<%BANNER%>

UFIR



Predictors of occurence and severity of first time low back pain
CITATION PDF VIEWER
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/IR00001280/00001
 Material Information
Title: Predictors of occurence and severity of first time low back pain
Physical Description: Journal Article
Creator: George, Steven
Childs, John D.
Teyhen, Deydre S.
Wu, Samuel S.
Wright, Alison C.
Dugan, Jessica L.
Robinson, Michael E.
Publisher: PLoS ONE
Publication Date: February 15, 2012
 Subjects
Genre:
Spatial Coverage:
 Notes
Abstract: Primary prevention studies suggest that additional research on identifying risk factors predictive of low back pain (LBP) is necessary before additional interventions can be developed. In the current study we assembled a large military cohort that was initially free of LBP and followed over 2 years. The purposes of this study were to identify baseline variables from demographic, socioeconomic, general health, and psychological domains that were predictive of a) occurrence; b) time; and c) severity for first episode of self-reported LBP. Baseline and outcome measures were collected via web-based surveillance system or phone to capture monthly information over 2 years. The assembled cohort consisted of 1230 Soldiers who provided self-report data with 518 (42.1%) reporting at least one episode of LBP over 2 years. Multivariate logistic regression analysis indicated that gender, active duty status, mental and physical health scores were significant predictors of LBP. Cox regression revealed that the time to first episode of LBP was significantly shorter for Soldiers that were female, active duty, reported previous injury, and had increased BMI. Multivariate linear regression analysis investigated severity of the first episode by identifying baseline predictors of pain intensity, disability, and psychological distress. Education level and physical fitness were consistent predictors of pain intensity, while gender, smoking status, and previous injury status were predictors of disability. Gender, smoking status, physical health scores, and beliefs of back pain were consistent predictors of psychological distress. These results provide additional data to confirm the multi-factorial nature of LBP and suggest future preventative interventions focus on multi-modal approaches that target modifiable risk factors specific to the population of interest.
Acquisition: Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by Steven George.
Publication Status: Published
Funding: The POLM trial was supported by the Peer-Review Medical Research Program of the Department of Defense (PR054098). All authors were independent from this funding program. Publication of this article was funded in part by the University of Florida Open-Access Publishing Fund. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
 Record Information
Source Institution: University of Florida Institutional Repository
Holding Location: University of Florida
Rights Management: All rights reserved by the submitter.
Resource Identifier: doi - doi:10.1371/journal.pone.0030597
System ID: IR00001280:00001

Downloads

This item is only available as the following downloads:

POLM_LBP_Predictors_journal.pone.0030597 ( PDF )


Full Text

PAGE 1

PredictorsofOccurrenceandSeverityofFirstTimeLow BackPainEpisodes:FindingsfromaMilitaryInception CohortStevenZ.George1* ,JohnD.Childs2,DeydreS.Teyhen2,SamuelS.Wu3,AlisonC.Wright2,JessicaL. Dugan2,MichaelE.Robinson41 DepartmentofPhysicalTherapyandCenterforPainResearchandBehavioralHealth,UniversityofFlorida,Gainesville,Florida,UnitedStatesofAm erica, 2 U.S.ArmyBaylorUniversityDoctoralPrograminPhysicalTherapy(MCCS-HGE-PT),ArmyMedicalDepartmentCenterandSchool,FortSamHouston,Texas,UnitedS tatesofAmerica, 3 DepartmentofBiostatistics,UniversityofFlorida,Gainesville,Florida,UnitedStatesofAmerica, 4 DepartmentofClinicalandHealthPsychologyandCenterforPain ResearchandBehavioralHealth,UniversityofFlorida,Gainesville,Florida,UnitedStatesofAmericaAbstractPrimarypreventionstudiessuggestthatadditionalresearchonidentifyingriskfactorspredictiveoflowbackpain(LBP)is necessarybeforeadditionalinterventionscanbedeveloped.Inthecurrentstudyweassembledalargemilitarycohortthat wasinitiallyfreeofLBPandfollowedover2years.Thepurposesofthisstudyweretoidentifybaselinevariablesfrom demographic,socioeconomic,generalhealth,andpsychologicaldomainsthatwerepredictiveofa)occurrence;b)time;and c)severityforfirstepisodeofself-reportedLBP.Baselineandoutcomemeasureswerecollectedviaweb-basedsurveillance systemorphonetocapturemonthlyinformationover2years.Theassembledcohortconsistedof1230Soldierswho providedself-reportdatawith518(42.1%)reportingatleastoneepisodeofLBPover2years.Multivariatelogisticregression analysisindicatedthatgender,activedutystatus,mentalandphysicalhealthscoresweresignificantpredictorsofLBP.Cox regressionrevealedthatthetimetofirstepisodeofLBPwassignificantlyshorterforSoldiersthatwerefemale,activeduty, reportedpreviousinjury,andhadincreasedBMI.Multivariatelinearregressionanalysisinvestigatedseverityofthefirst episodebyidentifyingbaselinepredictorsofpainintensity,disability,andpsychologicaldistress.Educationleveland physicalfitnesswereconsistentpredictorsofpainintensity,whilegender,smokingstatus,andpreviousinjurystatuswere predictorsofdisability.Gender,smokingstatus,physicalhealthscores,andbeliefsofbackpainwereconsistentpredictorsof psychologicaldistress.Theseresultsprovideadditionaldatatoconfirmthemulti-factorialnatureofLBPandsuggestfuture preventativeinterventionsfocusonmulti-modalapproachesthattargetmodifiableriskfactorsspecifictothepopulationof interest.Citation: GeorgeSZ,ChildsJD,TeyhenDS,WuSS,WrightAC,etal.(2012)PredictorsofOccurrenceandSeverityofFirstTimeLowBackPainEpisodes:Findings fromaMilitaryInceptionCohort.PLoSONE7(2):e30597.doi:10.1371/journal.pone.0030597 Editor: JosH.Verbeek,FinnishInstituteofOccupationalHealth,Finland Received August24,2011; Accepted December23,2011; Published February15,2012 Copyright: 2012Georgeetal.Thisisanopen-accessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense,whichpermits unrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited. Funding: ThePOLMtrialwassupportedbythePeer-ReviewMedicalResearchProgramoftheDepartmentofDefense(PR054098).Allauthorswereindependent fromthisfundingprogram.PublicationofthisarticlewasfundedinpartbytheUniversityofFloridaOpen-AccessPublishingFund.Thefundershadno rolein studydesign,datacollectionandanalysis,decisiontopublish,orpreparationofthemanuscript. CompetingInterests: Theauthorshavedeclaredthatnocompetinginterestsexist. *E-mail:szgeorge@phhp.ufl.eduIntroductionIngeneralpopulationslowbackpain(LBP)isthemostprevalent formofchronicmusculoskeletalpain[1]oftenleadingtodisability [2,3].Inmilitarypopulationsmusculoskeletalpainhasanadverse effectbyfrequentlycausingmedicalevacuation[4]andLBPin particularacommonreasonforlongtermdisability[5].Asaresult ofitsnegativeimpactpreventionofLBPhasremainedaresearch priorityforbothgeneral[6]andmilitarypopulations[4,7]. FactorsinvolvedinthetransitionfromacutetochronicLBP havebeenarecentfocusofdisabilitypreventionresearch.Suchan approachisconsistentwithsecondaryprevention[8],andstudies inthisareahaveprovidedimportantinformationoneffective managementofacuteLBP.Secondarypreventionstudieshave highlightedpsychologicalinfluenceonthedevelopmentofchronic LBP[9]andidentifiedpatientsubgroupsthathavelarger treatmenteffectswhenmatchedtreatmentisapplied[10,11]. Thefocusonsecondarypreventionhasbeenproductivein reducingdisabilityfromacuteepisodesofLBP,butthereremains thepotentialofprimarypreventionforlimitingthenegative impactofLBP. Thegoalinprimarypreventionistoreducetheoverallnumber ofLBPepisodesexperiencedbyapopulation[12].Incontrastto secondaryprevention,primarypreventionattemptstoreduce thosethattransitionfromapainfreestatetooneofexperiencing LBP.Backschools,lumbarsupports,andergonomicinterventions haveallbeenstudiedforprimarypreventionofLBP,butwith limitedsuccess[13,14,15].Theseprimarypreventionstudies suggestthatmoreworkneedstobecompletedindetermining whatfactorsarepredictiveofdevelopingLBPbeforeadditional preventativeinterventionscanbedeveloped.Forexample,if modifiablefactorsareidentifiedasbeingpredictiveofLBPthen theymayprovidelogicaltreatmenttargetsforfutureLBP preventiontrials[6]. Thepurposeofthispaperwastoreportpredictorsoffirsttime LBPepisodesself-reportedduring2yearsofmilitaryduty.This PLoSONE|www.plosone.org1February2012|Volume7|Issue2|e30597

