Association of pyschological distress with psychological factors in rescue workers within two months after a major earthquake

J Formos Med Assoc 2002 • Vol 101 • No 3 169

Psychological Distress in Major-Disaster Rescue Workers

(J Formos Med Assoc2002;101:169–76)

Key words:earthquakerescue workerspsychological distresspsychosocial factors

Departments of 1Psychiatry and 3Emergency Medicine, National Taiwan University Hospital, and 2Department of SocialMedicine, National Taiwan University Medical College, Taipei.Received: 20 January 2001. Revised: 2 May 2001. Accepted: 6 November 2001.Reprint requests and correspondence to: Professor Ming-Been Lee, Department of Psychiatry, National Taiwan UniversityHospital, 7 Chung-Shan South Road, Taipei, Taiwan.

ASSOCIATION OF PSYCHOLOGICAL DISTRESS WITH

PSYCHOLOGICAL FACTORS IN RESCUE WORKERS

WITHIN TWO MONTHS AFTER A MAJOR EARTHQUAKE

Shih-Cheng Liao,1 Ming-Been Lee,1,2 Yue-Joe Lee,1 Tei Weng,3 Fu-Yung Shih,3 andMatthew HM Ma3

In their efforts to help the victims, rescue workers inmajor disasters are placed at high risk of developingpsychological symptoms. Two studies of emergencyservices personnel responding to the 1989 Loma Prietaearthquake suggest that approximately 9% of workersshowed psychological symptoms at the level of those of

Background and Purpose: Studies of the health of rescue workers after a major disasterhave frequently focused on posttraumatic stress disorder. This study aimed todetermine the characteristics of psychological distress and its psychosocial predictorsin rescue workers within a 2-month period after an earthquake that struck centralTaiwan on September 21, 1999.Methods: A total of 1,104 rescue workers serving in the earthquake were enrolledin the study. Psychological distress was measured using the Brief Symptom RatingScale (BSRS), personality traits using the Maudsley Personality Inventory (MPI), andfamily function using APGAR (adaptability, partnership, growth, affection, andresolve) indexes. These measurements were performed within 2 months of theearthquake. Univariate and multivariate analyses were applied to examine theassociation between psychological distress and various psychosocial factors.Results: BSRS assessment revealed severe psychological distress in 137 (16.4%)subjects. The most common symptom dimension was phobic-anxiety (18.7%),followed by hostility (17.6%), obsessive-compulsive symptoms (16.2%), depression(14.9%), paranoid ideation (14.2%), interpersonal sensitivity (13.3%), psychoticism(11.9%), anxiety (10.8%), additional symptoms (8.5%), and somatization (6.2%).Pre-disaster major life events (R2 = 0.03) and most of the factor scores of the MPI(including moodiness, anxiety-prone, outgoing, conscientiousness, activity, andsociability factors; R2 = 0.25) predicted the severity of psychological distress. Timeof arrival at the scene, previous exposure, age, and family function had no or trivialpredictive power.Conclusion: The results of this study indicated that prevalence of general psychologicaldistress is high among rescue workers in the first 2 months after a major earthquake.Personality traits and pre-disaster life adjustment had a dominant predictive powerfor psychological distress. Immediate psychosocial intervention should be consideredto ameliorate the distress related to disaster rescue work.

a psychiatric outpatient population [1, 2]. Previousstudies have mainly focused on posttraumatic stressdisorder (PTSD) due to its high prevalence rate amongrescue workers involved in major disasters [3–5]. Alongitudinal study revealed that 21% of firefighterswho were exposed to the Ash Wednesday bushfires in

ORIGINAL ARTICLES

J Formos Med Assoc 2002 • Vol 101 • No 3

S.C. Liao, M.B. Lee, Y.J. Lee, et al

170

South Australia continued to experience disturbingimagery of the disaster between 4 and 29 months afterthe disaster [6]. Rescue workers exposed to traumaticdeath had increased intrusive and avoidant symptoms,hostility, and somatization that could persist for months[7]. Therefore, the identification of mental healthproblems and prompt intervention in rescue workersafter exposure to a major disaster are an important partof effective relief efforts.

