Predictors of web-based follow-up response in the Prevention of Low Back Pain in the Military Trial
http://www.biomedcentral.com/bmcmusculoskeletdisord/ ( Publisher's URL )
CITATION PDF VIEWER
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/IR00000568/00001
 Material Information
Title: Predictors of web-based follow-up response in the Prevention of Low Back Pain in the Military Trial
Series Title: PMID: 21668961
Physical Description: Journal Article
Creator: George, Steven
Childs, John D.
Teyhen, Deydre S.
Van Wyngaarden, Joshua J.
Dougherty, Brett F.
Ladislas, Bryan J.
Helton, Gary L.
Robinson, Michael E.
Wu, Samuel S.
Publisher: BioMed Central
Publication Date: June 13th, 2011
 Subjects
Genre: serial   ( sobekcm )
 Notes
Abstract: Background: Achieving adequate follow-up in clinical trials is essential to establish the validity of the findings. Achieving adequate response rates reduces bias and increases probability that the findings can be generalized to the population of interest. Therefore, the purpose of this study was to determine the influence of attention, demographic, psychological, and health status factors on web-based response rates in the ongoing Prevention of Low Back Pain in the Military (POLM) trial. Methods: Twenty companies of Soldiers (n = 4,325) were cluster randomized to complete a traditional exercise program including sit-ups (TEP) with or without a psychosocial educational program (PSEP) or a core stabilization exercise program (CSEP) with or without PSEP. A subgroup of Soldiers (n = 371) was randomized to receive an additional physical and ultrasound imaging (USI) examination of key trunk musculature. As part of the surveillance program, all Soldiers were encouraged to complete monthly surveys via email during the first year. Descriptive statistics of the predictor variables were obtained and compared between responders and non-responders using two sample t-tests or chi-square test, as appropriate. Generalized linear mixed models were subsequently fitted for the dichotomous outcomes to estimate the effects of the predictor variables. The significance level was set at .05 a priori. Results: The overall response rate was 18.9% (811 subjects) for the first year. Responders were more likely to be older, Caucasian, have higher levels of education and income, reservist military status, non smoker, lower BMI, and have received individualized attention via the physical/USI examination (p < .05). Age, race/ethnicity, education, military status, smoking history, BMI, and whether a Soldier received the physical/USI examination remained statistically significant (p < .05) when considered in a full multivariate model. Conclusion: The overall web based response rate during the first year of the POLM trial was consistent with studies that used similar methodology, but lower when compared to rates expected for standard clinical trials. One year response rate was significantly associated with demographic characteristics, health status, and individualized attention via additional testing. These data may assist for planning of future trials that use web based response systems. Trial Registration: This study has been registered at reports at http://clinicaltrials.gov (NCT00373009).
Acquisition: Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by Steven George.
Funding: Publication of this article was funded in part by the University of Florida Open-Access Publishing Fund.
 Record Information
Source Institution: University of Florida Institutional Repository
Holding Location: University of Florida
Rights Management: All rights reserved by the submitter.
Resource Identifier: sobekcm - IR00000568_00001
System ID: IR00000568:00001

Downloads

This item is only available as the following downloads:

Copy of manuscript ( PDF )


Full Text

PAGE 1

RESEARCHARTICLE OpenAccessPredictorsofweb-basedfollow-upresponsein thePreventionofLowBackPainintheMilitary Trial(POLM)JohnDChilds1*,DeydreSTeyhen1,JoshuaJVanWyngaarden1,BrettFDougherty1,BryanJLadislas1, GaryLHelton1,MichaelERobinson2,SamuelSWu3andStevenZGeorge4AbstractBackground: Achievingadequatefollow-upinclinicaltrialsisessentialtoestablishthevalidityofthefindings. Achievingadequateresponseratesreducesbiasandincreasesprobabilitythatthefindingscanbegeneralizedto thepopulationofinterest.Therefore,thepurposeofthisstudywastodeterminetheinfluenceofattention, demographic,psychological,andhealthstatusfactorsonweb-basedresponseratesintheongoingPreventionof LowBackPainintheMilitary(POLM)trial. Methods: TwentycompaniesofSoldiers(n=4,325)wereclusterrandomizedtocompleteatraditionalexercise programincludingsit-ups(TEP)withorwithoutapsychosocialeducationalprogram(PSEP)oracorestabilization exerciseprogram(CSEP)withorwithoutPSEP.AsubgroupofSoldiers(n=371)wasrandomizedtoreceivean additionalphysicalandultrasoundimaging(USI)examinationofkeytrunkmusculature.Aspartofthesurveillance program,allSoldierswereencouragedtocompletemonthlysurveysviaemailduringthefirstyear.Descriptive statisticsofthepredictorvariableswereobtainedandcomparedbetweenrespondersandnon-respondersusing twosamplet-testsorchi-squaretest,asappropriate.Generalizedlinearmixedmodelsweresubsequentlyfittedfor thedichotomousoutcomestoestimatetheeffectsofthepredictorvariables.Thesignificancelevelwassetat.05a priori. Results: Theoverallresponseratewas18.9%(811subjects)forthefirstyear.Respondersweremorelikelytobe older,Caucasian,havehigherlevelsofeducationandincome,reservistmilitarystatus,nonsmoker,lowerBMI,and havereceivedindividualizedattentionviathephysical/USIexamination(p<.05).Age,race/ethnicity,education, militarystatus,smokinghistory,BMI,andwhetheraSoldierreceivedthephysical/USIexaminationremained statisticallysignificant(p<.05)whenconsideredinafullmultivariatemodel. Conclusion: TheoverallwebbasedresponserateduringthefirstyearofthePOLMtrialwasconsistentwith studiesthatusedsimilarmethodology,butlowerwhencomparedtoratesexpectedforstandardclinicaltrials.One yearresponseratewassignificantlyassociatedwithdemographiccharacteristics,healthstatus,andindividualized attentionviaadditionaltesting.Thesedatamayassistforplanningoffuturetrialsthatusewebbasedresponse systems. TrialRegistration: Thisstudyhasbeenregisteredatreportsathttp://clinicaltrials.gov(NCT00373009). *Correspondence:childsjd@gmail.com Contributedequally1USArmy-BaylorUniversityDoctoralPrograminPhysicalTherapy(MCCSHMT),ArmyMedicalDepartmentCenterandSchool,3151ScottRd.,Rm. 2307,FortSamHouston,TX78234,USA FulllistofauthorinformationisavailableattheendofthearticleChilds etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 2011Childsetal;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommons AttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,andreproductionin anymedium,providedtheoriginalworkisproperlycited.

