Available online at www.sciencedirect.com

49 (2008) 393–398www.elsevier.com/locate/comppsych

Comprehensive Psychiatry

Are patients after severe injury who drop out of a longitudinal studyat high risk of mental disorder?

Daisuke Nishia,b,c,d,e, Yutaka Matsuokaa,b,c,e,⁎, Satomi Nakajimac,e,Hiroko Noguchib,c,e,f, Yoshiharu Kimc,e, Shigenobu Kanbad, Ulrich Schnyderg

aDepartment of Psychiatry, National Disaster Medical Center, Tokyo 190-0014, JapanbClinical Research Institute, National Disaster Medical Center, Tokyo 190-0014, Japan

cDepartment of Adult Mental Health, National Institute of Mental Health, National Center of Neurology and Psychiatry, Tokyo 187-8553 JapandDepartment of Neuropsychiatry, Graduate School of Medical Science, Kyushu University, Fukuoka 812-8581, Japan

eJapan Science and Technology Agency, CREST, Saitama 332-0012, JapanfDepartment of Human Science and Culture, Graduate School of Musashino University, Tokyo 202-8585, Japan

gDepartment of Psychiatry, University Hospital of Zurich, Zurich CH-8091, Switzerland

Abstract

In longitudinal studies of traumatic stress, it is particularly important to examine the data for any differences between those who drop outand those who continue to participate, because reluctance to participate might reflect symptoms of avoidance frequently seen in posttraumaticstress disorder (PTSD). However, whether those who drop out are at high risk of PTSD remains unclear. Over a 25-month period,188 consecutive patients with motor vehicle accident (MVA)–related injuries admitted emergently were enrolled and followed for 4 to6 weeks. Baseline characteristics were compared between subjects who did and did not participate in the follow-up study. At 4 to 6 weeks,66 (35.1%) of the participants had dropped out. Bivariate analyses revealed that those who dropped out were likely to be men, alcoholdrinkers, smokers, and unconscious just after MVA and to have fewer years of education, less severe injuries, less posttraumatic symptoms,and lower cooperativeness as assessed by the Temperament and Character Inventory. Logistic regression analysis revealed that male sex,unconsciousness during MVA, low cooperativeness, and less severe injuries were significant predictors of dropout. The literature says thatmale sex and unconsciousness just after MVA might be protective factors against MVA-related PTSD, whereas low cooperativeness is a riskfactor for general mental problems. To summarize, it is expected that those who drop from the follow-up are unlikely to have MVA-relatedPTSD, but might have mental problems independent of injury.© 2008 Elsevier Inc. All rights reserved.

1. Introduction

Longitudinal studies of posttraumatic stress provide theoptimal method for exploring the course of posttraumaticstress disorder (PTSD) and examining its predictors. Recentlongitudinal studies of PTSD have included many differentgroups of subjects, such as exposed disaster workers [1],refugees [2], a community sample [3], survivors of terroristattacks [4], soldiers hospitalized after serious combat injury

⁎ Corresponding author. Department of Adult Mental Health, NationalInstitute of Mental Health, National Center of Neurology and Psychiatry,Kodaira, Tokyo 187-8553, Japan. Tel.: +81 42 346 1975; fax: +81 42 3461986.

E-mail address: yutaka@ncnp.go.jp (Y. Matsuoka).

0010-440X/$ – see front matter © 2008 Elsevier Inc. All rights reserved.doi:10.1016/j.comppsych.2008.02.003

[5], patients with myocardial infarction [6], and victims ofsexual and physical assault [7]. Above all, longitudinalstudies of patients after severe injury, especially thoseinvolved in motor vehicle accidents (MVAs), comprise oneof the largest number of population samples studied [8-20].