PAGE 2

purposeisconsistentwithprimarypreventionprioritieshighlightedintheliterature[6,14].TobeststudythedevelopmentofLBPit isnecessarytorecruitagroupofhealthysubjectsandfollowthese subjectsuntilsomedevelopLBP.Furthermore,developmentof LBPisbelievedtobemulti-factorialinnaturesoconsiderationofa rangeofpotentialpredictorsiswarranted.Inthecurrentstudywe assembledalargemilitarycohortthatwasinitiallyfreeofLBPand includedpotentialpredictorsfromdemographic,socioeconomic, generalhealth,andpsychologicaldomains.Theprimarypurposes ofthisstudyweretoidentifyvariablesfromthesedomainsthat werepredictiveofa)occurrenceandb)timetofirstepisodeof LBP.Oursecondarypurposesweretoidentifyvariablesthatwere predictiveofa)higherpainintensity;b)disability;orc) psychologicaldistressduringthefirstLBPepisode.Methods EthicsStatementTheinstitutionalreviewboardsattheBrookeArmyMedical Center(FortSamHouston,Texas)andtheUniversityofFlorida (Gainesville,FL)grantedethicalapprovalforthisproject.All subjectsprovidedwritteninformedconsentpriortotheir participation.OverviewThisstudywaspartofthePreventionofLowBackPaininthe Military(POLM)clusterrandomizedtrial[16].ThePOLMtrial hasbeenregisteredatClinicalTrials.gov(http://clinicaltrials.gov) underNCT00373009. TheprimaryaimofthePOLMtrialwastodetermineifcorestabilizationexerciseandpsychosocialeducationresultedin decreasedLBPincidenceduring2yearsofmilitaryduty.POLM trialresultsindicatedthatpsychosocialeducationwasfoundtobe preventativeofseekinghealthcareforLBP[17].Thisstudyreports onasecondaryaimofthePOLMtrialwhichwastodetermine whatfactorswerepredictiveofself-reportedLBPforSoldiersthat respondedtoaweb-basedorphonesurveytools.SubjectsConsecutivesubjectsenteringatrainingprogramatFortSam Houston,TXtobecomeacombatmedicintheU.S.Armywere consideredforparticipationfromFebruary2007toMarch2008. ResearchstaffatFortSamHouston,Texasintroducedthestudyto individualcompaniesofSoldiersandscreenedpotentiallyeligible Soldiers. Subjectswererequiredtobe18–35yearsofage(or17yearold emancipatedminor),participatingintrainingtobecomeacombat medic,andbeabletospeakandreadEnglish.Subjectswitha priorhistoryofLBPwereexcluded.Inthisstudyapriorhistoryof LBPwasoperationallydefinedasapreviousepisodeofLBPthat limitedworkorphysicalactivity,lastedlongerthan48hours,and causedthesubjecttoseekhealthcare.Subjectswerealsoexcluded iftheywerecurrentlyseekingmedicalcareforLBP;unableto participateinunitexerciseduetoothermusculoskeletalinjury; hadahistoryoflowerextremityfracture(stressortraumatic);were pregnant;oriftheyhadtransferredfromanothertraininggroup. OtherpossibleexclusionsincludedSoldierswhowerebeing acceleratedintoacompanyalreadyrandomizedorSoldierswho werebeingre-assignedtoadifferentoccupationalspecialty.ExerciseandEducationProgramsCompaniesofSoldierswererandomlyassignedtoexerciseand/ oreducationprogramsaspartoftheclusterrandomizedtrial[16]. Theassignedexerciseandeducationprogramsarenotafocusof thiscurrentpaper,butarebrieflyreviewedasweincludedtheseas predictivevariablesinourstatisticalanalyses.Allexercise programswereperformedinagroupsettingunderthedirect supervisionoftheirdrillinstructorsaspartofdailyunitphysical training.Thetraditionalexerciseprogram(TEP)wasselected fromcommonlyperformedexercisesthattargettherectus abdominusandobliqueabdominalmuscles.Thecorestabilization exerciseprogram(CSEP)wasselectedfromexercisesthattarget deepertrunkmusclesthatattachtothespine;suchasthe transversusabdominus,multifidus,andtheerectorspinae.The TEPandCSEParedescribedinmoredetailinpreviousPOLM publications[18,19]. Thebriefpsychosocialeducationprogram(PSEP)involved attendanceat1sessionduringthefirstweekoftraining.The sessioninvolvedaninteractivelectureledbystudypersonnel (ACW,JLD)lastingapproximately45minutes.Thelecture consistedofavisualpresentationfollowedbyaquestionand answersession.ThePSEPprovidedSoldierscurrent,evidence basedinformationonLBPsuchasstressingthatanatomicalcauses ofLBParenotlikelyandencouragingactivecopinginresponseto LBP.EducationalmaterialwasalsoprovidedtotheSoldiersby issuingeach TheBackBook ashasbeendoneinotherLBPtrials [20,21,22].ThePSEPisdescribedinmoredetailinaprevious publication[23].MeasuresBaselinemeasureswerecollectedundersupervisionofstudy personnel.Outcomemeasureswerecollectedviaweb-based surveillancesystemorphonetocapturemonthlyinformation over2years.Baselinemeasures.Soldierscompletedstandardquestionnairestoassessvariablesconsistentwithdemographicand socioeconomicdomains.Theinformationcollectedincludedsuch variablesasage,sex,education,incomelevel,smokinghistory, previousactivitylevel,previousinjury,physicalfitnessscores,and militarystatus.Soldiersalsocompletedself-reportmeasuresto assessgeneralhealthandpsychologicaldomains.TheMedical OutcomesSurvey12-ItemShort-FormHealthSurveywasusedas aself-reportofhealthstatusforphysical(SF-12PCS)andmental function(SF-12MCS)[24].TheBackBeliefsQuestionnaire (BBQ)wasusedtoquantifybeliefsaboutLBPrelatedto managementandoutcome[25].TheState-TraitAnxiety Questionnaire(STAI)[26]andBeckDepressionInventory(BDI) [27,28,29]wereusedtomeasurenegativeaffectfromgeneralized anxietyandgeneralizeddepression,respectively.Finally,9items fromtheFearofPainQuestionnaire(FPQ-III)wereusedto measurefearaboutspecificsituationsthatnormallyproducepain [30,31,32].Outcomemeasures.Soldiersweretrainedinacomputerlab onhowtousethePOLMweb-basedoutcomesurveillancesystem andallassessmentswereprovidedthroughasecureweb-sitethat protectedSoldierconfidentiality.Accesstothesystemwas promptedbyanemailwhichwassenttotheSoldier’sofficial militaryemailaddressonthe1stofeachmonth.Additionalemails weresentonthe3rdofthemonth,andagainonthe7thofthe monthiftheSoldierstillhadnotresponded.Soldierswerequeried whethertheyhadexperiencedanyLBPinthelastcalendarmonth byemail,andthisinformationwasusedtodeterminetheinitial episodeofLBPaftercompletingtraining.Soldiersnotresponsive tomultipleemailrequestswerecontactedbyphoneattheendof 12and24monthstodetermineifLBPhadoccurredinthepast year.ThoseSoldiersrespondingtothephoneinterviewwere includedwiththeemailsurveyresultsbecausethestructureofthe phoneinterviewwasparalleltotheemailsurvey.SoldiersPredictorsofFirstTimeLowBackPain PLoSONE|www.plosone.org2February2012|Volume7|Issue2|e30597