The risk factors for development of psychosomaticdistress following exposure to a major disaster havebeen extensively examined. McFarlane found that neu-roticism and a history of treatment for psychologicaldisorder were better predictors of posttraumatic mor-bidity than the degree of exposure to the disaster or thelosses sustained [8]. These results raise doubts aboutthe postulated central etiologic role a traumatic eventplays in the onset of morbidity [6]. Other studiesreplicated the results that personality traits predictedpost-disaster distress in rescue workers [9, 10]. Inaddition, several other predictors, mainly related toPTSD, in rescue workers were identified, such as socialsupport [9–11], years of experience on the job [9],pathologic identification with disease [12],peritraumatic dissociation [2, 9], intensity of impact orexposure [2, 10], even identity of their profession [4].In addition, identification of risk factors helps to pre-dict which mental health problems are most likely tooccur after disasters and enables the development ofintervention programs.

Owing to the methodologic limitations anddifficulties in data collection, most studies on thesymptomatic distress of rescue workers after a majordisaster are conducted during the recovery stage, longafter the immediate impact. Therefore, the require-ments for immediate mental health intervention andthe profiles of acute symptomatology are hard to assess.Previous studies have also focused more on PTSD andless on global symptomatology, which may have led tounderestimation of the importance of some types ofdistress not caused by PTSD and the general mentalhealth condition of rescue workers in major disasters.Anxiety, depression, and other psychosomatic symp-toms may also be prevalent during the acute stage aftera major disaster and necessitate immediate interven-tion [6].

The present study aimed to determine the profilesof psychological distress in rescue workers within a 2-month period after a major earthquake using a seriesof standardized questionnaires to assess psychologicalsymptoms, personality trait, and family function. Thepredictive effect of personality traits, family function,and other psychosocial variables on the severity ofpsychological distress was also examined. The latentdimensions of personality assessment using data reduc-

tion methods, which may reveal more detailed contentof personality structure, were also examined.

Materials and Methods

SubjectsA total of 1,104 rescue workers involved in a majorearthquake were enrolled in this study. The earthquake,measuring 7.2 on the Richter scale, struck centralTaiwan at 1:47 am on September 21, 1999, causingsevere damage to property in several counties near theepicenter and resulting in the death of more than2,300 people. Study subjects were recruited from thoseenrolled in a collaborative program to survey mentalhealth problems conducted by the Taipei Bureau ofFire Safety and National Taiwan University Hospital.All assessments were completed within 2 months afterthe earthquake.

All subjects were male. Data from the Taipei Bureauof Fire Safety showed an average educational level forour sample. To explore the possible conditioning ef-fect of trauma, experience with previous disaster res-cue work (defined as more than 15 casualties) wasrecorded (yes or no). Major life events defined as deathof, or separation from, close relatives, major systemicdisease, and problems at work in the past 6 monthswere also documented. Time of arrival at the disastersite was noted for its possible effect on cognitive re-hearsal of the disaster experience.

Psychological symptomsThe frequencies and severity of psychologicalsymptoms were measured using the Brief SymptomRating Scale (BSRS) [13]. The BSRS is a self-reportingmeasure with 50 items rated from 0 to 4 on the basis ofthe degree of distress caused over the past week. Eachitem was scored as 0 = not at all, 1 = a little bit, 2 =moderately, 3 = quite a bit, or 4 = extremely. The BSRShas been shown to be a reliable and valid psychiatricself-rating scale for use in medical practice [14, 15]. Itwas modified from the widely used Derogatis’ Symp-tom Check List-90-R (SCL-90-R) [16] and designed tobe used as a shorter form. Like SCL-90-R, the BSRScovers nine dimensions of psychopathology:somatization, obsessive-compulsive, interpersonalsensitivity, depression, anxiety, hostility, phobic-anxiety,paranoid ideation, and psychoticism. Additional symp-toms included vegetative signs and suicidal ideation.The General Symptom Index (GSI) is essentially amean score of all BSRS items. The BSRS has beenshown in different populations to have an excellentsplit-half reliability as well as good internal structure



according to factor analysis [17–19]. In addition, BSRSscores are highly correlated with the parental form ofSCL-90-R among the medical population for each symp-tom dimension [13].