PAGE 2

BackgroundAchievingadequatefollow-up inclinicaltrialsisessentialtoestablishthevalidityofstudyfindingsandreduce bias,helpingtoinsurethatthefindingscanbegeneralizedtothepopulationofinterestandmoreaccurately informclinicaldecision-making.Studieswithlowfollow-upratespotentiallyconfoundinterpretationofthe resultssincesubjectswhodropoutmaybematerially differentfromthosewhocompletethestudy(i.e.attritionbias)[1].Lowsubjectresponseratescanfurther threatenexternalvaliditybyimpairingtheabilityof researcherstomakeclearscientificconclusionsbased ontheirdata[1].AccordingtoStrausetal,follow-up ratesthatexceed95%minimizethepotentialforattritionbiastoexistwhereasfollow-uprateslowerthan 80%poseathreattoexternalvalidity[2,3].Evensmall lossestofollow-upcanbiasastudy sresultsiffewindividualshavetheoutcomeofinterest.Collectively,these issuesmakeitimperativethatclinicaltrialsbeconductedinamannertomaximizeretention. Oneofthemostcommonlyreportedfactorstobe associatedwithimprovingretentionandfollow-upisthe attentionaffordedtosubject sduringtheirparticipation inthestudy[4].Diasetalfoundthatincreasedattentionintheformofstafffriendliness,responsiveness,and subjectencouragementposit ivelyinfluencedlong-term follow-up,withretentionratesof98.5%fortheir3year study[4].Alternatively,Loftinetalfoundthatfailingto follow-upwithsubjectsconsistentlyanddevelopcaring andtrustingrelationshipswithstudyparticipantsnegativelyimpactedretention[5].Onemightpresumethat increasedattentionatanindividuallevel(ie,physical examination,interview,etc.)mighttranslateinto improvedretentionandfollow-upcomparedtogroupbasedattention(ie,educationalclass)becauseofthe potentialtoformadeeperconnectionwithsubjectsina one-to-oneenvironmentcomparedtoagroupsetting. TheexperimentalgroupsintheLoftinstudiesreceived bothgroupandindividualattentionthroughdietary classesandweeklyphonecalls,respectively,hencethey wereunabletodeterminewhe therincreasedindividual attentionissuperiortogroup-basedattention[5]. Furtherstudiesareneededtodeterminetheinfluenceof attention,especiallyanalysesthatallowforcomparison ofdifferentformsofattention. Anumberofotherfactorshavealsobeenpurported topositivelyinfluencelong-termfollow-up.These includeageover60,thosewithlowerbaselineself-efficacy,andaparticipant sbeliefinthemeritsofthestudy [6].Loftinetalfoundthatsubjectswithhigherratesof follow-uphadstrongerbeliefsabouttheextenttowhich thestudysignificantlycontributedtothecommunity andtheadvancementofscience[5].Conversely,a numberoffactorshavebeenshowntonegativelyinfluenceretentionintrials.Jansonetalconductedastudy on35subjectswhohadvoluntarilywithdrawnfroma large,multi-centerrandomizedtrial[7].Theprimary factorfoundtobeassociatedwithdecreasedretention wasaperceivedlackofsensitivityonthepartofthe researchstaff.Therewerealsoafewdemographiccharacteristicscommonlyassociatedwithsubjectwithdrawal toincludeyoungerindividualsandethnicminorities. Whilethesefactorstendedtoinfluencedrop-outrates, theydidnotachievestatisticalsignificancesecondaryto lackofpowerasaresultofthesmallsamplesizeof35 [7].Otherstudieshavereinforcedthenotionthatdemographicfactorsarenothighlypredictiveofdrop-out rates.Forexample,alargeRCTwithover2,311subjects failedtodetectarelationshipbetweenBMI,sex,ethnicity,andretentionatoneyearfollowup[6]. Furtherresearchisneededtoidentifypotentially importantfactorsthatinfluencefollow-uprates.Then thesefactorscouldbeappropriatelyconsideredwhen designingclinicaltrials.AspartoftheongoingPreventionofLowBackPainintheMilitary(POLM)trial,we utilizedanovelweb-basedsurveillancesystemtotrack subjectresponserateandrecordincidenceandseverity oflowbackpain(LBP)episodesamongagroupofgeographicallydispersedSoldiersintheU.S.Armyovera 2-yearperiod[8].Aspartofthetrial,wehadaccessto manybaselinevariablespr eviouslyfoundtobeassociatedwithfollow-upratesintrials.Therefore,thepurposeofthissecondaryanalysiswastodetermine predictorsduringthefirstyearofweb-basedresponse ratesinthePOLMtrial.Wehypothesizedthatsubjects receivingincreasedattentionviaarandomlyselected educationprogramorphysicalexaminationsession wouldhavehigherfollow-upratesthanthosereceiving lessattention.Wealsosoughttodeterminetheinfluenceofvariousdemographic,psychological,andhealth statusfactorsonweb-basedresponserates.MethodsDesignOverviewThisstudyreportsaplannedsecondaryanalysisinthe PreventionofLowBackPainintheMilitaryclinicaltrial (NCT00373009)whichhasbeenregisteredathttp://clinicaltrials.gov[8].Consecutivesubjectsenteringa16-week trainingprogramatFortSamHouston,TXtobecomea combatmedicintheU.S.Armywereconsideredfor participation.Intheprimarytrial,20companiesofSoldierswereclusterrandomizedtocompleteoneof4 trainingprograms:atraditionalexerciseprogramincludingsit-ups(TEP)with(n=945)orwithout(n=1,212) apsychosocialeducationalprogram(PSEP)oracore stabilizationexerciseprogram(CSEP)with(n=1,049)Childs etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page2of11