Any longitudinal study risks being compromised byselective dropout. In longitudinal studies of traumatic stress,it is particularly important to examine the data for anydifferences between those who drop out and those whocontinue to participate, because reluctance to participate mayreflect symptoms of avoidance frequently seen in PTSD.Studies of severe injury have, to date, identified manydeterminants of dropout [9-11,15,16,19,20], but with partlycontradictory results; some suggested that those who dropout are at low risk of injury-related PTSD, whereas others

Table 1Dropout rates of longitudinal cohort studies after severe injury (mainly MVA)

Author (year) No. of subjects Follow-up Assessment Dropout rate (%) Difference Notes

Mayou et al [8] 200 3 month Self-report 13 No mention12 month 14.5

Blanchard et al [9] 158 12 month Structured interview 16.5 Ethnic minority (nonwhite) Subjects sought careEhlers et al [10] 1181 3 month Self-report 24.8 Men, younger, without

bone injuries, etcOnly 26% admitted

12 month 33.9Shalev et al [11] 270 1 month Structured interview 21.9 Fewer PTSD symptoms 85.8% MVA

4 month 21.9Ursano et al [12] 122 6 month Structured interview 18.9 No mention

12 month 29.5Mayou et al [13] 1148 3 month Self-report 24.7 No mention

12 month 32.7Schnyder et [14] 121 12 month Structured interview 12.4 None 60.4% MVAMason et al [15] 210 6 week Self-report 39.0 Unmarried, shorter hospital

stay, psychoticism17.6% MVA

6 month 45.718 month 54.3

O'Donnell et al [16] 363 3 month Structured interview 7.2 Younger, shorter hospital stay,higher levels of anxietyand depression

74 MVA12 month 15.4

Vaiva et al [17] 130 6 week Structured interview 16.9 No mentionHamanaka et al [18] 100 6 month Structured interview 18.0 No mention Telephone interviewWu and Cheung [19] 596 1 month Self-report 28.9 Men, a driver or a passenger,

having a history oftraumatic experience

3 month 43.16 month 54.6

Schnyder et al [20] 323 6 month Structured interview 21.1 Lower sense of coherence 31% MVA

394 D. Nishi et al. / Comprehensive Psychiatry 49 (2008) 393–398

suggest that they are at high risk. Longitudinal studies ofsevere injury that involved more than 100 subjects recruitedfrom emergency departments are shown in Table 1. On theone hand, Ehlers et al [10] and Wu and Cheung [19] showedthat men were likely to drop out, whereas it is wellknown that women are at high risk of PTSD [21]. Also,Shalev et al [11] showed that those who dropped out hadfewer PTSD symptoms. On the other hand, O'Donnell et al[16] showed that those who dropped out had higher levels ofanxiety and depression, and Mason et al [15] showed thatthey tend toward psychoticism. This apparent contradictionrequires elucidation.

To this end, we used the Tachikawa Cohort of MVAs(TCOM) study, a prospective longitudinal 3-year follow-upstudy, to investigate the incidence, prediction, protection,and natural course of psychiatric illness in a sample ofpatients severely injured in MVAs [16a]. The aims of thepresent study were to identify variables predicting patients atrisk of dropping out of the TCOM study and to examinewhether or not those who drop out are at high risk of injury-related PTSD.

2. Methods

Participants were selected from the TCOM study, whichwas approved by the institutional review board and ethicscommittee of the National Disaster Medical Center (NDMC),Tokyo. The inclusion criteria were (1) MVA-related severephysical injury causing a life-threatening or critical condition,

(2) consecutive admittance to the emergency department'sintensive care unit at NDMC, (3) age between 18 and69 years, and (4) a native Japanese–speaking ability. Theexclusion criteria were (1) diffuse axonal injuries, braincontusion, or subdural or subarachnoidal bleeding detectedby computed tomography and/or magnetic resonance ima-ging; (2) cognitive impairment, defined as a score of less than24 on the Mini-Mental State Examination [22,23]; (3)ongoing treatment of any neuropsychiatric disorder, such asschizophrenia, bipolar disorders, substance use disorders, orepilepsy, before the MVA; (4) marked clinical signs orsymptoms such as suicidal ideation, self-harm behavior, or asevere psychosomatic condition that may prevent the patientfrom tolerating the interview; and (5) living or working at alocation more than 40 km from the NDMC.