Personality traitsPersonality traits were measured using a short-form ofthe Maudsley Personality Inventory (MPI), a 26-itemself-reporting test to measure subscales of neuroticism-stability and extroversion-introversion [20]. Each itemwas rated as 0 = no, 1 = uncertain, or 2 = yes. The MPIwas originally developed under the personality con-struct proposed by Hans Eysenck, who hypothesizedthat individual differences in the dimension of intro-version-extraversion, neuroticism-stability, orpsychoticism determined the ease with which individu-als could acquire conditioned responses or be suscep-tible to stress, which in turn determined the form ofpsychopathology to which they were prone [20]. Itspsychometric properties and time stability have beendemonstrated elsewhere [21], and it has been usedwidely in both community and medical settings inTaiwan since the 1970s [21, 22]. The short-form MPIused in this study included 14 items of neuroticism-stability and 12 items of extroversion-introversiondimensions. The items of psychoticism and lie scalewere excluded.

Family functionThe family APGAR (adaptability, partnership, growth,affection, and resolve) index, which contains five struc-tured questions about family interaction, assessed fam-ily function. The scoring was 0 = seldom, 1 = sometimes,2 = most of time, or 3 = always. A higher score indicateshigher quality of family support [23].

Data analysisAll data were analyzed using SPSS for Windows (Version7.0, SSPS Inc, Chicago, IL, USA). Cases not included inthe analysis owing to missing data in any measurementwere examined using logistic regression with the for-ward procedure and Wald test. To examine the severityof psychological symptoms in our sample, an adjustedT score was determined from data obtained frommedical inpatients without formal diagnosis of psychi-atric disorder based on in-depth interview by seniorpsychiatrists. Control group data were obtained from aprevious study of 1,638 subjects randomly selectedfrom the Psychiatric Outpatient Clinic, the FamilyMedicine Clinic, and nonpsychiatric medical inpa-tients [24]. After adjustment, a T score of 50 wasconsidered identical to the mean of the referencegroup and the standard deviation (SD) was set at 10.One-sample t-test was used to determine statistical

significance. A p value of less than 0.05 was consideredsignificant.

Significant severity was defined as a symptom scoregreater than the mean score of the reference group bytwo SDs (> adjusted T score of 70). To explore thelatent structure of personality assessment, explorativefactor analysis was applied using the principle compo-nent method and orthogonal equamax rotation. Thecriterion of factor extraction was an eigenvalue greaterthan 1 and the factor scores were estimated by theregression method for further analysis. Since previousstudies have found that concurrent psychological symp-toms might influence cross-sectional measurement ofpersonality traits [25, 26], we further examined thishypothesis in our samples using a multiple regressionmodel that included 10 symptom dimensions as inde-pendent variables and each factor score as a dependentvariable. A multiple regression model that employedthe enter method was used in this test. The correlationbetween basic data, personality factor scores on theMPI, and dimensional scores of symptoms in the BSRSwere tested using Pearson’s product moment correla-tion coefficient. The stepwise regression model wasused to determine the predictors of psychologicalsymptoms. The independent variables in each regres-sion model included basic data (age, time of arrival atthe site, previous exposure to a major disaster, andprevious major life events) and factor scores of MPI.The dependent variables were the dimensional scoresof the BSRS.

Results

Due to scattered missing data from the BSRS andambiguous responses on the MPI, 268 cases were ex-cluded and 836 cases (75.72%) were included in theanalysis. From a comparison of the basic data (Table 1)for the 268 excluded and the 836 included cases, agewas the only significant predictor of exclusion in thefinal regression model (beta, .9571; 95% confidenceinterval, .9428 to .9716; Cox & Snell R square, .031).This suggested that younger subjects may have beenover-represented in our initial sample.

The most frequent items for which subjects were in‘quite a bit of’ or extreme distress were sleep distur-bance (9.8%), muscle ache (9.2%), loneliness (8.1%),depressive mood (7.9%), irritability (6.9%), inatten-tiveness (6.7%), tension (6.2%), and lack of interest(6.0%). ‘Quite a bit of’ or extreme suicidal ideation wasnoted in 3.0% of subjects. All T scores for the 10 symp-tom dimensions in BSRS were significantly greaterin subjects than in the reference group (p < 0.01).



172

In order to determine the percentage of subjects withsevere distress in BSRS symptom dimensions, we de-fined a T score greater than 70 as severely distressing(the probability of identical severity in the referencegroup was less than 2.5%). The most frequent distress-ing symptom dimension was phobic-anxiety (T score> 70 in 18.7% of subjects) followed by hostility (17.6%),obsessive-compulsive symptoms (16.2%), depression(14.9%), paranoid ideation (14.2%), interpersonalsensitivity (13.3%), psychoticism (11.9%), anxiety(10.8%), additional symptoms (8.5%), and somatiza-tion (6.2%) (Table 2). The Cronbach alpha of eachdimension ranged from 0.76 to 0.90, showing accept-able reliability.

