatic stress disorder [36], as well as suicidal behaviors [37, 38], which might deeply impair their functions in social, occupational and other significant fields [39]. Therefore, military psychologists were supposed to pay attention to how to distinguish anxiety-sensitive individuals and reduce the effect of anxiety disorders on military task effectiveness.

Above all, the current study was aimed to investigate the reliability, validity and norms of the ASI-3 in a sample of Chinese military personnel and address the following hypotheses (a) whether the three factor model could be replicated in current samples (factorial validity); (b) the relationship between AS and other anxiety constructs (construct validity); (c) the split-half model and Cronbach's alpha coefficient of ASI-3 (within-session reliability); (d) the cut-off score to predict anxious individuals (ROC curve); (e) the differences of the ASI-3 and its subscales in terms of services and positions.

Materials and methods


The participants were selected from nine military units using non-probabilistic (convenience) sampling, covering all five major components (Army, Navy, Air Force, Armed Police and Rocket Force) and five theater commands (Eastern, Southern, Western, Northern and Central)[40]. Finally, a total of 3,077 valid participants were included and were aged 16 to 36 years (M = 22.35, SD = 3.57); 94.0% of the participants were male. Additional demographic variables can be found in Table 1. The participants were directed to attend a large questionnaire-based study, which lasted approximately 15 min. They were instructed to read and sign the informed consent on the first page of the questionnaires, then were asked to complete the paper-and-pencil questionnaires in a small group.


The Anxiety Sensitivity Index-3 (ASI-3) [23] includes 18 items that assess anxiety-related physical (e.g., “When my stomach is upset, I worry that I might be seriously ill”), cognitive (e.g., “When my thoughts seem to speed up, I worry that I might be going crazy”), or social (e.g., “It scares me when I blush in front of people”) concerns. Participants were asked to rate how much they share these concerns on a 5-point scale (0 = I agree very little, 4 = I agree very much). The Chinese version showed a good reliability (Cronbach’s α = 0.95) in healthy Chinese women and correlated moderately with Beck Anxiety Inventory (Pearson’s r = 0.32 to 0.42)[32].

The Depression, Anxiety, and Stress Scales-21 (DASS-21)[41] is a self-report tool concluding 21 items to assess three constructs: Depression, Anxiety and Stress. Participants were asked to rate how much they share these concerns (e.g., “I was worried the about situations in which I might panic”) on a 4-point scale (0 = Did not apply to me at all, 3 = Applied to me very much or most of the time). The Chinese version showed high validity and reliability in Chinese college students[42].

The State-Trait Anxiety Inventory (STAI) [43] is one of the most widely used scales to assess individuals’ state and trait anxiety (e.g., ‘‘I worry too much over something that really doesn’t matter”). The items used a 4-point Likert scale ranging from 1 (almost never) to 4 (almost always). The Chinese version also showed high validity and reliability [44]. Only the trait anxiety inventory (TAI) was used in the current study and 2277 participants finished it.

Statistical analysis

The percentage of missing data was 3.6% and assumed to follow an arbitrary missing pattern. Therefore, Markov Chain Monte Carlo multiple imputation was used to replace missing data, which was reported to be a preferred technique for handling missing values due to several advantages over other approaches[45]. The main analysis included the factorial validity of the 18-item ASI-3, which was assessed through structural equations (SEM) using confirmatory factor analysis (CFA). The model was estimated using AMOS17.0. Goodness-of-fit was assessed with several robust fit indexes [46, 47], including the ratio of chi-square to the degree of freedom (χ2/df), the comparative fit index (CFI), the Tucker-Lewis index (TLI), and the root-mean-square error of approximation (RMSEA). The acceptability criteria for these indicators were as follows: χ2/df less than 5 [48], both CFI and TLI greater than 0.90 [47], and RMSEA less than 0.08 [49].