PAGE 3

orwithoutPSEP(n=1,089)[9,10].Subjectsineachof the4groupsperformedtheassignedexerciseprogram inagroupsettingunderthedirectsupervisionoftheir drillinstructorsaspartofdailyunitphysicaltraining [8,11,12].Subjectsarecurrentlybeingfollowedmonthly fortwoyearsusingaweb-basedsurveillancesystemto recordincidenceandseverityofsubsequentLBPepisodes.However,theprimarytrialresultsarenotyet available.Forthisanalysis,wecollapsedthestudyintoa singlecohortforthepurposeofdeterminingpredictors of1-yearresponseratestotheweb-basedfollow-up survey.SettingandParticipantsResearchstaffatFortSamHouston,Texasintroduced thestudytoindividualcompaniesofSoldiersand obtainedwritteninformedconsent.RefertoFigure1for aflowdiagramdescribingthenumberofcompaniesand Soldiersconsideredforthistrial,eventuallyenrolledinto thetrial,andcompletedthe1-yearweb-basedfollow-up survey,aspertheConsolidatedStandardsofReporting Trials(CONSORT)guidelines[9].Allsubjectswere recruitedduringatrainingor ientationsessionattended byallSoldiersaspartoftheirin-processingformedic training.For8consecuti vemonthssubjectswere screenedforeligibilityaccordingtotheinclusion/exclusioncriteria.Subjectswererequiredtobe18-35yearsof age(or17yearoldemancipatedminor),participatingin trainingtobecomeacombatmedic,andbeableto speakandreadEnglish.Subjectswithapriorhistoryof LBPwereexcluded.ApriorhistoryofLBPwasoperationallydefinedasLBPthatlimitedworkorphysical activity,lastedlongerthan48hours,andcausedthe subjecttoseekhealthcare.Subjectswerealsoexcluded iftheywerecurrentlyseekingmedicalcareforLBP; unabletoparticipateinunitexerciseduetoinjuryin foot,ankle,knee,hip,neck,shoulder,elbow,wrist,or hand;hadahistoryoffracture(stressortraumatic)in proximalfemur,hip,orpelvis;werepregnant;orifthey hadtransferredfromanothertraininggroup.OtherpossibleexclusionsincludedSoldierswhowerebeingacceleratedintoaCompanyalreadyrandomizedand recruitedforparticipationinthePreventionofLow BackPainintheMilitarytrialorSoldierswhowere beingre-assignedtoanoccupationalspecialtyother thanacombatmedic.EthicsApprovalTheinstitutionalreviewboardsattheBrookeArmy MedicalCenter(SanAntonio,TX)andtheUniversityof Florida(Gainesville,FL)grantedapprovalforthisproject.Allsubjectsprovidedwritteninformedconsent priortotheirparticipation.PotentialPredictorsofResponseRatestotheWeb-based SurveySelectdemographiccharacteristics,psychologicalvariables,healthstatusandphysicalactivity,injurystatus, andattention/relationshipe ffectvariableswereconsideredaspotentialpredictorsof1-yearresponserateson theweb-basedfollow-upsurvey.Thesemeasureswere collectedatbaselineusingavarietyofcommonlyutilizedandpreviouslyvalidatedself-reportquestionnaires andphysicalexaminationproceduresperformedby researchpersonnelunaware ofrandomizationassignmentatbaseline.Allmeasureswerescoredinamasked mannerbycomputeralgorithm.DemographicCharacteristicsDemographiccharacteristicswereconsideredasbotha) potentialpredictorsofresponserateandb)riskadjustmentvariables.Thesecharacteristicsincludedage,sex, race/ethnicity,levelofeducation,income,lengthofservice,militarystatus,andassignedCompanydrill instructors.PsychologicalVariablesTheBackBeliefsQuestionnaire(BBQ)isapreviously validatedself-reportquestionnaireusedtoquantify beliefsaboutthelikelyconsequencesofhavingLBP. HigherBBQscoresareindicativeofbetterLBPbeliefs andindicatethepotentialofabetterabilitytocopewith LBP[13].TheState-TraitAnxietyQuestionnaire(STAI) andBeckDepressionInventory(BDI)wereusedtomeasurenegativeaffectfromgeneralizedanxietyandgeneralizeddepression,respectively[13].Higherscoreson theseindiceswereindicatedofhigheranxietyand Figure1 Flowdiagramforsubjectrecruitmentandemail respondersoneyearaftertheconclusionofthestudy Childs etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page3of11

PAGE 4

depressivesymptoms.NineitemsfromtheFearofPain Questionnaire(FPQ-III)wereusedtomeasurefear aboutspecificsituationsthatnormallyproducepain [13].Higherscoresonthefearindicesindicatedhigher generalfearofpainandfearoflowbackpain.HealthStatusandPhysicalActivityTheMedicalOutcomesSurvey12-ItemShort-Form HealthSurvey(SF-12)wasusedasaself-reportofhealth statusforphysicalandmentalfunction.Thephysical andmentalcomponentsummaryscales(PCSandMCS) werereportedindividuallyinthisstudybecausetheyare validestimatesofphysicalandmentalhealth[13]. Aspartoftheintakequestionnaire,Soldierswere queriedastotheirlevelofphysicalactivitypriorto enteringtraining.Specifi cally,Soldierswereaskedto reporthowmanydaysperweekonaveragetheyperformedatleast30minutesofexerciseandhowmany yearsoverthecourseoftheirlifetimetheyhaveconsistentlyexercisedatleast3daysperweekpriortoenteringtraining.Soldierswerequeriedregardingtheir smokingstatus,andtheirbodymassindex(BMI)was calculated[8].Attention/RelationshipEffectPsychosocialEducationalProgram Soldierswhowere randomizedtoPSEP(n=1994)completedaneducationalsessionwithinagroupsettingduringthefirst14 daysofenteringtraining.Thesessionconsistedofan interactiveseminardesignedbythePOLMinvestigative teamandwasimplementedbystudypersonnel.The overallgoalofthe45minutesessionwastoemphasize currentscientificevidenceonLBPbasedonbiopsychosocialprinciplesthatpromotehealthybeliefsaboutLBP. TheseminarcoveredtopicsrelatedtothefavorablenaturalhistoryofLBP,lackofdefinitiveanatomicalcauses ofLBP,theimportanceofreturningtonormalactivity, anddecreasingfear-avoidancebeliefsandpaincatastrophizingwhenexperiencingLBP.Soldierswereinformed whyeducationalinformationonbestLBPcopingstrategieswasimportantdespitethefacttheydidnotcurrentlyhaveLBP.Aftertheseminar,Soldiersparticipated inaquestionandanswersessionandwereissuedThe BackBook[8].TheBackBookwasusedastheeducationalsupplementbecauseofourpriorexperiencewith itinaphysicaltherapyclinicaltrialanditspriorassociationwithpositiveshiftsinpatientLBPbeliefs[13]. PhysicalExaminationofTrunkMusculature Because itwouldbetimeandcostprohibitivetoperforman extensivephysicalexaminationonallsubjectsinatrial thislarge,asubgroupofSoldiersfromeachcompany(n =371)wererandomizedtoreceiveadditionaltestingin theformofaphysicalandultrasoundimaging(USI) examinationofkeytrunkmusculature.Thephysical examinationconsistedofmeasuringlowbackrangeof motion,straightlegraise,andbilateralhiprangeof motionmeasurements.Soldiersalsocompleted4trunk muscleendurancetests(extension,flexion,andbilateral sidesupports)bydetermininghowlongaspecificpositioncouldbemaintained.Separately,aUSIexamination wasperformedwhichincludedassessmentofthelateral abdominalmuscles(transverseabdominus,internaland externalobliquemuscles)du ringanactivestraightleg raiseandsymmetryofthemultifidimuscles[8].The examinationrequiredapproximately2hours.Soldiers whoreceivedthephysical/USIexaminationand/or receivethePSEPwereclassifiedashavingreceivedadditionalattentionforthepur poseofassessingthepotentialforincreasedattentiontoinfluenceresponserates.Web-basedFollow-upSurveysAttheendoftheinitial12weeksoftraining,Soldiers weretrainedinacomputerlabonhowtousethewebbasedsurveillancesystemtocompletethemonthlyfollow-upsurveys.Thepurposeofthefollow-upsurveys wastorecordincidenceandseverityofsubsequentLBP episodesinthepreviouscalendarmonth.Accesstothe web-basedsurveillancesystemwaspromptedbyan email,whichwassenttotheSoldier sofficialmilitary emailaddressonthe1stofeachmonth.Theweb-based surveystartedwithanemailpromptingtovisitthe studyhosted,confidential ,secureweb-site.Oncethe websitewasaccessed,Soldierswereaskedoneinitial screeningquestion- haveyouhadanybackpaininthe past30days? A no answerendedthesurveyandSoldierswerethankedfortheirparticipation.A yes answerpromptedtheSoldierstocompleteanadditional setof46itemsaboutthebackpainepisodeincluding duration,impactonworkactivities,whetherhealthcare wassought,andresponsetostandardquestionnaires(ie, NPRS,ODQ,FABQ,andPCS).Soldierswereprovided theirlogincredentials(usernameandpassword)during theinitialtrainingsessionattheendofthe12-week trial.Logincredentialswerealsoprovidedinthe monthlyemailreminders.IfaSoldierdidnotrespond tothefirstemail,anadditionalemailwassentonthe 3rdofthemonth,andagainonthe7thofthemonthif theSoldierstillhadnotresponded.DataAnalysisTheprimarydependentvariableforthispaperwasthe dichotomousoutcomeofwhetheraSoldierresponded toanyoneofthe12monthlysurveys.Theindependent variablesconsideredaspotentialpredictorsofresponse rateincludedpsychological variables(BDI,FPQ,BBQ, STAI),healthstatusandphysicalactivity(SF-12PCS andMCStotal,smokingstatus,levelofphysicalactivity, BMI),andtheattention/relationshipeffect(received physical/USIexaminationorPSEP).PotentialeffectsofChilds etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page4of11

