DTIC ADA514729: An Evidence-Based Public Health Approach to Injury Priorities and Prevention. Recommendations for the U.S. Military

0.23 MB·English

by  Defense Technical Information Center

An Evidence-Based Public Health
Approach to Injury Priorities
and Prevention
Recommendations for the U.S. Military
Bruce H. Jones, MD, MPH, Michelle Canham-Chervak, PhD, MPH, David A. Sleet, PhD
Abstract:InjuriesaretheleadingcauseofmorbidityandmortalityconfrontingU.S.militaryforces
inpeacetimeorcombatoperations.Notonlyareinjuriesthebiggesthealthproblemofthemilitary
services,theyarealsoacomplexproblem.Theleadingcausesofdeathsaredifferentfromthosethat
resultinhospitalization,whicharedifferentfromthosethatresultinoutpatientcare.Asaconse-
quence, it is not possible to focus on just one level of injury severity if the impact of injuries on
militarypersonnelistobereduced.Toeffectivelyreducetheimpactofaproblemasbigandcomplex
asinjuriesrequiresasystematicapproach.
The purpose of this paper is to: (1) review the steps of the public health process for injury
prevention;(2)reviewliteratureonevaluationofthescientifıcqualityandconsistencyofinformation
neededtomakedecisionsaboutpreventionpolicies,programs,andinterventions;and(3)summa-
rizecriteriaforsettingobjectiveinjurypreventionpriorities.Thereviewofthesetopicswillserveas
afoundationformakingrecommendationstoenhancetheeffectivenessofinjurypreventionefforts
inthemilitaryandsimilarlylargecommunities.Thispaperalsoservesasanintroductiontotheother
articlesinthissupplementtotheAmericanJournalofPreventiveMedicinethatillustratetherecom-
mendedsystematicapproach.
(AmJPrevMed2010;38(1S):S1–S10)PublishedbyElsevierInc.onbehalfofAmericanJournalofPreventive
Medicine
Introduction (OEF),nonbattleinjuriesaretheleadingcauseofhealth
conditions serious enough to require aero-medical
Injuries in the Military: a Large Problem
evacuationsoutofthetheaterofoperations.Nonbattle
I
njuries are the biggest health problem confronting injuriesaccountforabout35%ofsuchmedicalevacu-
U.S.militaryforcesinpeacetimeandcombatopera- ations,comparedto16%forbattleinjuriesand7%for
tions.1,2 Injuries result in over 1.8 million medical digestive diseases, the leading non-injury reason for
encounters annually across the military services and af- medical evacuation (Keith Hauret, U.S. Army Center
fectmorethan800,000individualservicemembers.3The for Health Promotion and Preventive Medicine, un-
secondleadingcauseofmedicalencounters,mentaldis- published data, 2008). Relative to other health prob-
orders,resultsinabout750,000encountersannually,af- lems,injurieshavethebiggestimpactonthehealthand
fectingabout190,000servicemembers.Historically,in- combatreadinessofmilitarypersonnel.
jurieshavebeenshowntobetheleadingcauseofdeaths, In the past, military surveillance of injuries and
disabilities, hospitalizations, and outpatient visits.3–7 “accidents” has focused primarily on fatalities, espe-
While battle injuries are the leading cause of death in cially motor vehicle and aviation fatalities. Since the
OperationsIraqiFreedom(OIF)andEnduringFreedom late 1990s, however, increasing attention has been di-
rected toward nonfatal injuries following establish-
ment of the Defense Medical Surveillance System in
FromtheInjuryPreventionProgram,U.S.ArmyCenterforHealthPromo-
1997.8Asaresultoftherecentfocusonnonfatalinju-
tionandPreventiveMedicine(Jones,Canham-Chervak),AberdeenProv-
ingGround,Maryland;andCentersforDiseaseControlandPrevention, ries,ithasbeenshownthatforeverynoncombatinjury
NationalCenterforInjuryPreventionandControl(Sleet),Atlanta,Georgia deathofamilitaryservicememberthereare33hospi-
Address correspondence and reprint requests to: Michelle Canham-
talizations and 3800 outpatient clinic visits for inju-
Chervak,PhD,MPH,U.S.ArmyCenterforHealthPromotionandPreven-
tive Medicine, ATTN: MCHB-TS-DI, Aberdeen Proving Ground MD ries.9 It has also been estimated that injuries result in
21010-5403.E-mail:[email protected].
about 25,000,000 days of limited duty among service
0749-3797/00/$17.00
doi:10.1016/j.amepre.2009.10.001 members annually.9 It is clear that injuries are a tre-
PublishedbyElsevierInc.onbehalfofAmericanJournalofPreventiveMedicine AmJPrevMed2010;38(1S)S1–S10 S1
Report Documentation Page Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it

does not display a currently valid OMB control number.