PAGE 3

successfullycontactedbyphonecompletedthesameinformation astheywouldhaveifusingtheweb-basedsystem.Theseself reportedincidencedatawereusedasoutcomesofinterestforour primarypurpose–determiningbaselinepredictorsforthe occurrenceandtimetofirstLBPepisode. SoldiersreportinganyLBPansweredadditionalquestionnaires sothattheseverityofthefirstepisodecouldbedetermined.These measuresincludedpainintensitywithanumerical0–10rating scale(NRS)[33],disabilitywiththeOswestryDisabilityQuestionnaire(ODQ)[34,35],physicalactivityandworkfearavoidancebeliefswiththeFear-AvoidanceBeliefsQuestionnaire (FABQ-PAandFABQ-W)[36],andpaincatastrophizingbythe PainCatastrophizingScale(CAT)[37].Soldierswerealsoaskedto reportdaysoflimitedmilitarydutyinthepast30daysassociated withtheinitialepisodeofLBP.Thesedatawereusedasoutcomes ofinterestforoursecondarypurpose–determiningbaseline predictorsfortheseverityofLBPepisodes.DataanalysisAllstatisticalanalyseswereperformedusingtheSASsoftware, version9(SASInstituteInc,1996)withatypeIerrorrateof0.05. Allauthorshadfullaccesstoallofthedatareportedinthestudy andcantakeresponsibilityfordataintegrityandaccuracy. Descriptivedataforthebaselinevariableswerecomputed.Then multivariatelogisticregressionanalysiswasusedtodetermine baselinepredictorsofreportingthefirstepisodeofLBPand multivariateCoxregressionanalysiswasusedtodetermine baselinepredictorsoftimetoreportingfirstepisodeofLBP.In theseverityanalyses,multivariatelinearregressionwasusedto determinebaselinepredictorsofpainintensity,disability,and psychologicaldistress.Intheregressionanalysesbaselinepredictorswereconsideredfromalltheindividualvariablesfromthe demographic,socioeconomic,generalhealth,andpsychological domains.Allvariableswereenteredsimultaneouslyintothe regressionmodelstodeterminewhichwerepredictiveofthe outcomeofinterestwhilecontrollingforotherpotentialvariables. Therefore,onlyadjustedestimatesarereportedintheresults.ResultsFigure1describestherecruitmentandfollowupforthePOLM cohort,with1230/3095(28.4%)respondingtothemonthly surveys.Ofthoserespondingtothesurveys,518/1230(42.1%) reportedatleastoneepisodeofLBPduringthe2-yearfollowup period.Specifically,420(48.6%)of865Soldierswhorespondedto themonthlyweb-basedsurveyhadatleastoneepisodeofLBP, andcorrespondingnumbersforthephonesurveywere129 (24.6%)outof525responders.Accompanyingdescriptivedatafor thePOLMcohortisreportedinTables1and2.Soldierswho respondedtothephoneorweb-basedsurveysoverthe2years differedfromthenon-respondersinage,race,educationlevel, militarystatus,timeinmilitary,negativeaffect(depressive symptomsandanxiety),backbeliefs,mentalfunction,smoking priortomilitaryservice,exercisingroutinelypriortomilitary service,andmilitaryfitnessscores(Tables1and2).Inanalyses directlyrelevantforthepurposesofthispaper,comparisonsof baselinecharacteristicsbetweenthosewhoreportedLBPand thosewhodidnotreportLBPrevealeddifferencesingender, activitydutystatus,BDI,FPQ,SF-12PCS,SF-12MCS,and reportingapreviousnonLBPrelatedinjury(Tables1and2).FactorsPredictiveofFirstEpisodeofSelf-ReportedLBPMultivariatelogisticregressionanalysisindicatedthatgender, activedutystatus,SF-12PCS,andSF-12MCSscoreswere significantpredictorsofLBP(Table3).Specifically,protective factorsfordevelopingLBPwerebeingmale(OR=0.644,95% CI=[0.490,0.846]),andhavingbetterSF-12PCSscores Figure1.Flowdiagramforinceptioncohortandresponderstoemailsurveys. InitialEntryTraining(IET),LowBackPain(LBP),Preventionof LowBackPainintheMilitary(POLM),n=totalnumberofsoldiers. doi:10.1371/journal.pone.0030597.g001 PredictorsofFirstTimeLowBackPain PLoSONE|www.plosone.org3February2012|Volume7|Issue2|e30597

