Contra Costa Emergency Medical Services EMS System Modernization Study conducted by  Fitch and Associates  EMCC Workshop Literature   October, 2013  A compilation of evidence‐based and best  practice literature on Emergency Medical  Services A D S “L -A ” C ESCRIPTIVE TUDY OF THE IFT SSIST ALL DavidC.Cone,MD,JohnAhern,ChristopherH.Lee,MD,MS,DorothyBaker,PhD, TerrenceMurphy,PhD,SandyBogucki,MD,PhD ABSTRACT evaluation.Keywords: emergencymedicalservices;geri- atrics;accidentalfalls Introduction.Responsesfor“liftassists”(LAs)arecommon PREHOSPITALEMERGENCYCARE2013;17:51–56 in many emergency medical services (EMS) systems, and result when a person dials 9-1-1 because of an inability to get up, is subsequently determined to be uninjured, and INTRODUCTION is not transported for further medical attention. Although LAsofteninvolverecurrentcallsandaregenerallynotreim- When elderly or disabled persons fall or are unable 13 bursable,littleisknownoftheiroperationaleffectsonEMS to move from an undesirable position to a preferred 05/ systems.WehypothesizedthatLAspresentanopportunity one, they may call 9-1-1 for assistance. Often there is on 09/ finojruerayrlpierervterenatitomnenintteorfvseunbttiloen-osninseatmpoepduiclaalticoonnadtithioignhsarinsdk no perceived injury or illness, so these individuals do c not want medical treatment or transport to the hospi- n forfalls.Objectives.ToquantifyLAcallsinonecommunity, US I describe EMS returns to the same address within 30 days tal. They simply want responders to physically help ma following an index LA call, and characterize utilization of them back to a bed, chair, or wheelchair. These calls ar EMS by LA patients.Methods. Data from the computer- arelocallyknownas“liftassists.” h s P aideddispatch(CAD)systemofasuburbanfire-basedEMS It is likely that in some percentage of cases, a lift- a ell system were retrospectively reviewed. All LAs from 2004 assist call represents a “sentinel event” or a marker Ast to 2009 were identified using “exit codes” transmitted by ofdeteriorationinfunctionofthepatient.1 Thiscould by paramedicsaftereachcall.Thenumberandnatureofreturn be due to an unapparent medical condition such as com nly. visits to the same address within 30 days were examined. a urinary tract infection or pneumonia, or could in- nformahealthcare.For personal use o RsTttLhopwAeeotsos.hnue-Festlaheotssdsira.rdmd(2F4rs1ree.84oso%asmfodetfsdoh2(frte01heE07seL44MssAeaotShdccrrcaiedonutlrluucresirrgsden(sh7dee2nvs2w6;ti0ss;3i0)i26tt9hst.65,o,i.n%8ti5h%t)33e;w0)5r5ewd6ad3saweifyropfeeseforrtetasehofsnt1eioetb,r0rnlae8eetdt7hu-tdteorhLrneiiAcnrscdodsamreleoelsxs--f. dcAaaolsildcsgzainhsattieeatilinmovacseeesnrcei’oanwsfpdaa(sciocsttrieaitvagyoisetneriegeooslofariotngoefrgsddatedaltaoioulcayakrcltlhhiodvrrfoei)intncisglisioc..ncIeditaicislonesuauplsdpeh,pyaosslsuircotcahhalneoardsr- m i Anecdotally, emergency medical services (EMS) o parepatientageandsexwiththoseassociatedwiththeini- d fr tialLA,revealingthat85%ofreturnvisitswerelikelyforthe providers report frequently returning to the same de samepatients.Ofthese,38.5%wereforanotherLA/refusal address in the days, weeks, or even hours following a o nl of transport, 8.2% for falls and other injuries, and 47.3% an initial lift assist, either for another lift assist, or for w o for medical complaints. Hospital transport was required in a more serious problem such as a fall with injury, or D e 55.5% of these return visits. The EMS crews averaged 21.5 a medical emergency, often resulting in transport to Car minutes out of service per LA call. Conclusion. Lift-assist the emergency department (ED). In the case of older erg calls are associated with substantial subsequent utilization patients,thereisconsiderableexpenseassociatedwith m sp E oinftEerMveSn,taionndss.hBoauselddtornigoguerrdfaaltlap,trheveseentciaolnlsamnadyobteheerarslayfeinty- atriptotheEDbecausethesepatientsreceiveagreater o number of diagnostic