Citation analysis of meta-analysis articles on posttraumatic stress disorder

1088Chin MedJ 2011;124(7): 1088-1093Meta analysisCitation analysis of meta-analysis articles on posttraumatic stressdisorderLIAO Xi-ming and CHEN Ping-yanKeywords: meta analysis; postraumatic stress disorder; bibliometric analysisBackground In the past two decades enormously scientific researches on postraumatic stress disorder (PTSD) havebeen undertaken and many related meta- analyses have been published. Citation analysis was used to getcomprehensive perspectives of meta-analysis articles (MA articles) on PTSD for the purpose of facilitating theresearchers, physicians and policy-makers to understand the PTSD.Methods MA articles on PTSD in any languages from January 1980 to March 2009 were included if they presentedmeta-analytical methods and received at least one citation recorded in the Web of Science (WoS). Whereas studies, inwhich any effect sizes of PTSD were not distinguished from other psychological disorders, were excluded. Citations toand by identified MA articles were documented basing on records in WoS. Citation analysis was used to examinedistribution patterns of characteristics and citation impact of MA articles on PTSD. Canonical analysis was used toexplore the relationship between the characterstics of MA articles and citation impact.Results Thirty-four MA aricles published during 1998 and 2008 were identified and revealed multiple study topics onPTSD: 10 (29.49%) were about epidemiology, 13 (38.2%) about treatment or intervention, 6 (17.6%) aboutpathophysiology or neurophysiology or neuroendocrine, 3 (8.8%) about childhood and 2 (5.9%) about psychosocialadversity. Two articles cited most frequently with 456 and 145 counts were published in Jouma/ of Consulting and ClinicalPsychology by Brewin (2000) and Psychological Bulletin by Ozer (2003), respectively. Mean cited count was 7.48+10.56and mean age (year 2009 minus article publication year) was (4.24+2.91) years. They had been cited approximately by67 disciplines and by authors from 42 countries or trritories. Characteristics of meta -analysis highly correlated withcitation impact and reflected by canonical correlation of 0.899 (P <0.000 01).Conclusions The age of MA articles predicted their citatin impact. Citation analysis would serve to capture the globalperspectives and topics of MA articles on PTSD.Chin Med J 2011;124(7):1088-1093Meta-analysis (MA) is a stistical technique forpolicy-makers by getting the whole pictures of existingamalgamating, summarizing, and reviewingscientific evidence on one specific field. Web of Scienceprevious quantitative researches on one topic to examine(WoS) enables the users to search curent and pasta particular question; it plays an important role inliteratures from approximately 8500 of the mostevidence-based medicine.' There are a wealth of primaryprestigious, high-impact research jourmals in the world. Itstudies from hybrid fields of medicine that have beenalso provides a unique search method as well as citedexplored through meta- analysis: 2 medical science,reference search. The advent of WoS benefits thepreventive medicine and clinical medicine,' includingapplication of citation analysis.diabetes, cardiovascular disease’ and critical carepractice." By collecting and reviewing a serial of MAPosttraumatic stress disorder (PTSD) is a form of anxietyarticles on a particular field from the perspective ofrelated to trauma exposure which is the cause of severebibliometrics, a panoramic evolution and development ofpsychological, psychosocial, biological and physicalthat field could be better understood. Citation analysis,consequences and symptoms. There is recently athe examination of the frequency, patterms and graphs ofincreasing awareness of the harmful effects of PTSD;citations in articles and books, uses citations in academicworks to establish links to other works or other research.DO: 0.760/cmij.s.036-6999.20 1.07.025Department of Biostatistics, School of Publie Health and TropicalIt has been utilized as a tool to track the progress ofMedicine, Southern Medical University, Guangzhou, Guangdongresearch in the medical community.' In an increasingly510515, China (Liao XM and Chen PY)competitive research environment, aspects of citationCorres中国煤化工Tropicep MedicineT -1g-yan, Department ofanalysis have been advocated as simple proxy, objectiveBiostalmeasures to appraise the quality of joumals, publishedSouthe一u, Guangdong 