An estimated internal consistency reliability (Cronbach’s alpha and item-factor correlation) was assessed in the current study, which meant the degree as to which a score or sub-score of a questionnaire reflected a narrow/ closely-confined or rather broad/widely-ranged construct[50, 51]. Besides, a stability test was conducted based on the Guttman split-half coefficient, which was also important to determine psychometric quality and test reliability[52]. What’s more, Multi-Trait-Multi-Method (MTMM) was conducted to assess the construct validity of the ASI-3 against the DASS-21 and TAI. According to Campbell and Fiske[53], MTMM could provide a practical methodology that researchers and practitioners could actually use for purposes of construct validation, especially convergent and discriminant validation. Previously, correlations from 0.40 to 0.70 have been considered satisfactory for convergent validity[54]. If correlation is too high (>0.70), it can be questioned whether the scales really assess different concepts and whether the application of two highly correlated scales adds useful information compared to if only one of the scales in included. Conversely, by discriminate validity one assumes that scales assessing different constructs should show low correlation (<0.40)[54, 55]. Additionally, a receiver operating characteristic (ROC) analysis was conducted to ascertain the cut-off scores for correctly identifying soldiers with anxiety (defined as a DASS-21 Anxiety score >7) [42]. The Delong Clarke-Pearson method was used to compute the areas under the ROC curves (AUCs) [56]. Lastly, one-way analysis of variance followed by LSD post-hoc test for multiple groups comparison was used to compare the AS differences in the services and positions of the participants. SPSS for Windows version 23.0 (SPSS Inc., Chicago, IL, USA) was used to conduct the above statistical analyses.

Ethics statement

The present study protocol was reviewed and approved by the institutional review board of the Second Military Medical University (IRB No. 20152049). All the participants were adults and signed the informed consent approving the use of their data for research purposes. For participants from 16 to 18 years old, the ethics committee has approved the lack of parent or guardian consent.


Descriptive analysis

Population parameters (M, SD, skewness, and kurtosis) of ASI are displayed in Table 2. According to Kline, the cut-off of the absolute values of skewness and kurtosis were 3.0 and 8.0, respectively [57]. In the current study, the univariate skewness for scores on the measurement instruments ranged from 0.784 to 2.175, and the univariate kurtosis for the instruments ranged from 0.037 to 5.756. Both met the acceptability criteria of normal distribution, so no adjustments were made to the scores on the variables in the following analysis.


Our study aimed to establish the factorial validity of the Chinese version of the Anxiety Sensitivity Index 3 in a sample of military personnel. Therefore, a confirmatory factor analysis was performed to examine the three-factor model. Overall, the fit indices mainly supported the scale structures: χ2/df = 16.583 (P<0.001), CFI = 0.925, TLI = 0.901, RMSEA = 0.071 and managed to meet the standard values.

A further detailed examination of the factor loadings showed the information on the analytical fit of the model, complementing the overall fit information. All indicators significantly loaded (P<0.01) on the hypothesized factor, supporting the adequacy of the three-factor model. As Fig 1 shows, the standardized factor loadings ranged from a minimum of 0.51 (item 1: It is important to me not to appear nervous) to a maximum of 0.79 (item 16: When I have trouble thinking clearly, I worry that there is something wrong with me). Most of the factor loading was better than those in a previous study using a sample of healthy adult women [32].

Fig 1. Factor structure of the ASI-3—The values between item and subscale is factor loading; The values between different scales is correlation coefficient.

Additionally, the MTMM correlation matrix is shown in Table 3. The total score and subscale scores correlated with each other (r = 0.700–0.913, p<0.001) and were also significantly associated with DASS-21 in positive ways (r = 0.432–0.597, p<0.001). This finding indicated that the convergent validity of the ASI-3 with related variables met the standard values in the current sample. One the other hand, correlation between ASI-3 and TAI was relatively low (r = 0.169–0.290, p<0.001), which indicated there was an empirical discrimination between anxiety sensitivity and trait anxiety.

Table 3. Multi-Trait-Multi-Method correlation matrix among ASI-3, DASS-21 and TAI.


Cronbach’s alpha coefficients were computed and are shown in the brackets of Table 2. Alpha coefficients had a value of 0.828 for the Physical concerns dimension (95% CI 0.818–0.837), 0.835 for the Social concerns dimension (95% CI 0.826–0.844), and 0.841 for the Cognitive concerns dimension (95% CI 0.833–0.850). Moreover, the internal consistency of the entire scale was 0.926 (95% CI 0.922–0.930). The Guttman split-half coefficient of ASI-3 was 0.881, indicating that it has adequate reliability.