The rotated solution of the explorative factor analy-sis for the short-form MPI showed a simple factorialstructure (Table 3). For factor loading greater than 0.4, no double loading or omission of each indicatorvariable was found. Six factors were extracted and theirorder of explained variance was as follows: factor 1 =moodiness factor; factor 2 = anxiety-proneness factor;factor 3 = outgoing factor; factor 4 = activity factor;

Table 1. Basic data of rescue workers (N = 836)

Age 28.32 ± 8.06Time of arrival at the scene

Never 67 (8.0%)< 6 h 333 (39.8%)6–12 h 123 (14.7%)12–24 h 73 (8.7%)1–3 d 86 (10.3%)> 3 d 154 (18.4%)

Major life event in the past 6 monthsYes 148 (17.7%)No 688 (82.3%)

Previous rescue work in a major disasterYes 440 (52.6%)No 396 (47.4%)

Table 2. Percentile of adjusted T score* and reliability test of Brief Symptom Rating Scale (BSRS) items (N = 836)

Percentile Adjusted T scores of BSRS†

PHO HOS OBS DEP PAR SEN PSY ANX ADD SOM GSI

Mean 57.76 57.50 56.53 54.86 58.29 55.13 57.74 51.12 51.98 49.31 56.15SD 20.64 17.81 15.87 15.94 17.94 13.90 19.86 13.11 12.92 10.70 19.64% with score > 70 18.7% 17.6% 16.2% 14.9% 14.2% 13.3% 11.9% 10.8% 8.5% 6.2% 16.4%Cronbach alpha .89 .86 .89 .86 .85 .85 .87 .90 .76 .80 .84

*Raw BSRS score adjusted by the means and standard deviations (SDs) derived from physically ill patients [13]. †Sorted by 70% case numbersexcept general symptom index (GSI). PHO = phobic-anxiety; HOS = hostility; OBS = obsessive-compulsive symptoms; DEP = depressivesymptoms; PAR = paranoid tendency; SEN = interpersonal sensitivity; PSY = psychoticism; ANX = anxiety symptoms; ADD = additionalsymptoms; SOM = somatic complaints.

factor 5 = conscientiousness factor; and factor 6 =sociability factor. Regarding the influence of symp-toms of the cross-sectional personality assessment,although the ANOVA test was significant for eachmodel, the R2 values were minimal: (moodinessfactor R2 = 0.095; anxiety-proneness factor R2 = 0.096;outgoing factor R2 = 0.022; activity factor R2 = 0.035;conscientiousness factor R2 = 0.057; sociability factor R2

= 0.022). Pearson’s product moment correlation showedsignificant correlation between symptom dimensionsand personality dimensions (Table 4), except for theoutgoing factor, which was only significantly correlatedwith interpersonal sensitivity, psychoticism, and GSI.None of the symptom dimensions was correlatedwith time of arrival at the site and the correlationswith family APGAR score only had marginal or nosignificance. Age was only correlated with anxiety,additional symptoms, and GSI.

In the stepwise regression models (Table 5), theanxiety-proneness factor was always the first predictorentered into the model except in predicting the depres-sive dimension. It had the largest predictive power forthe variance of symptom dimensions and reflected thatthe nature of distress after the rescue work came largelyfrom anxiety-related traits. Other personality dimensionsalso had predictive power for symptom dimensions,except for the outgoing factor, which was removed fromthe models of phobic-anxiety reaction, hostility, paranoidideation, and additional symptoms. Previous major lifeevents had a surprisingly higher ranking than somepersonality predictors, suggesting that pre-disaster lifeadjustment affected the subjects’ psychological well-being after rescue work. Previous exposure to a majordisaster only predicted the severity of the dimensions ofhostility and obsession, and family APGAR score onlypredicted obsession and anxiety. Age positively predictedthe severity of the symptom dimensions of anxiety, addi-tional symptoms, and somatization. Time of arrival at thescene was removed from all models.