510515,ChinaHarticles, research institutions and even individualchenpy99@ 126.com)CN M H G2-2061604832. Emal:researcher. Therefore using citation analysis on MATThis study was supported by the grants form the National Naturalarticles would facilitate the traditional concept ofScience Foundation of China (No. 30972554) and the Guangdongcitation analysis for researchers, physicians and Natural Science Foundation (No, 915180200400001).Chinese Medical Jourmal 2011;124(7;:1088 10931089including impaired verbal memory, behavioral, structuralimpact was explored with CCA. The correlation wasbrain abnormalities and sleep abnormalities. A called the canonical correlation and the linearburgeoning of original articles and MA articles havecombinations were the canonical variates. Thecontributed to the investigation of the basal cardiovascularfundamental principle behind CCA was the creation of aactivity and neurobiological pathophysiology of PTSDnumber of canonical variables, each consisting of a linearduring the last two decades. T0 The current project utilizedcombination of one set of variables (X), which had thecitation analysis to delve into the evolution andform: U1=ajX+...+a.+..+apXp and a lineardevelopment of MA articles on PTSD from 1980 to 2009.combination of the other set of variables (Yn), which hadPhenomena that arose from our investigation, along withthe form: V1=byY/+... +b;Y+...+bpYp. The goal was tothe limitations and implication for researchers aredetermine the coefficients, or canonical weights (ap anddiscussed.b1p), that maximize the correlation between the canonicalvariables U and Y, the first canonical correlation. CorrMETHODS(U, Y) was the highest possible correlation between anylinear combination of the variables in the MAs' charactersSearch strategy for identification of MA articles onset and any linear combination of the variables in thePTSDMAs citation impact set. Further pairs of maximallyMA articles on PTSD covering any languages from 1980correlated linear combinations were chosen in turm, and(the year that PTSD formally entered the psychiatrichey were orthogonal to those already identified.nosology) to March 2009 (the current study terminated)Structure correlation coefficients, also called canonicalwere selected by screening the WoS applying the searchloadings, were used to interpret the importance of eachterms of meta-analysis or systematic review, plus PTSDoriginal variable in the canonical variates. Variablesor postraumatic stress disorder, in the title, abstract orhighly correlated with a canonical variate should bekeyword. Potentially relevant MA articles were initiallyconsidered more important when deriving a meaningfulidentifed by title, then by abstract and full text.interpretation of the related canonical variate. Statisticalanalysis was performed with SPSS 17.0 for windows.Inclusion criteriaTwo-tailed P value less than 0.05 was consideredStudies were included if they included meta-analyticalstatistically significant.methods, presented the use of statistical methodology toRESULTScombine the results of primary studies on PTSD,presented the effect sizes calculating PTSD outcomeThe literature search rendered 119 articles that could bemeasure related variables of interest, and received at leastpotentially included in the analysis. Among them, 53 MAone citation recorded in WoS during the defined studyaricles contained a form of meta-analysis on PTSD. Wefound 34 from these 53 MA articles cited at least once inWoS within the defined study period and were included inExclusion criteria1245citation analysis (see References list)The mostMA articles on psychological disorders where any effectfrequently cited MA article was“Meta -analysis of risksizes of PTSD were not distinguished from otherfactor for posttraumatic stress disorder in trauma-exposedpsychological disorders were excluded.adults" in the Jourmal of Consulting and Clinical's with456 counts. The top six cited MA articles had high citesMeasureover 100 ranging from 104 to 145 counts (Table 1).Two sets of variables, either documented or calculatedbased on WoS were considered. First, the citationDescriptive on citation impact and characteristicscharacteristics for describing MA articles, including cited(Table 2)age (year 2009 minus article publication year) of MAThirty-four identified MA articles were published in 23aricles, cited subject areas, average references, and citedjournals. The American Journal of