1. REPORT DATE 3. DATES COVERED
2009 2. REPORT TYPE 00-00-2009 to 00-00-2009
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
An Evidence-Based Public Health Approach to Injury Priorities and
5b. GRANT NUMBER
Prevention. Recommendations for the U.S. Military
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
U.S. Army Center for Health Promotion and Preventive Medicine,ATTN: REPORT NUMBER
MCHB-TS-DI,Aberdeen Proving Ground,MD,21010-5403
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT
NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
14. ABSTRACT
see report
15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF
ABSTRACT OF PAGES RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE Same as 10
unclassified unclassified unclassified Report (SAR)
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
S2 Jonesetal/AmJPrevMed2010;38(1S):S1–S10
mendous drain on military manpower during peace- torvehiclecrashes.15Furthermore,becauseanabundance
timeandtimesofarmedconflict. ofevidenceforfurtherpreventionofmotor-vehiclecrashes
isavailablefromestablishedgovernmentorganizationssuch
Injuries in the Military: a Complex Problem astheNationalHighwayTraffıcSafetyAdministration,the
CDC, the Insurance Institute for Highway Safety, and
In addition to being a large problem, injuries are also a
academic organizations, the military can adopt civilian
complex problem for the military. Among other com-
approaches already demonstrated to be effective.16 But
plexities,aswithciviliancommunities,theleadingcauses
formostinjuryproblems,evenlarge,potentiallyserious
ofinjuryvarywidelydependingonthelevelofseverityof
injuryproblemssuchasfalls,verylittlepreventioninfor-
injuries.Forinstance,historically,motorvehiclecrashes
mationisavailable.Researchneedstobeconductedbe-
have been the leading cause of unintentional, nonbattle
fore policies and programs are implemented when sys-
injurydeathsacrossallthemilitaryservices,accounting
tematic reviews determine that scientifıc information is
for 55% to 64% of all unintentional injury deaths,10 re-
scant and where gaps in knowledge about prevention
sulting in fıve to ten times as many deaths as the next
exist.
leading specifıc injury causes (drowning, fıres/burns, or
falls,dependingontheservice).5Ontheotherhand,the
Purpose
topthreecausesofinjuryhospitalizationofmilitaryper-
sonnel have been documented to be falls, athletics
Aswithanylargecommunityoroccupationalgroup,for
(sports),andmotorvehiclecrashes.6In2006,theleading
apublichealthproblemasbigandcomplexasinjuriesin
cause of injury hospitalizations for military personnel
themilitary,asystematicapproachtoplanningandset-
wasfallsandnearfalls(slipsandtrips),whichaccounted
tingprioritiesisneededforpreventionactivitiestosuc-
for 17.5% of such hospitalizations, followed by motor
ceed.Becauseresourcesforpreventionarescarce,apro-
vehicle mishaps at 15.4%, and sports and athletics at
cessisneededforsettingprioritiesthatidentifiesnotonly
13.1%(MichelleCanham-Chervak,U.S.ArmyCenterfor
effective countermeasures, but also those strategies that
Health Promotion and Preventive Medicine, unpub-
affectthehealthofthelargestnumberofpersonnelatthe
lisheddata,2007). lowestcosts.17,18Whenscientifıcevidenceisavailable,a
Thesamethreecausesofinjuries—falls,sports,andmo-
process for evaluating the quality of individual studies
torvehiclemishaps—arealsotheleadingcausesofmedical
and a mechanism for making recommendations for the
evacuationsfromoperationsinIraqandAfghanistan.2The
aggregate fındings on a particular injury problem are
only data readily available on causes of outpatient-treated
needed.Thepurposeofthispaperistooutlineasystem-
injuries come from army fıeld investigations. That data
aticprocessforidentifyingthelargest,mostsevereinjury
indicatedthatphysicaltraining(i.e.,exercise)isthelead-
problems for which effective prevention strategies have
ingcauseofoutpatientinjuryvisits,accountingfor25%
beendeveloped,andtargetingthoseproblemsforinter-
to 40% of such injuries.11,12 On the other hand, motor
vention. An evidence-based approach to identifying
vehicle mishaps account for less than 5% of all injuries
problemsforwhicheffectivepreventionstrategiesexistis
treatedinoutpatientclinicsandranknohigherthan7th
described and used as a foundation for making recom-
or 8th when compared to other causes. It is clear from
mendationsthatcouldbeappliedtothemilitaryorother
data such as these that if priorities for military injury
similarlylargepopulations.
prevention were based on fatalities, the major causes of
themajorityofinjuries,physicaltrainingandfalls,would
A Systematic Process for Injury Prevention
notbeaddressed.
Adding further to the complexity of the problem of The public health approach. Modern epidemiology
injuries,thecircumstancesofinjuriesresultingfromsim- hasshownthatinjuriesarenot“accidents”;theyarepre-
ilarcausescanbequitedifferent.Asanexample,fallscan dictableandpreventable.Acomprehensivepublichealth
occurfromstairs,ladders,andotherheights,andonlevel approach for the prevention of injuries has previously
surfaces during garrison or combat conditions.13 Like- been recommended for the military services.4,19–21 The
wise,athleticsresultinfrequent,sometimesserious,inju- primary recommendation entailed establishing the fıve
ries associated with a variety of sports occurring under functional elements of the public health approach for
variedcircumstances.2,14 injurypreventionlistedinTable1.Foralargecommunity
Withregardtoprevention,successhasbeenachieved or organization such as the military to successfully pre-
where attention has been focused and surveillance sys- ventinjuries,itisnecessaryforeachofthefıvefunctional
tems are in place. For example, just as in the civilian elements to be operating. Although the approach does
community,themilitaryhashadgreatsuccesspreventing notnecessarilyneedtobecarriedoutinsequentialorder,
injuriesanddeathsassociatedwithprivatelyownedmo- all of the steps are necessary in order to successfully
www.ajpm-online.net
Jonesetal/AmJPrevMed2010;38(1S):S1–S10 S3
Table 1. Functional steps of the public health approach to injury preventiona
Functionalstepof Descriptionoffunction
preventionprocess
1.Surveillance Medicalandsafetysurveillanceroutinelytracksfrequencies,rates,andtrendsininjuriesand
otherhealthproblems.Thedataareusedtoidentifyongoingandemergentproblemsand
tohelpsetpriorities.Surveillancecanalsohelpmonitorpreventionpolicyandprogram
effectiveness.
2.Researchandfield Researchandtosomeextentpublichealthfieldinvestigationsprovideinformationonthe
investigationsonrisk incidenceofinjuriesandotherhealthproblemsanddeterminecausesandriskfactorsfor
factorsandcauses healthproblems.