tests, remain in the ED longer, Preh dicators of medical problems that require more aggressive have higher ED charges,2 and are more likely to be admitted to acute or intensive care units.3 Perhaps the lift-assist call can be used to trigger interventions to help prevent the “next call,” thereby improving Received January 10, 2012 from the Section of EMS, Department the quality of care and reducing the use of financial of Emergency Medicine (DCC, JA, CHL, SB) and the Section of resources. Screening of elders has been advocated Geriatrics, Department of Internal Medicine (TM, DB), Yale by national organizations as a key component of UniversitySchoolofMedicine,NewHaven,Connecticut.Revision receivedMay29,2012;acceptedforpublicationJune14,2012. high-quality geriatric emergency care;4 perhaps such screening should begin in the field, at the patient’s Theauthorsreportnoconflictsofinterest. home. Reprintsarenotavailable. Sinceatleastonevehicleandcrewmustrespondto each lift-assist call in order to locate and assess the Address correspondence to: David C. Cone, MD, Yale Univer- patient and resolve the problem, lift assists consume sity School of Medicine, Section of Emergency Medicine, 464 Congress Avenue, Suite 260, New Haven, CT 06519. e-mail: EMS resources. In some cases, multiple vehicles and [email protected] personnel must respond to these calls. Examples in- doi:10.3109/10903127.2012.717168 cludebariatricpatientswhocannotbesafelyliftedby 51 52 PREHOSPITALEMERGENCYCARE JANUARY/MARCH2013 VOLUME17/NUMBER1 twoproviders,orpatientsbehindlockeddoorsrequir- dence, and describe other characteristics of lift-assist ing forced entry. There is generally no way for EMS calls. agenciestorecoverthecostsassociatedwithlift-assist calls, since with few exceptions, EMS efforts are re- StudyProtocol imbursed only when patients are transported to the hospital. TheCADsystemdatabasewasqueriedfortherecords While there are published studies of the EMS re- ofall lift-assist calls (as identified byexit code) occur- sponse to falls,5,6 including analysis of patients who ringbetween2004and2009,inclusive.Thesedatawere are not transported, no research describing lift-assist exported to the investigators in a single Excel spread- calls, the patients involved, or the workload for the sheet (Microsoft Inc., Redmond, WA). The system be- EMS system have been reported to date. We hypoth- gan using ePCRs in July 2007; no PCRs were avail- esized that lift assists present an opportunity for ear- ableforthestudyperiodpriortoJuly2007.However, lier treatment of subtle-onset medical conditions and dispatchers sometimes enter free-text “CAD notes” injurypreventioninterventionsinapopulationathigh such as “80 y/o male unable to get off floor.” The 13 risk for falls. As the first step of a multiphase project, combination of address, age, and sex can be consid- 5/ 0 we conducted a descriptive study of lift-assist calls in ered identifying information; therefore, relevant data 9/ n 0 one EMS jurisdiction. Specific objectives were: 1) to privacy practices were followed for the entire study o c quantifythefrequencyanddemographicsoflift-assist period. n S I callsinonesuburbanEMSsystem,2)todescribeEMS ThespreadsheetswerescannedmanuallyandbyEx- U a returns to the same address within 30 days following cel utilities for data inconsistencies, such as calls with m ar an index lift-assist call, and 3) to grossly estimate the the lift-assist exit code that also reported transport h P resources used by the EMS system by these lift-assist timesanddestinationcodesindicatingthatthepatient s a ell responses. was taken by ambulance to a hospital, or calls with Ast multiple exit codes. In each case found to have such by METHODS internalconflicts,theePCRwasreviewedtodetermine com nly. StudySettingandPopulation whetherthecallwasindeedaliftassist,inwhichcase mahealthcare.personal use o