PAGE 5

additionalattentioninthe formofanindividualized examinationandgroupattentionfromthePSEPwere examinedseparately.Otherexplanatoryvariablesof interestandforriskadjustmentincludeddemographic characteristicssuchasage,sex,race/ethnicity(White/ Caucasian,Black/AfricanAmerican,Hispanic,and others),levelofeducation(Collegeormore,Somecollege,Highschoolorless),income($>20,000ormore, lengthofservice(<5months,5months-1year,>1year), militarystatus(Activeduty,Reservist,orNational Guard),andassignedCompanydrillinstructors. Descriptivestatisticsofthedemographicandclinical variableswerecomparedbetweentherespondersand non-respondersusingtwosamplet-testsorchi-square test,asappropriate.Agen eralizedlinearmixedmodel wasthenfittedforthedichotomousoutcometoestimatetheeffectsofpotentialpredictorsandtheother explanatoryvariableslistedabove.Arandomcompany effectwasincludedinthemodelstoaccommodatefor thecorrelationamongSoldierswithinthesamecompany.Furthermore,toassesstheresponsedifference overtime,wefittedasecondgeneralizedlinearmixed modelusingthelongitudinalbinaryoutcomesof whetheraSoldierrespondedtoeachoneofthe12 monthlysurveysasdependentvariable,withquadratic timeeffectinadditiontothesamepredictor/explanatory variablesasinthefirstmodel.Thesignificancelevelwas setat.05apriori,andallanalyseswereperformedwith theuseofSASsoftware,version9.1.ResultsAmongthe4,325SoldierswhocompletedPOLMtrial, 4,295Soldiers(99.3%)hadcompletedatainallpredictor variablesandincludedinthefinalanalyses(Figure1). Amongthe4,295Soldiers,71%weremale,72%were White/Caucasian,55%hadatleastsomecollegeor moreeducation,51%had$20,000ormorehousehold income,63%hadbeenenlistedintheArmyforless than5months,and15%formorethan1year.The studypopulationhadameanageof22.0years(SD= 4.2)(Table1).Theoverallresponseratetothewebbasedsurveywas18.9%(811subjects)forthefirstyear ofthePOLMtrial. Non-respondersandresponderssignificantlydiffered inage,race/ethnicity,education,income,militarystatus, lengthofservice,depression,backbeliefs,anxiety,health status,smokinghistory,BMI,andwhetheraSoldier receivedindividualattentionfromthephysical/USI examination(allwithp<.05,Table1).Basedonthe adjustedmodel(Table2),theoddsofresponseincreased by5%foreveryoneyearincreaseinage.Black/African AmericanSoldiershad.76timesoddsofresponsecomparedtoWhite/Caucasian.ComparedwithSoldierswith collegeorhighereducation,theoddsofresponsewere .54and.70timesforthosewithhighschoolorlessand thosewithsomecollegeeducation,respectively.Full-time activedutyservicemembershad.68timesoddsof responsecomparedtothosefromaReserveorNational Guardunit.Theoddsofresponsedecreasedby3%for everyoneunitincreaseinBMI.Thosewhodidnot smokehad1.69timesoddsofresponsecomparedto thosewhosmokedpriortoenteringtheArmy.Inaddition,thosewhodidnotreceivethephysical/USIexaminationhad.70timesoddsofresponsecomparedtothose whoreceivedtheexamination.Therewasnodifference inresponseratebasedonw hetherSoldiersreceived groupattentionviathePSEP.Thefollowingfactors: income,lengthofservice,BDI,BBQandSF-12became statisticallynon-significantafteradjustingthepreviously statedfactors(Table2).Inaddition,theaboveeffects remainedstatisticallysignificantinthesecondgeneralizedlinearmixedmodelthatincludedthequadratictime effect,whichindicatedthattheresponseratessignificantlydecreasedoverthefirst12monthsofthetrial(p< .001,Figure2).DiscussionTheresultsofthisanalysisdemonstratedthatresponse ratetotheweb-basedsurveywassignificantlyassociated withdemographiccharacteristics,healthstatus,and individualizedattentionv iaadditionaltesting.Our responseratewaslowcomparedtostandardrandomizedclinicaltrialsthatincorporateface-to-facecontact tosecurefollow-updata(whichtypicallyrangefrom8095%[2,3])andcomparedtoatleastonesimilarly designedstudythatdependedheavilyonweb-basedsurveillancesystemswithoutdirectface-to-facecontact withthesubjectduringthefollow-upphaseofthestudy [14].Theoveralllowerresponseratesobservedwith web-basedsurveillancesystemscomparedtomoretraditionalfollow-upstrategies(ie,phone,face-to-face,etc.) islikelyattributabletolesssubjectaccountabilityduring thefollow-upphaseofthestudy.Soldiersdidnothave face-to-facecontactandacc ountabilityforsurveycompletionfollowingtheinitial trainingphaseofthestudy, placingmoreresponsibilityontheindividualSoldiersto completetheonlinesurveys.Althoughdifficulttoconfirm,itislikelythatthegeographicdispersionofSoldiersaroundtheworld,deploymentstoausterepartsof theworldwithlimitedinternetaccess(ie,Iraq/Afghanistan),andsubsequentdischargefromtheArmymay havealsocontributedtotheoveralldecreasedresponse rate.Whenusingweb-basedsurveillancesystems,follow-upratesmaybefurtherenhancedbysupplementing withtraditionalmethodssuchasphonecallcentersand queryingavailabledatabasesforhealthcareutilization relatedtoLBP.ThesecombinationsofmultiplefollowupstrategieshavethepotentialtoincreaseoverallChilds etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page5of11