PAGE 4

(OR=0.960foreachadditionalpoint,95%CI=[0.935,0.987]) andbetterSF-12MCSscores(OR=0.964foreachadditional point,95%CI=[0.943,0.985]).IncreasedriskofreportingLBP wasnotedforSoldiersonactiveduty(OR=1.441timestheodds forsoldiersonreserve,95%CI=[1.094,1.899]).Figure2 presentstheROCcurveforthefittedlogisticregressionmodel, whichhasanareaundercurve(AUC)of0.64.Duetocorrelation betweenMCSscore,STAIandBDI(withcorrelationcoefficients approximatelyequal0.60),wecomparedtheaboveresultswith thosefromareducedmodelattainedthroughbackwardvariable selection.Allfourfactorsremainedstatisticallysignificantwith ORsof0.626,0.960,0.973and1.392,respectively.Thereduced modelalsoidentifiedBMIasariskfactor(OR=1.044foreach additionalpoint,95%CI=[1.005,1.084]),andithasAUCof 0.61foritsROCcurve.Furthermore,similarsensitivityand specificitywerefoundfrom100repetitionsofcross-validations thatleftoutone-thirdrandomlyselectedsamplesinthemodel fitting. CoxregressionrevealedthatthetimetofirstepisodeofLBPwas significantlyshorterforfemale,activeduty,profiledsoldiersand increasingBMI,withcorrespondinghazardratiosof1.497(95% CI=[1.238, 2 1.809]),1.366(95%CI=[1.120,1.665]),1.271 Table1. ComparisonofbaselineinnateandpsychologicalcharacteristicsbetweenthosewhohadLBPandthosewhohadnoLBP.VariableLabel RespondersLBPNoLBP P-Value* Nonresponders P-Value # n=1230n=518n=712n=3095 InnateCharacteristics Age(N=4319)22.3 6 4.522.6 6 4.622.1 6 4.30.08021.9 6 4.1 0.001 GenderMale852(69.6%)329(63.9%)523(73.7%) 0.0002 2230(72.3%)0.068 Female373(30.4%)186(36.1%)187(26.3%)853(27.7%) RaceBlackorAfrica109(8.9%)58(11.2%)51(7.2%)0.059311(10.1%) 0.020 Hispanic100(8.1%)37(7.1%)63(8.9%)326(10.6%) WhiteorCaucas927(75.5%)381(73.6%)546(76.9%)2263(73.3%) Other92(7.5%)42(8.1%)50(7.0%)187(6.1%) EDUCATIONHighschoolorlower448(36.4%)193(37.3%)255(35.8%)0.2041487(48.1%) 0.0001 Somecollege622(50.6%)249(48.1%)373(52.4%)1376(44.5%) Collegeorhigher160(13.0%)76(14.7%)84(11.8%)231(7.5%) INCOMELessthan$20,000609(49.7%)249(48.3%)360(50.6%)0.4301516(49.1%)0.738 Greaterthan$20,000617(50.3%)266(51.7%)351(49.4%)1571(50.9%) ActiveDutyActive568(46.2%)263(50.8%)305(42.8%) 0.012 1964(63.5%) 0.0001 Reserve660(53.7%)255(49.2%)405(56.9%)1122(36.3%) Other2(0.2%)2(0.3%)6(0.2%) TimeInArmy 5months684(55.6%)287(55.4%)397(55.8%)0.9682007(64.9%) 0.0001 5months–1year315(25.6%)132(25.5%)183(25.7%)654(21.2%) Morethan1year231(18.8%)99(19.1%)132(18.5%)430(13.9%) CompanyInstructorDelta295(24.0%)117(22.6%)178(25.0%)0.482674(21.8%)0.286 Foxtrot130(10.6%)49(9.5%)81(11.4%)393(12.7%) Echo194(15.8%)85(16.4%)109(15.3%)471(15.2%) Alpha185(15.0%)79(15.3%)106(14.9%)442(14.3%) Charlie169(13.7%)68(13.1%)101(14.2%)441(14.2%) Bravo257(20.9%)120(23.2%)137(19.2%)674(21.8%) Height68.3 6 3.968.0 6 3.968.4 6 3.90.08068.3 6 3.90.585 Weight164.0 6 27.8164.0 6 28.0164.0 6 27.70.992165.2 6 27.60.207 BMI24.7 6 3.124.8 6 3.124.6 6 3.20.18424.8 6 3.10.194 PsychologicalCharacteristics BDITotal6.0 6 6.16.6 6 5.95.5 6 6.2 0.002 6.6 6 6.7 0.003 FPQTotal18.2 6 5.718.5 6 5.917.9 6 5.5 0.040 18.0 6 6.00.476 BBQTotal43.9 6 7.243.8 6 7.244.0 6 7.10.61043.2 6 7.0 0.002 STAI35.3 6 9.035.9 6 9.434.9 6 8.60.05536.3 6 9.2 0.001 Responders=thoserespondingtoonlinesurvey,Non-responders=thosenotrespondingtosurvey,LBP=lowbackpain, *=p-valuesforcomparisonofthosewithLBPandthosewithoutLBP(respondersonly),#=p-valuesforcomparisonofthoserespondingtosurveyandthosenotrespondingtosurvey, BMI=BodyMassIndex,BDI=BeckDepressionInventory,FPQ=FearofPainQuestionnaire(9items),BBQ=BackBeliefsQuestionnaire,STAI=StateTraitA nxiety Inventory(stateportiononly).Boldfontindicatesp-valuelessthan0.05. doi:10.1371/journal.pone.0030597.t001 PredictorsofFirstTimeLowBackPain PLoSONE|www.plosone.org4February2012|Volume7|Issue2|e30597

PAGE 5

(95%CI=[1.026,1.573])and1.031(95%CI=[1.002,1.060]), respectively.FactorsPredictiveofLBPSeverity–PainIntensityDescriptivestatisticsforpainintensityreportedduringfirst LBPepisodewere2.0(sd=1.9)forthecurrentpainintensity rating,3.2(sd=2.5)forthehighestpainintensityratinginpast 24hours,and0.8(sd=1.4)forthelowestpainintensityratingin thepast24hours.Multivariatelinearregressionanalysis identifiedbaselinepredictorsofcurrent(R2=7.4%),highest (R2=9.9%),andlowest(R2=9.7%)NRSpainintensityratings duringthefirstepisodeofLBP.Onlyeducationlevelandlast physicalfitnessscorewerepredictorsofallNRSpainintensity ratings.DurationofserviceandbetterSF-12MCSscorewere alsopredictorsforhighestNRSpainintensityratings.Specifically,highestpainintensityrati ngsforsoldierswithhighschool orlowereducationlevelswere1.121higher(95%CI=[0.239, 2.003]onaveragethanratingsforsoldierswithcollegedegrees, 0.948higher(95%CI=[0.186,1.710])forsoldierswith5 monthsto1yearofservicecomparedtothosewithmorethan1 yearofservice,and5.560higher(95%CI=[0.117,11.003])on averagethanratingsforsoldierswithphysicalfitnessscores below150whencomparedtothoseabove300.SF-12MCS scoreswereprotectiveofhighestpainintensityratings,witha decreaseof0.043foreachadditionalpoint(95%CI=[ 2 0.086, 0.001]).FactorsPredictiveofLBPSeverity–DisabilityDescriptivestatisticsfordisabilityreportedduringfirstLBP episodewere9.8(sd=11.7)forODQscoreand1.3(sd=4.5)for numberofdayswithlimitedworkdutiesinthepast30days. Multivariatelinearregressionanalysisrevealedgender,smoking status,andpreviousinjurystatusasbaselinepredictorsofODQ score(R2=11.4%)anddaysoflimitedworkduty(R2=6.2%). Specifically,menonaveragescored3.118pointsloweronthe ODQthanwomen(95%CI=[ 2 5.646, 2 0.590])whileSoldiers whosmokedpriortoenteringtheArmyscoredonaverage2.671 higherthanthosewhodidnot(95%CI=[0.0270,5.315]). SoldierswhohadapreviousnonLBPrelatedinjuryscoredon average3.394higherontheODQthansoldierswhohadnot (95%CI=[0.505,6.283]).Thenumberofdaysoflimitedduty, onaverage,was1.6lessforSoldiersintheTEPonlygroupthan forsoldiersreceivingcombinedCSEP + PSEPgroup(95% CI=[ 2 2.858, 2 0.342]). Table2. Comparisonofbaselinehealthstatus,activity,andattentioneffectsbetweenthosewhohadLBPandthosewhohadno LBP.variableLabel RespondersLBPNoLBP P-Value* Non-responders P-Value # n=1230n=518n=712n=3095 BaselineHealthStatus&PhysicalActivity PCSTotal53.6 6 5.053.0 6 5.554.0 6 4.6 0.001 53.3 6 5.20.191 MCSTotal49.8 6 8.049.0 6 8.450.4 6 7.5 0.002 48.9 6 8.8 0.002 SmokePriortoArmyYes333(27.1%)150(29.0%)183(25.7%)0.2051219(39.4%) 0.0001 No897(72.9%)368(71.0%)529(74.3%)1874(60.6%) ExerciseRoutinelyYes670(54.5%)278(53.7%)392(55.1%)0.6291550(50.1%) 0.010 No560(45.5%)240(46.3%)320(44.9%)1542(49.9%) LastAPFTScoreBelow1504(0.3%)1(0.2%)3(0.4%)0.19120(0.6%) 0.021 150–200260(21.2%)125(24.2%)135(19.0%)750(24.2%) 200–250567(46.2%)233(45.1%)334(47.0%)1461(47.2%) 250–300369(30.0%)149(28.8%)220(30.9%)808(26.1%) Above30028(2.3%)9(1.7%)19(2.7%)55(1.8%) ProfiledYes231(18.8%)118(22.8%)113(15.9%) 0.002 664(21.5%)0.050 No999(81.2%)400(77.2%)599(84.1%)2430(78.5%) Attention/RelationalEffect Physical/USIExamNo1102(89.6%)458(88.4%)644(90.4%)0.2492849(92.1%) 0.010 Yes128(10.4%)60(11.6%)68(9.6%)246(7.9%) PSEPNo642(52.2%)269(51.9%)373(52.4%)0.8741670(54.0%)0.294 Yes588(47.8%)249(48.1%)339(47.6%)1425(46.0%) ExerciseGroupTEPonly334(27.2%)148(28.6%)186(26.1%)0.622882(28.5%)0.747 TEP + PSEP277(22.5%)119(23.0%)158(22.2%)675(21.8%) CSEP308(25.0%)121(23.4%)187(26.3%)788(25.5%) CSEP + PSEP311(25.3%)130(25.1%)181(25.4%)750(24.2%) Responders=thoserespondingtoonlinesurvey,Non-responders=thosenotrespondingtosurvey,LBP=lowbackpain, *=p-valuesforcomparisonofthosewithLBPandthosewithoutLBP(respondersonly),#=p-valuesforcomparisonofthoserespondingtosurveyandthosenotrespondingtosurvey, PCS(SF-12)=PhysicalComponentSummaryScorefromtheShortFormMedicalSurvey(12items),MCS(SF-12)=MentalComponentSummaryScorefromtheSho rt FormMedicalSurvey(12items),APFT=ArmyPhysicalFitnessTest,Profiled=injuredduringtraining,USI=ultrasoundimaging,PSEP=psychosociale ducationprogram, TEP=traditionalexerciseprogram,CSEP=corestabilizationexerciseprogram.Boldfontindicatesp-valuelessthan0.05. doi:10.1371/journal.pone.0030597.t002 PredictorsofFirstTimeLowBackPain PLoSONE|www.plosone.org5February2012|Volume7|Issue2|e30597