Descriptive statistics, item homogeneity (item-factor correlation), and alpha if item deleted are presented in Table 4. Overall, the estimated internal consistency of the scale can be considered adequate.

Table 4. Means, standard deviations, item-factor correlation, alpha if item deleted for the items of the ASI-3.

Cut-off scores for anxiety

Individuals with a score of 7 or higher on the DASS-21 Anxiety subscale were defined as anxious, based on the diagnostic performances of ASI-3, which are illustrated in Fig 2 and Table 5. Maximal discrimination between anxiety and non-anxiety was obtained at the cut-off score of 16, accompanied by the improvement of positive predictive value and negative predictive value. This result means that a score of 16 or higher indicates the presence of anxiety. When 16 was used as the cut-off value, the sensitivity, specificity, positive predictive value, negative predictive value, and Area Under the Curve were 71.1%, 76.7%, 51.8%, 88.3% and 0.800 (P<0.001; 95% CI: 0.757–0.843), respectively.

Fig 2. Receiver operating characteristics curve for the ASI-3 to detect anxious individuals.

Table 5. Sensitivity, specificity, positive predictive value, and negative predictive value of ASI-3 at various cut-off scores for anxiety.

Differences according to services

As Table 6 shows, there were significant differences in the ASI and subscales among five services (F = 45.934–123.389, P<0.001). In the multiple comparisons, total points and subscales were different from each other (P<0.05), except for physical concerns between navy and air force (LSD-t = -0.345, P = 0.068), army and rocket force (LSD-t = -0.261, P = 0.180), social concerns between navy and rocket force (LSD-t = 0.123, P = 0.643), and cognitive concerns between navy and air force (LSD-t = -0.291, P = 0.144).

Table 6. Comparison of ASI and subscales among five services (Mean±SD).

Differences based on positions

As Table 7 shows, there were significant differences in the ASI and subscales among four positions (F = 19.544–102.758, P<0.001). In the multiple comparisons, no significant differences between soldier and cadet were observed in total points and subscales (P>0.05). Officers had higher scores in total points and subscales than individuals on the other three positions (P<0.001). Additionally, social concerns values were lower in sergeants than those in soldiers (LSD-t = -0.991, P<0.001) and cadets (LSD-t = -1.180, P<0.001). Soldiers scored higher on the ASI-3 than sergeants (LSD-t = 1.501, P = 0.001).

Table 7. Comparison of ASI and subscales among four positions (Mean±SD).


As the first Chinese large-sample norm, the current study aimed to provide evidence for the validity of the Chinese version of the ASI-3 using robust methods. More than three thousand soldiers in various services and positions were included, and the findings are summarized as follows.

First, the CFA results indicated that the Chinese version of the ASI-3 met basic psychometric quality criteria. Particularly, the three-factor (physical, social, and cognitive concern) model provided a good fit of the data structure, which indicated that a high construct validity was obtained in the current samples. In contrast, the fit indices tended to be poorer than those of several other cross-cultural studies [23]. This result may be due to the weakness of some items in the social and cognitive concerns subscale that showed lower loading in the Chinese version of the ASI-3 compared with the English version. However, the fit indices did not significantly improve, and these items were removed.

Besides, sufficient coefficient alpha values of ASI-3 and subscales (Table 2) indicated high internal consistency. Since test length has an effect on coefficient alpha values[50], r > 0.8 is quite high for the subscale with only 6 items. Meanwhile, the Guttman split-half coefficient of ASI-3 was 0.881, indicating that it has adequate reliability.

Additionally, slight correlations (r = 0.169–0.290) were seen between anxiety sensitivity and trait anxiety, which is another important neuroticism subcomponent to experiencing anxiety, which showed there is an empirical discrimination between anxiety sensitivity and trait anxiety. On the other hand, positive correlations between ASI and DASS-21, especially anxiety subscale, indicating high criterion-related validity of this instrument.