Participants in the present study were recruited betweenMay 31, 2004, and June 22, 2006. During this period, 213 ofthe 560 newly admitted patients with MVA injuries met thestudy criteria and were asked to participate in the study.After a complete description of the study, written informedconsent was obtained from 188 (88.3%) patients. The25 patients who refused to participate did not differsignificantly from those who participated in sex, age, injuryseverity score (ISS) [24], Glasgow Coma Scale, or situationsurrounding injury in the MVA. The mean number of daysbetween the accident and initial assessment was 3.9 days(SD = 4.3 days). Initial follow-up assessment was performedan average of 40.4 days (SD = 7.9 days) postaccident.Dropout was defined as refusal to participate or no responseup to 6 weeks after the MVA.

Table 2Demographic characteristics of dropouts and participants who completedfollow-up assessment

Dropout(n = 66)

Participant(n = 122)

t df P

Mean (SD) Mean (SD)

Heart rate (beats/min) 84.5 (15.8) 83.5 (17.3) −0.37 186 .71Systolic bloodpressure (mm Hg)

140.4 (23.3) 136.4 (26.3) −1.03 186 .30

Diastolic bloodpressure (mm Hg)

78.1 (17.9) 76.9 (21.6) −0.38 185 .71

Dropout(n = 66)

Participant(n = 122)

P

Median(Range)

Median(Range)

Age 30.5 (18-64) 35.5 (18-69) .14ISS 5.0 (1-24) 9.0 (1-41) b.01GCS 15.0 (3-15) 15.0 (3-15) .74Education (years) 12.2 (7-16) 12.0 (6-25) .03Total IES-R score 14.0 (0-59) 18.0 (0-62) .04Total HADS score 7.0 (0-26) 9.0 (0-35) .13Total PDEQ score 6.0 (0-33) 7.0 (0-26) .63

Dropout(n = 66)

Participant(n = 122)

χ2 df P

n (%) n (%)

Sex (male) 57 (86) 88 (72) 4.92 1 .03Injury sustainedat MVA (driver)

48 (73) 78 (64) 1.50 1 .22

Married or havinga partner

25 (38) 52 (43) 0.40 1 .53

Employmentstatus (full-time)

45 (68) 73 (60) 1.28 1 .26

Alcohol (once aweek or more)

41 (62) 57 (47) 4.07 1 .04

Smoker 44 (67) 61 (50) 4.83 1 .03Exercise (once aweek or more)

16 (24) 35 (29) 0.43 1 .51

Prior MVA 43 (65) 68 (56) 1.57 1 .21Any physical problembefore MVA

20 (30) 47 (39) 1.26 1 .26

Sense of life-threatat MVA

21 (32) 35 (29) 0.20 1 .65

Retrograde amnesia 13 (20) 22 (18) 0.08 1 .78Unconsciousnessjust after MVA

45 (68) 65 (53) 3.92 1 .05

Loss of MVA memory 25 (38) 34 (28) 1.99 1 .16Self-reproach feeling 37 (56) 67 (55) 0.02 1 .88Pain 57 (86) 115 (94) 3.43 1 .06PS (0 or 1) 26 (39) 26 (21) 7.00 1 b.01

PS was defined by the Eastern Cooperative Oncology Group criteria.PS-0 indicates fully active, able to carry out all predisease performancewithout restriction. PS-1 indicates restricted in physically strenuous activitybut ambulatory and able to carry out work of a light or sedentary nature, forexample, light housework and office work. PS-2 indicates ambulatory andcapable of all self-care but unable to carry out any work activities up to andmore than 50% of waking hours. PS-3 indicates capable of only limited self-care, confined to bed or chair more than 50% of waking hours. PS-4indicates completely disabled, cannot carry out any self-care, and totallyconfined to bed or chair.GCS indicates Glasgow Coma Scale; HADS, Hospital Anxiety and De-pression Scale; PDEQ, Peritraumatic Dissociative Experience Questionnaire.