Table 3. Factor loadings of explorative factor analysis* in selected Maudsley Personality Inventory (MPI) items (N = 836)

MPI Items Factors†

1 2 3 4 5 6

MPI02 Fluctuation of mood between depression and happiness .68 .05 –.20 .08 .18 –.01MPI09 Frequent ups and downs .68 .28 –.05 –.06 .22 –.13MPI03 Poor concentration .58 .29 –.12 -.14 .06 –.03MPI08 Frequently getting depressed .56 .21 –.06 -.02 .33 –.20MPI06 Easily distracted when talking to others .47 .27 –.17 .12 .09 –.05MPI12 Sometimes feeling sad without reason .47 .23 –.29 -.04 .37 –.10MPI17 Intermittently feeling lonely .46 .27 –.23 -.24 .15 .14MPI23 Frequent insomnia resulting from excessive worrying .17 .68 –.04 -.04 .17 –.05MPI26 Intermittent restlessness .25 .66 –.17 -.03 .22 –.05MPI24 Bothered by inefficient overthinking .28 .64 –.19 -.05 .14 .00MPI20 Frequent fatigue and lack of initiation .32 .50 –.26 -.24 .10 .08MPI25 Frequently get angry .00 .48 –.09 .02 .17 –.20MPI21 Not self-inhibited to talk in a group –.08 -.18 .75 .00 –.06 .15MPI10 Not passive at parties –.09 .00 .69 -.02 –.16 .17MPI18 Not retiring in front of the opposite sex –.11 -.08 .64 .16 –.03 –.22MPI05 Preferring action to planning for actions –.09 -.06 .03 .68 –.10 .12MPI07 Self-rating as a lively individual –.15 -.08 –.04 .60 –.17 .30MPI01 Happiest when involved in projects calling for rapid action .33 -.08 –.05 .60 .03 –.03MPI04 Take the initiative to make new friends .07 .08 .33 .54 .03 .10MPI15 Not too serious towards work –.04 .11 –.07 -.03 .82 .16MPI14 Frequently bothered by guilty feelings .19 .35 –.08 .05 .55 –.12MPI11 Unable to become fully relaxed at a happy social gathering .27 .10 –.37 .01 .43 –.24MPI13 Easygoing in social circles –.14 -.16 .10 .07 .38 .68MPI16 Enjoy numerous social activities .25 -.15 .06 .01 –.02 .65MPI22 Enjoy parties a lot –.15 .15 .10 .32 –.17 .56MPI19 Carefree and optimistic –.17 .05 –.05 .38 –.14 .51Eigenvalue 5.89 2.34 1.43 1.29 1.16 1.09% Of total variance 22.65 8.99 5.52 4.98 4.45 4.21Cumulative % of total variance 22.65 31.64 37.15 42.13 46.57 50.78

*Extraction method: principle component analysis; rotation method: equamax rotation; inclusion criteria: factor eigenvalue > 1; factorloading > 0.4 in bold type. †Factor 1 = moodiness factor; factor 2 = anxiety-proneness factor; factor 3 = outgoing factor; factor 4 = activityfactor; factor 5 = conscientiousness factor; factor 6 = sociability factor.

Table 4. Pearson’s product moment correlation coefficients between dimensions of the Maudsley Personality Inventory,APGAR index, age, and adjusted subscores of the Brief Symptoms Rating Scales (BSRS) (N = 836)

Dimensions of BSRS


Personality traitMoodiness factor .15** .24** .23** .27** .21** .23** .20** .22** .22** .19** .24**

Anxiety-proneness factor .26** .27** .30** .26** .28** .27** .26** .30** .31** .29** .31**

Outgoing factor –.06 –.06 –.09 -.07 -.05 –.12** –.08* –.05 –.04 -.06 –.08*

Activity factor -.11** –.11** –.16** –.17** –.11** –.17** –.12** –.13** –.11** –.14** –.15**

Conscientiousness factor .21** .20** .20** .22** .22** .20** .24** .23** .22** .17** .23**

Sociability factor –.15** -.13** –.14** –.15** –.16** –.14** –.18** –.17** –.13** –.12** –.16**

Neuroticism scores .36** .43** .45** .45** .42** .44** .42** .45** .43** .39** .47**

Extroversion scores –.15** –.15** –.19** –.20** –.15** –.21** –.18** –.17** –.14** –.17** –.19**

APGAR –.07* –.05 –.06 –.12** –.08* –.08* –.08* –.06 -.06 –.07* –.08*

Age .03 .05 .05 .01 .00 –.03 .01 .09** .09* .10** .06Time to arrival .04 .03 .03 .04 .05 .00 .04 .04 .04 .02 .04

*p < 0.05; **p <0.01; ***p <0.001. APGAR = adaptability, partnership, growth, affection, and resolve; PHO = phobic-anxiety; HOS= hostility; OBS = obsessive-compulsive symptoms; DEP = depressive symptoms; PAR = paranoid tendency; SEN = interpersonal sensitivity;PSY = psychoticism: ANX = anxiety symptoms; ADD = additional symptoms; SOM = somatic complaints; GSI = general symptom index.