Ta29hs,e0h0os0rteualdinnydeawtonawsanrceoawnoidtfhuacabteopduotpi2nu2laBstrqiaounnafrooerfdma,piClpeosron(xn5i7emcktaimtceu2l)ty,. totinhhrceauitddnceaeantxlaltpswlwtahaiensareteedoxincciclncuulcudrordenedesddifs.artoetIfnmacadintdehsreeepPssesCteurRssdisoywt.feaIdpnsuauabfdnltiedacrvistraieeoitlvntaii,bnealwgelsl,, nforFor Thetownhasahigherpopulationofresidentsoverthe such as businesses or parks were excluded from the m i ageof65years(19.9%)thaneithertherestofthestate study. Thus, all lift-assist calls included in this study o d fr (14.2%)ortheUnitedStatesasawhole(13%).7 involvedresponsestoprivateresidences. de The town has a fire department–based EMS system a o wnl that provides both basic and advanced life support DataAnalysis o first response and transport, and responds to approx- D e imately 4,000 EMS calls and 1,600 fire calls per year. The lift-assist calls for each year and for the entire ar C Thepublicsafetyansweringpoint,includingEMSdis- study interval were recorded both as total numbers g er patch,ismaintainedbythetown’spolicedepartment, andasproportionsofallEMScalls.Poissonregressions m p E usingtheMedicalPriorityDispatchSystem(SaltLake with robust estimation of standard errors were used s o City,UT). toassessforlineartrendsincallvolumeandlift-assist h e Pr At the conclusion of each call, in addition to com- volume over time, and the Cochrane-Armitage trend pletingelectronicpatientcarereport(ePCR)documen- testwasused toassess fortrendsintheproportion of tation,theEMScrewverballytransmitsan“exitcode” the department’s call volume that consisted of lift as- tothedispatcherbyradio.Theexitcodeindicatesthe sistsovertime.Theproportionofresponsestounique natureoftheincidentactuallyfoundbytheproviders addressesforliftassistswasalsocomputed. onscene,andisdistinctfromthe“complaint”thatwas WeexaminedallEMSresponsestothesameaddress described by the caller at the time of the 9–1-1 call. within30daysafteranindexlift-assistcalland,where These codes are included in the permanent record of possible,basedonmanuallyenterednotesintheCAD each call in the computer-aided dispatch (CAD) sys- file and ePCR data, assessed whether the return visit tem.“Liftassist”isoneof126possibleEMSexitcodes was for the same patient. In the absence of a criterion inthissystem. standardintheliterature,30dayswaschosenbycon- Branford was chosen for the study for two reasons. sensus of the authors as a time frame within which it First,ithasasingleEMSprovideragency,ratherthan seemed likely that return calls could be related to the one for first response and another for transport. Sec- indexlift-assistcall.Iftheindexlift-assistcallandsub- ond, inclusion of the exit codes in the CAD database sequentreturnvisitwithin30dayscontainedmatching provides a reliable way to identify cases, report inci- names, or in the absence of names (when only CAD Coneetal. LIFT-ASSISTCALLS 53 notes were available), matching age and sex data, it TABLE1. LiftAssistsandTotalEmergencyMedical wasanalyzedasbeingforthesamepatient.Forexam- ServicesCallsperYear,2004–2009 ple,intheyearspriortotheuseofePCRs,aCADnote Percentageof of“75y/oM”fortheindexcallcombinedwithaCAD CallsThatWere note of “75 year old male” for the return visit a few Year LiftAssists∗ EMSCalls† LiftAssists‡ days later was considered a return visit for the same 2004 152 3,367 4.5 patient. 2005 175 3,640 4.8 Caseswithconflictingageandsexdata,suchas“78 2006 129 3,829 3.4 2007 195 3,807 5.1 F” and a “69 y/o male,” or incomplete data, such as 2008 218 3,891 5.6 age but no sex, were excluded from the return-visit 2009 218 4,062 5.4 analysis.Acallatthesameaddresswithin30daysaf- TOTAL 1,087 22,596 4.8 teranindexcallwasconsideredareturn-visitcall,not ∗Increaseyeartoyearissignificant,p=0.017. another index call, and return visits were not them- †Increaseyeartoyearissignificant,p<0.0001. selvessearchedforadditionalreturncalls.Likethein- ‡Increaseyeartoyearissignificant,p=0.003. 