PAGE 6

FactorsPredictiveofLBPSeverity–PsychologicalDistressDescriptivestatisticsforpsychologicaldistressreportedduring firstLBPepisodewere9.7(sd=6.1)forFABQ-PAscores,10.6 (sd=9.1)forFABQ-Wscores,and5.5(sd=9.5)forCATscores. Multivariatelinearregressionanalysisindicatedthatgender, smokingstatus,andSF-12PCSscoreswerepredictiveofFABQPAscores(R2=14.2%)whileBMI,SF-12PCS,andBBQscores werepredictiveofFABQ-Wscores(R2=10.6%).FortheFABQPAmenonaveragescored1.639lowerthanwomen(95% CI=[ 2 2.952, 2 0.326])andSoldierswhosmokedscoredon average1.700higherthanthosewhodidnotsmoke(95% CI=[0.332,3.068]).FABQ-PAscoresdecreasedonaverageby 0.183pointsforeachadditionalpointofSF-12PCS(95% CI=[ 2 0.303, 2 0.0634]).FABQ-Wscoresdecreasedonaverage by0.590foreachunitincreaseinBMI(95%CI=[ 2 0.911, 2 0.269])andby0.224pointsforeachadditionalpointofSF-12 PCS(95%CI=[ 2 0.406, 2 0.0417]).FABQ-Wscoresalso decreasedonaverageby0.133foreachadditionalpointofBBQ (95%CI=[ 2 0.260, 2 0.0056]). Multivariatelinearregressionanalysisindicatedthatexercising routinelybeforeenteringthemilitary,SF-12PCS,andtheBBQ werepredictiveofCATscores(R2=8.6%)duringfirstepisodeof LBP.Specifically,weestimatedthatCATscoresare2.380points higheronaverageforsoldierswhoexercisedroutinely(95% CI=[0.347,4.413]),whileCATscoresdecreasedby0.239for eachadditionalpointofSF-12PCS(95%CI=[ 2 0.433, 2 0.0450],p=0.016).CATscoresalsodecreasedby0.143for eachadditionalpointofBBQscore(95%CI=[ 2 0.280, 2 0.0058],p=0.040).DiscussionRigorousstudiesreportingpredictorsforthedevelopmentof LBParerarelyreportedintheliteraturebecauseofthedifficultyof assemblingpainfreecohortsandfollowingthemuntilLBPoccurs. Strengthsofthecurrentstudyweretherecruitmentofalarge inceptioncohortofSoldierswithoutprevioushistoryofLBP, considerationofarangeofpotentiallyrelevantbaselinepredictors, andcollectionoffollowupdataovera2yearperiod.Resultsfrom thisstudysuggestthatactiveduty(i.e.notreserveornational guard)statusincreasedriskofLBPoccurrence,whilebettermental andphysicalhealthscoresandbeingmalewereprotective.Being femaleandactivedutystatuswerealsopredictiveofshortertime tofirstepisodeofLBP,aswellasexperiencingapreviousnonLBP relatedinjuryandhavinghigherBMI.Instudiesthatuseda similarprospectivedesignwefoundagreementwithbetter physicalhealth[38,39,40]asbeingprotectiveforthedevelopment ofLBP.Otherstudiesreportedriskfactorsthatwerenotpredictive ofoccurrenceinthiscohortincludingage[41,40],bodyweight [39,40],smokingstatus[42],andpsychologicalorpsychosocial factors[43,42,44].Ourmilitarycohortwasofarelatively homogenousagerange,withlowbaselinelevelsofpsychological andpsychosocialfactors.Obviouslycohortdifferencescould accountforthediscrepanciesinidentifiedpredictorsofLBP occurrence,asthereislimitedconsistencyamongthesestudies. Figure2.ROCcurveforthemultivariatelogisticregressionmodelpredictinginitiallowbackpainoccurrence. doi:10.1371/journal.pone.0030597.g002 PredictorsofFirstTimeLowBackPain PLoSONE|www.plosone.org6February2012|Volume7|Issue2|e30597