174

Table 5. Stepwise multiple regression model of subscores of the Brief Symptom Rating Scale (BSRS)

Independent variables Dependent variables


Moodiness factor Beta 2.83 4.11 3.68 4.16 3.53 3.07 3.79 3.02 2.83 2.05 4.50Order 4 2 2 1 3 2 3 3 2 2 2

Anxiety-proneness factor Beta 5.21 4.71 4.62 3.95 4.88 3.63 4.93 3.90 3.81 3.01 5.84Order 1 1 1 2 1 1 1 1 1 1 1

Outgoing factor Beta X X -1.41 -1.05 X -1.55 -1.42 -.80 X -.72 -1.39Order 7 7 7 7 9 8 7

Activity factor Beta -2.10 -1.92 -2.64 -2.66 -1.89 -2.25 -2.14 -1.85 -1.48 -1.50 -2.79Order 6 6 5 4 6 4 6 6 6 4 5

Sociability factor Beta -2.80 -2.10 -2.19 -2.26 -2.61 -1.80 -3.26 -1.98 -1.41 -1.03 -2.94Order 5 4 6 5 4 5 4 4 4 6 4

Conscientiousness factor Beta 4.13 3.42 3.26 3.44 3.93 2.68 4.62 3.04 2.68 1.73 4.43Order 2 3 3 3 2 3 2 3 3 3 3

Time to arrival Beta X X X X X X X X X X XOrder

APGAR Beta X X .32 X X X X .26 X X XOrder 8 8

Age Beta X X X X X X X .14 0.12 .13 XOrder 7 7 7

Previous life event Beta 7.91 4.73 5.62 4.90 6.45 4.57 7.33 4.30 3.73 3.09 7.13Order 3 5 4 6 5 6 5 5 5 5 6

Previous exposure Beta X 3.14 1.89 X X X X X X X XOrder 7 9

R Square .19 .22 .26 .26 .23 .24 .24 .27 .24 .21 .28

p < 0.05 for all coefficients. X = independent variable removed from stepwise regression model. Order = sequence dependent variableentered regression model. PHO = phobic reaction; HOS = hostility; OBS = obsessive-compulsive symptoms; DEP = depressive symptoms;PAR = paranoid tendency; SEN = interpersonal sensitivity; PSY = psychoticism: ANX = anxiety symptoms; ADD = additional symptoms;SOM = somatic complaints; GSI = global severity index; APGAR = adaptability, partnership, growth, affection, and resolve.

Discussion

This study found a high prevalence of psychologicalsymptoms in rescue workers within the first 2 months aftera major earthquake. The most frequent symptoms orsymptom dimensions were trauma related, which reflectsthe specific psychopathological processes of intrusion,hyperarousal, and avoidance. However, other symptomswithout a direct theoretical correlation with trauma alsocaused prominent distress. Our study was limited in itsability to estimate the prevalence of psychiatric disorders,because the BSRS was originally designed as a dimen-sional approach for the assessment of general psychologi-cal well-being. However, this investigative strategy waschosen to emphasize that other dimensions of symptomsthan just the discrete syndromes of PTSD should bemonitored after disaster exposure and may be useful asindicators of specific psychological intervention such asdebriefing work or group counseling.

Our study confirmed the previous hypothesis ofpersonality traits as valid predictors of stress-related

symptomatology after a major disaster and even inthe early stage. The factor score of each personalitydimension has an independent predictive power ofpost-disaster distress. Factors with more homogeneouscontent belonging to traditionally defined neuroticismhad a higher rank in the predictive models. This findingis comparable with previous studies [2, 6–10]. McFarlanestudied a group of 469 firefighters at 4, 11, and 29 monthsafter extreme exposure to a bushfire disaster andconcluded that neuroticism and a history of treatment fora psychological disorder were better predictors of post-traumatic morbidity than the degree of exposure to thedisaster or the losses sustained [6, 8]. In our study, thesimple structure of explorative factor analysis demon-strated the construct validity of the personality assessmentin rescue workers. This finding needs to be confirmedwith other samples. Determination of the correlation ofeach factor with psychiatric diagnosis and the need forintervention will require further external validation. Inthis study, except for the outgoing factor, all factors hadsignificant predictive power. Lack of discriminative poweramong these factors may arise from the turmoil in theacute stage. Further studies of the long-term effects of



personality factors on predicting psychiatric morbiditiesmay determine the specificity of these factors to diagnosisand the need for intervention.