Psychiatry, Britishcountries/territories. Second, the citation impact forJourmnal of Psychiatry, International Clinicaldescribing the articles that cited the MA articles,Psychopharmacology and Jourmnal of Traumatic Stressincluding cited times, ratio of cited to citing, averagecombined to publish 12 MA articles, three in each.citations per year, times cited before year two (TC2)."Among the 34 MA articles; 10 were about epidemiology,citing subject areas, and citing counties/erritories.3 about treatment or intervention, six aboutpathophysiology or neurophysiology or neuroendo-Statistical analysiscrinology, three_ about childhood, and two aboutCanonical correlation analysis was used to explore thepsychc中国煤化工relationship between the two sets of variables asdescribed above. Canonical correlation analysis (CCA) isCanonHC N M H Gracteristics anda multidimensional exploratory method that can highlightcitation impact (Table 3)correlations between the two sets of variables. In theThe first canonical correlation was 0.899 and wascurrent study the relation between variables describingstatistically significant (x*=89.623, P <0.000 01), .MAs characters and variables describing MAs citationsuggesting the two sets of variables were highly1090Chin Med J 2011;124(7): 1088-1093Table 1. Top six cited MA articlesFirst authorTile。Published yearTotal ciationBrewin"6Meta analysis of risk factors for postraumatic stress disorder in trauma- exposed adults2000)zer"Predictors of posttraumatic stress disorder and symptoms in adults: a meta-analysis2003Van Etten*Comparative eficacy of treatments for post-traumatic stress disorder: a meta-analysis998143radley'00511HuNeuroimaging findings in post-traumatic stress disorder - systematic review200van Emmerik41Single ssion debreing after psychological trauma: a meta-analysis200210Table 2. Descriptive on variables of citation impact andDISCUSSIONcharacteristicsQuartilePublication variablesMean MedianRangeThe idea underlying the exploration of publicationCitation impact(Q1-03)patterms of MA articles on PTSD is that, the influence ofImpact factor6.045.74.4 0.6 24.22.6 6.8the ideas contained in an academic publication is relatedTolal ciations51.9+92.7 12.5 1.0 456.0 3.0-59.0 .to the number of scholars who read the publication andReference counts74.0+32.4 65.0 30 -15346.0 -98.5use the ideas. An influential idea is one that stimulatesCitations/references0.7+1.004.20-0.9thoughts and discussions by others.t One way to assessAVC7.5+10.63.0.4 45.6 1.0 -9.0TCOthe academic influence of a publication is therefore to30.2+51.61.0-240.0 3.0-36.3count the number of times the publication has been cited15.7+14.712.51.0-67.05.8 _23.0Country or trritory10.7+10.3 7.0 1.0 42.01.8 -3.0in subsequent publications by other authors.47CharacteristicsAge of MA aricles(year) 4.2.93.0 1.0-11.02.0- -6.0Characteristics highly correlated with citation impactCited subject1.4+0.61.1.0-3.01.0- 3.0Citing subject areasReferences/MA articles2.2+1.00.9-4.51.4 2.9The analysis of subject areas where citations to MAAffliaions2.7+2.02.1.0-10.0 1.8 -3.0articles on PTSD were found identified a number ofOriginal studies2675+9853364 1276 4170 2586 -3759Self-ciation rate (%)10.6+18.82.50-75.00-16.7citing disciplines. We found that the 34 MA articles wereStatus of authorship3.5+1.6included in the reference lists from dozens of disciplines.AVC: average citation per year. TC2: times cited ill year of 2007.New findings from the specific area which are assimilatedwithin the wide ranges of scientific communities wouldTable 3. Canonical structures of the first pair of canonical variatesbe the impetus on advanced development for both partiesCiuation impactwith particular perspectives provided by various subjectsVariablesL oading VariablesLoading Cross loading- Loadingsparking the novel idea. To understand PTSD, one shouldAge of MA articles-0.005Muli-disciplinarity-0.156 Ciution counts 0.6320.556consult not only the most relevant subject, psychology,ReierencesMA aricles 0.256 Citedleiting 0.6390.572but also general and intermnal medicine， public,Afliations-0.3390.5150.462environmental and occupational health, family studies,Self-citation rate-0.342TC20.5850.524and social work as well.Status of aulborship-0.519 Cing subject 0.7930.710CGL0.7120.638Citing country or territoryAVC: average citation per year. TC2: times cited ill year of 2007. CGL: cingThe extent of research into PTSD is intermational orauthor geological locatin. Fist pair RC:=0.899 (x2=89.623. Pc0.0001).global in scope. Citations of MA articles from the mostcorrelated. The first pair of canonical variates showed: U,cited counties or territories were the USA, England,variables of characteristics, mainly represented age ofNetherlands, Germany, Canada and Australia. Thecited MA articles and status of authorship (Table 4) andcitation pattem for this group of countries was similar forV，variables of citation impact, mainly representedother categories of medical researches such as generalcitingsubject area, citing countyterritory， ratio ofand internal medicine, clinical psychology and psychiatry.cited/citing. Findings of cross loadings (column 4 ofThe USA dominated with the largest proportion ofTable 3) examining whether U1 would predict V1 were ascitations.follow: 0.710 for citing subject area, 0.638 for citingauthor geological location (CGL), 0.95 for age, whichBridging gap between countries in studies on PTSDindicated the age of MA articles played a major influenceA striking gap in the published papers on PTSD betweenon the citation effect of MA articles defined by thedeveloped and developing countries may representnumber of citing subject areas and citing country/territory.countries ripe for new researches on PTSD. The presentIn summary, characteristics of MA articles would predictstudy indicated that, although the research field of PTSDciting. subject area, citing country/territory， ratio ofis gaining popularity with the scientific community with acited/citing and TC2.growing intemational acceptance of the PTSD diagnostic中国煤化工Table 4. Distribution in age of MA articles acrosYHCNMH(1998Total200820072006 ._2005Age (years,Frequency (n)5634Percentage (9%)14.717.620.65.,9，00Cumularive percentage (%)32.4。52.994.1Chinese Medical Joumal 201:;1247):1088 1093091category," the concentration of publication in a fewevents need to be factored. Their role in peri-traumaticcountries reflects the infancy of PTSD research.dissociation during or very shortly after the event, alongBroadening the focus to include other countries besideswith pre-morbid vulnerabilities or their interaction withthe most economically developed countries such as USAthe effects of singular and/or cumulative trauma exposureand European countries, may enhance the understandingawait further research.of PTSD. Exploring crOS8-culural factors and thedifference of epidemiology of PTSD between rich andPathophysiology, neurophysiology or neuroendocrine ofpoor countries may better predict the trajectory of PTSD.PTSDThe increasing awareness of the harmfulness of PTSDA broad range of topics about studies on PTSDhas impelled researchers to a better understanding of thePTSD epidemiologyneurobiology of PTSD through neuroimaging; such asMA articles contain the core ideas of studies of PTSDmagnetic resonance techniques, computed tomographyform the basic background for the PTSD research areastudies, magnetic imaging, functional magnetic resonanceand cover some key study topics in the epidemiology ofimaging (MRI),"' magnetic resonance spectroscopyPTSD. These include etiology, effects of intervention,(MRS), positron emission tomography (PET), and singlediagnosis and prognosis, screening instruments, riskphoton emission computed tomography (SPECT)."factors detection, prevalence or incidence of PTSD afterFindings from neuroimaging studies showed changes inspecific traumatic event or in specific population, andthe bilateral hippocal, hippocampal and amygdalaso indetection strategies of high risk population. According tosuffers with PTSD and suggested that after psychologicalthe most referenced MA articles,' ”.. the type of traumatrauma, biological changes were not restricted toexposure, clinical features, course, and response todysregulation of neurochemical systems,' . ”but includetreatment were highly confirmed as independent riskaltered brain structure. Understanding the neurobiologyfactors. The heterogeneous nature and diversity ofof PTSD is important for development of new treatmentssymptoms inherent in PTSD inspires researchers to delvefor this disorder.into the enigma of PTSD from many perspectives and totry and find treatments for improving the health ofTreatment or intervention on PTSDvictims suffering from this disorder.Knowledge about causality and the psychologicalprocesses that account for PTSD boost its visibility andIdentifying the groups that resolved PTSD spontaneously,acceptance within the medical communities, the generalthose where