Street Medicine: A Program Evaluation

Georgia State University Georgia State University

ScholarWorks @ Georgia State University ScholarWorks @ Georgia State University

Public Health Capstone Projects School of Public Health

1-6-2017

Street Medicine: A Program Evaluation Street Medicine: A Program Evaluation

Ariel L. Edwards Georgia State University

Follow this and additional works at: https://scholarworks.gsu.edu/iph_capstone

Recommended Citation Recommended Citation Edwards, Ariel L., "Street Medicine: A Program Evaluation." , Georgia State University, 2017. doi: https://doi.org/10.57709/9452085

This Capstone Project is brought to you for free and open access by the School of Public Health at ScholarWorks @ Georgia State University. It has been accepted for inclusion in Public Health Capstone Projects by an authorized administrator of ScholarWorks @ Georgia State University. For more information, please contact [email protected].

StreetMedicine 1

ABSTRACT

STREETMEDICINE:APROGRAMEVALUATIONBy

ARIELL.EDWARDS

DECEMBER8,2016

INTRODUCTION:Homelessindividualshavepoorerhealthoutcomesthantheirhousedcounterparts,yettherearemanybarrierstoreceivingconsistentmedicalandbehavioralhealthcare.StreetMedicineisamethodofhealthcaredeliveryinwhichamultidisciplinarygroupofhealthcareprovidersbringhealthcaretopeoplelivingonthestreets.StreetMedicinecouldbeapromisingsolutiontomeetingtheunmethealthneedsofpeopleexperiencinghomelessness. AIM:ThepurposeoftheevaluationwastoexaminetheeffectivenessofMercyCare’sStreetMedicineprogramatengagingtheirhomelesspatientsinconsistentcareanddecreasingthenumberofpatientsthatseekunnecessaryhospitalservices. METHODS:Aretroactivemedicalrecordsreviewwasconductedfor284patients.Ofthe284patients,26patientshadtobeexcludedfromdataanalysisbecausethepatientshadnotbeenseenduringtheStreetMedicineroundsandarenotconsideredStreetMedicinepatients.Allofthedatathatwereextractedwerecollectedusingacomputer-baseddatacollectionform.Thedatawasanalyzedusingfrequencytablesandt-testsinSAS. RESULTS:Resultssuggestthat54.26%oftheStreetMedicinepatientsareconnectedtoMercyCarethroughStreetMedicineand/orclinicvisitsandengaginginconsistentprimarycareandbehavioralhealthcareservices.TheaveragenumberoftotalStreetMedicineencountersperpatientwaslowerthantheaveragenumberofclinicencounters.ForbothStreetMedicineandclinicencounters,thepatientsaccessedmoreprimarycareservicesthanbehavioralhealthservicesandcasemanagementservices.TherewasnosignificantdifferenceinEDvisitsbeforeoraftertheinitialMercyCareencounter.Theaveragenumberoftotalhospitaladmissiondayssignificantlyincreasedfrom0.98daysbeforethepatient’sinitialMercyCareencounterto1.84daysaftertheinitialMercyCareencounter. DISCUSSION:TheStreetMedicineprogrammaybeapromisingsolutionforgettingandkeepingpeopleexperiencinghomelessnessengagedinhealthcareanddecreasingthenumberofpatientsthatturntothehospitalforavoidableandcostlyhealthcareservices.TheStreetMedicineprogrameliminatesmanyofthebarrierstocarethatpeopleexperiencinghomelessnessfaceandcouldpotentiallydecreasetheratesofmorbidityandmortalityinthisvulnerablecommunity.

StreetMedicine 2

STREETMEDICINE:APROGRAMEVALUATIONby

ARIELL.EDWARDS

B.S.,SPELMANCOLLEGE

ACapstoneSubmittedtotheGraduateFacultyofGeorgiaStateUniversityinPartialFulfillment

oftheRequirementsfortheDegree

MASTEROFPUBLICHEALTH

ATLANTA,GEORGIA30303

StreetMedicine 3

APPROVALPAGE

STREETMEDICINE:APROGRAMEVALUATIONby

ARIELL.EDWARDS

Approved:AshliA.Owen-Smith,PhDSMCommitteeChairLaraFrye,MDMPHCommitteeMemberDecember5,2016Date

StreetMedicine 4

Acknowledgments

IwanttothankDr.FryefortheopportunitytoworkwiththeStreetMedicineprogramatMercyCareandallofthesupport.Ilearnedsomuchaboutthecommunityofpeopleexperiencing

homelessnessinthemetroAtlantaareaandhealthdisparities.IwouldliketothankDr.Owen-Smithforallofthesupportandguidanceduringthecapstoneprocess.Iwould,also,liketothankGeorgiaStateUniversityforprovidingmewiththeopportunitytogainvaluablepublic

healthexperience.