3.Researchoninterventions Researchmayalsoentailconductinginterventiontrials,bothrandomizedandnonrandomized,
todeterminewhatworkstopreventinjuriesandotherhealthproblems.Interventiontrials
provideinformationontheefficacyofpreventionstrategies.
4.Programandpolicy Policymakers,worksitesupervisors,militarycommanders,andotherauthoritiesdirect
implementation implementationofinjurypreventionandotherpublichealthpolicies,programs,and
strategiestoprotectpopulationsandcommunities.
5.Evaluationandmonitoring Oncepolicies,programs,andstrategiesareimplementedtheeffectivenessofthoseactivities
ofprogramsandpolicies shouldbeevaluatedtodeterminetheeffectivenessoftheactions.Surveillancedatacan
alsobeusedtomonitorongoingeffectiveness.
aAdaptedfromotherpublishedreports.16,19,20ThepublichealthapproachwasfirstdescribedbyMercyetal.56
preventinjuriesovertime.Greatstrideshavebeenmade the Guide to Clinical Preventive Services in 1989.28,29
since the initial recommendation of the fıve-step public WiththedevelopmentoftheGuidetoCommunityPre-
health approach to the Armed Forces Epidemiological ventive Services in 2000, that process was extended to
Boardin1996.20Routinemedicalsurveillanceofinjuries communitypublichealth.27–30Theprocessforidentify-
resulting in hospitalization of military service members ing successful evidence-based prevention strategies and
andalsothosetreatedinoutpatientclinicshasbeenim- setting public health and safety priorities has gained
plemented. Additionally, the means to evaluate public enoughcredencetorecommendwideimplementation.
healthpracticesimplementedtopreventinjurieshasbeen Thekeystepoftheevidence-basedprocessistheeval-
demonstrated.22,23Thestepsforwhichtheleastprogresshas uationofthequalityofindividualpreventionstudiesor
been made are research. While occasional ad hoc injury investigations(Table2).31,32Thenextstepoftheprocess
researchinitiativesarise,atthistimethereisnodedicated is development of recommendations for prevention
injurypreventionresearchobjectiveorprogramforthemil- based on the overall quality and consistency of the evi-
itary.Inaddition,injurycausecodingofoutpatientdatais dencesupportingorrefutingtheeffectivenessofpreven-
needed to prioritize and guide prevention efforts. Despite tion strategies.31,32 Earlier steps of the evidence-based
great progress, for injury prevention in the military to be decision-making process begin with defıning the prob-
effective,allofthestepsoftheprocessneedtobeimproved lemandsystematicreviewsofthescientifıcliterature,or
foreachoftheservices. meta-analyses. The fınal steps are to set priorities for
prevention and research based on the magnitude or se-
The evidence-based process. In addition to the fıve
verityofpublichealthproblems;thestrengthandconsis-
stepsofthepublichealthapproach,cost-effectiveinjury
tencyoftheevidencethateffectivesolutionsfortheprob-
anddiseasepreventionrequiresanevidence-basedmech-
lems exist; and a determination that the identifıed
anismforprioritizingpreventionactivitiesandallocation
preventionstrategiesarefeasibletoimplement.24,25,33
ofpublichealthandpreventionresources(Table2).Pri-
ority should be given to problems for which there is Surveillance and the evidence-based process. The
scientifıc evidence of effective prevention policies, pro- evidence-basedprocessbeginswithidentifıcationofthe
grams, or interventions.24–26 Great progress has been biggestandmostseverehealthproblemsaffectingacom-
made in evidence-based decision making in preventive munityorpopulation.34,35Healthandsafetysurveillance
medicine and public health over the last 20 years in the andsurveysarethelogicalmeansforidentifyingthemost
U.S.,startingwiththeU.S.PreventiveServicesTaskForce commonandsevereinjuryandotherhealthproblemsof
(USPSTF)inthelate1980s.27,28AsdescribedbyBrissand acommunity.Inthepast,thepublichealthimportanceof
McGinnis, the USPSTF fırst applied the evidence-based injuriesandotherhealthproblemshasbeenestablished
processtotheevaluationofclinicalpreventiveservicesin primarily using fatality data.36 In the military services,
January2010
S4 Jonesetal/AmJPrevMed2010;38(1S):S1–S10
Table 2. Steps of evidence-based public health decision-making process
Stepofprocess Descriptionofstep
1.Identificationofbiggestor Thefirststepoftheevidence-basedpublichealthprocessutilizesmedical,safety,andother
mostsevereproblems surveillanceandsurveydatasourcestoidentifycausesortypesofinjurywithhighratesor
indicatorsofseveritytotargetforpotentialprevention.
2.Searchforevidenceof Thesecondstepoftheprocessusesknowledgeofthemostsignificantinjuryproblems
effectiveprevention confrontingapopulationfromStep1tofocussystematicreviewsofthescientificliterature
onthoseproblemstodeterminewhatevidenceexistsfortheirprevention.
3.Evaluationofqualityof Thethirdstepoftheprocessevaluatesthequalityofindividualresearchstudiesusing
evidenceforprevention predeterminedcriteriatoassessstrengthsandweaknessesofdesign,execution,and
analysis.
4.Recommendationsbasedon Thefourthstepoftheprocessassessesthestrengthandconsistencyoftheoverall
strengthandconsistencyof evidencethatinterventionsworktopreventtheproblemsidentifiedasafoundationfor
evidence recommendations.Note:Noonestudydesignaddressesallthequestionsrequiring
answersabouteffectiveness,harms,andreal-worldfeasibility.Onestudyisnotsufficient
tomakeevidence-basedrecommendations.
