USF Libraries
USF Digital Collections

Children of Iniki


Material Information

Children of Iniki effects of evacuation and intervention : final report submitted to Natural Hazards Research and Applications Information Center
Series Title:
Quick response research report ;
Physical Description:
22 p. : ; 28 cm.
Hamada, Roger Saburo
University of Colorado, Boulder -- Natural Hazards Research and Applications Information Center
The Center
Place of Publication:
Boulder, Colo.
Publication Date:


Subjects / Keywords:
Disasters -- Psychological aspects   ( lcsh )
Hurricane Iniki, 1992 -- Psychological aspects   ( lcsh )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Includes bibliographical references (p. 18-19).
Additional Physical Form:
Also issued online as part of a joint project with the Louis de la Parte Florida Mental Health Institute (FMHI) Research Library's disaster mental health initiative.
Statement of Responsibility:
by Roger S. Hamada.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001985077
oclc - 31876997
usfldc doi - F57-00022
usfldc handle - f57.22
System ID:

This item is only available as the following downloads:

Full Text


#JrItzrdUJff),0'Ntt..{iPnst1}'1'l:IAZARD HOUSE Children ofIniki: Effects of Evacuation and InterventionByRoger S. HamadaQUICK RESPONSE RESEARCH REPOe 1994Theviews expressedinthis report are thoseofthe authors and not necessarily thoseofthe Natural Hazards Centerorthe UniversityofColorado.


Children of Iniki: Effects of Evacuation and InterventionByRogerS.HamadaQUICK RESPONSERESEARCH REPOe 1994This pUblication is partofthe Natural Hazards &Applications Information Center's ongoing QUIck Response Research Report Series.


CHILDRENOFINIKI:EFFECTSOFEVACUATION AND INTERVENTION Final Report SubmittedtoNatural Hazards Research and Applications Information Center by RogerS.Hamada, Ph.D. Kapiolani Medical Center for Women and Children and John A. Burns SchoolofMedicine University of HawaiiatMano'aAugust25,1994


ChildrenofIniki Final Report Hamada, Roger S. -2 INTRODUCTION Background Effectsofdisaster on adults' postdisaster psychosocialadjustmenthaye beenwelldocumented.'Agrowingliterature also documentstheeffectsofnatural disaster on children's emotional and behavioraladjustmentfollowingtheevent. Vogel and Vernberg, in a comprehensivereviewofthe literature, concludethatcommon responsesofchildrentonatural disaster include specific fears, separation difficulties, sleep problems, andsymptomsassociatedwithposttraumaticstress disorder (PTSD).2Also reported aresymptomsofanxiety, depression, somatization, and behavioral disturbance.3A recent large-scalestudybyLonigan and his colleagues4,5studied5,687children in grades five through12whowere exposedtoHurricane Hugo in South Carolina in1989.Using the Frederick Reaction IndexforChildren (RIlB ,theyfoundthat5.42%ofchildren in their samplemetdiagnostic criteriaforself-reported PTSD. Further,theyfoundthatratesofself-reported PTSDwerehigher in younger children than in older children(9.2%in9-12year-olds;4.2%in13-15year-olds; 3.1 % in16-19year-olds). Variousfactorsbesides age have been hypothesizedtoaccountfordifferences in children's reactionstodisaster. These include degreeofexposuretothe evene, race and gender4 ,premorbid mental health8 ,the abilityofthecommunitytooffersupport9 ,andwhetherornotchildren were separated from their parents3 Data regardingtherelevanceofthese factors for postdisaster