Another powerful predictor of psychological dis-tress was a previous life event. Although the sensitiza-tion effects of previous trauma are well established[27], life event as defined in our study was a moreattenuated form of relational or financial changes ineveryday life. However, its effect was strong in everysymptom dimension. In addition, previous rescue workexperience in a major disaster was removed from mostof our stepwise multiple regression models. This find-ing suggested not only that immediate debriefing workis needed after a disaster, but also that adjustment ineveryday life and personality traits play an importantrole in psychological well-being when facing a majordisaster. In our study, the lack of a baseline assessmentof symptomatology diminished the strength of thisinference. We do not know whether any of the subjectshad high levels of pre-existing distress from life eventsprior to the rescue work or if the distress was allgenerated after the disaster. However, the adjusted Tscore of each symptom dimension, as the dependentvariable in the regression model, was comparable tothat of physically ill inpatients in stressful life situations.Thus, the effect of trauma per se does not appear to havebeen diluted and the predictive power of previous lifeevents seems valid. Further prospective studies areneeded to determine the preventive effect of mentalhealth work in reaction to severe stress.

Family function based on APGAR index was not asignificant predictor of psychological distress in ourstudy, although it had a marginal effect on the severityof the symptom dimensions of obsession and anxiety.Marmar et al [4] and Weiss et al [9] reported thepredictive power of social support on distress in rescueworkers. However, their studies were performed in thechronic phase and the nature of the distress might havediffered from that in the acute stage. Furthermore, thesubculture of rescue work, which emphasizes loyalty,group cohesiveness, and masculinity, may re-direct theeffects of social support toward the peer relationship atleast in the acute stage. Further research focusing onthe influence of the peer relationship and supportfrom colleagues as variables of social recourse is needed.Because this study was performed in the acute stage,the requirement to provide psychiatric services limitedthe scope of data collection, and comprehensive assess-ment of impact and other demographic data were notincluded. However, the available data showed that agehad a marginal predictive power for anxiety, additionalsymptoms, and GSI. Age had a weakly positive correla-tion with symptom severity. The effect of age has beenextensively studied in PTSD of the general populationwith controversial results [28]. Marmar et al also showed

non-significant power of rescuer’s age in predictingperitraumatic dissociation that occurred just after di-saster exposure [4]. The low variance of age in oursubjects and overwhelming nature of the traumaticexperience in the acute stage may have blurred itssignificance. Time of arrival at the scene also failed topredict any symptom dimension. This suggests thattraumatic impact is a highly subjective phenomenon.

In summary, our findings have confirmed that per-sonality traits are predictors of post-disaster symptoma-tology in rescue workers. Pre-disaster adjustment to lifeevents also had dominant predictive power for symp-tomatic distress. In addition to the theoretical plausi-bility of the neuroticism-stress model, promotion ofmental health through the practice of stress manage-ment techniques in rescue workers prior to disasterexperience by way of self-help groups or mentoringprograms seems critical in ameliorating the distressrelated to future disaster rescue work.

References

1. Marmar CR, Weiss DS, Metzler TJ, et al: Stress responsesof emergency services personnel to the Loma Prietaearthquake Interstate 880 freeway collapse and controltraumatic incidents. J Trauma Stress 1996;9:63–85.

2. Marmar CR, Weiss DS, Metzler TJ, et al: Longitudinalcourse and predictors of continuing distress followingcritical incident exposure in emergency servicespersonnel. J Nerv Ment Dis 1999;187:15–22.

3. Lesaca T: Symptoms of stress disorder and depressionamong trauma counselors after an airline disaster. PsychiatrServ 1996;47:424–6.

4. Marmar CR, Weiss DS, Metzler TJ, et al: Characteristics ofemergency services personnel related to peritraumaticdissociation during critical incident exposure. Am J Psy-chiatry 1996;153:94–102.

5. Epstein RS, Fullerton CS, Ursano RJ: Posttraumatic stressdisorder following an air disaster: a prospective study. AmJ Psychiatry 1998;155:934–8.