PAGE 7

Anovelaspectofthecurrentstudywasthatwewerealsoableto identifypredictorsofseverityforthefirstLBPepisodebycollecting dataonpainintensity,disability,andpsychologicaldistress. Collectivelyourresultsindicatedthatlowereducationalandfitness levelswerepredictiveofhigherpainintensityscores.Smokingand historyofpreviousnonLBPrelatedinjurywerepredictiveof havingmoredisability,whilebeingmalewaspredictiveofless disability.Onlythetypeofexerciseprogramwaspredictiveof numberofdaysofmissedduty.Lowereducationandfitnesslevels consistentlypredictedhigherpsychologicaldistress,whilebetter physicalhealthscoresandbeliefsaboutbackpainwerepredictive oflowerpsychologicaldistress.Thereweretwocounterintuitive findingsforpsychologicaldistressnoted.HigherBMIwas predictiveoflowerworkfear-avoidancebeliefsandexercising routinelywaspredictiveofhigherpaincatastrophizing.Overall theseresultsconsideringseverityoffirstLBPepisodemayprovide importantnewdataonpreventionstrategies,aspreviousstudiesin thisareahavefocusedexclusivelyonincidenceoroccurrenceof LBP[39,43,41,42,38,44,40].Datafromthecurrentstudyextends previousworkbyhighlightingfactorsthatmaypredictseverityof firstLBPepisode.Thesefactorsmaybeespeciallyimportantfor futurepreventionstudiesbecauseseverityappearstobeakey factorinthedecisiontoseekhealthcareforLBP[45,46,47]. Comparisonofthesefactorstoothersreportedintheliteraturewas notpossiblebecauseincidencestudieshavenottypicallyincluded severitymeasures.Thereforethesepredictorsofseverityshouldbe consideredaspreliminarywithfutureresearchnecessaryto replicatethesefindings. AlthoughwearereportingbaselinepredictorsofthedevelopmentandseverityofLBP,severalcaveatsshouldbeconsidered wheninterpretingtheseresults.First,therewereveryfew consistentpredictorsidentifiedinthisstudy.Manyofthe predictorsappearedtobespecifictotheoutcomemeasure. Second,weidentifiedbaselinepredictorsthatwerestatistically significant,butthemagnitudeofthesepredictorswasoftenlow. Forexample,theincreasedriskofbeingonactivedutywas associatedwithanORof1.44(lowerboundof95%CI=1.09), whiletheprotectivefactorofbeingmalewasassociatedwithan Table3. Logisticregressionforbaselinepredictionofreportingfirstepisodeoflowbackpain.FactorLavel OddsRatioEstimates P-value ComparetoPointEstimate95%ConfidenceLimits GenderMaleFemale0.6440.490.846 0.002 RaceBlackorAfricaOther1.3240.7292.4050.357 HispanicOther0.7610.411.4140.389 WhiteorCaucasOther0.9490.5921.5220.829 EducationHighschoolorlowerCollegeorhigher1.0130.651.580.954 SomecollegeCollegeorhigher0.8450.5731.2470.398 INCOMELessthan$20,000Greaterthan$20,0001.0480.8071.3610.727 ActiveDutyStatusActiveReserve1.4411.0941.899 0.009 SmokePriortoenteringArmyYesNo1.0980.8331.4480.508 TimeInArmy 5monthsMorethan1year0.9140.6541.2790.601 5months–1yearMorethan1year1.0750.7461.5510.697 ExerciseRoutinelyYesNo1.0740.8341.3820.582 LastAPFTScoreBelow150Above3000.3570.0294.4390.423 150–200Above3001.4710.6033.590.397 200–250Above3001.1220.4752.6540.793 250–300Above3001.1630.4912.7560.731 PreviousInjuryYesNo1.2240.8911.6830.213 Age 1.0120.9781.0460.498 BMI 1.0390.9981.0820.061 PCS(SF-12) 0.960.9350.987 0.003 MCS(SF-12) 0.9640.9430.985 0.001 FPQ 1.0080.9861.0310.468 STAI 0.9820.9611.0030.089 BDI 1.010.9821.0390.497 BBQ 10.9831.0170.966 ExerciseandEducationGroupsTEPCSEP1.3720.9851.910.061 TEP + PSEPCSEP1.2970.9111.8450.149 TEPTEP + PSEP1.0580.7511.4890.748 LBP=lowbackpain,BMI=bodymassindex,BDI=BeckDepressionInventory,FPQ=FearofPainQuestionnaire(9items),BBQ=BackBeliefsQuestionnaire, STAI=State TraitAnxietyInventory(stateportiononly),PCS(SF-12)=PhysicalComponentSummaryScorefromtheShortFormMedicalSurvey(12items),MCS(SF-1 2)=Mental ComponentSummaryScorefromtheShortFormMedicalSurvey(12items),TEP=traditionalexerciseprogram,PSEP=psychosocialeducationprogram,CS EP=core stabilizationexerciseprogram.Boldfontindicatesp-valuelessthan0.05. doi:10.1371/journal.pone.0030597.t003 PredictorsofFirstTimeLowBackPain PLoSONE|www.plosone.org7February2012|Volume7|Issue2|e30597

PAGE 8

ORof0.64(upperboundof95%CI=0.85).Third,weincluded allparticipantsinourpredictionmodelsinsteadofincludingonly thoseinacontrolconditionasismorecommonlydonein prognosticstudies.Thisdecisionwasmadebecauseinthissetting allSoldierswereundergoingtrainingwhichincludedrequired exercisesotherewasnooptionfora‘‘true’’controlconditionofno exercise.Last,notmanyoftheidentifiedpredictivefactorswere modifiableinnature.Thiswasespeciallytrueforthefactors predictingtheoccurrenceofLBP,withmoreopportunityfor modifiablefactorsnoticedfortheseverityoutcomes.Datafrom thiscohortprovidefurtherindicationofthemulti-factorialnature ofthedevelopmentofLBPasinconsistencyinpredictorsseemsto bethenormintheliterature,[39,43,41,42,38,44,40].Intheevent consistentfactorsareidentifiedtheyoftenlackthemagnitudetobe consideredas‘‘definitive’’predictorsforthedevelopmentofLBP. EffectivestrategiesforpreventingLBPremainelusive.Back schools,lumbarsupports,andergonomicinterventionshave limitedevidenceforprevention[14,13,15]Educationforprimary preventionofLBPhasreceivedmixedsupportintrials;with supportforpsychosocialeducation[17],butnotforbiomedicalor biomechanicalbasededucationprograms[14].Collectivelythe resultsfromthecurrentstudyandothersinvestigatingpredictors ofLBP[39,43,41,42,38,44,40]provideguidanceforfuture primarypreventioninterventionstrategies.Itisclearfromthe assembleddatathatsinglemodalityapproachesarenotlikelyto