StreetMedicine 5

Author’sStatementPage

Inpresentingthiscapstoneasapartialfulfillmentoftherequirementsforanadvanceddegree fromGeorgia StateUniversity, I agree that the Libraryof theUniversity shallmake itavailableforinspectionandcirculationinaccordancewithitsregulationsgoverningmaterialsofthistype.Iagreethatpermissiontoquotefrom,tocopyfrom,ortopublishthiscapstonemaybegrantedbytheauthoror,inhis/herabsence,bytheprofessorunderwhosedirectionitwaswritten,or inhis/her absence,by theAssociateDean, SchoolofPublicHealth. Suchquoting,copying, or publishing must be solely for scholarly purposes and will not involve potentialfinancial gain. It is understood that any copying from or publication of this capstone whichinvolvespotentialfinancialgainwillnotbeallowedwithoutwrittenpermissionoftheauthor.

ArielL.EdwardsSignatureofAuthor

StreetMedicine 6

TABLEOFCONTENTS

ACKNOWLEDGMENTS............................................................................................................4

LISTOFTABLES/FIGURES……………………………….……………………………………………………………….....7

I.INTRODUCTION....................................................................................................................9HealthDisparitiesAmongHomelessPopulation…………………..........9

HomelessnessinGeorgia……………………………………………………………..9StreetMedicine………………………………………………………………..………..10PurposeoftheEvaluation…………………………………………………………..11

II.REVIEWOFTHELITERATURE.............................................................................................12BarrierstoHealthCare……………………………………………………………….12HomelessPatientsandEDUse/Hospitalizations…………………………13ImprovingAccesstoPrimaryCareandMentalHealthServices…..15

III.PROGRAMDESCRIPTION……………………………………………………………………………………………...20AboutMercyCare.……………………………………………………………………..20StreetMedicineProgram……………………………………………………………20

IV.METHODS…………………......................................................................................................22EvaluationDesign……………………………………………………………………….22DataCollectionProcedures…………………………………………………..……22Measures…………............................................................................23Analysis..........................................................................................25

V.RESULTS............................................................................................................................26Sample………………………..……………………………………………………………..26StreetMedicineEncounters……………………………………………………….26ClinicEncounters............................................................................27StreetMedicinevs.ClinicEncounters…………………………………………28ConnectiontoMercyCare………………………………………………………….28EDVisits……………………………………………………………………………………..29HospitalAdmissions……………………………………………………………………30

VI.DISCUSSIONANDCONCLUSION......................................................................................33Discussion........................................................................................33Limitations.......................................................................................37Recommendations...........................................................................38Conclusion………………..…………..........................................................39

VII.REFERENCES....................................................................................................................41

VIII.APPENDICES………………………………………………………………………………………………………………45

StreetMedicine 7

ListofTables

Table1.SociodemographicsoftheStudyPopulation

Table2.StreetMedicineEncounters

Table3.ClinicEncounters

Table4.CaseManagement

Table5.StreetMedicineEncountersvs.ClinicEncounters

Table6.PatientConnectionStatus

Table7.EDVisits

Table8.PairedSampleTTestforEDVisitsBeforeandAfterMercyCare

Table9.HospitalAdmissionsBeforeandAfterMercyCare

Table10.PairedSampleTTestforTotalHospitalAdmissionDaysBeforeandAfterMercyCare

Table11.PairedSampleTTestforPsychiatricAdmissionsBeforeandAfterMercyCare

Table12.PairedSampleTTestforTotalHospitalAdmissionDaysBeforeandAfterMercyCare

StreetMedicine 8

ListofFiguresFigure1.LogicModel

StreetMedicine 9

I.Introduction

HealthDisparitiesAmongHomelessPopulation

Homelessnessisanurgentpublichealthproblem.Peopleexperiencinghomelessness

havepoorerhealthoutcomesthannon-homelesspeople(Oppenheimer,Nurius,&Green,

2016).Forexample,homelesspeoplehaveincreasedratesofacuteandchronicdiseasesand

mortality(Henwood,Cabassa,Craig,&Padgett,2013).AstudyfromMassachusettsassertsthat