13 dex lift-assist calls, the natures ofall return calls were 5/ 0 derived from the exit codes in the CAD file and on n 09/ theePCR,andrecordedbybroadcategory.Thesecate- Over the study period, there were 563 EMS re- o sponses to the same address within 30 days after an c gorieswereformedbygroupingtheexitcodesintoone S In of the following: cardiac, gastrointestinal, pain with- index lift-assist call. For 214 of these 563 returns to U the same address (38%), the name or the age and sex a out history of trauma, respiratory, altered mental sta- m of the patient for both initial lift assist and return re- ar tus, miscellaneous medical, cardiac arrest, falls and h sponse could be determined; CAD notes did not pro- P other injuries, burns, other, and refusal of treatment/ as videthesedatafortheremaining349(62%)duringthe ell transport. Ast Resource use was estimated by examining the years when ePCRs were not yet in use. Of these re- by CAD database for the vehicles that responded (e.g., sponses to the same address within 30 days after an mahealthcare.com personal use only. ppatnhmoaluertoisacfiumreenedewntdgtsoihicfwneteaeidmmriecsbeopnumasolpttapcienfihanceentedryotoohnnpfalltttyuhh,tseehorerrdeyesepspwppaouorenantrysmeseeaetcvd(ohfairitcieohlafmae)bmtliatembhnfueedolrattihnatmhecnyeee- istctneoaardmlaqlesuehxfaooplrisrtafpitttl-hieiateeasrnlsastin(as6(mgt56ec8.e3a%[%lpIlQ,aff1tReei8me]m2n7aatr8lel(e–e5t;9;u50mmr.)5n,ee%saddn)wiiaadreennr1seua0ag1algteteeotdr8f8i0bti4hnuyeytetsereaedaarnrsrte,ssot,pIuQtoihrnrRne-t nforFor othercall). 72–88) (Fig. 1). The 182 returns for the same patients m i followed157indexliftassists;thenaturesofthese182 ed fro HumanSubjectsApproval rTewtuernntyc-athllrseeariendsehxolwifntaisnsiFstisgu(1r4e.62%, )anredsuvlaterdyiwnimdeolrye. d oa than one return for the same patient within 30 days nl The study was submitted to the Yale University Hu- w (21 with two return calls and two with three return o man Investigation Committee, which deemed it ex- D e emptfromfullrevieworapproval. calls). Car Crews averaged 21.5 minutes (standard deviation Emerg RESULTS [toSDa]n±d1a7rrmiviinnugteosn,rtahnegsec3entoe2o1f8lmifti-nausstiesst)craelslps.onTdhiensge sp responses totaled over 366 hours of responder out-of- o There were 1,087 lift-assist calls over the six-year pe- h e service time over the duration of the study. One hun- Pr riod, accounting for 4.8% of all EMS incidents. As dred twelve (10.3%) of the lift-assist calls required, in shown in Table 1, both the total number of EMS calls (p < 0.0001) and the number of lift assists (p = 0.017) additiontotheambulance,additionalfiredepartment apparatus.Inmost(79%)ofthesecases,theadditional increasedoverthestudyyears,asdidtheproportionof thedepartment’scallsthatwereliftassists(p=0.003). unit(s) were dispatched simultaneously with the am- bulance, suggesting overtriage by the dispatcher. In a This suggests that lift assists represent a substantial minorityofthesecases,theadditionalunit(s)weredis- andgrowingproblemintermsofincidenceinthecom- patched after arrival of the ambulance on scene, indi- munityandinrelativeprevalencewithinEMScallvol- cating that the ambulance crew requested additional ume. help, e.g., for lifting a bariatric patient or for forcing The 1,087 lift assist responses were made to 535 entry. unique private addresses. One hundred seventy-four addresses (32.5%) accounted for 726 lift-assist re- sponses (66.8%). Almost all of these calls were to ad- DISCUSSION dresseswherethepatientswerelivingindependently; there were two lift-assist responses to addresses with ThisdescriptivestudyusedCADdatatoevaluatethe full-timemedicalstaffonduty(hospice). frequency and system burden of lift-assist calls in a 54 PREHOSPITALEMERGENCYCARE JANUARY/MARCH2013 VOLUME17/NUMBER1 Total EMS calls: 22596 Li(cid:2) assist calls: 1087 No return Return visit in visit in 30 30 days: 563 days: 524 3 1 5/ 0 9/ 0 on Names or Names or nc age/sex not age/sex S I available: 349 available: 214 