ChildrenofIniki Final Report Hamada, Roger S. -3 adjustment is only beginningtoaccrue, and someofthis evidence is contradictory. For example, Milne10foundthatchildrenwhowereevacuatedaftercyclone Tracy in Australia in1974andwhodid not returnweremore detrimentallyaffectedthan childrenwhodidnotevacuate orwhoevacuated and returned. These results are consistentwithan hypothesisthatevacueeswhodidnotaccess the community support availabletonon-evacuees fared worse. In contrast, Najarian11foundthatchildrenwhoexperienced the Armenian earthquakeof1988and remained in the earthquake zone manifested a higher incidenceofPTSDthandid childrenwhoexperienced the earthquake but relocatedoutofthe disaster zone. A numberoffactorsmaycontributetothese contrasting findings includingwhetherchildrenwereseparated from their parents andwhetherandwhatkindofpsychosocial interventions children participated inafterthe disaster. In their comprehensivereviewofpsychological interventionswithchildren after natural disaster, Vernberg and Vogel12conclude,"Solittleoutcomeresearch exists in the disaster literaturethatthere islittlesolid evidencetosupport'(or question) thedifferenttreatmentrecommendations in termsofefficacy,or eventodemonstratethatanyofthe interventions haveimportanteffectson child and adolescentadjustment."(p.496)Nevertheless, interventions are appliedtochildren and adolescents after disasters in anefforttomitigate postdisaster psychological morbidity. School-based interventions have included story-telling, art projects, role playing, group projects, coloring books, structured play activities, and "debriefing." 13-16


ChildrenofIniki Final Report Hamada, Roger S. -4 We are awareofonlytwointervention studieswhichhave compared treatedtountreated groupsofchildrenfollowingdisaster. Yule17comparedtwenty-four14-year-old girls receiving small group debriefing interventionswithgirlsnotreceiving the intervention. Both groups had witnessed, afewweeksearlier, the deathsoftwoseamen in a cruise ship accident, and experienced an ensuing chaotic evacuation.Atafivemonthfollowupassessment, Yule17foundthattreated girls had significantlylowerscores on theImpactofEvent Scale18and a fear survey than untreated girls,butdidnotdifferonmeasuresofanxiety and depression. In astudyofschool-based interventionbymental health professionals, Galante and Foa19provided seven one-hour group therapy sessions over the courseofa yeartofirstthroughfourthgraders in one Italian villagefollowingan earthquake. Following treatment, these children earnedlowerscores on a measureof"riskforantisocial or neurotic tendencies" than did children from a neighboring villagewhohad experienced the earthquakebuthadnotreceived intervention. The CurrentStudyOn September11,1992Hurricane Iniki, a category 4 storm,struckthe islandofKauai, Hawaii.Thehurricane carried sustainedwindsof145miles per hourwithgusts upto175mph. Seventy-one percentofhomes on the islandweredamaged or destroyed. Damagewasestimatedtobeclosetotwobillion dollars. The entirecommunitywasaffected. Research reported herewasdesignedtoaddress three questions. (1)Whatwasthe incidenceofPTSD symptoms among elementary-school-aged


ChildrenofIniki Final Report Hamada, RogerS.-5 children on KauaifollowingHurricane Iniki? Wasitcomparabletorates reported by Lonigan and his colleagues?Whatwasthe incidence amongfirstthroughfourthgraders, a groupnotstudiedbyLonigan? (2) Did incidenceofPTSDsymptomsdifferbetweenchildrenwhowereevacuated from the island postdisaster and non-evacuees?Ifyes, inwhatdirection? (3) Did the large-scale school-wide interventions implemented on Kauai intheweeksfollowingthe hurricane mitigate incidence and severityofPTSDsymptomsin childrenwhoparticipated in the interventions comparedtothosewhodid not?Itwashopedthatdata relevanttoquestions (2) and (3)mightcontributetodecisions regarding evacuation and psychological interventionsfollowingfuturenatural disasters. METHOD Subjects Four hundredfifty-twochildren from grades one throughsixcomprised the sample. All the children had experienced Hurricane Iniki on Kauai, the island overwhichthe eyeofthestorm had passed. The children ranged in agefromsixto12 yearsofage (Mean=8.9,S.D.=1.77).Withpermission from the Hawaii DepartmentofEducation(DOE)and the principalsofthe individual schools, data were obtained from students intwoschools on Kauaitowhichinvestigators had access (Koloa Elementary: N=236(52.2%ofsample) and Kapaa Elementary: N =181(40.0%ofsample)) and from20schools on the islandsofOahu and Hawaii