peopleexperiencinghomelessnessaremorelikelytodiefromdrugoverdoseandcancerand

heartdiseaseatratesof16-to24-foldand2-to3-foldhigher,respectively,thanthegeneral

population(Baggettetal.,2013).Thesepoorhealthoutcomesareduetoinadequateliving

conditions,violenceandtrauma,anddrugandalcoholabuseamongmanyotherfactors.Even

thoughhomelesspeopleareinmostneedofhealthcare,theyaretheleastlikelytoreceive

adequatehealthcareifanyhealthcareatall.Whentheydoaccessthehealthcaresystem,it

usuallythroughexpensive,andoftenavoidable,EmergencyDepartment(ED)visitsthatarepaid

forthroughpublicfunds(Kuetal.,2014).Itisimperativethatwehaveprogramsavailableto

homelesspeoplethatwillprovidethemwithcost-effective,appropriate,andmeaningfulhealth

careservices.

HomelessnessinGeorgia

InGeorgia,therearealmost14,000documentedhomelessindividuals.Ofthe14,000

homelesspeople,65%areAfricanAmerican,31%areWhite,and4%areMulti-racialorother

(GeorgiaDepartmentofCommunityAffairs,2015).Accordingtoaneedsassessmentconducted

byMercyCareinthemetropolitanAtlantaarea,homelessclientsassertedthatthetopunmet

healthcareneedswereprimarycare,dentalcare,andmentalhealthtreatment(Laswell,2015).

StreetMedicine 10

ThehomelessclientsalsoreportedthatthemainbarrierstocareinAtlantaarelackofincome,

lackofinsurance,transportation,lackofinformationaboutwheretoaccessservices,wait

times,andlimitedcoordinationamongcareproviders(Laswell,2015).Becauseoftheexpressed

unmethealthcareneedsandbarrierstocare,manyofthehomelesspeopleinthemetro

Atlantaareadonothaveaconsistentsourceofhealthcarethatcaterstotheirneeds.

StreetMedicine

StreetMedicineismodelofservicedeliverythatemploystheuseofmultidisciplinary

mobileoutreachteamsthatprovidecaretohomelesspopulationslivingonthestreet(Howe,

Buck,&Withers,2009).ThemaingoalofStreetMedicineistoincreaseaccesstocarefor

homelesspatientsbydecreasingbarriersthatoftencausehomelessindividualstoresistthe

healthcaresystem.Byprovidinghealthcaretounshelteredhomelesspeopleonthestreets,in

homelessencampments,abandonedbuildings,etc.,StreetMedicineeliminatestheissuesof

transportation,lackofinsuranceorfinancialresourcesandlongwaittimes.Theunconventional

settinghelpstocreateasenseoftrustbetweenthehealthcareprovidersandthepatientsand

offersamorepositiveexperiencewiththehealthcaresystem.WhenStreetMedicineprograms

collaboratewithothercommunityhospitals,clinics,andproviders,theyeliminatetheproblems

oflackofcoordinationofcareandlackofinformationaboutwheretoaccessservices.

Therefore,StreetMedicinemaybeapromisingmodelofservicedeliverythatwillconnect

homelesspatientstoconsistenthealthcareservices.Itmayalsobeeffectiveindecreasingthe

numberofhomelesspatientsthatseekavoidableEDcare.

MercyCaredevelopeditsStreetMedicineprogramin2013.Theprogramconsistsofa

multidisciplinaryteamthatprovidesprimarycare,psychiatricservices,andsocialsupport

StreetMedicine 11

servicesforhomelesspatientslivingonthestreets.Thegoaloftheprogramistoengage

homelessindividualsinconsistentmedicalcareservicesandcreateamedicalhomefor

homelesspatientsinhopesthatitwilldecreasethenumberofpatientsthatseekavoidableED

andhospitalservices.ThoughtheStreetMedicineprogramhasbeenprovidingservicesto

homelesspatientsforover3years,nodatahasbeencollectedtoreportontheeffectivenessof

theprograminengaginghomelesspatientsinconsistentmedicalcareservicessuchasprimary

careandbehavioralhealthservices.