Table 3TCI score of dropouts and participants

Dropout(n = 35)

Participant(n = 97)

P

Median (Range) Median (Range)

Novelty seeking 20.0 (9-34) 20.0 (8-36) .56Harm avoidance 19.0 (8-31) 19.0 (4-32) .38Reward dependence 15.0 (4-21) 16.0 (3-24) .17Persistence 4.0 (1-8) 4.0 (1-8) .08Self-directedness 27.0 (12-39) 27.0 (7-41) .23Cooperativeness 27.0 (13-35) 29.0 (14-39) .04Self-transcendence 10.0 (3-25) 12.0 (3-27) .13

395D. Nishi et al. / Comprehensive Psychiatry 49 (2008) 393–398

Initial assessments were conducted after cessation ofintravenous narcotic administration. Repeated assessmentsof cognitive performance using the Mini-Mental StateExamination were carried out by trained research nurseswho were postgraduates in clinical psychology or trainedpsychiatrists. The assessments included general sociodemo-graphics, detailed information about theMVA, vital signs andphysical condition on admission, past medical and psychia-tric history, psychiatric family history, sense of the threat ofdeath during the MVA, unconsciousness just after the MVA,loss of memory regarding the MVA referring to Gil's study[25], retrograde amnesia, feelings of self-reproach, physicalpain, prior traumatic events, involvement in previous MVAs,consumption of alcohol before the MVA, and dailyperformance status (PS) as defined by the Eastern Coopera-tive Oncology Group [26]. Psychometric instrumentsincluded the Impact of Event Scale–Revised (IES-R)[27,28], the Hospital Anxiety and Depression Scale[29,30], the Peritraumatic Dissociative Experience Ques-tionnaire [31], and the Temperament and Character Inventory(TCI) [32,33]. All participants were given a gift voucherfor their participation after assessment (¥1000 for theinitial interview).

Table 4Multivariate model of predictors of dropout in a prospective study ofsurvivors after severe injury

Characteristics β SE Adjusted OR 95% CI P

Sex (male) 1.80 0.88 6.03 1.08-33.7 .04Unconsciousnessjust after MVA

1.43 0.53 4.16 1.47-11.7 b.01

Smoker 0.37 0.55 1.45 0.49-4.28 .51IES-R score −0.03 0.02 0.97 0.94-1.01 .15Cooperativeness −0.10 0.05 0.91 0.83-0.99 .04ISS −0.13 0.04 0.88 0.81-0.96 b.01Education (years) −0.13 0.09 0.88 0.73-1.05 .15Alcohol(once a weekor more)

−0.39 0.49 0.67 0.26-1.75 .42

Potential variables associated with dropout (P ≦ .05) were included in themultivariate analysis. However, performance status was significantlyassociated with ISS and was not included in the multivariate model.SE indicates standard error; OR, odds ratio; CI, confidence interval.

396 D. Nishi et al. / Comprehensive Psychiatry 49 (2008) 393–398

2.1. Statistical analyses

Intergroup comparisons of categorical, nonparametric,and continuous variables were examined using the χ2 test(or Fisher exact test), the Wilcoxon 2-sample test, and theStudent t test for unpaired data, respectively. In a logisticregression analysis, we estimated the adjusted relative risksfor dropout. All statistical analyses used 2-tailed tests. For allstatistical evaluations, P values of .05 or less wereconsidered significant. All data analyses were performedusing SPSS version 14.0J for Windows statistical software(SPSS Inc, Tokyo, Japan).

3. Results

Of the 188 patients enrolled, 66 (35.1%; 95% confidenceinterval, 28.6-42.2) had dropped out at the initial follow-up,15 refused to participate, 46 made no response to a telephonecall or letter, and 5 had moved to an unknown address.