5.Prioritizationof Thefifthstepappliespredeterminedcriteriatorankpreventionstrategiesforallocationof
interventions resourcesandimplementationbasedonthemagnitudeorseverityofaproblem,its
preventability(evidenceofeffectiveinterventions),andthefeasibilityofimplementation.
6.Identificationofresearch Thesixthandfinalstepoftheevidence-basedpreventionprocesscantakeplace
gaps concurrentlywiththefifth.Thisstepidentifiesgapsinknowledgeofwhatpreventsthe
mostsignificanthealthproblemsconfrontingapopulationandtargetsthemformore
research.
topprioritiesforinjurypreventionarestillpredicatedon processeshavebeenadoptedbyothergroupsandorgani-
theleadingcausesofdeaths—motorvehicleandaviation zations.31,41,42Mostrecently,asimilarprocesshasbeen
mishaps.Theprioritizationprocessshouldincludemea- adoptedbytheGuideforCommunityPreventiveSer-
suresnotjustofmortality,butalsomorbiditymeasures vices.28,30,43 What the Guide for Community Preven-
such as disabilities, hospital discharges, and visits for tiveServicesandothersuchevidence-basedprocesses
emergency and other outpatient treatment, among oth- haveincommonistheinitiationoftheprocesswitha
ers.9,30,37Usingfatalitydataforsettinginjuryprevention systematicreviewoftheliteratureusingawell-defıned,
prioritiescanbeparticularlymisleadingasthemostfre- pre-established approach. Once the most prevalent
quent causes of injuries do not necessarily result in and severe public health problems have been identi-
death.38Also,ashighlightedearlier,becausetheleading fıed,thesecondstepservestoidentifypotentialinter-
causes of injury deaths are different than the leading ventions/countermeasuresthathavebeenscientifıcally
causesofthemorenumerousnonfatalinjuries,reducing evaluatedandfoundeffective.
theleadingcausesofdeathsmayhavelittleimpactonthe Aftertheliteraturesearcheshavebeencompleted,the
overall burden of injuries on a population. As a conse- nextstepistoassessthequalityofthescienceforidenti-
quence,intheinitialphaseofidentifyingthemostimpor-
fıedstudiesandtocharacterizethehealthoutcomesasso-
tant injury problems of a community, both magnitude
ciatedwiththeinterventionsstudiedandthesizeofthe
andseverityofinjuriesshouldbeconsideredusingfatal-
effectsduetotheinterventions.Thereisagrowingcon-
ity,disability,hospitalization,andoutpatientdata.
sensusthat,whetheroneisassessingtheeffectivenessofa
Systematic reviews and the quality of evidence. In medication,clinicalpreventiveservices,oracommunity
targetingandconductingeffectiveinjuryprevention,itis preventiveservice,morethanjustthebenefıcialeffectof
notenoughtoknowwhatthebiggestinjuryproblemsare. aninterventionmustbeconsideredwhenmakingrecom-
It is also necessary to know which are preventable. A mendations for prevention. The potential harms of an
process for identifying and evaluating the evidence for interventionmustbeassessedtoo.31,32,41,42Toworkwell,
whatworkstopreventinjuriesisalsoessential(Table2). theprocessmusthave(1)astandardizedmethodoffınd-
Anumberofapproacheshavebeenestablishedforeval- ingevidencetoassess,(2)astandardsetofconsiderations
uating the effectiveness of interventions to treat or pre- in evaluating the quality of individual scientifıc studies,
vent health problems. The best known is the process and(3)amethodofarrivingatacompositescoreforeach
established by the U.S. Preventive Services Task Force, study on a particular prevention strategy that can be
which has been well-described elsewhere.32,39,40 Similar comparedtootherstudies.Suchsystematicreviews(lit-
www.ajpm-online.net
Jonesetal/AmJPrevMed2010;38(1S):S1–S10 S5
Table 3. Levels of recommendations for injury prevention strategiesa
Recommendation Reasonsforrecommendation
Stronglyrecommend Gooddataoneffectivenessexist,someofithighquality,andfindingsacrossstudiesareconsistent.
Effectsizesaresubstantial.
Recommend Atleastfairevidenceofinterventioneffectivenessexists,andfindingsofeffectivenessaremostly
consistent.Effectsizesmaybemodest.
Norecommendation Benefitsandharmstooclosetomakearecommendation.
Recommendagainst Datafromstudiesofadequatesamplesizetoshowinterventioneffectsofmodestmagnitudedonot
use indicatethattheinterventioniseffectiveorthattheharmsoftheinterventionoutweighthebenefits.
Insufficientevidence Insufficiencyofevidencemayresultfromacompletelackofdata,fewstudies,orinconsistencyof
results.
Expertopinion Intheabsenceofscientificevidenceontheeffectivenessofinterventions(e.g.,insufficientevidence),
utilizationofexpertopinionorconsensusopinionsonrecommendationsforpreventionmaybe
warrantedforurgenthealthproblems.Whenthebasisforaninterventionisexpertopinion,the
interventionimplementedshouldberigorouslyevaluatedandcloselymonitoredforeffectiveness.
aAdaptedfromotherpublishedreports.31–33
erature reviews coupled with quality assessments) are implementation.26,29,31,32,41,42 In making recommen-
nowviewedasacriticalpartofthepublichealthdecision- dations,informationonthequalityandconsistencyof
makingprocess.44,45 evidence that a strategy works must be balanced
Completing the literature search and evaluation pro- againstpotentialharmsandthecostsimplementation
cessistimeconsumingandrigorous.Asaconsequence, mayimpose.