PAGE 6

Table1Statisticalanalysisofweb-basedrespondersandnon-respondersviaduringfirstyearfollowupfromPOLM studyVariables Overall (n=4,295) NoResponse (n=3,484) Response (n=811) P-value Age 22.0 (4.2) 21.8 (4.1) 22.8 (4.7) <.001 Gender Female 1233 (28.7%) 978 (28.1%) 255 (31.4%) .056 Male 3062 (71.3%) 2506 (71.9%) 556 (68.6%) Race/Ethnicity African American 444 (10.3%) 367 (82.7%) 77 (17.3%) .043 Caucasian 3,094 (72.0%) 2,508 (81.1%) 586 (18.9%) Other 757 (17.7%) 609 (80.4%) 148 (19.6%) Education Highschoolorlower 1,952 (45.4%) 1,667 (85.4%) 285 (14.6%) Somecollege 1,955 (45.5%) 1,549 (79.2%) 406 (20.8%) <.001 Graduatedfromcollegeorhigher 388 (9.0%) 268 (69.1%) 120 (30.9%) Income <$20,000 2,119 (49.3%) 1,750 (82.6%) 369 (17.4%) .015 $20,000ormore 2,176 (50.7%) 1,734 (79.7%) 442 (20.3%) MilitaryStatus Active 2,518 (58.6%) 2,125 (84.4%) 393 (15.6%) <.001 Reserve 1,777 (41.4%) 1,359 (76.5%) 418 (23.5%) LengthofService<5months 2,684 (62.5%) 2,232 (83.2%) 452 (16.8%) 5 months-1year 964 (22.4%) 750 (77.8%) 214 (22.2%) <.001 >1year 647 (15.1%) 502 (77.6%) 145 (22.4%) Depression (BDI) 6.4 (6.6) 6.5 (6.7) 6.0 (6.1) .039 FearofPain (FPQ) 18.1 (5.9) 18.0 (5.9) 18.2 (5.6) .452 BackBeliefs (BBQ) 43.4 (7.1) 43.3 (7.0) 44.0 (7.4) .005 Anxiety (STAI) 36.0 (9.2) 36.2 (9.2) 35.2 (9.1) .004 PhysicalHealthStatus (PCSTotal) 53.4 (5.1) 53.4 (5.1) 53.5 (5.2) .400 MentalHealthStatus (MCSTotal) 49.2 (8.6) 49.1 (8.7) 49.6 (8.0) .099 SmokePriortoArmy Childs etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page6of11

PAGE 7

responseratescomparedtoanysinglestrategyalonebut needfurthertestingbefore firmrecommendationscan bemade. Amonghealthstatusfactors,aSoldier ssmokingstatuswasasignificantpredictorofresponseratestothe web-basedsurveillancesystem.Thosewhodidnot smokepriortoenteringtheArmyhada1.7timesodds ofresponsecomparedtothosewhosmokedpriorto enteringtheArmy(p<.001).ThismeansSoldierswho smokedhad42.0%loweroddsofresponsecomparedto thosewhodidnotsmoke.Itispossiblethatsmoking statusmayberelatedtoothermeasuresofhealth,yet smokingstillemergedasanindependentpredictorof responseratesdespitecontrollingforthesefactors.PerhapsSoldierswhosmokearelessinclinedtoappreciate theimportanceofhealth-relatedresearch.Thisfinding isparticularlyrelevantforthePOLMtrialbecause 35.8%ofSoldiersinthisstudyclassifiedthemselvesas smokers,definedasindividualswhohadsmokedatleast 100cigarettesintheirlifetime.ThePOLMtrialdataare consistentwitharecentstudydemonstratingthat32.2% ofmilitarypersonnelaresmokers[15].Incontrast, Table1Statisticalanalysisofweb-b asedrespondersandnon-respondersv iaduringfirstyearfollowupfromPOLM study (Continued)No 2,756 (64.2%) 2,151 (78.0%) 605 (22.0%) <.001 Yes 1,539 (35.8%) 1,333 (86.6%) 206 (13.4%) CompanyInstructor Alpha 621 (14.5%) 494 (14.2%) 127 (15.7%) Bravo 929 (21.6%) 766 (22.0%) 163 (20.1%) Charlie 607 (14.1%) 497 (14.3%) 110 (13.6%) Delta 957 (22.3%) 760 (21.8%) 197 (24.3%) .256 Echo 660 (15.4%) 531 (15.2%) 129 (15.9%) Foxtrot 521 (12.1%) 436 (12.5%) 85 (10.5%) BMI 24.8 (3.2) 24.8 (3.2) 24.6 (3.2) .027 PhysicalExamination No 3,924 (91.4%) 3,202 (81.6%) 722 (18.4%) .009 Yes 371 (8.6%) 282 (76.0%) 89 (24.0%) PsychosocialEducationalProgram (PSEP) No 2,301 (53.6%) 1,871 (81.3%) 430 (18.7%) .726 Yes 1,994 (46.4%) 1,613 (80.9%) 381 (19.1%) ExerciseGroup TEPonly 1,212 (28.2%) 990 (28.4%) 222 (27.4%) TEP+PSEP 945 (22.0%) 767 (22.0%) 178 (21.9%) .932 CSEPonly 1,089 (25.4%) 881 (25.3%) 208 (25.6%) CSEP+PSEP 1,049 (24.4%) 846 (24.3%) 203 (25.0%)POLM,PreventionofLowBackPainintheMilitary;BDI,BeckDepressionInventory;BBQ,BackBeliefsQuestionnaire;STAI,State-TraitAnxietyIndex;SF12,Medical OutcomesSurvey12-ItemShort-FormHealthSurvey;BMI,BodyMassIndex;PSEP,PsychosocialEducationalProgram.Thep-valuesarebasedont-testsorchisquaretest,asappropriate.Childs etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page7of11