Bivariate analyses of intergroup differences are presentedin Table 2. Because only 132 patients completed the TCI, theTCI analyses are separately presented in Table 3. Univariateanalyses revealed that those who dropped out were likely tobe men, alcohol drinkers, smokers, and unconscious duringMVA and to have fewer years of education, lower ISS scores,lower IES-R scores, lower PS scores, and lower coopera-tiveness as determined by the TCI.

The adjusted estimates from a multiple logistic regressionmodel are shown in Table 4. Because of missing data,regression analysis involved only 132 patients. Performancestatus was not included in the multivariate model because itwas significantly associated with ISS. Logistic regressionanalysis revealed that men, unconsciousness during MVA,low cooperativeness, and less severe injuries were significantpredictors of dropout.

4. Discussion

The attrition rate of the present study was relatively high.Recently, Scott et al [34,35] reported the effectiveness oftheir strategic approach focusing on frequent contact tominimize attrition in longitudinal studies. Because theprimary aims of the TCOM study was to investigate thenatural course of patients injured in MVAs, we contactedparticipants only when interviews were scheduled. Althoughmuch more frequent contact might reduce attrition, theattrition rate of the present study is comparable with those ofprevious studies (Table 1).

We found that male sex was an independent predictor ofdropout, a result consistent with that of previous studies[10,19]. No studies have shown that female sex is anindependent predictor. Because it is well known that womenare at high risk of PTSD [21], these results might indicatethat those who drop out are at relatively low risk of injury-related PTSD. It is unclear why men were more likely to dropout of the present study.

Experiencing unconsciousness during MVA was anindependent predictor of dropout. To our knowledge, noappropriate comparable data are available in the medicalliterature. Those patients who experienced unconsciousnessduring MVAwere unlikely to fulfill stressor criterion A-2 ofthe PTSD for Diagnostic and Statistical Manual of MentalDisorders, Fourth Edition, which defines the person'sresponse as involving intense fear, helplessness, or horror[36]. Furthermore, Gil et al [25] reported that subjectswithout memory of the traumatic event were less likely todevelop PTSD. Thus, those patients who drop out might wellbe at relatively low risk of injury-related PTSD, althoughloss of memory was not a significant predictor of dropout inthe present study. A study by Shalev et al [11] that showedfewer PTSD symptoms among those who dropped outcoincides with this view. However, O'Donnell et al [16]found higher levels of anxiety and depression among thosewho dropped out. The fact that the attrition rate of the studyof O'Donnell et al was one of the lowest among all studiesafter severe injury may make a difference to the results, butfurther research is required to confirm such a suggestion.

To our knowledge, this is the first report amongcomparable studies to identify low cooperativeness as apredictor of dropout and is an important finding. Cloninger,who developed the TCI, described uncooperative indivi-duals as socially intolerant, disinterested in other people,unhelpful, and revengeful [32]. He also pointed out that lowcooperativeness on the TCI was a substantial contributor tothe likelihood of concomitant personality disorder. Manyother studies have shown that low self-directedness and lowcooperativeness are core features of personality disorder[37-39] and, moreover, associated with immature defense[39], attention deficit hyperactivity disorder, and autismspectrum disorder such as Asperger syndrome [40,41],alexithymia [42], sex offenses [43], pathologic gambling[44], and depression [45]. Although the results of this studymight indicate that those who drop out are at relatively highrisk of personality disorder and other behavioral andemotional problems independent of MVA, we should becareful about such an interpretation because self-directed-ness was not a significant predictor of dropout in the presentstudy. However, previous studies showing dropout to beassociated with psychoticism [15] and a lower sense ofcoherence [20] seem to support this notion.

Finally, less severe injury was found to predict dropout.Injury severity was not associated with the development oflater psychopathology according to previous studies[13,46,47], and thus, it seems that those who drop out areneither at high risk nor at low risk of injury-related PTSD.This finding is similar to that of Mason et al [15] andO'Donnell et al [16], who reported shorter hospital stay as apredictor of dropout because those with less severe injuriestend to be discharged shortly after MVA. Participants withless severe injury might consider the MVA unimportant and,therefore, might tend to be indifferent to this kind of follow-up study.