The AUC for identifying individuals with anxiety showed a high diagnostic accuracy on the ASI-3 (0.800). At a cut-off score of 16, the sensitivity and specificity levels were 71.1% and 76.7%, respectively, where the sum of sensitivity and specificity becomes the maximum, accompanied with improvement of PPV and NPV. Above results indicated that the ASI-3 had good criterion validity.

Interestingly, differences of AS in services and positions were found in the current study. Individuals in the Air Force tended to score higher than others on the ASI-3, especially on the social concerns subscale. According to a U.S. Air Force report, the suicide rate increased from 8.9 per 100,000 active duty Airmen in 2004 to 15.5 in 2012 [58]. Perceived lack of social support [59], low unit cohesion [60], and greater psychosocial difficulties influenced the severity of suicidal ideation among those Airmen affected [61]. Due to facing heavier workloads, more dangerous military tasks, and longer family separations, the Airmen suffered more from anxiety problems. Moreover, a strict elimination system in the AF led to the fear of showing anxiety in the presence of others, which also raised Airmen’s anxiety sensitivity and social concerns [62]. Among other military personnel, the officers reported much more anxiety sensitivity than others, and the sergeants reported the least, which was consistent with the findings reported by Jiang et al. of a sample of Chinese military personnel participating in an earthquake rescue [63]. Officers of ground soldiers’ shouldered many responsibilities, such as organizing trainings, ensuring safety, and executing commands, which required them to stay alert at all times. This responsibility directly gave rise to their anxiety sensitivity in the cognitive, physical, and social dimensions. In contrast, most sergeants were experienced soldiers who had adapted to military life. Their brain work was lighter than that of officers, and their manual work was lighter than that of soldiers. Therefore, they reported the lowest scores on the ASI-3 and its subscales.

There were still several limitations worthy of consideration. First, sample constructs remained to be improved. There were much important differences among sub-samples, thus future work should examine whether the three factor model could be replicated across difference samples, especially some special forces, such as airborne troops, submarine troops, and missile troops, who have to face many more stressors and might show poor mental health status. Second, the current study managed to investigate the reliability and validity of the 3-factor ASI in a sample of Chinese military personnel but failed to modify the items and constructs based on Chinese culture. Therefore, a more complex application of confirmatory factor analysis and cross-cultural methodology should be used in future research. Third, the cross-sectional design made it difficult to establish a causal relationship between the data. It remains to be determined whether high anxiety sensitivity individuals in the military would be more likely to develop an anxiety disorder in the battlefield environment or during retirement.

Nevertheless, the findings of the current study indicated that the Chinese version of the ASI-3 met the basic psychometric criteria. High construct validity, criterion validity, and internal consistency reliability suggested that the scale is useful in a sample of military personnel. Finally, the present findings provided strong support to focus on the emotional state of Airmen and officers and to take actions to improve their mental health.


The authors would like to thank our adorable students: Yingnan Ge, Kaixuan Ma, Jinchun Li, Zhuangzhuang Wei, Yue Yang, for their collecting data and careful check.