Harris et al.,32 in writing about the USPSTF, state that Inadditiontoweighingeffectiveness,harms,andcosts
“limited resources and time requires compromises in intheprocessofmakingpublichealthrecommendations,
the intensity of reviews.... One strategy is...topic considerationshouldbegiventowhatneedstobedonein
prioritization....Anotherstrategy...istofocusthereview situationswhenanurgentpublichealthorsafetyproblem
onthequestionsandevidencemostcriticaltomakingrec- exists,butthereisinsuffıcientscientifıcevidenceofinter-
ommendations.” This type of process has been applied to ventionsthatworktopreventtheproblem.41,46Inarating
settingprioritiesformilitaryinjuryprevention,andanex- schemeforrecommendationsbytheStrengthofRecom-
peditedprocessformorerapidevaluationbypublichealth mendationTaxonomy(SORT),Ebell31indicatesthatthe
and safety organizations has been recommended.9,33 To lowest level of evidence is consensus or usual practices
facilitatemorerapidtransmissionofevidence-basedin- (i.e., expert opinion). In his discussion of the USPSTF,
jurypreventioninformationtodecisionmakersandpoli- Harris warns that if evidence is deemed insuffıcient to
cymakers, an expert military panel established an expe- makearecommendationtoprovidepreventiveservices,
dited process for scientifıc study evaluation.33 Thus
then decision makers must rely on factors other than
several approaches can be used to facilitate more rapid science.32 Claxton41 addresses the issue directly, stating
transmissionofinformationfromevaluatorstodecision
thatamethodisneededforacquiring“judgmentsfrom
makers,includingfocusingthesystematicreviewprocess
expertswhennoevidenceisavailable.”
onthemostimportantinterventionsandexpeditingthe
Withtheaboveconsiderationsinmind,asetofratings
reviewprocessitself.
forrecommendationshasbeenmadethatissuitablefor
Systematicreviewsandthestrengthandconsistency use in a large population that frequently confronts new
of evidence. Following the identifıcation of evidence andsignifıcantpublichealthandsafetyproblems,many
sources (studies) and the evaluation of the quality of of which may be of an urgent nature (Table 3). The
individualstudies,thenextstepistranslationofthebody proposedcategoriesofrecommendationsareasfollows:
of the evidence as a whole into recommendations. This stronglyrecommend,recommend,norecommendation,
stepoftheevidence-basedprocessentailstheassessment insuffıcientevidence,andexpertopinion.Toaccommo-
oftheoverallstrengthandconsistencyoftheevidencefor date the inevitable situation in which the military and
a particular intervention.29,31,32 As mentioned earlier, otherorganizationsneedrecommendationsforinterven-
not only must the process evaluate how effective an tion when no evidence of preventability exists, an addi-
interventionstrategyisatpreventinginjuriesorother tionalcategoryforexpertopinionwasadded.Itshouldbe
healthoutcomesbutalsoifanyharmsmightarisefrom noted, however, that when interventions without clear
January2010
S6 Jonesetal/AmJPrevMed2010;38(1S):S1–S10
evidenceofeffectivenessareimplemented,theyshouldbe that “policymakers should be skeptical of evidence de-
carefullyevaluated. rivedfromasinglestudy.”Thefactthatnoonestudyor
study type is an adequate basis for policy and public
Studydesignandtrade-offsinvalidityofevidence. An
health decisions argues for greater use of systematic re-
issue of importance to the process of evaluating the
viewsthatmakeuseofallavailableevidenceregardlessof
quality and strength of evidence supporting preven-
studytype,publishedorunpublished.46–48Weighingev-
tionisthatofstudytypeordesign.Inthepast,theonly
idence from multiple studies of different types provides a
acceptablestandardwasanRCT.Thereis,however,a
greateropportunitytobalanceeffectivenessagainstharms
growing consensus that RCTs are not necessarily the
andcosts.WhileRCTsprovideconfıdencethatstudyfınd-
only acceptable evidence or even the gold standard,
ingsarenotduetochanceorbiasresultingfrominadequate
especiallyfornonpharmaceutical,nonclinical,community-
studydesign(i.e.,internalvalidity),othertypesofwell-de-
basedinterventions.26,32,40,42,47,48EventheUSPSTFaccepts
signedstudiesmaybemorepractical,lesscostly,andmore
studytypesotherthanRCTsasevidence.32Thisconsen-
generalizable(i.e.,externallyvalid).
sushasarisenfromthegrowingawarenessoftheshort-
comingsofRCTsindocumentingtheharmsoradverse Criteria for setting priorities. The need for a mecha-
outcomes of interventions, the inability to provide an nismforsettingprioritiesforallocationofresourcesfor
accurateassessmentofthemagnitudeofhealthbenefıts prevention of injuries and disease is widely recog-
of an intervention in a non-experimental setting, and nized.17,24–26,35,37,41,49–52 The Institute of Medicine’s
impracticalityofconductingrandomizedstudiesinmany CommitteeonInjuryPreventionandControlstatedthat,
circumstances. While RCTs may have greater internal forinjuryprevention,“Whatevertheoveralllevelofpub-
validity, they lack the external validity offered by other licinvestment...prioritiesforresearchandsocialaction
studydesigns.Atkinsetal.46specifıcallynotethatRCTs mustbeset.Thechallengefacingthefıeldisdeveloping
“maynotgiveanaccuratepictureoftheimpactofapolicy criteriaforsettingthesepriorities.”35
decisionunderreal-worldconditions.” Anumberofapproachestoestablishingpreventionpri-
Because of time and funding constraints, RCTs fre- oritiesemployingavarietyofcriteriahavebeensuggested,
quentlyemployintermediateoutcomes.Thisisnotaccept- includingusingtheburdenofdisease,37considerationofthe
able.Severalauthorscautionthat,indeterminingtheeffec- magnitude,severity,andcostsofproblems49,53orthesein
combination with preventability/effectiveness17 and feasi-
tiveness of an intervention, it is essential that the health
outcomesofinterestbeassessed,26,46unlessthelinkbetween bilityfactorssuchasacceptability,availableresources,and
legalauthority.24,25,50,54Runyan16,25andFowler24havede-
anintermediateoutcomeandtheoccurrenceofthehealth
scribedmorecomprehensivecriteriathanothersandcre-
outcomeofinterestiswellestablished,suchasthelinkbe-
ateddecisionmatricesspecifıcallyforsettinginjurypreven-
tweenseatbeltuseanddecreasedriskoftraffıcfatalities.