beeffectiveinpreventingLBP,echoingexpertandconsensus opinioninthisarea[6,48].Arecommendationfromthecurrent studyisthatLBPpreventionstrategieswilllikelyhavetobe contextualinnature,suchthateffectiveLBPpreventionstrategies foramilitarypopulationmaydifferfromoneusedforhospital nurses.Forexample,resultsfromthismilitarycohortsuggest preventionofoccurrenceofLBPmaybefutilebasedonthelack ofmodifiablepredictors,butthereweresomemodifiablefactors identifiedinastudyofnurses[44].Basedonourdata preventativeinterventionstailoredatdecreasingtheseverityof LBPorvisitsforhealthcareseemfeasibleforthemilitary, especiallythosethatutilizegeneralexerciseapproachestotarget improvingfitnesslevelsandeducationalprogramstoimprove backbeliefs[17,23]. Thereare,ofcourse,limitationstoconsiderwheninterpreting resultsfromthisstudy.Theprimarylimitationisthelowfollow upratetoourwebbasedsurveysystem(28.4%)whichdidnot allowustofollowtheentirecohort.Wehaveidentified differencesinrespondersinapreliminaryanalysisat1year followup[49]andthesedifferenceswereconfirmedinthispaper with2yearfollowup.Howeverwedonothavespecificreasons forthelackofresponse.Itcouldbeduetothewidegeographic areaofdeploymentfortheseSoldiersincludingareasthatwould nothavereadyemailaccess.Wedidconsiderusingestimationor imputationmethodstoaccountforthesemissingdata.However ourconfidenceinthevalidityofthesetechniquestoprovide additionalinformationwaslowduetothefollowuprate(not enoughcompletedatatoallowforestimationorimputation)and theaforementionedbaselinedifferencesbetweenresponders(data usedforestimatingorimputing)andnon-responders.Therefore, adecisionwasmadetoreportonlythecompleteddataandnote thelimitationsofdoingso,ratherthantakingonadditional limitationsinherentinestimatingmissingdatainasituationlike this. Theresultofthelowfollowupratehighlightsthedifficultyof collectingcompletedataoninceptioncohortsforthedevelopment ofLBP.Criticalinterpretationofourresultshingesonthe acknowledgmentthatresponderbiasmayhavehadaprofound impactontheseanalyses.Therearenumerousbaselinedifferences inrespondersandnon-respondersandthisisaconcern.Manyof thesestatisticaldifferencesweresmallinmagnitudeandlikelya functionofthehighstatisticalpowerthataccompaniesalarge sample.Forexampletheagedifferencebetweenrespondersand non-responderswas0.4yearscorrespondingtoaneffectsizeof 0.09(Cohen’sD).Suchasmalldifferenceinagebetween respondersandnonrespondersisnotlikelytoreflectresponder biasforage,andthereweresimilarlysmalldifferencesformanyof theothervariables(Tables1and2).Anotherissuetoconsider whendeterminingtheimpactofthelowfollowuprateisthatour methodologydifferedfromothersurveystudiesinthatwehad accesstotheentiresampleatbaseline(Figure1).Wewerethen abletodeterminedifferencesbetweensurveyrespondersandnon responders.Thistypeofcomparisonisnotanoptioninthemore commonlyincorporateddesignofsurveyingalargergroupwith theresearchersonlyabletoanalyzedatafromresponders.Inthe currentstudywehadtheadvantageofbeingabletoidentify characteristicsofnon-responderssothatthereadercanmakehis/ herownconclusionsabouttheimpactofresponderbias. Unfortunatelythedefinitequantitativeimpactofthisresponder biasforthisstudyisimpossibletoestimate.Inpracticaltermsit meansthatthesedatacanonlybegeneralizedtoindividualsthat respondtosurveyrequests,whichmakesitquiteconsistentwith otherstudiesintheliteraturethatusingthismethodology. Anotherconsequenceofthislowfollowuprateisthatour estimatesofLBPoccurrenceshouldnotbemistakenforatrue incidenceestimateforthispopulation.Itisquitelikelythattherate ofLBP(42.1%)inthiscohortisanoverestimateoftheactual incidencerate.Anotherpotentiallimitationisthattheseanalyses focusedonlyonthefirsttimeepisodeofLBP.Wedidnotconsider recurrenceintheseanalyses,butacknowledgethatLBPisoftena recurrentproblem[50].Finallyanotherlimitationtoconsideris thevalidityofself-reportofLBPhasbeenquestionedformilitary populationsasdiscrepancieswereidentifiedinindicationofever havingLBPandscoresonvalidatedquestionnaires(liketheODQ) [51].Thisconcernismitigatedsomewhatinthecurrentstudy becausewedidn’trelysolelyononeaspectofself-report,and insteadincludedmultiplevalidatedquestionnaires. Inconclusion,thismilitaryinceptioncohortprovidedinformationonbaselinefactorsthatwerepredictiveofoccurrenceofLBP andseverity.Ourresultsconfirmthemulti-factorialnatureofLBP aspredictorsofLBPoccurrenceandseveritywereneither consistentnordefinitive.Futureattemptsatdevelopingpreventativestrategiesshouldconsidertheseresultsandfocusonmultimodalapproachesthattargetmodifiableriskfactorsspecifictothe populationofinterest.AcknowledgmentsTheleadership,cadre,andstudentsfromthe232ndMedicalBattalionand the32ndMedicalBrigadefortheirsupportandstudyparticipation. ChristopherBarnes,YangLi,andErikHenricksonforcreationand managementofthewebsiteanddatabaseandJessicaNeffforassistance withdataentryandconfirmation.YunfengDaiforassistancewith statisticalanalyses. DonnaCunninghamforheradministrativeassistanceandvarious physicaltherapystudentsfromtheU.S.Army-BaylorUniversity, UniversityofTexasHealthScienceCenteratSanAntonio,TexasState University,UniversityofPugetSound,EastTennesseeStateUniversity, andUniversityofColoradoatDenverandHealthScienceCenter.AuthorContributionsConceivedanddesignedtheexperiments:SZGJDCDSTMER. Performedtheexperiments:JDCDSTACWJLD.Analyzedthedata: SSW.Wrotethepaper:SZGJDCDSTSSWMER. PredictorsofFirstTimeLowBackPain PLoSONE|www.plosone.org8February2012|Volume7|Issue2|e30597