  1. 1. Clark DM. Anxiety disorders: Why they persist and how to treat them. Behaviour research and therapy. 1999;37:S5–S27. pmid:10402694
  2. 2. Clark DM, Wells A. A cognitive model of social phobia. Social phobia: diagnosis, assessment, and treatment. New York: Guilford Press; 1995. p. 69–93.
  3. 3. Clark DM. A cognitive approach to panic. Behaviour research and therapy. 1986;24(4):461–70. pmid:3741311
  4. 4. Berman NC, Wheaton MG, McGrath P, Abramowitz JS. Predicting anxiety: The role of experiential avoidance and anxiety sensitivity. Journal of Anxiety Disorders. 2010;24(1):109–13. pmid:19819106
  5. 5. Reiss S, Mcnally RJ. The expectancy model of fear: New York, NY, Academic Press; 1985.
  6. 6. Zvolensky MJ, Schmidt NB, Bernstein A, Keough ME. Risk-factor research and prevention programs for anxiety disorders: a translational research framework. Behaviour Research and Therapy. 2006;44(9):1219–39. pmid:16867299
  7. 7. Schmidt NB, Buckner JD, Keough ME. Anxiety sensitivity as a prospective predictor of alcohol use disorders. Behavior Modification. 2007;31(2):202–19. pmid:17307935
  8. 8. Taylor S, Cox BJ. An expanded anxiety sensitivity index: evidence for a hierarchic structure in a clinical sample. Journal of anxiety disorders. 1998;12(5):463–83. pmid:9801964
  9. 9. McNally RJ. Is anxiety sensitivity distinguishable from trait anxiety? Reply to Lilienfeld, Jacob, and Turner (1989). Journal of abnormal psychology. 1989;98(2):193–4. pmid:2708664
  10. 10. Reiss S. Trait anxiety: it's not what you think it is. Journal of Anxiety Disorders. 1997;11(2):201–14. pmid:9168342
  11. 11. Donnell CD, Mcnally RJ. Anxiety sensitivity and panic attacks in a nonclinical population. Behav Res Ther. 1990;28(1):83–5. pmid:2302152
  12. 12. Reiss S. Expectancy model of fear, anxiety, and panic. Clinical Psychology Review. 1991;11(2):141–53.
  13. 13. McNally RJ. Psychological approaches to panic disorder: a review. Psychological Bulletin. 1990;108(3):403–19. pmid:2270235
  14. 14. Taylor S, Cox BJ. Anxiety sensitivity: multiple dimensions and hierarchic structure. Behaviour Research & Therapy. 1998;36(1):37–51.
  15. 15. Çakmak SŞ. Psychometric properties of Anxiety Sensitivity Index-Revised and the relationship with drinking motives and alcohol use in Turkish university students and patients. M.Sc. Thesis, Middle East Technical University. 2006. Available from:
  16. 16. Reiss S, Peterson RA, Gursky DM, McNally RJ. Anxiety sensitivity, anxiety frequency and the prediction of fearfulness. Behaviour research and therapy. 1986;24(1):1–8. pmid:3947307
  17. 17. Peterson RA, Heilbronner RL. The anxiety sensitivity index:: Construct validity and factor analytic structure. Journal of Anxiety Disorders. 1987;1(2):117–21.
  18. 18. Telch MJ, Shermis MD, Lucas JA. Anxiety sensitivity: Unitary personality trait or domain-specific appraisals? Journal of Anxiety Disorders. 1989;3(1):25–32.
  19. 19. Zinbarg RE, Barlow DH, Brown TA. Hierarchical structure and general factor saturation of the Anxiety Sensitivity Index: Evidence and implications. Psychological Assessment. 1997;9(3):277–84.
  20. 20. Stewart SH, Taylor S, Baker JM. Gender differences in dimensions of anxiety sensitivity. Journal of Anxiety Disorders. 1997;11(2):179–200. pmid:9168341
  21. 21. Deacon BJ, Abramowitz JS, Woods CM, Tolin DF. The Anxiety Sensitivity Index-Revised: psychometric properties and factor structure in two nonclinical samples. Behaviour research and therapy. 2003;41(12):1427–49. pmid:14583412
  22. 22. Zvolensky MJ, Arrindell WA, Taylor S, Bouvard M, Cox BJ, Stewart SH, et al. Anxiety sensitivity in six countries. Behaviour Research and Therapy. 2003;41(7):841–59. pmid:12781249