PAGE 8

approximately21%ofthegeneralpublicamongasimilaragegrouparesmokers[15].Thisindicatesthat smokingstatusmayneedtobeconsideredduringstudy planning,particularlyforpoweringlargeclinicaltrialsin whichtheprimaryoutcomemaybeaninfrequent occurrenceorwhenperformingstudieswithahighpreponderanceofsmokers. Anothersignificantpredictorofresponseratesinthe POLMtrialwasmilitarystatus,definedaswhetherthe Soldierwasinan activeduty or reservist status. ActivedutySoldiershad15.6%responseratecompared to23.5%ofthoseinthereserves(<.001).Althoughthis considerationmayhavelimitedapplicabilitybeyondthe militarypopulation,thisdistinctionappearstobean importantconsiderationfordesigningtrialsthatinclude militarysubjects.Thereasonforthediscrepancyin responseratesbetweenactivedutyandreservistsis unclear;however,thereareseveralpossibleexplanations forthisfinding.Manyofthetrainingrequirementsfor reservistsarecompletedindividuallyonlineviaavariety ofdistance-basedtrainingplatformsgiventheirpart timestatusandgeographicseparationfromtheiractive dutyArmycounterparts.Asaresult,theincreasein theirresponseratescouldbepartiallyexplainedbytheir increasedfamiliaritywithonlinetraining.Although Table2Statisticallysignificantpredictorsofweb-basedresponsefromgeneralizedlinearmixedmodel*Variable OddsRatio (95%CI) P-value Age 1.05 (1.02;1.07) <.001 Race/Ethnicity AfricanAmericanvs.Caucasian.76 (.57;.99) .046 Education Highschoolorlowervs.Graduatedfromcollegeorhigher.54 (.40;.71) <.001 Somecollegevs.Graduatedfromcollegeorhigher.70 (.54;.91) .008 Militarystatus Activevs.Reserve.68 (.56;.81) <.001 SmokePriortoArmy Novs.Yes 1.69 (1.41;2.03) <.001 BMI Increasing1unit .97 (.94;1.00) .027 Physicalexamination Novs.Yes .70 (.54;.90) .006*Thefittedmodelincludedallpotentialpredictorsandotherpre-specifiedexplanatoryvariables(seetextfordetails);resultsforthestatisticallysignificant predictorsarereportedhere. POLM,PreventionofLowBackPainintheMilitary;BDI,BeckDepressionInventory;BBQ,BackBeliefsQuestionnaire;STAI,State-TraitAnxietyIndex;PCS,Physical ComponentSummaryoftheSF12,MCS,MentalComponentSummaryoftheSF12,MedicalOutcomesSurvey12-ItemShort-FormHealthSurvey;BMI,BodyMass Index;PSEP,PsychosocialEducationalProgram. Figure2 Monthlyresponserateduringthefirstyearfollowup. Childs etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page8of11

PAGE 9

purelyspeculative,perhapsreservistpersonnelalsotend tobemoreself-motivatedtocompletetrainingrequirementsbecausetheyaremoreaccustomedtonothaving significantday-to-dayoversightandaccountabilityfor completingtheirtrainingrequirements,whichisclosely alignedwiththemethodologyusedtoadministerthe web-basedfollowupforthePOLMstudy.Alternatively, activedutySoldierstendtocompletetrainingrequirementsingroupssettingswithinenvironmentsthathave moredirectmonitoringandaccountability.Thesedifferencesmayinfluencethisgrouptobelesslikelyto respondtothefollow-upsurveysintheabsenceof directaccountability. Inaddition,highereducationallevelswereassociated withincreasedresponseratesonthefollow-upsurvey. Specifically,20.8%ofSoldierswithatleastsomecollege educationrespondedcomparedto14.6%ofthosewho onlycompletedhighschool(p<.001).Collegegraduates hadthehighestresponseratesat30.9%.Individualswith ahighschooleducationorlowerwereonly.5timesas likelytorespondasthosewithsomecollege,whereas thosewithsomecollegeeducationwere.7timesas likelytorespondascollegegraduates.Theseresultsare notsurprisingsinceonemightsuspectthatindividuals withhigherlevelsofeducationmayhavemoreintrinsic motivationandaretheref oremorelikelytorespond [16].Itisalsopossiblethattheseindividualshaveabetterappreciationforthevalueofhealth-relatedresearch andimportanceofsubjectparticipation.Furthermore, theincreasedresponseratesamongSoldierwhohad completedatleastsomecollegemayberelatedto increasedcomputerliteracy,whichcouldcertainlyinfluenceresponseratesgiventheweb-basedplatformutilizedtoassessfollow-upinthePOLMtrial. Previousresearchhasdem onstratedthatincreased subjectattentionmayenhancefollow-upratesinclinical trials,regardlessofthefollow-upmechanismthatis used[4].ToexaminethepotentialforincreasedattentiontoenhanceresponseratesinthePOLMtrial,we examinedgroupandindividualizedattention.Subjects inthePSEPgroupwhoreceivedtheadditionalback educationclassinagroupsettingwereclassifiedashavingreceivedadditionalgroupattention,whereasSoldiers randomizedtoreceivethephysical/USIexamination wereclassifiedashavingreceivedadditionalindividual attention.Theresultsofthisstudyreinforceconclusions fromtheexistingliteraturethatincreasedattentionduringtrialsmayenhanceresponserates,evenwhenthe extraattentionisnotdirectlyrelatedtocompletingfollow-upprocedures.However,astatisticallysignificant enhancementinresponserateswasonlyobserved amongSoldierswhoreceived increasedindividualized attention.Forexample,Soldiersreceivingindividualized attentionhadresponseratesof24.0%comparedto 18.4%amongthosewhodidnot(p=.009).Incontrast, receivinggroupattentionwasnotassociatedwithsignificantimprovedresponserates.SoldiersreceivingPSEP hadresponseratesof19.1%comparedto18.7%among thosewhodidnot(Table1).Soldiersreceivingboth PSEPandUSIhadresponseratesof26.4%comparedto 21.4%amongthosereceivingUSIonly,butthisdifferencewasnotstatisticallysignificantduetosmallsample sizes. Ourresultsareincontrasttothefindingsfrompreviousstudiesthathavefoun dincreasedattentionin groupsettingstobeassociatedwithincreasedfollow-up rates[5].Onepossibleexplanationforthisdiscrepant findingisthatalargemajorityofthetrainingcompleted bySoldiersinthemilitaryisdoneingroupsettings.Thus itispossiblethatSoldiersinthePSEPgroupmayhave perceivedthebackeducationclassasanadditionaltrainingburden,asopposedtovalueaddedtrainingdesigned toimprovetheirabilitytocopewithbackpain.Additionally,thebackeducationclasswascompletedonaSaturdaymorningoutsideofthenormaltrainingsyllabus, whichcouldhavebeenperceivedinamorenegative light.Ontheotherhand,thephysicalexaminationswere substitutedforanothertrainingrequirementratherthan additive,increasingthelikelihoodthatSoldiersperceived receivingtheexaminationasa gooddeal becausethey wereexemptfromthatmorningsphysicaltraining.Additionally,individualizedatte ntionfromtheexamination mayhavepeakedtheSoldiersinterestandpersonal appreciationforthestudy,furtherbuildingrapport betweentheSoldiersandstudystaff,increasingtheir buy-in tothestudy.Designingtrialsthatincludeindividualizedattentionisanimportantconsiderationfor improvingresponseratesintrials,whichhelpsto improveprecisionoftheresultsandincreaseoverallgeneralizabilityofthefindings.However,moreattention mustbepaidtothetypeofattentionthatprovidesmaximalimprovementinresponserate,insteadofthe assumptionthatanyadditionalattentionisvalueadded. Severallimitationsforthisanalysisshouldbeconsidered. Despiteachievingstatisticalsignificance,it spossiblethat someofthefindingsmaybespurious,asevidencedbythe questionablemeaningfulnessoftheeffectsizesamong someofthesignificantfindings,predominantlyage,race/ ethnicity,andBMI.Theconfidenceintervalsoftheodds ratiosaccordingtoourdataapproximatedavalueof1.0, whichisequivalenttonoincreaseordecreaseinoddsof response,thusnegatingthepo tentialmeaningfulnessof thesefindings.Asanexample,ageemergedasasignificant predictorofresponserates,yetthemeanageamong responderswas22.8comparedto21.8yearsofageamong nonresponders,resultinginanoddsratioof1.1.Although thisresultwasstatisticallysignificant,oneyearinagedifferencedoesnotappeartobeamaterialfindingthatChilds etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page9of11