PurposeoftheEvaluation

ThepurposeoftheevaluationistoexaminetheeffectivenessofMercyCare’sStreet

Medicineprogramatengagingtheirhomelesspatientsinongoingcareanddecreasingthe

numberofpatientsthatseekunnecessaryhospitalcareandEDservices.Theoutcomesof

interestinclude(1)thenumberofpatientsthatwereseenduringStreetMedicinerounds;(2)

thepercentofStreetMedicinepatientsthatwereseenataMercyClinicormobileclinic

location;(3)thepercentofStreetMedicinepatientswereseenconsistently;(4)thetypesof

servicesinwhichthepatientsengaged;(5)thenumberofEDvisits;and(6)thenumberof

hospitaladmissions.Thedatawascollectedthrougharetrospectivemedicalrecordreviewat

MercyCare,withaccesstoGradyHospitalrecordsthroughMercyCareaswell,andanalyzed

usingfrequencytablesandt-testsinSAS.Thefindingswillbeusedasaguidetomakequality

improvementchangestotheprogram.

StreetMedicine 12

II.ReviewoftheLiterature

BarrierstoHealthCare

Inordertoprovidemorepositivehealthcareexperiencesforthehomelesspatients,itis

importanttoexplorethebarrierstohealthcarethroughtheeyesofthehomelesspatients.

NickaschandMarnocha(2009)attemptedtoinvestigatetheseexperiencesbyconducting

interviewswithhomelessindividualsinnortheasternWisconsinwhoareovertheageof18

years.Theresearchersconcludedthatmostofthehomelesspeoplethatwereinterviewed

believedthattheyhadnocontrolovertheirownlivesandhealth,andmostofthebarrierswere

duetoexternalforces(Nickasch&Marnocha,2009).Thebarriersthatwerefrequently

mentionedduringtheinterviewsincluded:1)lackofattainmentofphysicalneeds;2)lackof

affordability;3)lackofavailableresources;and4)lackofcompassionofhealthcareproviders

(Nickasch&Marnocha,2009).

Martens(2009)alsoexaminedthebarrierstoadequatehealthcareoftenfacedby

homelessindividuals.Theresearcherconductedaliteraturesearchonhealthcareaccessfor

homelesspeople.Thesearchincludedsubgroupsofhomelesspeoplesuchaschildrenand

adolescents,women,families,veterans,andmentallydisordered(Martens,2009).Thedata

collectedfromtheliteraturesearchwerefromresearcharticlespublishedbetween1988and

2008.Martensfoundthatlackofrespectandjudgementalismfromprovidersandstigma

negativelyaffecttreatmentseekingbehaviorforhomelesspeople(2009).Theresearchalso

concludedthatlackofinsuranceorunderinsurance,immobilityduetosickness,transportation

problems,andconfusionandinadequacyofhealthcaresystemalsohaveanegativeimpacton

homelessindividuals(Martens,2009).

StreetMedicine 13

HomelessPatientsandEDUse/Hospitalizations

Homelesspeopleoftendonothaveconsistentfollow-upwithprimarycareanduse

emergencyroomandinpatienthospitalservicesathigherrateswhencomparedtonon-

homelesspeople.Lebrun-Harrisetal.(2013)conductedastudyassessingthehealthstatusand

healthcareexperiencesofpatientswhoarehomelessversustheirhousedcounterpartsin

federallysupportedhealthcenters.Theresearchersanalyzedcross-sectionaldataon2,683

adulthomelesspatientsfromthe2009HealthCenterPatientSurvey,whichisanationally

representativesurveyfundedbytheHealthResourcesandServicesAdministration.Thissurvey

wasconductedusingpersonalinterviewswithquestionsthatfocusedonhomelessstatus,

sociodemographiccharacteristics,healthstatus,medicalconditions,accesstocareand

utilizationofservices.Thestudyconcludedthathomelesspatientsweretwiceaslikelyas

housedpatientstohaveunmethealthcareneeds(medicalcare,prescriptionmanagement,

dentalcare,andmentalhealthcare)andanemergencyroomvisitwithinthepastyear(Lebrun-

Harrisetal.,2013).Lebrun-Harrisetal.foundthathomelesspatientshadthreetimestheodds

ofreportingEDvisitsasausualsourceofhealthcarewhencomparedtohousedpatientsand

twicetheoddsofbeingheavyusersofEDservices(2013).