  23. 23. Taylor S, Zvolensky MJ, Cox BJ, Deacon B, Heimberg RG, Ledley DR, et al. Robust dimensions of anxiety sensitivity: development and initial validation of the Anxiety Sensitivity Index-3. Psychological assessment. 2007;19(2):176–88. pmid:17563199
  24. 24. Mantar A, Yemez B, Alkin T. The validity and reliability of the Turkish version of the anxiety sensitivity index-3. Turk Psikiyatri Dergisi. 2010;21(3):1–9.
  25. 25. Michałowski JM, Holas P, Zvolensky MJ. Polish Adaptation and Psychometric Validation of the Anxiety Sensitivity Index-III. Journal of Individual Differences. 2014;32:79–86.
  26. 26. Ebesutani C, Kim M, Park H-H. The utility of the bifactor model in understanding unique components of anxiety sensitivity in a South Korean sample. Asian Journal of Psychiatry. 2016;22:116–23. pmid:27520911
  27. 27. Ghisi M, Bottesi G, Altoè G, Razzetti E, Melli G, Sica C. Factor Structure and Psychometric Properties of the Anxiety Sensitivity Index-3 in an Italian Community Sample. Frontiers in psychology. 2016;7:1–13. pmid:26858668
  28. 28. Yan Y, Lin R, Tang X, Liu M, You L. Anxiety Sensitivity and Sleep Quality in Young Students:The Relationship and Its Internal Mechanism. Journal of Psychological Science. 2011;34(4):987–92.
  29. 29. Liu YH, Huang XO, Wang Q, Wu JJ. Study of Route Analysis of the Relationship between Mental Health and State-trait Anxiety,Anxiety Sensitivity of the Military Pilots. Clinical Journal of Medical Officers. 2009;4:703–5.
  30. 30. Chen W, Cheng C, Yang L, Liu XC, Liu HL. Relationship between Anxiety Sensitivity Cognitive Concerns, Depression and Suicidal Ideation. Chinese Journal of Clinical Psychology. 2016;24:833–6.
  31. 31. Li Q, Guo L-T, Feng Z-Z. Reliability and validity of Chinese revision of the anxiety sensitivity index-revised. Chinese Mental Health Journal. 2006;20(10):675–8.
  32. 32. Wang L, LIU W, Zhu X, Wang Y, LI L, Yang Y, et al. Validity and reliability of the Chinese Version of the Anxiety Sensitivity Index-3 in healthy adult women. Chinese Mental Health Journal. 2014;28(10):767–71.
  33. 33. Van Dam NT, Earleywine M, Forsyth JP. Gender bias in the sixteen-item Anxiety Sensitivity Index: An application of polytomous differential item functioning. Journal of Anxiety Disorders. 2009;23(2):256–9. pmid:18786807
  34. 34. Bartone PT, Adler AB, Vaitkus MA. Dimensions of psychological stress in peacekeeping operations. Military Medicine. 1998;163(9):587–93. pmid:9753982
  35. 35. Castro CA, Adler AB. OPTEMPO: Effects on soldier and unit readiness. Parameters. 1999;29(3):86–95.
  36. 36. Hoge CW, Castro CA, Messer SC, McGurk D, Cotting DI, Koffman RL. Combat duty in Iraq and Afghanistan, mental health problems, and barriers to care. New England Journal of Medicine. 2004;351(1):13–22. pmid:15229303
  37. 37. Kuehn BM. Soldier suicide rates continue to rise. Jama. 2009;301(11):1111–3. pmid:19293405
  38. 38. Anglemyer A, Miller ML, Buttrey S, Whitaker L. Suicide rates and methods in active duty military personnel, 2005 to 2011: a cohort study. Annals of internal medicine. 2016;165(3):167–74. pmid:27272476
  39. 39. Sharp S, Meyer WJ III. Acute stress disorder and post traumatic stress disorder in individuals suffering from burn injury. Handbook of Burns: Springer; 2012. p. 3–13.
  40. 40. Saunders P. Chinese Military Reform in the Age of Xi Jinping: Drivers, Challenges, and Implications. Washington, D.C.: Government Printing Office; 2017.
  41. 41. Lovibond PF, Lovibond SH. The structure of negative emotional states: comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav Res Ther. 1995;33(3):335–43. pmid:7726811
  42. 42. Gong Y, Xie XY, Xu R, Lu YJ. Psychometric Properties of the Chinese Versions of DASS-21 in Chinese College Students. Chinese Journal of Clinical Psychology. 2010;18(4):443–6.
  43. 43. Spielberger CD. State‐Trait anxiety inventory: Wiley Online Library; 2010.