tion priorities. Their criteria for setting injury prevention
Avarietyofvalidalternative,nonrandomizedstudyde-
prioritiescanbeaggregatedintoseveraloverarchingcatego-
signsmaybeusefulinassessingtheeffectivenessofinterven-
riesincludingeffectiveness(preventabilityconsideringben-
tionsintendedtobeimplementedonawidescaleincom-
efıtsandharms),costs,feasibility(funding,infrastructure,
munities and populations. These study types include
personnel,legalauthority),acceptability(socialandpoliti-
nonrandomizedprospectiveandretrospectivecohortstud-
cal),andsustainability.
ies,pre–poststudies,time-series,case–control,andnatural
experiments,aswellasotherquasi-experimentalorobser- Processforsettingprioritiesbasedoncriteria. Once
vationaltypesofstudies.26,47,48Suchstudiesmaybebetterat criteria are established for setting priorities, a process
determiningtheeffıcacyofinterventionsinreal-worldcom- forrankingthepotentialprioritiesmustbedeveloped.
munity settings than RCTs. Teutsch notes that data from Fowler24 described a qualitative process using rankings
RCTs is scarce for many interventions, so investigators of “high,” “moderate,” and “low” for each criterion in a
should not be deterred from using other more practical decisionmatrix.Runyan25suggestedthateitherqualita-
studydesigns.26Grouprandomizedtrialsprovideanother tiveorquantitativemethodsforapplyingcriteriacanbe
alternative to RCTs.47,48 What is important in choosing a employedtosetpriorities,providedthatdecisionmakers
studydesignisthatitbeabletodeterminewhetherimple- considerthemostimportantfactorslikelytodetermine
mentationofaninterventionchangedtheincidenceofthe policy,program,orinterventionsuccess.Claxtonetal.41
healthoutcomeofinterest. expressed a preference for a quantitative approach to
Furthermore,Merceretal.47state“Noonestudyestab- suchdecisionmaking,stating:“Inparticular,evidentiary
lishescausality.”Rather,consistentoutcomesfrommul- criteriaarenottiedformallyandquantitativelytobene-
tiplestudiesmakeabetterfoundationforevidence-based fıts,risks,andcostsassociatedwithaninterventionandas
health and public health policy. Atkins et al.46 caution a result do not maximize health benefıts.” As a conse-
www.ajpm-online.net
Jonesetal/AmJPrevMed2010;38(1S):S1–S10 S7
Table 4. Priority-setting criteria employed by military injury prevention working groups
Criterion Scoring
A.Programorpolicyisconsistentwithmissionoftheworkinggroup/organization IfYES—Continuewithscoring.
IfNO—Stophere.
B.Importanceofproblemtopublichealthorworkplaceproductivity (10points;1(cid:1)low,10(cid:1)high)
Considerations:
1. Magnitudeoftheproblem(e.g.frequency,incidence)
2. Severityofproblem(e.g.,degreeofeffectonpersonnelhealthandperformance)
3. Costoftheproblem(e.g.,training,property,andpersonnelcosts)
4. Sizeofpopulationatrisk
5. Degreeofconcern(e.g.,leadershipconcern,publicandServicememberconcern,
visibilityofproblem)
C.Preventabilityofproblem (10points;1(cid:1)low,10(cid:1)high)
Considerations:
1. Cause(s)areidentifiable
2. Riskfactorsaremodifiable
3. Provenpreventionstrategiesthatreduceexistinginjuryratesexist
4. Preventionstrategiesthatreduceexistinginjuryratescanbedesigned
5. Effectsize
D.Feasibilityofprogramorpolicy (10points;1(cid:1)low,10(cid:1)high)
Considerations:
1. Existenceofinfrastructuretosupportimplementationandsustainabilityofthe
programorpolicy(e.g.,medicalstaffandfacilities,safetystaffandresources,
manpoweravailability)
2. Perceivedadequacyoffundingtosupportimplementationandsustainability
3. Authoritytoimplementandsustaintheprogramorpolicyisheldorobtainableby
theimplementingorganization(s)
4. Programorpolicywillnotundermineessentialmissions
5. Politicalandculturalacceptabilityofprogramorpolicy
6. Accountabilityandresponsibilityforimplementationandsustainabilityexistsorcan
beestablished
E.Timeliness (5points;1(cid:1)low,5(cid:1)high)
Considerations:
1. Timetoimplementation
2. Timetoresults
F.Evaluationofprogramorpolicy (5points;1(cid:1)low,5(cid:1)high)
Considerations:
1. Abilitytoevaluateeffectsofprogramorpolicy(i.e.,availabilityofmetrics)
2. Benefitsofprogramorpolicyoutweighthecostsofimplementationand
sustainability
3. Collateralbenefitsasaresultofimplementation(i.e.,increasedreadiness,
decreasedattrition,anddecreasedotherhealthproblems)
quence of this kind of thinking, several military injury Asimilarsetofcriteriatothoseforsettingprevention
preventionworkinggroupshaveadoptedaquantitative prioritiescanbeusedtosetresearchpriorities.In2002,a
approach to setting priorities.9,33,55 The criteria em- preliminarysetofsuchcriteriawasdevelopedbytheU.S.