ChildrenofIniki Final Report Hamada, Roger S. -6(N=35(7.7%ofsample)).AtKoloa Elementary, all studentswerepotential participants.AtKapaa Elementary,twoclassrooms from eachofgrades 26, and one classroomfromgrade 1 volunteered participation. Potential participants on Oahu and Hawaiiwereidentified fromDOElistsofstudents from Kauai schoolswhohad transferredtooff-Kauai schoolsafterthe hurricane. A passive consent procedurewasutilized wherein letters describing thestudywere senttoparents under school principal and project principal investigator signatures. Parentswereaskedtoreturn a form included in the letterwithin10daysifthey didnotwanttheir childrentoparticipate in the study.Ifno formwasreturned,itwasassumedthatthe parent consentedtotheir child's participation. A totalof42parents,40from Kapaa Elementary,withheldconsent. Slightly over half the childrenwereboys(56.9%).Information onethnicity,asreportedbytheir classroom teachers,wasavailablefor440children. The primary groups represented were Caucasian(21.8%),Hawaiian or part-Hawaiian(20.7%),Filipino(19.8%),Japanese(8.9%),and a category comprised of childrenwithmixedethnicity(24.5%).The remaining19children comprised such small samplesoftheir ethnic groupsthattheywerenotincluded in analyses involving ethnicity. Other ethnicities reported were Other Asian (3), African American (2), Hispanic (6), and Other (8). In ordertoaddress the hypothesesofinterest, the452childrenweredivided into a groupwhohad been evacuatedoffKauai in thetwoweeksfollowingthe hurricane(N=35) and a groupwhoremained on Kauai(N=417).The sample


Chi'ldrenofIniki Final Report Hamada, Roger S. -7wasfurther divided into a groupthatreceived hurricane-related classroom interventions intheweekfollowingtheir returntoschool(N=399;88.5%)and a groupwhodidnot(N=52).Mostofthe evacuated children didnotreceive intervention(N=26).Information regarding interventionwasnotavailableforone evacuated child. Children in the no-intervention groupwhoremained on Kauai were,forthemostpart, absent on the days classroom interventions were conducted. Interventionswereunsystematic and varied in intensity, modality, and modeofdelivery. Possible interventions included allowing childrentodescribe andtalkabout their hurricane experiences in class, drawingwhathappened andhowthey felt, utilizing a hurricane coloring book, other art projects, singing about the disaster, researching andwritingabout hurricanes, and actingoutvarious vignettes about the hurricane experience. In one school (Koloa Elementary) the school counselor personally conducted the classroom exercises, in another (Kapaa Elementary) classroom teachers performed this function in consultationwithcounselors. On Oahu and Hawaii, counselorsmetindividuallywithchildren if, in their assessment,thechildren evinced emotional or behavioral difficulties. Measures In ordertomaximize comparability across studies,weused instruments and procedures in the current researchwhichwere similartoones usedbyLonigan and his colleagues4,5in their large-scale evaluationofchildrenfollowingHurricane Hugo. Therefore, in additiontoinformation regarding age, gender, grade, and teacher-


ChildrenofIniki Final Report Hamada, RogerS. -8 rated ethnicity, thefollowingtwomeasures were administered. Revised Children'sManifestAnxietyScale {RCMAS)20. The RCMAS is a37-itemtrue/false self-report measureoftraitanxiety.Itwasobtainedasan estimateofchildren's pre-hurricane (i.e., trait) levelsofanxiety. This instrument has been demonstratedtohave adequate psychometric properties andiswidelyused. Child Reaction Inventory (CRI). Children completed a child adaptationofthe selfreport versionofthe Frederick's Reaction Index6 Wordingofthe instrumentwasslightly alteredforthisstudytomakeitspecifictoHurricane Iniki andtofacilitate comprehensionbythe relatively young children assessed in thisstudy.The resultwasa 34-item scale designedtomeasure presence or absenceofsymptomsofPTSDasdefined in the DSM-III-R21. Each itemwasratedona five-point scale (i.e.,1=No,2=Alittleofthe time, 3=Someofthe time, 4=Muchofthe time, 5=Almostall the time). Frederick6reportedthatthe correlation between Reaction Index scores and diagnosed casesofPTSDwas0.91forchildren and0.95foradults. Using aninterviewversionoftheRI,pynoos7foundthatscoresofchildrenwhowerevictimsofor had witnessed violence were correlatedtotheir degreeofexposuretothe traumatic event. Lonigan4 Sadministered Reaction Index questionnairestoover5,600fifth-through eleventh-graderswhohad experienced Hurricane Hugo and foundthatself-reported PTSDsymptomswere relatedtolevelsoftraitanxiety, emotionalreactivityduring the hurricane, and exposuretothe hurricane and itseffects.In the currentstudy,theCRIshowedhigh overall internal consistency