it persisted long term, and those with morepublic, and among article-citing investigators. This willcomplicatedlapsing/relapsing pattem would behelp in the design of more efficient treatment strategies ofinvaluable in guiding the implementation of effectivePTSD and more efficient interventional management.interventionstrategies. Socalaneconomicconsequences associated with short term and long termThirteen of 34 MA articles reviewed the effects of manycourse of PTSD and the course and profile oftreatments and interventions including psychotherapy'traumatic-related symptomsshould be examined.and pharmacological treatment for PTSD. PsychotherapyFurthermore, the process of contagion may be a usefulmodalities included eye movement desensitization andconcept in understanding the symptomatic course andreprocessing, psychodynamic psychotherapy and groupprofile of symptoms; such a process may maintainCBT, relaxation techniques, and social skills training.s2symptoms that might have otherwise remitted in theEvaluations on pharmacological treatment for PTSD,333absence of mutually reinforcing responses. Futuresuch as antipsychotics, anticonvulsants, adrenergic-researches need to highlight the specific factors whichinhibiting agents, opioid antagonists, benzodiazepinesmay play a role in the course and profile of directand other agents are also popular with physicians.specific-traumatic event exposure, and in indirectexposure as well as cumulative trauma exposure on PTSDPTSD in children and adolescentssymptomatology.Gleaning the thematic information from these 34 MAarticles, only three of them qualitatively reviewedLongitudinal studies on PTSDchildren and adolescents with PTSD. Resorting to theIt should be noted that it remains unclear, due to theirelectronic database for medicine found a handful ofcorrelative nature, whether risk factors reflect a original articles and case studies about PTSD in childrenvulnerability to or a result of PTSD, or both. Hence,and adolescents. Future work in the field calls forfundamental theoretical trajectory of PTSD with morededication from researchers and practitioners tcstringent design, such as longitudinal cohort studies andsynth中国煤化工，about children andlarge samples of individuals, is required to study riskadole: :ft such a delicatefactors for PTSD that might also explain the relationpopulMYHCNMHGbetween psychosocial adversity and PTSD. Peri- traumaticdissociation and posttraumatic stress," delayed-onsetThe current study was subject to limitation. For futurePTSD，personal characteristics and medical andstudies, literature retrieval of MA articles should bepsychological bistory prior to the potentially traumatizingidentified using a variety of strategies. Conclusions1092Chin MedJ 2011;124(7): 1088-1093exclusively based on published studies, therefore, can bebasal cardiovascular activity in postraumatic stress disorder.misleading. In order to minimize the multiple bias inPsychosom Med 2001; 63: 585-594.publication, citation, language, location and reference,10. Figueira I, Luz Md, Braga RJ, Cabizuca M, Coutinho E,relevant articles were also located by searching theMendlowicz MV. The increasing intemnationalization ofreferences of the located articles. WoS data weremainstream postraumatic stress disorder research: aincomplete for many types of cited material, includingbibliometric study. J Trauma Stress 2007; 20: 89-95.legal citations and websites. Without these citations, the11. Chuang KY, Huang YL, Ho YS. A bibliometric and citationresults would have been skewed towards joumal articlesanalysis of stroke-related research in Taiwan. Scientometricsand books and away from gray literatures sources. Using2007; 72: 201-212.automated data sources for bibliometrical analysis,12. Andrews B. Brewin CR, Philpott R, Stewart L. Delayed-onsetparicularly in gray literature- dependent fields, shouldposttraumatic stress disorder: a systematic review of thethus be viewed with some skepticisms or caution.evidence. Am J Psychiat 2007; 164: 1319-1326.13. Bisson J, Ehlers A, Matthews R, Plling s, Richards D,In conclusion, 34 MA articles published during 1998 andTumer S. Psychological treatments for chronic post-traumatic2008 were identified and revealed multiple study topicsstress disorder-systematic review and meta-analysis. Br Jon PTSD: epidemiology, treatment or intervention,Psychiatry 2007; 190: 97-104.pathophysiology, neurophysiology or neuroendocrine,14. Bradley R, Greene J, Russ E, Dutra