Wen-Chieh,Bharel,Jianying,O’Connell,andClark(2015)assessedfactorsthatare

associatedwithfrequenthospitalandEDutilizationamong6,494homelesspeoplewith

MedicaidinMassachusetts.Theresultsofthestudyshowthatmorethan70%ofthe

hospitalizationswereacquiredbyonly12%ofthesample.Morethan70%oftheEDvisitswere

acquiredbyonly21%ofthesample.Homelesspeoplewithco-occurringmentalillnessand

substanceusedisorderspresentthegreatestriskforfrequenthospitalizationsandEDvisits

StreetMedicine 14

(Wen-Chiehetal.,2015).Theresearchersbelievethattheincreasedriskforhomelesspeople

withmentalillnessandsubstanceusedisordersstemsfromchallengesinlocatingbehavioral

healthservicesandincoordinationofcare,lackofrelationshipwithproviders,andashortage

ofbehavioralhealthproviders.

NegativeperceptionsofhealthcareprovidersalsoincreaseEDuseamonghomeless

patients.ManyEDprovidersbelievethathomelessindividualswillnotfollowupwitharegular

primarycareproviderduetotheassumptionthathomelesspeopleare“highlymobile”or

transient.BecausemanyEDsoftenadvisehomelesspatienttoreturnfornon-emergenthealth

careneeds,thisincreasesdependenceonEDuseandleadstoovercrowding(Parker&Dykema,

2013).Theresearcherswantedtochallengethisnotionofhighmobilityandtransienceby

conductingacrosssectionalstudywithasampleof674homelessadultsrecruitedfromalarge

homelessshelterinSouthCarolina.ParkerandDykema'sresearchwascounterintuitivetothe

ideathathomelesspatientsarehighlymobileandtransient.Theresearchersconcludedthat

manyhomelesspeopletendtostayinthesamestateorcityinwhichtheyfirstbecame

homeless(2013).Theresearchersfurtherconcludedthatbecausehomelesspatientsarenot

highlymobile,EDphysiciansshouldputmoreeffortintoprovidinghomelesspatientswith

primarycarereferralstoreduceEDuse(Parker&Dykema,2013).ParkerandDykemaassert

thatreferralstonon-EDsourcesofcaremayproduceconsiderablecostsavingsforthe

organizationandthehealthcaresystemaswhole(2013).

HomelesspatientsrepresentasmallsubgroupoffrequentEDusers,howeverthese

patientsincurexpensivehealthcarebillsforacutecareservicesthataresubsidizedbythe

public(Kuetal.,2014).Iftheresearchersareabletoquantifytheeconomicburdenthat

StreetMedicine 15

frequentEDusersplaceonhealthcareexpenditures,policymakerscanproperlyaddresshigh

EDexpenditureswithtargetedinterventions.TheneedforthiskindofinformationledKuetal.

(2014)tostudytheEDcostsoffrequentusers,bothhomelessandnon-homeless,byconducting

aretrospectivecrosssectionalmedicalrecordreviewforEDvisitsatanurbanacademicmedical

center.Kuetal.foundthat74homelesspatientsthatwereconsideredfrequentusersaccessed

themedicalcenterED845timesinoneyear(2014).HomelessfrequentusersutilizedtheED

moreoftenthannon-homelessfrequentuserswithchargesthatadduptoalmost$5millionin

oneyear,whichisestimatedtobe$64,000perhomelesspatientthatisafrequentuser(Kuet

al.,2014).Theamountinhospitalchargesforthetopfivehomelessfrequentuserswasalmost

$2millionforoneyearwiththetopuserincurringalmost$500,000alone(Kuetal.,2014).