ployedbythesemilitaryworkinggroupsandthescoring ArmyCenterforHealthPromotionandPreventiveMed-
ofcriteriaareprovidedinTable4.Thesynthesisofcrite- icineandtheJohnsHopkinsCenterforInjuryResearch
riaforpreventionprioritiesenumeratedbythesemilitary andPrevention(Table5).55Insettinginjuryprevention
workinggroupsactuallybeganattheCDCin2000with andotherpublichealthpreventionpriorities,aprimary
workstartedtherebyoneoftheauthors(BHJ). criterionisscientifıcevidencethateffectiveinterventions
January2010
S8 Jonesetal/AmJPrevMed2010;38(1S):S1–S10
Table 5. Suggested priority-setting criteria for military necessary. While the infrastructure needed to address
research
eachstepofthepublichealthapproachexistswithinthe
DepartmentofDefense,italsoneedstobestrengthened.
A.Programorpolicyisconsistentwithmissionofthe
workinggroup/organization Currently, the strongest element of the U.S. military’s
injury prevention system is medical and safety surveil-
B.Importanceofproblemtoforcehealthandreadiness
lance.Whileimprovementssuchastheadditionofout-
Considerations:
patient injury cause coding are needed, these data are
1. Magnitudeandseverityofproblem currentlyadequatetoidentifysignifıcantmilitaryinjury
2. Highcostsofproblem
problemsandtomonitorchangesinratesofinjuriesover
3. Sizeand/orvulnerabilityofpopulationatrisk
time following implementation of interventions, pro-
4. Degreeofconcern(commandorpublic)
5. Gapsinknowledgeofeffectivepreventionstrategies,or grams,orpolicies.
modifiablecausesandriskfactorsexist In regard to getting prevention information to those
6. Militaryuniqueness
who need it, the infrastructure for disseminating injury
C.Potentialvalueofresearch preventioninformationisreadilyavailablethroughthemil-
Considerations: itaryservicesafetycentersandchainsofcommand.Like-
wise, once the effectiveness and feasibility of an injury
1. Cross-cutting(cutsacrosstypesofinjury)
prevention strategy has been demonstrated, the infra-
2. Likelihoodofidentifyingdiscretemodifiablerisk
factors structure and mechanisms exist within the military to
3. Demonstratedpreventabilityincivilianpopulation rapidlyimplementthestrategy.
D.Feasibilityofresearchprogramorproject The ability to evaluate programs and document suc-
cess at the installation and service level has also been
Considerations:
demonstrated,22,23,33buthumanandfıscalresourcesfor
1. Publichealthandmedicalinfrastructureexiststo
thisessentialpublichealthservicearecurrentlylimited.
supportresearchefforts
2. Researchpartnersexist In addition, despite the fact that an evidence-based ap-
3. Technologicfeasibilityofdoingresearch(abilityto proach to making recommendations for injury preven-
collectdata)
tionandsettingprioritieshasbeenemployedinthepast,
4. Adequacyofresources
the process has not been institutionalized in the
military.9,33
Finally, the weakest step in the process for the De-
exist.Ontheotherhand,aprimarycriterionforsetting partmentofDefenseisresearch.Eventhoughinjuries
research priorities is that adequate evidence of effective arethesinglebiggesthealthproblemofallofthemili-
interventionsdoesnotexist.Thus,themostobviousevi- tary services, there is no specifıc injury-prevention
dence that research is needed is when a big problem is scientifıcortechnicalobjectivetowhichresourcescan
identifıed, but no research is found to support preven- be routinely applied, with the exception of occasional
tion. Insuffıcient evidence is also another indicator that monies directed to address ad hoc problems, such as
research is needed, as multiple studies are typically re- traumatic brain injuries. Without an injury-research
quiredtoestablishthevalidityofasuggestedprevention programguidedbydata-drivenprioritiesandwithout
strategy. The most effıcient way to set both prevention identifıcation of gaps in prevention knowledge,
and research priorities may be to conduct both at the progress with military injury prevention will cease
sametime,becausetheprocessforidentifyingimportant,
onceoff-the-shelfsolutionshavebeenexhausted.
preventable injury or other health problems will be the
Withtheaboveconsiderationsinmind,thefollow-
same,andthecriteriaforsettingprioritieswillbesimilar
ingrecommendationsaremadetoestablishacompre-
withtheexceptionthattherewillbeevidencetosupport
hensive, evidence-based approach to military injury
interventionsintheformerandnotforthelatter.
prevention:
● Use readily available military surveillance databases
Conclusion and Recommendations
(deaths, disabilities, hospitalization, outpatient, and
Theproblemofinjuriesforthemilitaryislargeenoughand safety)toidentifythelargestandmostseveremilitary
costlyenoughtowarrantthetimeandresourcesneededto injuryproblems.4,19
conductasystematic,data-driven,andevidence-basedpro- ● Commission systematic reviews of prevention and
cessofdefıningpreventionpriorities.Toeffectivelyim- safetyliteraturetodeterminewhathasbeenshownto
plement such a process, all the functional capabilities/ work for prevention of the largest, most serious mili-
stepsofthepublichealthapproachlistedinTable1are taryinjuryproblems.
www.ajpm-online.net
Jonesetal/AmJPrevMed2010;38(1S):S1–S10 S9
● Establish committees of medical and safety subject References
matterexpertstoroutinelyaccessandsetprioritiesfor
both injury prevention research and program/policy 1. BrundageJF,JohnsonKE,LangeJL,RubertoneMV.Compar-
ingthepopulationhealthimpactsofmedicalconditionsusing
implementation.
routinely collected health care utilization data: nature and
● Implementoradaptprovenpreventionstrategiesina
sourcesofvariability.MilMed2006;171(10):937–42.
prioritizedmanner.