PAGE 10

mightinformthedesignoffuturetrials.Similarfindings wereobservedforbothrace/ethnicityandBMI.These smallbutstatisticallysignificanteffectscanlikelybeattributedtofactthattheoriginalPOLMstudywaspoweredon theprimaryaimofdetectingfutureepisodesofbackpain inthe2yearsfollowingcompletionoftraining.Thismay haveresultedinanincrease dchanceforTypeIerrorin thissecondaryanalysis. Thisstudyreportedpredictorsofresponsetoawebbasedsurveyusingadichotomousoutcometorepresent responserate.Thisdecisionwasmadebecausetheprimaryoutcomeofthetrialisadichotomousmeasure (occurrenceoflowbackpain)andwewantedtheseanalysestobeparallel.Ouradditionalanalysisshowedthat responseratessignificantlydecreasedovertime,which wasanexpectedoutcomethatistypicalinclinicaltrials. Also,itwouldbeinterestingtoassesswhetherinternet accesswasabarrierforsomeoftheSoldiersinthis study,inparticularthosewhoweredeployedinremote settingsaroundtheworld.However,thisinformation wasnotavailabletous,hencewecanonlyspeculate thatresponseratesmaybelowerforthoseSoldierswho didnotinternetaccessduringthefollow-upphaseof thestudy.Futurestudiesmi ghtalsoexaminewhether othercontemporarymethodsofcommunication(ie, SMStextmessaging,socialmedia,etc.)mightbemore effectivethanemailinsecuringfollow-up[17]. Anotherlimitationisthatthesubjectsinthistrial weremorehomogenouscomparedtothegeneralpopulation.ManyoftheSoldiers eatinghabits,activitylevels, andworkenvironmentsare nearlyidenticalbecauseof themorecontrolledenvironmentwithinthemilitary. Similarly,subjectsinthemilitaryhavebeenshownto havesimilarpsychologicalprofiles[18].Asaresult, thesefactorswouldnothavehadtheopportunityto competeforexplainingadditionalvarianceinthe responserates,evenifsome relationshipmightexistin amoreheterogeneousgeneralpublic.Thisispotentially thereasonwhythepsychologicalfactorsdidnotremain inthefinalregressionmodelaspredictorsofresponse rates.Finally,theparticipantsinthisstudyweretraining tobecomecombatmedics.Onemightexpectthattheir responseratewouldbehigherthanSoldiersinnonmedicalfields,similartohowmedicalpersonneldemonstratehigherresponseratescomparedtosubjectsinthe generalpopulation[19].Howeverwehadnocomparisongroupinthecurrentstudysowecanonlyspeculate thatthesefollowupratesmi ghtbehigherthanifthis studytargetedsubjectsinthegeneralpopulation.ConclusionsUnderstandingwhichfactorsareassociatedwith responseratescanhelptoimprovefollow-upby informingthedesignofclinicaltrialsandimprovingour understandingoftheeffectivenessofweb-basedsurveillancesystemsinlargeclinicaltrialsamongahighlygeographicallydispersedsubjec tpool.Additionalattention duringatrialmayimproveresponserates,butoptimal strategieshaveyettobeidentified.Futurestudiesshould considerhowtobestincorporateindividualizedattentionwithinclinicaltrialstoincreaseresponserates. Researchersshouldalsomonitorotherpredictorsoffollow-upratesidentifiedinthisanalysiswithintheirclinicaltrialssothatanydeferentialinfluenceofthese factorsinresponseratescanbeconsideredwheninterpretingtheresultsoftheirstudies.Listofabbreviationsused POLM:PreventionofLowBackPainintheMilitaryTrial;BDI:BeckDepression Inventory;BBQ:BackBeliefsQuestionnaire;STAI:State-TraitAnxietyIndex; SF12:MedicalOutcomesSurvey12-ItemShort-FormHealthSurvey;BMI: BodyMassIndex;CSEP:CoreStabilizationExerciseProgram;TEP:Traditional ExerciseProgram;AIT:AdvancedIndividualizedTraining;PSEP:Psychosocial EducationalProgram;USI:UltrasoundImaging;LBP:lowbackpain;RCT: randomizedcontrolledtrial;FPQ-III:FearofPainQuestionnaire,FABQ:FearavoidanceBeliefsQuestionnaire;PCS:PhysicalComponentSummaryscale; MCS:MentalComponentSummaryscale;ODI:OswestryDisabilityInventory; NPRS:NumericalPainRatingScale;SD:standarddeviation Acknowledgements CongressionallyDirectedPeerReviewedMedicalResearchProgram (#W81XWH-06-1-0564) ChristopherBarnes,YangLi,andErikHenricksonforcreationand managementofthewebsiteanddatabase. DonnaCunninghamforheradministrativeassistance. VariousstudentswithinthephysicaltherapyprogramsattheUniversityof Florida,U.S.Army-BaylorUniversityDoctoralPrograminPhysicalTherapy, UniversityofTexasHealthScienceCenteratSanAntonio,EastTennessee StateUniversity,UniversityofColoradoatDenverandHealthSciences Center,TexasStateUniversity,andUniversityofPugetSound Fundingattributionwillbeacknowledgedinthepaperusingthestatement: PublicationofthisarticlewasfundedinpartbytheUniversityofFlorida Open-AccessPublishingFund. Authordetails1USArmy-BaylorUniversityDoctoralPrograminPhysicalTherapy(MCCSHMT),ArmyMedicalDepartmentCenterandSchool,3151ScottRd.,Rm. 2307,FortSamHouston,TX78234,USA.2DepartmentofClinicalandHealth Psychology,POBox100165,HealthSciencesCenter,UniversityofFlorida, Gainesville,FL32610,USA.3DepartmentofBiostatistics,1329SW16thSt., Rm.5231,POBox100177,UniversityofFloridaGainesville,FL32610-0177, USA.4DepartmentofPhysicalTherapy,CenterforPainResearchand BehavioralTreatment,POBox100154,UniversityofFlorida,Gainesville,FL 32610-0154,USA. Authors contributions JDCcontributedbydevelopingstudydesign,interpretingdata,and composingthemanuscript.DSTparticipatedinstudydesignanddata interpretation.JJV,BFD,BJL,andGLHaidedindatainterpretationand manuscriptcomposition.SSWperformedthestatisticalanalysisandhelped indatainterpretation.SZGaidedinstudydesignanddatainterpretation.All authorsreadandapprovedthefinalmanuscript. Competinginterests Theauthorsdeclarethattheyhavenocompetinginterests. Received:18October2010Accepted:13June2011 Published:13June2011Childs etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page10of11