ChildrenofIniki Final Report Hamada, Roger S. -9 (alpha=.848),butlowerinternal consistencyforitem clusters reflecting the Intrusion, Avoidance, and Arousal symptomsofthe DSM-III-R PTSD diagnosis (alpha=.728,.401,and.581,respectively). Procedure Questionnaireswereadministered approximately threetofourmonthsafter Hurricane Iniki during December1992and January1993.For children assessed on the islandofKauai, participating homeroom teachers administered the RCMAS and theCRItotheir entire classrooms. Teachers read each itemtotheir students, and the children marked their responses on sheetswhichincluded the questions. A visual analoguewasusedtoprovide childrenwitha concrete representationofthe five response categoriesforeach question on theCRI.The questionnaires from those studentswhoseparents had declined their participationwerelater removed and destroyed by researchstaffwithoutfurther examination. Since there were sofewsuch students,wefeltthatthiswayofhandling the datawouldbeless likelytosingle out and possibly stigmatize the nonparticipants in the class thanwouldexcluding them from taking the questionnaires. For children assessedoffKauai (the Evacuation group), researcherswenttothechildren'snewschools and administered questionnairestothe children in groupsofonetofourstudents. This procedural variationwasnecessaryforthe evacuated children becausewefeltthe small numberofsuch children in each off-Kauai school made classroom-wide administration untenable. Procedures were otherwise identicalforbothevacuated and nonevacuated groups.


ChildrenofIniki Final Report Hamada, RogerS.--10Children's responses on theCRIwere scored intwoways.Onewayfollowed the scoringofLoniganetaJ.4.5wherein individualCRIitems were identified for their correspondencetoDSM-III-R symptomsofPTSD.Twogroupsofsixitems each were judgedtocorrespondtoCriterionBand0 symptoms, and a groupofseven itemstoCriterion C symptoms. A symptom was judgedtobe presentifthe child endorsed"Muchofthetime"or"Almostall thetime."Asymptomwasjudgedtobeabsentifthe child endorsed"No,""Alittleofthetime,"or "Someofthetime."From this scoring and using thecutoffsdictated by DSM-III-R, children were classifiedasmanifesting a "self-reported posttraumatic-stress-disorder-like syndrome" ("self-reported PTSD") or not. Because structured clinical diagnostic interviews werenotused and because intensity and durationofsymptomsparameters were not obtained, PTSD diagnoses could notbemade.Analternative scoring method assigned valuesofone through fivetoresponsesof"No"through"Almostall thetime,"respectively. These values were then summed across the19items correspondingtoDSM-III-R symptomstogive a continuous posttrauma score or scoresforeach groupingofitems representing reexperiencing, avoidance, and arousal dimensionsofsymptoms. RESULTS Subjects Children in thetwoEvacuation groups didnotdiffersignificantly in grade (t=1.9,df=450,p>.06),totalrawscore on the RCMAS(t=1.62,df=450.


ChildrenofIniki Final Report Hamada, Roger S. --11P>.10), grade distribution (X2 =3.06,df=2,p>.21),or gender (X2 =1.06,df=1, P>.30).Thetwogroups differed in age (t=2.34,df=450,P<.05),withmean ageforevacuated children being slightly lower, andethnicity (X2 =11.49,df=4,P<.05),withthe evacuated group having proportionately more part-Hawaiian and mixedethnicitychildren, andfewerCaucasian and Japanese children (Table 1). Children in thetwoIntervention groups didnotdiffersignificantly in age (t=1.16,df=449,P>.24),grade (t=0.9,df=449,P>.36), total RCMASrawscore (t=0.25,df=449,p>.80),grade distribution (X2 =1.85,df=2, p>.39),gender (X2 =1.90,df=1, P>.16)or ethnicity (X2 =7.54,df=4, P>.10) (Table 1). Insert Table 1AboutHere Dichotomous Self-Reported PTSD Some children gave no responsetosome items on the Children's Reaction Index. For those childrenwithmissing dataforwhomsufficientinformationwasavailabletoclassify themashaving self-reported PTSDbythe criteria outlined above, they were included in the PTSD group. Other caseswithmissing data were included in the no-PTSD group.Itis possible, therefore,thatsome childrenwithmissing datawhowouldotherwise havemetPTSD criteriawereincluded in the no-PTSDgroup.