PAGE 9

References1.MantyselkaP,KumpusaloE,AhonenR,KumpusaloA,KauhanenJ,etal. (2001)Painasareasontovisitthedoctor:astudyinFinnishprimaryhealthcare. Pain89:175–180. 2.WalkerBF,MullerR,GrantWD(2004)LowbackpaininAustralianadults: prevalenceandassociateddisability.JManipulativePhysiolTher27:238–244. 3.StewartWF,RicciJA,CheeE,MorgansteinD,LiptonR(2003)Lostproductive timeandcostduetocommonpainconditionsintheUSworkforce.JAMA290: 2443–2454. 4.CohenSP,BrownC,KuriharaC,PlunkettA,NguyenC,etal.(2010)Diagnoses andfactorsassociatedwithmedicalevacuationandreturntodutyforservice membersparticipatinginOperationIraqiFreedomorOperationEnduring Freedom:aprospectivecohortstudy.Lancet375:301–309. 5.LincolnAE,SmithGS,AmorosoPJ,BellNS(2002)Thenaturalhistoryandrisk factorsofmusculoskeletalconditionsresultingindisabilityamongUSArmy personnel.Work18:99–113. 6.BurtonAK,BalagueF,CardonG,EriksenHR,HenrotinY,etal.(2005)How topreventlowbackpain.BestPractResClinRheumatol19:541–555. 7.CohenSP,NguyenC,KapoorSG,Anderson-BarnesVC,FosterL,etal.(2009) Backpainduringwar:ananalysisoffactorsaffectingoutcome.ArchInternMed 169:1916–1923. 8.FrankJW,BrookerAS,DeMaioSE,KerrMS,MaetzelA,etal.(1996)Disability resultingfromoccupationallowbackpain.PartII:Whatdoweknowabout secondaryprevention?Areviewofthescientificevidenceonpreventionafter disabilitybegins.Spine21:2918–2929. 9.NicholasMK,LintonSJ,WatsonPJ,MainCJ(2011)EarlyIdentificationand ManagementofPsychologicalRiskFactors(‘‘YellowFlags’’)inPatientsWith LowBackPain:AReappraisal.PhysTher91:737–753. 10.FosterNE,HillJC,HayEM(2011)Subgroupingpatientswithlowbackpainin primarycare:arewegettinganybetteratit?ManTher16:3–8. 11.FritzJM,ClelandJA,ChildsJD(2007)Subgroupingpatientswithlowbackpain: evolutionofaclassificationapproachtophysicaltherapy.JOrthopSportsPhys Ther37:290–302. 12.FrankJW,KerrMS,BrookerAS,DeMaioSE,MaetzelA,etal.(1996)Disability resultingfromoccupationallowbackpain.PartI:Whatdoweknowabout primaryprevention?Areviewofthescientificevidenceonpreventionbefore disabilitybegins.Spine21:2908–2917. 13.LintonSJ,vanTulderMW(2001)Preventiveinterventionsforbackandneck painproblems:whatistheevidence?Spine26:778–787. 14.BurtonAK,BalagueF,CardonG,EriksenHR,HenrotinY,etal.(2006) Chapter2.Europeanguidelinesforpreventioninlowbackpain:November 2004.EurSpineJ15Suppl2:S136–S168. 15.BigosSJ,HollandJ,HollandC,WebsterJS,BattieM,etal.(2009)High-quality controlledtrialsonpreventingepisodesofbackproblems:systematicliterature reviewinworking-ageadults.SpineJ9:147–168. 16.GeorgeSZ,ChildsJD,TeyhenDS,WuSS,WrightAC,etal.(2007)Rationale, design,andprotocolforthepreventionoflowbackpaininthemilitary(POLM) trial(NCT00373009).BMCMusculoskeletDisord8:92. 17.GeorgeSZ,ChildsJD,TeyhenDS,WuSS,WrightAC,etal.(2011)Briefy psychosocialeducation,notcorestabilization,reducedincidenceoflowback pain:resultsfromthePreventionofLowBackPainintheMilitary(POLM) clusterrandomizedtrial.BMCMed9:128. 18.ChildsJD,TeyhenDS,BenedictTM,MorrisJB,FortenberryAD,etal.(2009) Effectsofsit-uptrainingversuscorestabilizationexercisesonsit-up performance.MedSciSportsExerc41:2072–2083. 19.ChildsJD,TeyhenDS,CaseyPR,Coy-SinghKA,FeldtmannAW,etal.(2010) Effectsoftraditionalsit-uptrainingversuscorestabilizationexercisesonshorttermmusculoskeletalinjuriesinUSArmysoldiers:aclusterrandomizedtrial. PhysTher90:1404–1412. 20.BurtonAK,WaddellG,TillotsonKM,SummertonN(1999)Informationand advicetopatientswithbackpaincanhaveapositiveeffect.Arandomized controlledtrialofanoveleducationalbookletinprimarycare.Spine24: 2484–2491. 21.CoudeyreE,TubachF,RannouF,BaronG,CoriatF,etal.(2007)Effectofa simpleinformationbookletonpainpersistenceafteranacuteepisodeoflow backpain:anon-randomizedtrialinaprimarycaresetting.PLoSONE2:e706. 22.GeorgeSZ,FritzJM,BialoskyJE,DonaldDA(2003)Theeffectofafearavoidance-basedphysicaltherapyinterventionforpatientswithacutelowback pain:resultsofarandomizedclinicaltrial.Spine28:2551–2560. 23.GeorgeSZ,TeyhenDS,WuSS,WrightAC,DuganJL,etal.(2009) Psychosocialeducationimproveslowbackpainbeliefs:resultsfromacluster randomizedclinicaltrial(NCT00373009)inaprimarypreventionsetting.Eur SpineJ18:1050–1058. 24.WareJ,Jr.,KosinskiM,KellerSD(1996)A12-ItemShort-FormHealthSurvey: constructionofscalesandpreliminarytestsofreliabilityandvalidity.MedCare 34:220–233. 25.SymondsTL,BurtonAK,TillotsonKM,MainCJ(1996)Doattitudesand beliefsinfluenceworklossduetolowbacktrouble?OccupMed(Lond)46: 25–32. 26.SpielbergerCD,GorsuchRL,LusheneRE,VaggPR,JacobsGA(1983) Manualforthestateandtraitanxiet yinventory(formY)Consulting PsychologistsPress,PaloAltoCA. 27.WhismanMA,PerezJE,RamelW(2000)FactorstructureoftheBeck DepressionInventory-SecondEdition(BDI-II)inastudentsample.JClin Psychol56:545–551. 28.SchotteCK,MaesM,CluydtsR,DeDonckerD,CosynsP(1997)Construct validityoftheBeckDepressionInventoryinadepressivepopulation.JAffect Disord46:115–125. 29.ChibnallJT,TaitRC(1994)TheshortformoftheBeckDepressionInventory: validityissueswithchronicpainpatients.ClinJPain10:261–266. 30.McNeilDW,RainwaterAJ(1998)DevelopmentoftheFearofPain Questionnaire–III.JBehavMed21:389–410. 31.AlbaretMC,MunozSastreMT,CottencinA,MulletE(2004)TheFearofPain questionnaire:factorstructureinsamplesofyoung,middle-agedandelderly Europeanpeople.EurJPain8:273–281. 32.OsmanA,BreitensteinJL,BarriosFX,GutierrezPM,KopperBA(2002)The FearofPainQuestionnaire-III:furtherreliabilityandvaliditywithnonclinical samples.JBehavMed25:155–173. 33.JensenMP,TurnerJA,RomanoJM(1994)Whatisthemaximumnumberof levelsneededinpainintensitymeasurement?Pain58:387–392. 34.FairbankJC,PynsentPB(2000)TheOswestryDisabilityIndex.Spine25: 2940–2953.35.RolandM,FairbankJ(2000)TheRoland-MorrisDisabilityQuestionnaireand theOswestryDisabilityQuestionnaire.Spine25:3115–3124. 36.WaddellG,NewtonM,HendersonI,SomervilleD,MainCJ(1993)AFearAvoidanceBeliefsQuestionnaire(FABQ)andtheroleoffear-avoidancebeliefs inchroniclowbackpainanddisability.Pain52:157–168. 37.SullivanMJL,BishopSR,PivikJ(1995)ThePainCatastrophizingScale: developmentandvalidation.PsychologicalAssessment7:524–532. 38.StevensonJM,WeberCL,SmithJT,DumasGA,AlbertWJ(2001)A longitudinalstudyofthedevelopmentoflowbackpaininanindustrial population.Spine26:1370–1377. 39.CroftPR,PapageorgiouAC,ThomasE,MacfarlaneGJ,SilmanAJ(1999) Short-termphysicalriskfactorsfornewepisodesoflowbackpain.Prospective evidencefromtheSouthManchesterBackPainStudy.Spine24:1556–1561. 40.EldersLA,BurdorfA(2004)Prevalence,incidence,andrecurrenceoflowback paininscaffoldersduringa3-yearfollow-upstudy.Spine29:E101–E106. 41.ParkH,SprinceNL,WhittenPS,BurmeisterLF,ZwerlingC(2001)Riskfactors forbackpainamongmalefarmers:analysisofIowaFarmFamilyHealthand HazardSurveillanceStudy.AmJIndMed40:646–654. 42.PowerC,FrankJ,HertzmanC,SchierhoutG,LiL(2001)Predictorsoflow backpainonsetinaprospectiveBritishstudy.AmJPublicHealth91: 1671–1678. 43.PapageorgiouAC,MacfarlaneGJ,ThomasE,CroftPR,JaysonMI,etal.(1997) Psychosocialfactorsintheworkplace–dotheypredictnewepisodesoflowback pain?EvidencefromtheSouthManchesterBackPainStudy.Spine22: 1137–1142. 44.YipVY(2004)Newlowbackpaininnurses:workactivities,workstressand sedentarylifestyle.JAdvNurs46:430–440. 45.CoteP,CassidyJD,CarrollL(2001)Thetreatmentofneckandlowbackpain: whoseekscare?whogoeswhere?MedCare39:956–967. 46.MortimerM,AhlbergG(2003)Toseekornottoseek?Care-seekingbehaviour amongpeoplewithlow-backpain.ScandJPublicHealth31:194–203. 47.IJzelenbergW,BurdorfA(2004)Patternsofcareforlowbackpaininaworking population.Spine29:1362–1368. 48.GuzmanJ,HaydenJ,FurlanAD,CassidyJD,LoiselP,etal.(2007)Keyfactors inbackdisabilityprevention:aco nsensuspanelontheirimpactand modifiability.Spine(PhilaPa1976)32:807–815. 49.ChildsJD,TeyhenDS,VanWyngaardenJJ,DoughertyBF,LadislasBJ,etal. (2011)Predictorsofweb-basedfollow-upresponseinthePreventionofLowBack PaintheMilitaryTrial(POLM).BMCMusculoskeletDisord12:132. 50.CareyTS,GarrettJM,JackmanA,HadlerN(1999)Recurrenceandcare seekingafteracutebackpain:resultsofalong-termfollow-upstudy.North CarolinaBackPainProject.MedCare37:157–164. 51.CarrageeEJ,CohenSP(2009)Lifetimeasymptomaticforbackpain:thevalidity ofself-reportmeasuresinsoldiers.Spine(PhilaPa1976)34:978–983.PredictorsofFirstTimeLowBackPain PLoSONE|www.plosone.org9February2012|Volume7|Issue2|e30597