397D. Nishi et al. / Comprehensive Psychiatry 49 (2008) 393–398

The present study has some limitations. First, the resultswere obtained from only one institution and were based on asmall number of participants. Therefore, the study samplemight not be representative of Japanese patients with severeinjuries. Second, we may not yet have sufficient evidence toposit that those who drop out are at high risk of personalitydisorder because few previous studies have revealed anassociation between low cooperativeness alone and person-ality disorder. Further studies are needed to confirm anassociation between low cooperativeness alone and person-ality disorder.

Overall, this study suggests that those who drop out oflongitudinal studies after severe injury seem to be atrelatively low risk of developing injury-related PTSD, butat relatively high risk of having mental problems indepen-dent of injury, such as impulsivity. These findings shouldbenefit researchers of longitudinal studies of patients aftersevere injury to develop strategies to improve dropout rateand thereby advance our knowledge of the impact of severeinjury on patients' lives.

Acknowledgments

The authors thank Doctors Henmi and Tomoyasu for theirgenerous support of this study; Miss Kawase, Sano,Hasegawa, and Takahashi for careful recruitment andcommunication with the participants; and Miss Akutsu andSuzuki for data management.

References

[1] Fullerton CS, Ursano RJ, Wang L. Acute stress disorder, posttrau-matic stress disorder, and depression in disaster or rescue workers.Am J Psychiatry 2004;161(8):1370-6.

[2] Sondergaard HP, Theorell T. Language acquisition in relation tocumulative posttraumatic stress disorder symptom load over time in asample of re-settled refugees. Psychother Psychosom 2004;73(5):320-3.

[3] Perkonigg A, Pfister H, Stein MB, Hofler M, Lieb R, Maercker A,et al. Longitudinal course of posttraumatic stress disorder andposttraumatic stress disorder symptoms in a community sample ofadolescents and young adults. Am J Psychiatry 2005;162(7):1320-7.

[4] Pollack MH, Simon NM, Fagiolini A, Pitman R, McNally RJ,Nierenberg AA, et al. Persistent posttraumatic stress disorderfollowing September 11 in patients with bipolar disorder. J ClinPsychiatry 2006;67(3):394-9.

[5] Grieger TA, Cozza SJ, Ursano RJ, Hoge C, Martinez PE, Engel CC,et al. Posttraumatic stress disorder and depression in battle-injuredsoldiers. Am J Psychiatry 2006;163(10):1777-83 quiz 1860.

[6] Ginzburg K, Solomon Z, Dekel R, Bleich A. Longitudinal study ofacute stress disorder, posttraumatic stress disorder and dissociationfollowing myocardial infarction. J Nerv Ment Dis 2006;194(12):945-50.

[7] Kaysen D, Simpson T, Dillworth T, Larimer ME, Gutner C, ResickPA. Alcohol problems and posttraumatic stress disorder in femalecrime victims. J Trauma Stress 2006;19(3):399-403.

[8] Mayou R, Bryant B, Duthie R. Psychiatric consequences of roadtraffic accidents. BMJ 1993;307(6905):647-51.

[9] Blanchard EB, Hickling EJ, Barton KA, Taylor AE, Loos WR, Jones-Alexander J. One-year prospective follow-up of motor vehicleaccident victims. Behav Res Ther 1996;34(10):775-86.

[10] Ehlers A, Mayou RA, Bryant B. Psychological predictors of chronicposttraumatic stress disorder after motor vehicle accidents. J AbnormPsychol 1998;107(3):508-19.

[11] Shalev AY, Freedman S, Peri T, Brandes D, Sahar T, Orr SP, et al.Prospective study of posttraumatic stress disorder and depressionfollowing trauma. Am J Psychiatry 1998;155(5):630-7.