U a m ar h P s a ell Not the same Same Ast pa(cid:3)ent: 32 pa(cid:3)ent:182 y b m y. conl mahealthcare.personal use o transN8po1otr ted: Trans1p0o1rted: nforFor m i o FIGURE1. Patientflowdiagramofemergencymedicalservices(EMS)calls. d fr e ad suburban setting. We found that nearly 5% of EMS Our findings are similar to those of the London o nl incidents in this jurisdiction are for lift assists, that Ambulance Service, which found that 47% of elderly w Do EMS returned to the same address within the next (age ≥65 years) fall patients not transported to the are 30daysfairlyoften,andthatabouthalfofsubsequent hospital summoned EMS again within two weeks, g C calls for the same patient were for conditions that re- with47%ofthesepatientscallingmorethanonce.5 In mer quiredambulancetransporttothehospital.Significant the subset of these repeat-call patients transported to E p municipalresourcesarebeingexpendedonthesenon- studyhospitalsinLondon,41%wereadmitted—arate hos reimbursablecalls. 4.7 times that of the general ≥65-year-old population. e Pr FIGURE 2. Natureof182returncallsconfirmedtobeforthesamepatient.AMS=alteredmentalstatus;GI=gastrointestinal;Misc.Med= miscellaneousmedical;Resp=respiratory;w/o=without. Coneetal. LIFT-ASSISTCALLS 55 This London study examined patients to whom EMS of Health Services,” to “ensure that the care provided was dispatched for an incident (complaint) code by EMS does not occur in isolation, and that positive of “fall,” and is thus different from our population effectsareenhancedbylinkagewithothercommunity identifiedonthebasisofaconfirmedliftassist. healthresourcesandintegrationwithinthehealthcare The similar high rate and varied nature of return system.”14 The Agenda’s authors, specifically citing a calls we observed in the 30 days following an index role for EMS in prevention of falls, noted that EMS lift-assistcallsuggestthatneithertheEMSsystemnor must “expand its public health role and develop on- the patients are best served bysimply returning apa- goingrelationshipswithcommunitypublichealthand tient to bed or chair without additional medical and socialservicesresources”inordertomaximizetheben- social services evaluation. Halter et al. in the United efits to patients, both present and future. We believe Kingdom have recently developed a decision tool to thatitisfeasible,practical,andimportantforfieldper- assistEMSproviderswhoareconsideringwhetherel- sonneltoserveastheeyesandearsofthecommunity derlypatientsshouldbetransportedtothehospitalun- healthsysteminapreventiverole. der these circumstances.8 Likewise, our findings sug- 5/13 gest that lift-assist calls could offer an opportunity to LIMITATIONS 0 screenpatientsandinitiateinterventionstomitigateat 9/ n 0 leastsomefutureclinicalproblems. The study was conducted in a single EMS system, o c EMS screening of elderly patients has been shown in a suburban New England town, and may not be S In to be feasible,9–11 and there are several studies sug- generalizable to other types of systems (e.g., third- U a gesting that such intervention can be beneficial. A service) or other types of municipalities (e.g., large m ar study conducted in Nottinghamshire, UK, examined citiesorruralareas).Inparticular,thistownhasahigh h P patientsovertheageof60yearswhohadsummoned proportion of residents who are 65 years of age and s a ell EMS because of a fall, but were not transported to older.AlloftheEMSagenciesinourarea(covering11 Ast the hospital.12 The patients were randomized to re- other cities and towns) report having lift-assist calls, by ceive either standard care or an individualized multi- but response patterns differ somewhat from town to com nly. disciplinaryfall-interventionprograminvolvinghome town, and data analysis would likely reveal different mahealthcare.personal use o