ChildrenofIniki Final Report Hamada, RogerS.--12Forty-threeof452children(9.5%)metcriteria for self-reported PTSD. This prevalence ratewasnotsignificantly relatedtogender (X2 =1.32,df=1,P>.25) orethnicity (X2 =2.23,df=4,P>.69). Rateofself-reported PTSDwassignificantly relatedtograde (X2 =13.49,df=2, p<.002).Post hoc 2x2 chisquare analyses revealedthatchildren in grades 5 and 6 experiencedlowerrates of self-reported PTSD than did children in either grades 1 and 2 or grades 3 and 4(3.3%vs.14.5%and12.8%,respectively). Thetwoyounger grade groups did notdiffersignificantly from each other. Correspondingly, childrenwithselfreported PTSD were significantly younger than childrenwithoutself-reported PTSD(8.00(s.d.=1.53)vs.9.09(s.d.=1.76),respectively; t=13.32,df=409,P=.0003).The self-reported PTSD group reported significantly moretraitanxiety on the RCMAS than did the non-PTSD group(13.47(s.d.=6.49)vs.11.43(s.d.=6.16),respectively; t=4.21,df=451,P<.05). Countertoexpectations, classroom intervention wasnotsignificantly relatedtoprevalenceofself-reported PTSD (X2 =0.34,df=1, P>.55)with9."0% (36/399)ofthe childrenwhoreceived and11.5%(6/52) of thosewhodid not receive classroom interventions reporting PTSD-like syndrome. Neitherwasevacuation significantly relatedtoself-reported PTSD (X2 =1.00,df=1,P>.32)with9.1 %(38/417)ofthe non-evacuees and14.3%(5/35)ofthe evacuees reporting PTSD-like syndrome. As a check, childrenwithmissing datawhowere classifiedasnon-PTSD weredroppedfrom the sample and the same analyses were recalculated. Because


ChildrenofIniki Final Report Hamada, Roger S. --13of the decreased sample size, rateofself-reported PTSD increasedto14.2%(43/303).The resultsofthe analyses reported abovewereotherwise unchanged. Continuous ScoringofSelf-Reported PTSD SymptomsTwo-by-twoanalysesofvariancewithIntervention (Yes/No) and Evacuation (Yes/No)asindependent variables were conducted on the total scoreforthe Children's Reaction Indexaswellason the scoresforeachofthe threeCRIsubscales. Becauseofmissing data, degreesoffreedom varied depending on the dependent measures being analyzed. Noneofthe maineffectsor interactions on anyCRIscores were statistically significant (for Intervention and Evacuation, respectively, Total CRI: F(1,299)=0.001,P>.97;F(1,299)=0.535,p>.46;Intrusion:F(1,329)=0.668,p>.41;F(1,329)=0.014,P>.90;Avoidance: F(1,413)=0.831,p>.36;F(1,413)=3.54,p=.06;Arousal: F(1,419)=0.300,p>.58;F(1,419)=0.138,P>.71).The maineffectfor EvacuationonAvoidance scoreswasnear significant(p=.06).Because thetwoevacuation groups differed in age,ananalysisofcovariancewithageasa covariate was performed in ordertostatistically control for the possible confoundingeffectofageonEvacuation. The ANCOVA revealedthatwhenagewastaken into account, the maineffectforEvacuation on Avoidance scoreswasnotsignificant(F(1,412)=1.484,P>.22).Mirroring the resultsofanalyses conducted on the dichotomous data, totalCRIscores were significantly correlatedwithage (r=-.25,p<.001;-.28,p<.001;-.32,p<.001;-.20,p<.001 for Total score and Intrusion, Avoidance,