2. U.S. Army Medical Surveillance Activity. Frequencies and
● Evaluateeffectivenessofallimplementedpolicies,pro- characteristicsofmedicalevacuationsofsoldiersbyair(with
grams,andinterventions/countermeasures. emphasisonnonbattleinjuries),OperationsEnduringFree-
● For the largest, most serious injury problems, where dom/Iraqi Freedom (OEF/OIF), January–November 2003.
MedicalSurveillanceMonthlyReport2004;10(3):8–12.
informationoneffectivenessdoesnotexistorisinsuf-
3. U.S.ArmyMedicalSurveillanceActivity.Estimatesofabsolute
fıcienttomakeevidence-basedrecommendations,em-
and relative health care burdens attributable to various ill-
panel subject matter experts to make best-practice
nessesandinjuries,U.S.ArmedForces,2005.MedicalSurveil-
recommendations. lanceMonthlyReport2006;12(3):2–23.
● Forthelargest,mostseriousinjuryproblemsforwhich 4. JonesBH,PerrottaDM,Canham-ChervakML,NeeMA,Brund-
evidenceforpreventiondoesnotexistorisinadequate, ageJF.Injuriesinthemilitary:areviewandcommentaryfocused
onprevention.AmJPrevMed2000;18(3S):S71–84.
commissionresearch.
5. PowellKE,FingerhutLA,BrancheCM,PerrottaDM.Deathsdue
● Establishroutinechannelsfordisseminatinginformation
toinjuryinthemilitary.AmJPrevMed2000;18(3S):S26–32.
fromeachofthestepsofthepublichealthandevidence- 6. SmithGS,DannenbergAL,AmorosoPJ.Hospitalizationdue
baseddecision-makingprocessestoensurethatkeystake- to injuries in the military. Evaluation of current data and
holdersreceivetheinformationandtrainingnecessaryto recommendations on their use for injury prevention. Am J
PrevMed2000;18(3S):S41–53.
effectivelyreducetheimpactofinjuriesonthehealthand
7. SongerTJ,LaPorteRE.Disabilitiesduetoinjuryinthemili-
readinessofmilitarypersonnel.
tary.AmJPrevMed2000;18(3S):S33–40.
● Work with partners in injury prevention (e.g., the
8. Rubertone MV, Brundage JF. The Defense Medical Surveil-
CDC,CDCInjuryResearchCenters,NIH,Agencyfor lanceSystemandtheDepartmentofDefenseserumreposi-
HealthcareResearchandQuality)tocompletethesys- tory:glimpsesofthefutureofpublichealthsurveillance.AmJ
tematicreviewsandresearchneededtoadvanceinjury PublicHealth2002;92(12):1900–4.
9. Ruscio B, Smith J, Amoroso P, et al. DoD Military Injury
prevention.
PreventionPrioritiesWorkingGroup:leadinginjuries,causes,
In following these recommendations, the military andmitigationrecommendations.Washington:Offıceofthe
Assistant Secretary of Defense for Health Affairs, Clinical
servicesandDepartmentofDefensehaveanopportu-
andProgramPolicy,2006.www.stormingmedia.us/75/7528/
nity not only to substantially reduce the incidence of
A752854.html.
injuries to service members, but also to establish a
10. WortleyWH,FeiersteinG,LillibridgeA,ParliR,MangusG,
modelforpublichealthpracticeformilitaryandcivil- Seibert JF. Chapter 3. Fatal and nonfatal accidents/mishaps:
iancommunities. safetycenterdata.MilMed1999;164(8S):S1–88.
Thepapersthatfollowthisintroductiontothesystem- 11. Knapik JJ, Darakjy S, Jones SB, et al. Injuries and physical
fıtness before and after deployment by the 10th Mountain
atic evidence-based process of injury prevention illus-
Division to Afghanistan for Operation Enduring Freedom.
trate:(1)howprioritiescanbesetusingmilitarysurveil-
AberdeenProvingGroundMD:U.S.ArmyCenterforHealth
lanceandresearchdata;(2)howsurveillancedatacanbe
Promotion and Preventive Medicine, 2007. Report No. 12-
usedtodefıneandmonitorinjuryproblemsforthemili- MA-05SD-07.
tary; (3) how systematic reviews can be employed to 12. KnapikJJ,JonesSB,DarakjyS,etal.Injuryratesandinjuryrisk
providemilitary-relevantinformationonwhatworksto factorsamongU.S.Armywheelvehiclemechanics.MilMed
2007;172(9):988–96.
prevent injuries; and (4) what have been some of the
13. SenierL,BellNS,YoreMM,AmorosoPJ.Hospitalizationsfor
results of military injury–prevention program evalua-
fall-relatedinjuriesamongactive-dutyArmysoldiers,1980–
tionsandresearch. 1998.Work2002;18(2):161–70.
14. LauderTD,BakerSP,SmithGS,LincolnAE.Sportsandphys-
ical training injury hospitalizations in the army. Am J Prev
Theviewsexpressedhereinaretheviewsoftheauthor(s)
Med2000;18(3S):S118–28.
anddonotreflecttheoffıcialpolicyoftheDepartmentof
15. KrullAR,JonesBH,DellingerAM,YoreMM,AmorosoPJ.
the Army, the Department of Defense, or the U.S.
Motor vehicle fatalities among men in the U.S. Army from
Government. 1980to1997.MilMed2004;169(11):926–31.
Nofınancialdisclosureswerereportedbytheauthors 16. Doll LS, Bonzo SE, Mercy JA, Sleet DA, eds. Handbook
ofthispaper. of injury and violence prevention. New York: Springer,
2007.
January2010