vtnchiieaseelirdtassepebdirysvtrisenc,feueasrr,nsreaasalns,sdspteohscysothmsmiecemanplutrtinhomieftaryhar-ypocemicsnatetsre,ehraapngzhdraoyrousdcipscc,iuaswpneasiottshiiroonsnaoass--l. esaosbAtmliemseradeptleieaesstcdi.reionbntesdm,,aatnnhdueatChlAoresDceonCrdoAitneDsgnwoofetpreeastnitehontattaivwnafeoilrraembalaevtiafooilnr- nforFor Theincidenceratesofself-reportedfallsperyearwere bythedispatchers.Furthermore,theEMSsystemstud- m i 3.46 in the intervention group and 7.68 in the control ied did not begin using ePCRs until July 2007. Thus, o d fr group(incidencerateratio0.45,95%confidenceinter- age and sex data were available for only a subset of de val [CI] 0.35 to 0.58, p < 0.001), and the number of patients,somatchinginitialliftassistswithreturncalls a o nl times an ambulance was called because of a fall was was frequently not possible during early study years. w o significantly lower among the intervention group (in- We cannot determine whether this subset is represen- D e cidencerateratio0.60,95%CI0.40to0.92,p=0.018). tative of the entire sample; the analysis of the return ar C The intervention group also demonstrated increased calls must therefore be viewed with caution. How- g er levels of activities of daily living and reduced fear of ever, our data seem reasonable given the anecdotal m p E falling. reports from the paramedics regarding the frequency os A similar study from Rochester, New York, found withwhichtheyreturntothesamepatientinthedays h e Pr that volunteer EMS personnel in a rural community and weeks afteralift-assist call.Our dataalso appear could effectively screen elderly patients for fall risk, reasonable in the context of at least one prior study depression, and medication-management problems, of patients who summoned EMS at least once in the andreferthemtocasemanagersforin-homevisitsfor yearafteranindexEMScall,showingthatelderly(age further assessment and intervention.13 This parallels 65 years or greater) “repeaters” accounted for 18% of theapproachthatourprogramisnowtaking,withfire elderly EMS patients and 40% of elderly EMS trans- department paramedics providing their screening in- ports to the ED.15 Similarly, Shah et al. found that in formation to involve primary care physicians and the astudyofruralEMSuse,25%to30%ofcasesinvolv- localvisitingnurseservice. ingpatientsaged65yearsandoverwererepeatcallers, This move away from reflexive, response-based in- suggestingthatafairlyhighproportionofreturnvisits terventionandtowardproactive,preventivemeasures toelderlyEMSpatientsisnotunusual.16 reflectsgeneralinterestintheEMScommunityinmov- Ourstudydidnotinvolvereviewofhospitalrecords, ingtowardatrulyintegrated,community-basedhealth so while we know how many of the repeat callers system, as suggested by the 1998 Emergency Medical weretransportedtothehospital,wedonotknowhow Services Agenda for the Future.14 The first EMS sys- many of these were admitted to the hospital. Such tem attribute discussed in the Agenda is “Integration data would have allowed comparison with the 41% 56 PREHOSPITALEMERGENCYCARE JANUARY/MARCH2013 VOLUME17/NUMBER1 admission rate for the repeat-call patients in the Lon- 6. ClawsonJ,OlolaC,ScottG,etal.Associationbetweenpatient donstudy.17 unconsciousornotalertconditionsandcardiacarrestorhigh- Finally, while we were able to calculate resource acuityoutcomeswithintheMedicalPriorityDispatchSystem “Falls”protocol.PrehospDisasterMed.2010;25:302–8. use in terms of man-hours, we did not conduct a ro- 7. UnitedStatesCensusBureau.Connecticut.Availableat:http:// bustcostanalysisofthefinancialburdenthatlift-assist quickfacts.census.gov/qfd/states/09000.html.AccessedAugust calls impose on EMS systems. A cost analysis, using 8,2012. the methods described by Lerner and colleagues,18,19 8. HalterM,VernonS,SnooksH,etal.Complexityofthedecision- wouldprovidemoreaccuratedata,butwasbeyondthe making process of ambulance staff for assessment and refer- ralofolderpeoplewhohavefallen:aqualitativestudy.Emerg scopeofthisstudy. MedJ.2011;28:44–50. 