L, Westen D. Achildhood, and psychosocial adversity. Characteristics ofmultidimensional meta- analysis of psychotherapy for PTSD.MA articles highly associated with citation impact,Am J Psychiat 2005; 162: 214-227.especially the relationship among age, citing subject areas15. Brewin CR. Systematic review of screening instruments forand citing country or territory were considered. Theadults at risk of PTSD. J Trauma Stress 2005; 18: 53-62.current study elucidates how to use citation analysis to16. Brewin CR, Andrews B, Valentine JD. Meta- analysis of riskexamine dimensional topics of MA articles on PTSD,factors for postraumatic stress disorder in trauma-exposedwhich could help to capture a more global perspective ofadults. J Consult Clin Psychol 2000; 68: 748-766.the findings from PTSD investigations. It is hoped that17. Bruce M. A systematic and conceptual review ofthis study would give the readers a better knowledge ofposttraumatic stress in childhood cancer urvivors and theirPTSD and inspire others to continue biliometricparents. Clin Psychol Rev 2006; 26: 233-256.researches into PTSD. It also opens an avenue to18. Davydow DS, Gifford JM， Desai SV, Needham DM,exploring findings in other scientific domains." CitationBienvenu OJ. Posttraumatic stress disorder in generalanalysis can help capture a global perspective by MAintensive care unit survivors: a systematic review. Gen Hosparticles on PTSD.Psychiatry 2008; 30: 421-434.19. Erkin A, Wager TD. Functional neuroimaging of anxiety: aREFERENCESmeta-analysis of emotional processing in PTSD, socialanxiety disorder, and specific phobia. Am J Psychiat 2007;1. Sauerland S, Seiler CM. Role of systematic reviews and164: 1476-1488.meta-analysis in evidence-based medicine. World J Surg 2005;20. Fazel M, Wheeler J, Danesh J. Prevalence of serious mental29: 582-587. .disorder in 7000 refugees resettled in westem countries: a2Gerber s, Tallonn D, Trelle s, Schneider M, Jini P, Egger M.systematic review. Lancet 2005; 365: 1309-1314.Bibliographic study showed improving methodology of21. Frewen PA， Dozois DJA, Neufeld RWJ, Lanius RA.meta-analyses published in leading joumnals 1993- -2002. JMeta-analysis of alexithymia in posttraumatic stress disorder.Clin Epidemiol 2007; 60: 773-780.J Trauma Suress 2008; 21: 243-246.Shou ZF. Efficacy and safety of induction therapy with22. Fu SS, McFall M, Saxon AJ, Beckham JC, Carmody TP,alemtuzumab in kidney transplantation: a meta-analysis. ChinBaker DG, et al. Post-traumatic stress disorder and smoking:Med J 2009; 122: 1692-1698.a systematic review. Nicotine Tob Res 2007; 9: 1071-1084.Zhang XM, Wang HY, Luo YY, Ji LN. HLA-DQ, Dr alele23. Criffths J, Fortune G Bartber V, Young J. The prevalence ofpolymorphism of type I diabetes in the Chinese population: apost traumatic stress disorder in survivors of ICU treatment: ameta- analyis. Chin Med J 2009; 122: 980-986.systematic review. Intensive Care Med 2007; 33: 1506-1518.5. Cleophas TJ, Zwinderman AH. Meta analysis. Circulation24. Hull AM. Neuroimaging findings in post-traumatic stess2007; 115: 2870-2875.disorder: systematic review. Br J Psychiatry 2002; 181:6Moran J, Solomon P, Warm D. Methodology in meta-analysis:102-110.a study from critical care meta-analytic practice. Health Serv25. Ipser J, Seedat S， Stein DJ. Pharmacotherapy forOutcomes Res Methodol 2004; 5: 207-226.post-traumatic stress disorder: a systematic review and7. Hasbrouck LM, Taliano J, Hirshon JM, Dannenberg AL. Usemeta-analvsis S Afr Mad 1 206 06: 1088-1096.of epidemiology in clinical medical publicaions, 1983-1999:26中国煤化工:1 D, Rohleder N, Wernera citation analysis. Am J Epidemiol 2003; 157: 399-408.1 abnormalities in PTSD.8Yue W, Wilson CS. Measuring the citation impact of researchYH. CN MH G6a1031joumals in clinical neurology: a structural equation modelling27. Kitayama N, Vaccarino V, Kutner M, Weiss P, Bremner JD.analysis. Scientometrics 2004; 60: 317-332.Magnetic resonance imaging (MRI) measurement of9Buckley TC, Kaloupek DG. A meta analytic examination ofhippocampal volume in postraumatic stress disorder: aChinese Medical Joumal 2011;124(7):1088 10931093meta- analysis. J Affect Disord 2005; 88: 79-86.psychological rauma: a meta-analysis. Lancet 2002; 360:28. Meewisse ML, Reitsma JB, De Vries GJ, Gersons BPR, Olff766-771.M. Cortisol and post-traumatic stress disorder in adults:42. Van Etten ML, Taylor S. Comparative efficacy of teatmentssystematic review and meta- analysis. Br J Psychiatry 2007;for post-raumatic