[12] Ursano RJ, Fullerton CS, Epstein RS, Crowley B, Kao TC, Vance K,et al. Acute and chronic posttraumatic stress disorder in motor vehicleaccident victims. Am J Psychiatry 1999;156(4):589-95.

[13] Mayou R, Bryant B, Ehlers A. Prediction of psychological outcomesone year after a motor vehicle accident. Am J Psychiatry 2001;158(8):1231-8.

[14] Schnyder U, Moergeli H, Klaghofer R, Buddeberg C. Incidence andprediction of posttraumatic stress disorder symptoms in severelyinjured accident victims. Am J Psychiatry 2001;158(4):594-9.

[15] Mason S, Wardrope J, Turpin G, Rowlands A. The psychologicalburden of injury: an 18 month prospective cohort study. Emerg Med J2002;19(5):400-4.

[16] O'Donnell ML, Creamer M, Pattison P, Atkin C. Psychiatricmorbidity following injury. Am J Psychiatry 2004;161(3):507-14.

[16a] Matsouka Y, Nishi D, Nakajima S, Kim Y, Homma M, Otomo Y.Incidence and prediction of psychiatric morbidity after a motorvehicle accident in Japan: the Tachikawa Cohort of Motor VehicleAccident Study. Crit Care Med 2008;36(1):74-800.

[17] Vaiva G, Thomas P, Ducrocq F, Fontaine M, Boss V, Devos P, et al.Low posttrauma GABA plasma levels as a predictive factor in thedevelopment of acute posttraumatic stress disorder. Biol Psychiatry2004;55(3):250-4.

[18] Hamanaka S, Asukai N, Kamijo Y, Hatta K, Kishimoto J, Miyaoka H.Acute stress disorder and posttraumatic stress disorder symptomsamong patients severely injured in motor vehicle accidents in Japan.Gen Hosp Psychiatry 2006;28(3):234-41.

[19] Wu KK, Cheung MW. Posttraumatic stress after a motor vehicleaccident: a six-month follow-up study utilizing latent growthmodeling. J Trauma Stress 2006;19(6):923-36.

[20] Schnyder U, Wittmann L, Friedrich-Perez J, Hepp U, Moergeli H.PTSD following accidental injury: rule or exception in Switzerland?Psychother Psychosom 2008;77(2):111-8

[21] Kessler RC, Sonnega A, Bromet E, Hughes M, Nelson CB.Posttraumatic stress disorder in the National Comorbidity Survey.Arch Gen Psychiatry 1995;52(12):1048-60.

[22] Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” Apractical method for grading the cognitive state of patients for theclinician. J Psychiatr Res 1975;12(3):189-98.

[23] Mori E, Mitani Y, Yamadori A. Usefulness of a Japanese version ofthe Mini-Mental State Test in neurological patients. (in Japanese)Shinkeishinrigaku 1985;1:82-90.

[24] Baker SP, O'Neill B. The injury severity score: an update. J Trauma1976;16(11):882-5.

[25] Gil S, Caspi Y, Ben-Ari IZ, Koren D, Klein E. Does memory of atraumatic event increase the risk for posttraumatic stress disorder inpatients with traumatic brain injury? A prospective study. Am JPsychiatry 2005;162(5):963-9.

[26] Oken MM, Creech RH, Tormey DC, Horton J, Davis TE, McFaddenET, et al. Toxicity and response criteria of the Eastern CooperativeOncology Group. Am J Clin Oncol 1982;5(6):649-55.

[27] Weiss DS, Marmar CR. The Impact of Event Scale–Revised.New York: Guilford Press; 1997.

[28] Asukai N, Kato H, Kawamura N, Kim Y, Yamamoto K, Kishimoto J,et al. Reliability and validity of the Japanese-language version of theImpact of Event Scale–Revised (IES-R-J): four studies of differenttraumatic events. J Nerv Ment Dis 2002;190(3):175-82.

[29] Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale.Acta Psychiatr Scand 1983;67(6):361-70.