9. Weiss SJ, Chong R, Ong M, Ernst AA, Balash M. Emergency CONCLUSIONS medicalservicesscreeningofelderlyfallsinthehome.Prehosp EmergCare.2003;7:79–84. In this descriptive study of a mid-sized EMS system, 10. Shah MN, Caprio TV, Swanson P, et al. A novel emer- gency medical services-based program to identify and assist lift-assistcallsarecommon,appeartobeincreasingin olderadultsinaruralcommunity.JAmGeriatrSoc.2010;58: 5/13 frequency, and create a substantial number of nonre- 2205–11. 0 imbursedresponsesforthemunicipalEMSsystem.We 11. ShahMN,LernerEB,ChiumentoS,DavisEA.Anevaluation 9/ n 0 found that EMS personnel often return to care for the ofparamedics’abilitytoscreenolderadultsduringemergency c o same patients within 30 days, by which time approx- responses.PrehospEmergCare.2004;8:298–303. n 12. Logan PA, Coupland CAC, Gladman JRF, et al. Community US I imately half require transport to an ED. These find- falls prevention for people who call an emergency ambu- a ingsrequirefurtherstudy;however,theysuggestthat lance after a fall: randomised controlled trial. BMJ. 2010;340: m ar it may be feasible to use the lift-assist call as the trig- c2102. h P ger for additional assessment and intervention in an 13. Shah MN, Clarkson L, Lerner EB, Fairbanks RJ, McCann s Astella etuffronrtEtMoSpcreavllesn,atnmdeadviocaidl dasesteorciioartaetdiosnu,ffienrjiunrgy,,danisdabriel-- Rto, SpcrhomneoidteerthSeMh.eAalnthemoferoglednecryamduedltisc.alJ sAermvicGeseripartorgrSaomc. com by nly. ity,andcosts. 14. 2FD0ue0tlu6b;rr5ei4d::gw9e5h6Te–Rr6e2,w.BaeilaeryeB..,.CwhhewereJLwJer,weatnatl.toEMbeS.PArgeehnodspafEomrethrge care.use o References 15. CWaeries.s1S9J9,8E;r2n:1s–t1A2.A,MillerP,RussellS.RepeatEMStransports althnal amongelderlyemergencydepartmentpatients.PrehospEmerg nformaheFor perso 1. ppWraiotlimbeenprttsS?eTmA,ceBargdlaeEnndmcyaerdMgeM,paGerdtem.rs2eo0nn0t6Lv;1Wi3s:i6.ts8D0a–on2ed.safdumncitsisoinonalindeoclldineer 16. CgSheanarehcy.M2m0N0e2,d;6Sic:w6a–la1ns0es.rovnicPe,sRuasjeasbeykaorladnerKa,dDuoltzsieirnAa.rRueraplecaotmemmeur-- m i 2. Singal BM, Hedges JR, Rousseau EW, et al. Geriatric patient nity: impact on research methods and study length. Prehosp o ed fr eymouenrggeenrcpyatvieisnittss..APanrntIE:mCeormgpMaerids.o1n9o92f;v2i1s:i8t0s2b–y7.geriatricand 17. ESnmoeorkgsCHaAre,.H20a0lt9e;r13M:1,7C3–lo8s.eJC,CheungWY,MooreF,Roberts d a 3. BaumSA,RubensteinLZ.Oldpeopleintheemergencyroom: SE.Emergencycareofolderpeoplewhofall:amissedoppor- o nl age-relateddifferencesinemergencydepartmentuseandcare. tunity.QualSafHealthCare.2006;15:390–2. w o JAmGeriatrSoc.1987;35:398–404. 18. LernerEB,GarrisonHG,NicholG,etal.Aneconomictoolkit D e 4. Carpenter CR, Heard K, Wilber S, et al. Research priorities foridentifyingthecostofEMSsystems:detailedmethodology Car forhigh-qualitygeriatricemergencycare:medicationmanage- oftheEMSCostAnalysisProject(EMSCAP).AcadEmergMed. erg ment, screening, and prevention and functional assessment. 2012;19:210–6. m AcadEmergMed.2011;18:644–54. 19. Lerner EB, Nichol G, Spaite DW, Garrison HG, Maio RF. E p 5. Snooks HA, Halter M, Close JCT, Cheung W-Y, Moore F, A comprehensive framework for determining the cost of s ho RobertsSE.Emergencycareofolderpeoplewhofall:amissed an emergency medical services system. Ann Emerg Med. Pre opportunity.QualSafHealthCare.2006;15:390–2. 2007;49:304–13. Building the evidence base in pre-hospital urgent and emergency care A review of research evidence and priorities for future research by the University of Sheffield Medical Care Research Unit Research funded by the Department of Health Policy Research Programme (cid:1) Building the evidence base in pre-hospital urgent and emergency care A review of research evidence and priorities for future research by the University of Sheffield Medical Care Research Unit Janette Turner Evidence reviewers Mike Bjarkoy Patricia Coleman Steve Goodacre Emma Knowles Suzanne Mason Jon Nichol Alicia O’Cathain Colin O’Keeffe Janette Turner Richard Wilson