stress disorder: a meta-analysis. Clin191: 387-392.Psychol Psychother 1998; 5: 126-144.29. Mitte K, Steil R, Nachtigall C. The effects of a brief43. van Liempt S, Vermetten E, Geuze E, Westenberg HGintervention after acute traumatization in preventingPharmacotherapy for disordered sleep in post traumatic stresspostraumatic symptomatology: a meta-analysis using thedisorder: a systematic review. Int Clin Psychopharmacolrandom effects model. Z Klin Psychol Psychiatr Psychother2006; 21: 193-202.2005; 34: 1-9.Wethington HR, Hahn RA, Fuqua-Whitley DS, Sipe TA,30. Neria Y, Nandi A, Galea s. Post traumatic stress disorderCrosby AE, Johnson RL, et al. The effectiveness ofafter disasters: a systematic review. Psychol Med 2008; 38:interventions to reduce psychological harm from traumatic467- 480.events among children and adolescents: a systematic review.31. Orth U, Wieland E. Anger, hostilit, and postrumatic stressAm J Prev Med 2008; 35: 287-313.disorder in trauma-exposed adults: a meta- analysis. J Consult45. Wolfe DA, Crooks CV, Lee V, McIntyre Smith A, Jaffe PGClin Psychol 2006; 74: 698-706.The efects of children's exposure to domestic violence: a32. Ozer EJ, Best SR, Lipsey TL, Weiss DS. Predictors ofmeta-analysis and critique. Clin Child Fam Psychol Rev 2003;postraumatic stress disorder and symptoms in adults: a6: 171-187.meta-analysis. Psychol Bull 2003; 129: 52-73.46. Fan JC, McGhee CN. Citation analysis of the most influential3. Pae CU, Lim HK, Peindl K, Ajwani N, Seretti A, Patkar AA,authors and ophthalmology jourmals in the field of cataractet al. The atypical antipsychotics olanzapine and risperidoneand cormeal refractive surgery 2000 2004. Clin Experimentin the treatment of posttraumatic stress disorder: AOphthalmol 2008; 36: 54-61.meta-analysis of randomized, double-blind, placebo-47. Costanza R, Stem DI, Fisher BP, He L, Ma C. Influentialcontrolled clinical trials. Int Clin Psychopharmacol 2008; 23:publications in ecological economics: a citation analysis. Ecol1-8.Econ 2004; 50: 261-292.34. Paolucci EO, Genuis ML, Violato C. A meta- analysis of the48. Fava GA, Guidia J, Soninob N. How citation analysis canpublished research on the effects of child sexual abuse. Jmonitor the progress of research in clinical medicine,Psychol 2001; 135: 17-36.Psychother Psychosom 2004; 73: 31-333.335. Pole N. Moderators of PTSD-related psychophysiological49. Lensvel-Mulders G van der Hart O, van Ochten JM, van Soneffect sizes: results from a meta- analysis. Ann N Y Acad SciMJM, Steele K, Breeman L. Relations among peritraumatic2006; 1071: 422-424.dissociation and posttraumatic stress: a meta-analysis. Clin36. Sack M, Lempa W, Lamprecht F. Study quality andPsychol Rev 2008; 28: 1138-1150. .effect-sizes: a meta-analysis of emdr-treatment for50. Woon FL, Hedges DW. Amygdala volume in adults withposttraumatic stress disorder. Psychother Psychosom Medposttraumatic stress disorder: a meta-analysis. JPsychol 2001; 51: 350-355.Neuropsychiatry Clin Neurosci 2009;21: 5-12.37. Sherman JI. Effects of psychotherapeutic treatments for ptsd:51. Roberts NP, Kitchiner N, Kenardy J, Bisson JI. Systematica meta-analysis of controlled clinical trials. J Trauma Stressreview and meta-analysis of multiple-session early1998; 11: 413-435.interventions following traumatic events. Am J Psychiatry38. Smith ME. Bilateral hippocampal volume reduction in adults2009; 166: 293-301.with post-traumatic stress disorder: a meta-analysis 052. Mendes DD, Mello MF, Ventura P, Passarela CM, MariJJ. Astructural mri studies. Hippocampus 2005; 15: 798- 807.systematic review on the effectiveness of cognitive39. Stapleton AB, Lating J, Kirkhart M, Everly GS. Effects ofbehavioral therapy for posttraumatic stress disorder. Int Jmedical crisis intervention on anxiety, depression, andPsychiatry Med 2008; 38: 241-259.postraumatic stress symptoms: a meta-analysis. Psychiatr Q53. Milic NM, Parapid BJ, Ostoji MC, Nedeljkovic MA,2006; 77: 231-238.Marinkovic JM. Meta-analysis of stent thrombosis after40. Stein DJ, Seedat s, van der Linden GJH, Zungu-Dirwayi N.drug-eluting stent implantation: 4-year follow-up. Chin MedSelective serotonin reuptake inhibitors in the treatment ofJ 2010; 123: 3689-3696.post- uraumatic stress disorder: a meta-analysis of randomizedcontolld trials. Int Clin Psychopharmacol 2000; 15: 31-39.41. van Emmerik AA， Kamphuis JH， Hulsbosch AM,(Received December 17, 2010)Emmelkamp PM. Single session debriefng afterEdited by SUN Jing中国煤化工MYHCNMHG