ChildrenofIniki Final Report Hamada, RogerS. -14and Arousal subscales, respectively).ANOVA'sconducted onCRItotal and subscale scores revealedthatgrades differed significantly from each other on total and subscale scores (F(5,303)=4.67,P=.0004;F(5, 328)=5.79,p<.0001;F(5,412)=9.14,P<.0001;F(5,418)=3.77,P<.003forTotal, Intrusion, Avoidance, and Arousal, respectively). Post hoc comparisons using the Scheffetestat the.05significance level indicatedthatfor totalCRIand Intrusion subscale scores, grades one and three differed significantly from grade six. For the Avoidance subscale, grades one,two,and three differedfrom grade six, and gradetwoalso differed from grade 5 significantly. For the Arousal subscale, grade one differed significantly from gradesix(see Table 2). Insert Table 2 About Here Examinationofrelationships betweenCRIsymptom scores and demographic variables revealsthatbesides age, only Schoolwasa significant factor, and then onlyonthe Arousal subscale (F(2,421)=3.25,p<.04).ScheffetestindicatedthatKapaa Elementary students earned significantly higher Arousal scores than Koloa Elementary. Thetwoschools also differed in mean ageofstudents sampled (F(1,408)=4.88,P<.03)withstudentsatKapaa being, on average, older than those at Koloa. Therefore, ANCOVAwasusedtocompare Arousal subscale scores between Kapaa and Koloa schools while statistically controllingforage differences by using age as a covariate. Resultsofthis analysis revealthat,evenwithage


ChildrenofIniki Final Report Hamada, Roger S. --15differences accounted for, the schools differed significantly on theirstudents'Arousal subscale scores (F(1, 387)=7.99,p=.005;adjusted meansforKapaa:13.06,Koloa:11.74).DISCUSSION The overall prevalence rateforself-reported PTSD found in this sample,9.5%,is higher thanthatreported by Lonigan's4.5 group for Hurricane Hugo. The children assessed in Hawaii were younger than those studied by Lonigan. Whenwecompared the PTSD prevalence ratesforthe Hawaii and South Carolina childrenwhowere comparable in age (Le., 9-12-year-olds),wefound, in fact,thatHawaii'schildren had alowerrate of self-reported PTSD(6.0%in Hawaii vs.9.2%in South Carolina). This findingmayreflect the actual prevalence rates, oritmaybe associatedwithculture-specific response biases wherein Asian/Pacific Islander cultures underreport psychiatric symptoms.22Another difference between the current data and Lonigan'sstudyisthatno statistically significant relationships between gender or ethnic differences and PTSD prevalence or patterns of symptom were found in the Hawaii sample. However, our data replicated Lonigan's findingofa strong relationshipbetweenprevalence of PTSD and age,withyounger children reporting higher ratesofPTSD and higher intensityofsymptoms. The currentworkextended this finding downwardtoincludesixthrough eight-year-old children. Again, there is a question aboutwhetherthis is a valid finding orwhetheritis influencedbyresponse bias or


ChildrenofIniki Final Report Hamada, Roger S. --16limitationsofyounger childrentoadequately complete self-report instruments, evenwithteacher guidance. The resultsofcomparisonsbetweenevacuees and non-evacueesweresomewhatsurprising. We found no statistically significant differencesbetweenthese groups either on ratesofPTSD orwhenanalyzingCRIscores, although the data showed a trendforevacueestohave higher rates and scores. This trend suggeststhatevacuationafterHurricane Inikimayhave had a detrimentaleffecton children's postdisaster adjustment.Itis unclearwhetherthis could have been duetobeing deprivedofestablished social support in the homecommunity,separation from parents, ornotparticipating in the"honeymoon"recovery phase inthedays immediatelyfollowingthe hurricane. Similarly and contrarytoour expectation, no statistically significant differences in PTSD prevalence orCRIscores were foundbetweenchildrenwhoreceived school-based interventions and thosewhodidnot.Again, therewasa trend in the expected direction,butweexpected a much strongereffect.This lackofa finding certainly requires further exploration and replication.Itmayberelatedtoa numberoffactorsincludingbutnotlimitedto(1) perhaps the interventions employed indeed areineffectiveatalleviating postdisaster adjustment difficulties; (2) moresystematicallyplanned and implemented interventions employed by trained personnel (suchasthatusedbyLaGreca23and colleagues in FloridaafterHurricane Andrew)mayhave been moreeffective;(3) the comparison groups studied werenotrandomly assignedtotreatmentsothatsome preexisting