PAGE 11

References1.RobinerW: Enhancingadherenceinclinicalresearch. ContempClinTrials 2005, 26(1) :59-77. 2.StrausSE,GlasqiouP,HaynesRB: Evidence-BasedMedicine:HowtoPractice andTeachEBM ,32005,104-106. 3.FergussonD,AaronSD,GuyattG,HebertP: Post-randomisation exclusions:theintentiontotreatprincipleandexcludingpatientsfrom analysis. BMJ 2002, 325 :652-4. 4.DiasL,ThomasJ: Reasonsforhighretentioninpediatricclinicaltrials: comparisonofparticipantandstaffresponsesintheCorrectionof MyopiaEvaluationTrial. ClinTrials 2005, 2(5) :443-52. 5.LoftinWA,BarnettSK,BunnPS,SullivanP: Recruitmentandretentionof ruralAfricanAmericansindiabetesresearch:lessonslearned. Diabetes Educ 2000, 31(2) :251-9. 6.GlasgowRE,NelsonCC,KearneyKA,ReidR,RitzwollerDP,StrecherVJ, CouperMP,GreenB,WildenhausK: Reach,engagement,andretentionin anInternet-basedweightlossprograminamulti-siterandomized controlledtrial. JMedInternetRes 2007, 9(2) :e11. 7.JansonSL,AliotoME,BousheyHA: AsthmaClinicalTrialsNetwork. Attritionandretentionofethnicallydiversesubjectsinamulticenter randomizedcontrolledresearchtrial. ControlClinTrials 2001, 22(6) :236S-43S. 8.GeorgeSZ,ChildsJD,TeyhenDS,WuSS,WrightAC,DuganJL, RobinsonME: Rationale,design,andprotocolforthepreventionoflow backpaininthemilitary(POLM)trial(NCT00373009). BMCMusculoskelet Disord 2007, 8 :92. 9.CampbellMK,ElbourneDR,AltmanDG: CONSORTstatement:extensionto clusterrandomizedtrials. BMJ 2004, 328 :702-708. 10.MetzelthinSF,vanRossumE,deWitteLP,HendriksMR,KempenGI: The reductionofdisabilityincommunity-dwellingfrailolderpeople:design ofatwo-armclusterrandomizedcontrolledtrial. BMCPublicHealth 2010, 10(511) 11.ChildsJD,TeyhenDS,BenedictTM,MorrisJB,FortenberryAD, McQueenRM,PrestonJB,WrightAC,DuganJL,GeorgeSZ: Effectsofsituptrainingversuscorestabilizationexercisesonsit-upperformance. MedSciSportsExerc 2009, 41(11) :2072-83. 12.ChildsJD,TeyhenDS,CaseyPR,McCoy-SinghKA,FeldtmannAW, WrightAC,DuganJL,WuSS,GeorgeSZ: Effectsoftraditionalsit-up trainingversuscorestabilizationexercisesonshort-term musculoskeletalinjuriesinUSArmysoldiers:aclusterrandomizedtrial. PhysTher 2010, 90(10) :1404-12. 13.GeorgeSZ,FritzJM,BialoskyJE,DonaldDA: Theeffectofafearavoidance-basedphysicaltherapyinterventionforpatientswithacute lowbackpain:resultsofarandomizedclinicaltrial. Spine 2003, 28(23) :2551-60. 14.BuchbinderR,JolleyD,WyattM: 2001VolvoAwardWinnerinClinical Studies:EffectsofaCampaignonBackPainBeliefsandItsPotential InfluenceonManagementofLowBackPaininGeneralPractice. Spine 2001, 26(23) :2535-2542. 15.SmithEA,MaloneRE: EverywheretheSoldierWillBe":WartimeTobacco PromotionintheUSMilitary. AmericanJournalofPublicHealth 2009, 1595-1602. 16.BosnjakM,TutenTL: ClassifyingResponseBehaviorsinWeb-based Surveys. JournalofComputer-MediatedCommunication 2001, 6(3) 17.VilellaA,BayasJM,DiazMT,GuinovartC,DiezC,SimD,MuozA, CerezoJ: Theroleofmobilephonesinimprovingvaccinationratesin travelers. PrevMed 2004, 38(4) :503-9. 18.BrittTW,AdlerAB: Militarylife:Thepsychologyofservinginpeaceand combat. Militaryperformance;2006 1. 19.StellmanJM:In EncyclopediaofOccupationalHealthandSafety.VolumeIII.. 4 edition.OverviewofInfectiousDisease;1998.Pre-publicationhistory Thepre-publicationhistoryforthispapercanbeaccessedhere: http://www.biomedcentral.com/1471-2474/12/132/prepubdoi:10.1186/1471-2474-12-132 Citethisarticleas: Childs etal .: Predictorsofweb-basedfollow-up responseinthePreventionofLowBackPainintheMilitaryTrial (POLM). BMCMusculoskeletalDisorders 2011 12 :132. Submit your next manuscript to BioMed Central and take full advantage of: Convenient online submission Thorough peer review No space constraints or color gure charges Immediate publication on acceptance Inclusion in PubMed, CAS, Scopus and Google Scholar Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Childs etal BMCMusculoskeletalDisorders 2011, 12 :132 http://www.biomedcentral.com/1471-2474/12/132 Page11of11


University of Florida Home Page
© 2004 - 2011 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Powered by SobekCM