6. McFarlane AC: Long-term psychiatric morbidity after anatural disaster. Med J Aust 1986;145:561–3.

7. Ursano RJ, Fullerton CS, Kao TC, et al: Longitudinalassessment of posttraumatic stress disorder and depres-sion after exposure to traumatic death. J Ment Nerv Dis1995;183:36–42.

8. McFarlane AC: The aetiology of post-traumatic morbidity:predisposing, precipitating and perpetuating factors.Br J Psychiatry 1989;154:221–8.

9. Weiss DS, Marmar CR, Metzler TJ, et al: Predictingsymptomatic distress in emergency services personnel.J Consult Clin Psychol 1995;63:361–8.

10. Bartone PD, Ursano RJ, Wright KM, et al: The impact ofa military air disaster on the health assistance workers: aprospective study. J Nerv Ment Dis 1989;177:317–28.



176

11. Fullerton CS, McCarroll JE, Ursano RJ, et al: Psychologi-cal responses of rescue workers: fire fighter and trauma.Am J Orthopsychiatry 1992;62:371–8.

12. Ursano RJ, Fullerton CS, Vance K, et al: Posttraumaticstress disorder and identification in disaster workers. AmJ Psychiatry 1999;156:353–9.

13. Lee MB, Lee YJ, Yen LL, et al: Reliability and validity ofusing a Brief Psychiatric Symptom Rating Scale in clinicalpractice. J Formos Med Assoc 1990;89:1081–7.

14. Hwang SL, Chang Y, Ko WJ, et al: Stress-reducing effectof physician’s tape-recorded support on cardiac surgicalpatients in the intensive care unit. J Formos Med Assoc1998;97:191–6.

15. Lin LS, Lee MB, Lee YG, et al: Psychiatric disorders ininpatients with stroke. Chinese Psychiatry 1995;9:65–73.

16. Derogatis LR, Rickels K, Rock AF: The SCL-90 and theMMPI: a step in the validation of a new self-report scale.Br J Psychiatry 1976;128:280–9.

17. Derogatis LR, Lipman RS, Covi L: The SCL-90: an outpa-tient psychiatric rating scale. Psychopharmacol Bull 1973;9:13–8.

18. Peveler RC, Fairburn CG: Measurement of neurotic symp-toms by self-report questionnaire: validity of the SCL-90-R. Psychol Med 1990;20:873–9.

19. Derogatis LR, Rickels K, Rock AF: Factorial invariance ofsymptom dimensions in anxious and depressive neurosis.Arch Gen Psychiatry 1972;27:659–65.

20. Wretmark G, Astrom J, Eriksson M: The Maudsley Per-

sonality Inventory as a prognostic instrument. Br J Psy-chiatry 1970;116:21–6.

21. Lee MB, Lin HN, Lee YG, et al: Personality as an effectivepredictor of outcome of neurotic disorders. Chinese Psy-chiatry 1990;4:111–21.

22. Chen YY, Lee YG, Tseng MC, et al: Psychiatric morbidity andassociated psychosocial characteristics of women with preg-nancy loss. Taiwanese Psychiatry 1998;12:194–206.

23. Sprusinska E: The Family APGAR Index: study on rela-tionship between family function, social support, globalstress and mental health perception in women. Int J OccupMed Environ Health 1994;7:23–32.

24. Lee MB, Lee YJ: A cross-sectional epidemiological studyof psychiatric comorbidity in hospitalized medically ill.Chinese Psychiatry 1990;4:10–26.

25. Liebowitz MR, Stallone F, Dunner DL, et al: Personalityfeatures of patients with primary affective disorder. ActaPsychiatr Scand 1979;60:214–24.

26. Hare EH, Payne H, Laurence KM, et al: Effect of severestress on the Maudsley Personality Inventory score innormal subjects. Br J Soc Clin Psychol 1972;11:353–8.

27. Kessler RC, Sonnega A, Bromet E, et al: Posttraumaticstress disorder in the National Comorbidity Survey. ArchGen Psychiatry 1995;52:1048–60.

28. Kessler RC, Sonnega A, Bromet E, et al: Epidemiologicalrisk factors for trauma and PTSD. In: Yehuda R, ed. RiskFactors for Posttraumatic Stress Disorder. Washington DC:American Psychiatry Press, 2000:41.

Documents

Association of pyschological distress with psychological factors in rescue workers within two months after a major earthquake