ChildrenofIniki Final Report Hamada, Roger S. --17differences in the groups may have compromised the data. For example, perhaps the childrenwhodidnotreceive intervention becausetheydidnotattend school after the hurricane hadfeweror milder adjustment problems than childrenwhodid attend school; and (4) perhaps the childrenwhodidnotparticipate in school-based intervention received help from community-based agencies and programs. ChildrenatKapaa Elementary earned significantly higher scores on theCRIArousal subscale only. This finding issomewhatpuzzling and, becauseitdoes not generalize across subscales, is probablynotrelatedtoimplementation or intervention differences. Rather, this differencemaybe tothe more extensive hurricane-related damage sufferedbyKapaa and its school than by Koloa. Further research should addresswhetherposthurricane adjustmentwasmediated by exposuretothe hurricane, orfamilyand social support-related variables; by receiving any community-based interventions; should obtain larger samples of evacuees and assesswhichof these were separated from their parents andwhichevacuatedwiththeir parents,aswellaswhetherthereweredifferences between evacueeswhoreturnedtothe disaster site vs. thosewhonever returned; and should evaluate theeffectsofmoresystematicpostdisaster interventions employing trained personnel using a structuredtreatmentprotocol.


ChildrenofIniki Final Report Hamada, Roger S. --18References 1. Green, B.L.(1993).Disasters and posttraumatic stress disorder. In J.R. Davidson&E.B.Foa(Eds.) Posttraumatic Stress disorder: DSM-IV and Beyond (pp.75-97).Washington, D.C.: American Psychiatric Press. 2. Vogel,J.M.,&Vernberg, E.M.(1993).Children's psychological responsestodisaster. JournalofClinical Child Psychology, 22,464-484.3. Aptekar, L.,&Boore,J.A.(1990).The emotionaleffectsofdisaster on children: Areviewofthe literature. International JournalofMental Health,19,7790. 4. Shannon, M.P., Lonigan, C.J., Finch,A.J.,&Taylor, C.M.(1994).Children exposedtodisaster: I. Epidemiologyofposttraumatic symptoms and diagnostic profiles. Journal of the American AcademyofChild and Adolescent Psychiatry, 33,80-93.5. Lonigan, C.J., Shannon, M.P., Taylor, C.M.,&Finch,A.J.(1994).Children exposedtodisaster: II. Risk factors for the developmentofposttraumatic symptomatology. Journal of the American AcademyofChild and Adolescent Psychiatry,33,94-105.6. Frederick, C.J.(1985).Children traumatized by catastrophic situations.InS.Eth&R.S. Pynoos (Eds.), Post-traumatic Stress Disorder in Children (pp.73-99).Washington, D.C.: American Psychiatric Press. 7. Pynoos, R.S., Frederick, C., Nader, K.,etal. (1987). Life threat and posttraumatic stress in school-age children. ArchivesofGeneral PsychiatrY, 44,1057-1063.8. Chamberlin, B.C.(1980).Mayoseminars in psychiatry: The psychological aftermathofdisaster. Journal of Clinical Psychiatry,41,238.9. Ziv,A.,&Israeli,R.(1973).Effectsofbombardment on the manifest anxiety levelofchildren living in kibbutzim. JournalofConsulting and Clinical Psychology,40,287-291.10. Milne, G.(1977).Cyclone Tracy: II. Theeffectson Darwin children. Australian Psychologist,12,55-62.