  44. 44. Zheng X, Li Y. State-trait anxiety inventory. Chinese Mental Health Journal. 1997 (4):219–20.
  45. 45. Herbei R, Paul R, Berliner LM. Applying diffusion-based Markov chain Monte Carlo. Plos One. 2017;12(3):e0173453. pmid:28301529
  46. 46. Tanaka JS. Multifaceted conceptions of fit in structural equation models. Sage Focus Editions; 1993.
  47. 47. Lt Hu, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural equation modeling: a multidisciplinary journal. 1999;6(1):1–55.
  48. 48. Hooper D, Coughlan J, Mullen M. Structural equation modelling: Guidelines for determining model fit. Electronic Journal on Business Research Methods. 2008;6(1):141–146.
  49. 49. Marsh HW, Hau K-T, Wen Z. In search of golden rules: Comment on hypothesis-testing approaches to setting cutoff values for fit indexes and dangers in overgeneralizing Hu and Bentler's (1999) findings. Structural equation modeling. 2004;11(3):320–41.
  50. 50. Steinborn MB, Langner R, Flehmig HC, Huestegge L. Methodology of Performance Scoring in the d2 Sustained-Attention Test: Cumulative-Reliability Functions and Practical Guidelines. Psychological Assessment. 2018;30(3):339–57. pmid:28406669
  51. 51. Miller J, Ulrich R. Mental chronometry and individual differences: modeling reliabilities and correlations of reaction time means and effect sizes. Psychonomic Bulletin & Review. 2013;20(5):819–58.
  52. 52. Wu LF, Yang SH, Koo M. Psychometric properties of the Chinese version of Spiritual Index of Well-Being in elderly Taiwanese. Bmc Geriatrics. 2017;17(1):3. pmid:28052767
  53. 53. Campbell D, Fiske D. Validation by the multitrait/multimethod matrix. Psychological Bulletin. 1959;51:79–105.
  54. 54. Kaasa S, Bjordal K, Aaronson N, Moum T, Wist E, Hagen S, et al. The EORTC Core Quality of Life questionnaire (QLQ-C30): validity and reliability when analysed with patients treated with palliative radiotherapy. European Journal of Cancer. 1995;31A(13–14):2260–3. pmid:8652253
  55. 55. Fredheim OM, Borchgrevink PC, Saltnes T, Kaasa S. Validation and comparison of the health-related quality-of-life instruments EORTC QLQ-C30 and SF-36 in assessment of patients with chronic nonmalignant pain. J Pain Symptom Manage. 2007;34(6):657–65. pmid:17618079
  56. 56. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988:837–45. pmid:3203132
  57. 57. Kline RB. Principles and practice of structural equation modeling: Guilford publications; 2015.
  58. 58. Bryan CJ, Hernandez AM. The functions of social support as protective factors for suicidal ideation in a sample of air force personnel. Suicide and Life-Threatening Behavior. 2013;43(5):562–73. pmid:23829611
  59. 59. Bossarte RM, Knox KL, Piegari R, Altieri J, Kemp J, Katz IR. Prevalence and characteristics of suicide ideation and attempts among active military and veteran participants in a national health survey. American journal of public health. 2012;102(S1):38–40.
  60. 60. Mitchell MM, Gallaway MS, Millikan AM, Bell M. Interaction of combat exposure and unit cohesion in predicting suicide‐related ideation among post‐deployment soldiers. Suicide and life-threatening behavior. 2012;42(5):486–94. pmid:22934836
  61. 61. Pietrzak RH, Goldstein MB, Malley JC, Rivers AJ, Johnson DC, Southwick SM. Risk and protective factors associated with suicidal ideation in veterans of Operations Enduring Freedom and Iraqi Freedom. Journal of affective disorders. 2010;123(1):102–7.
  62. 62. Zou ZK, Zhang L, Liu QY, Xia Q, Liu W, Bao-Min JI, et al. Comparison of recruitment standards for cadet pilots between PLAAF and USAF. Military Medical Sciences. 2016;2:28–30.
  63. 63. Jiang XL, Gao YS, Wang XD, Guo M. Survey of anxiety and depression of workers engaged in rescue activity in Wenchuan earthquake. China Tropical Medicine. 2009;9:389–90.