[30] Kugaya A, Akechi T, Okuyama T, Okamura H, Uchitomi Y.Screening for psychological distress in Japanese cancer patients.Jpn J Clin Oncol 1998;28(5):333-8.

398 D. Nishi et al. / Comprehensive Psychiatry 49 (2008) 393–398

[31] Marmar CR, Weiss DS, Schlenger WE, Fairbank JA, Jordan BK,Kulka RA, et al. Peritraumatic dissociation and posttraumaticstress in male Vietnam theater veterans. Am J Psychiatry 1994;151(6):902-7.

[32] Cloninger CR, Svrakic DM, Przybeck TR. A psychobiological modelof temperament and character. Arch Gen Psychiatry 1993;50(12):975-90.

[33] Kijima N, Tanaka E, Suzuki N, Higuchi H, Kitamura T. Reliabilityand validity of the Japanese version of the Temperament andCharacter Inventory. Psychol Rep 2000;86(3 Pt 1):1050-8.

[34] Scott CK. A replicable model for achieving over 90% follow-up ratesin longitudinal studies of substance abusers. Drug Alcohol Depend2004;74(1):21-36.

[35] Scott CK, Sonis J, Creamer M, Dennis ML. Maximizing follow-up inlongitudinal studies of traumatized populations. J Trauma Stress2006;19(6):757-69.

[36] Association AP. American Psychiatric Association diagnostic andstatistical manual of mental disorders, fourth edition, text revision;DSM-IV-TR. Washington (DC): American Psychiatric Association;2000.

[37] Svrakic DM, Whitehead C, Przybeck TR, Cloninger CR.Differential diagnosis of personality disorders by the seven-factor model of temperament and character. Arch Gen Psychiatry1993;50(12):991-9.

[38] Bayon C, Hill K, Svrakic DM, Przybeck TR, Cloninger CR.Dimensional assessment of personality in an out-patient sample:relations of the systems of Millon and Cloninger. J Psychiatr Res1996;30(5):341-52.

[39] Mulder RT, Joyce PR, Sullivan PF, Bulik CM, Carter FA. Therelationship among three models of personality psychopathology:

DSM-III-R personality disorder, TCI scores and DSQ defences.Psychol Med 1999;29(4):943-51.

[40] Soderstrom H, Rastam M, Gillberg C. Temperament andcharacter in adults with Asperger syndrome. Autism 2002;6(3):287-97.

[41] Anckarsater H, Stahlberg O, Larson T, Hakansson C, Jutblad SB,Niklasson L, et al. The impact of ADHD and autism spectrumdisorders on temperament, character, and personality development.Am J Psychiatry 2006;163(7):1239-44.

[42] Picardi A, Toni A, Caroppo E. Stability of alexithymia and itsrelationships with the ‘big five’ factors, temperament, character, andattachment style. Psychother Psychosom 2005;74(6):371-8.

[43] Giotakos O, Markianos M, Vaidakis N, Christodoulou GN. Sexhormones and biogenic amine turnover of sex offenders in relationto their temperament and character dimensions. Psychiatry Res 2004;127(3):185-93.

[44] Martinotti G, Andreoli S, Giametta E, Poli V, Bria P, Janiri L. Thedimensional assessment of personality in pathologic and socialgamblers: the role of novelty seeking and self-transcendence. ComprPsychiatry 2006;47(5):350-6.

[45] Hansenne M, Reggers J, Pinto E, Kjiri K, Ajamier A, Ansseau M.Temperament and Character Inventory (TCI) and depression.J Psychiatr Res 1999;33(1):31-6.

[46] Koren D, Arnon I, Klein E. Acute stress response and posttraumaticstress disorder in traffic accident victims: a one-year prospective,follow-up study. Am J Psychiatry 1999;156(3):367-73.

[47] Schnyder U, Moergeli H, Trentz O, Klaghofer R, Buddeberg C.Prediction of psychiatric morbidity in severely injured accidentvictims at one-year follow-up. Am J Respir Crit Care Med 2001;164(4):653-6.