\ ChildrenofIniki Final Report Hamada, Roger S. -'19 11. Najarian, L., Pelcovitz, D., Goenjian, A.,&Najarian,B.(October22,1992).Post traumatic stress disorder (PTSD) in Armenia: Relocationafterthe earthquake. Paper presentedatthe annual meetingofthe American AcademyofChild and Adolescent Psychiatry, Washington, D.C. 12. Vernberg, E.M.,&Vogel,J.M.(1993).Interventionswithchildren after disasters. JournalofClinical Child Psychology,22,485-498.13. Ponton, L.E.(1989).Facts for Teachers: WorkingwithChildrenAfterthe Earthquake. Langley Porter Institute. 14. Corder, B.F.,&Haizlip, T. (no date). A Coloring BookAfterthe Hurricane for Children and Their Parents or Helpers. 15. Farberow, N.C.,&Gordon, N.S. (1986). Manual for Child Health Workers in Major Disasters. Washington, D.C.: National InstituteofMental Health. 16. Pynoos, R.S.,&Nader,K.(1988).Psychological first aid andtreatmentapproachtochildren exposedtocommunity violence: Research implications. JournalofTraumatic Stress,1,445-473.17. Yule, W.(1993).Technology-related disasters. In C.F. Saylor (Ed.), Children and Disasters (pp.105-122),NewYork: Plenum. 18. Horowitz,M.,Wilner, N.,&Alvarez, W.(1979).Impactofeventscale: A measureofsubjective stress. Psychosomatic Medicine,41,209-218.19. Galante, R.,&Foa,D.(1986).An epidemiological studyofpsychic trauma and treatment effectiveness for children after a natural disaster. Journalofthe American AcademyofChild and Adolescent Psychiatry,25,357-363.20. Reynolds, C.R.,&Richmond, B.D.(1985).Revised Children'sManifestAnxiety Scale (RCMAS). LA: Western Psychological Services. 21. American Psychiatric Association(1987).Diagnostic and Statistical ManualofMental Disorders (3rd ed., rev.). Washington, D.C.: Author. 22. Sue, S.,&Morishima,J.K.(1982).The Mental HealthofAsian Americans. San Francisco: Jossey Bass. 23. LaGrecaA.M.,Vernberg, E.M., Silverman,W.K.,Vogel,A.L.,&Prinstein,M.J.(no date). Helping children copewithnatural disasters: A manualforprofessionals workingwithelementary school children. Univ.ofMiami: Miami,FL.


ChildrenofIniki Final Report Hamada, RogerS.--20Table 1 Demographic characteristicsofIntervention x Evacuation groups I nterventionll Overaua Yes No Characteristic Evacuation Evacuation Intervention Evacuation No Yes No Yes No Yes No Yes Age Mean8.978. 8.988.26SO1.76 1.891.79 1.731.791.77 1.761.74Gender(%)Male58.35046.25048.1 58.157.648.6Female41.75053.85051.941.942.451.4Grade(%)112.337.511.519.215.412.8 12.222.92 14.1 015.423.119.213.814.1 17.1 315.312.53.815.49.615.314.614.3417.412.534.611.523.


ChildrenofIniki Final Report Hamada, RogerS.--21Table 1, cont. Intervention Overall Yes No Characteristic Evacuation Evacuation Intervention Evacuation No Yes No Yes No Yes No Yes Ethnicity(%)Caucasian 21.1 046.215.430.820.722.711.4Hawaiian20.337.511.530.821.220.719.831.4Filipino19.8026.919.223.119.420.214.3Japanese10.3000 0 10.19.60 Other Asian0.5003. African-0.500000.5 0.50 American Mixed23.762.515.40 23.124.523.240Other 2.1 00 0 0 2.1 2 0 Total RCMAS Mean11.5214.6311.0012.6211.8111.5811.47 13.26SD6.175.586.386.746.556.176.186.45 may notmatchbecause intervention status was unknownforone child.


ChildrenofIniki Final Report Hamada, Roger S. --22Table2Means and standard deviations onCRITotal, Intrusion. Avoidance. and Arousal Scores foreachgrade* GradeCRIScale1 2 3 4 5 6Total Mean43.5-40.844.2b39.640.4 34.rb SO11.99.911.212.011.710.4Intrusion Mean13.9a11.9b13.612.011.89.5a .bSO6.34.3 Mean16.6a17.3b c15.9d14.9 14.6c,3.0a .b dSO4. Mean13.5a12.812.812.9 12.310.6aSO4.34.1 letters indicate differences significantatthe.05level.

xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200253Ia 4500
controlfield tag 001 001985077
005 20090218143716.0
008 950124s1994 cou b s000 0 eng d
datafield ind1 8 ind2 024
subfield code a F57-00022
1 100
Hamada, Roger Saburo.
0 245
Children of Iniki :
b effects of evacuation and intervention : final report submitted to Natural Hazards Research and Applications Information Center /
by Roger S. Hamada.
[Boulder, Colo.] :
The Center,
22 p. ;
28 cm.
Quick response research report ;
v #70
Includes bibliographical references (p. 18-19).
Also issued online as part of a joint project with the Louis de la Parte Florida Mental Health Institute (FMHI) Research Library's disaster mental health initiative.
x Psychological aspects.
Hurricane Iniki, 1992
Psychological aspects.
2 710
University of Colorado, Boulder.
Natural Hazards Research and Applications Information Center.
t Natural Hazards Center Collection
H36 1994
4 856