ImprovingAccesstoPrimaryCareandMentalHealthServices

Homelesspeoplehavebeenshowntohavehigherratesofmortalityandmorbiditythan

theirhousedcounterpartsandpresentagreaterneedforprimarycare.However,thereare

manybarriersthathinderhomelessindividualsfromreceivingongoingprimarycareservices

suchasabsenceofinsurance,transportationissues,anddistrustofthehealthcaresystemto

nameafew.HealthQualityOntario(2016)performedasystematicreviewtostudythe

effectivenessofinterventionsthataimtoincreaseaccesstoprimarycareamonghomeless

peoplewholiveinurbanareasandlackseriousmentalillnesses.Theresearchersreviewed

4,047citationsandisolatedfivestudiesthatdiscussedinterventionstrategiesusedtoimprove

accesstoprimarycareservicesforhomelesspeople(HealthQualityOntario,2016).The

interventionstrategiesincludedclinicorientationwithorwithoutoutreachservices,integration

ofprimarycareserviceswithotherservicesforhomelessindividuals,andprovisionofhousing

StreetMedicine 16

andsupportiveservices.Oftheseinterventions,clinicorientationtoavailableserviceswithor

withoutoutreachservicesprovedtobethebestmethodforimprovingaccesstoprimarycare

services.

O’Connelletal.(2010)completedareviewontheBostonHealthCarefortheHomeless

Program(BHCHP).BHCHPbeganprovidingcomprehensiveprimaryhealthcareservicesin1985

tohomelessindividualsinBoston.Inadditiontoprovidingservicesinaclinicalsetting,health

careprovidersalsoprovideservicesdirectlytohomelesspatientsonthestreets.Asmall,

multidisciplinaryteamofhealthcareprovidersdeliveredcontinuousprimarycaretopatients

livinginalleys,underbridgesorinfrontofdoorwaysinordertobuildrapportwiththese

homelesscommunities.Theevidencesupportedthat“streetmedicine”roundswereeffective

fortreatingandpreventingchronicdiseases.Thestreetmedicineroundshelpedhomeless

patientstoregularlyreceivefluvaccines,TBskintests,screeningsforhypertension,and

screeningsfordiabetes(O’Connelletal.,2010).Theresearchersnotethat79%ofthehomeless

patientsobtainedorwererecommendedforafluvaccine.Ofthepatientsthatwerewomen,

45%hadPaptests,and56%hadmammograms(O’Connelletal.,2010).

HwangandBurns(2014)conductedastudyofinterventionsthatcanbeusedto

improvethehealthofhomelesspeople.Theresearcherstargetedhomelessindividualsinhigh-

incomecountries,likemanyotherstudies.Thestudyfocusedoninterventionsforprimary

healthcareservices,mentalhealthcareservices,permanentsupportivehousing,medical

respiteprograms,substanceusers,andhomelessyoungpeople(Hwang&Burns,2014).The

researchersperformedanobservationalstudyintheUSatfiveclinicalcaresites.Hwangand

Burnsconcludedthatprimarycareprogramsthatweretailoredtohomelesspatientswere

StreetMedicine 17

ratedhigherinqualityofcarethanprimarycareprogramsthatarenottailoredtohomeless

patients.Thetailoredprimarycareprogramsincludedactiveoutreach,casemanagement,

partnershipswithcommunityorganizations,andcommunityadvisoryboards(Hwang&Burns,

2014).

Activeoutreachhasalsobeenusedtoimproveaccesstomentalhealthcareforpeople

experiencinghomelessness.Bond,Drake,Mueser,andLatimer(2001)conductedaliterature

reviewtodiscusstheeffectivenessandcost-effectivenessofoneactiveoutreachmodelin

particular–assertivecommunitytreatment(ACT).TheACTmodelisacommunitycaremodelin

whichamultidisciplinaryteamofprofessionalshelpspeoplewithseverementalillness

successfullyintegrateintothecommunity(Bondetal.,2001).ThestudyconcludedthatACT

significantlyreducedpsychiatrichospitalutilization,increasedhousingstability,andimproved

theperceptionofqualityoflife.ACTengagedpatientsinmentalhealthtreatmentand

increased1-yearretentioninservicesfrom54%to84%(Bondetal.,2001).Whencompared

withstandardaftercareandlowintensitycasemanagement,ACTreducedhospitalizationsby

78%and58%,respectively.ThoughintensiveACTserviceswerecostly,thereductionin

hospitalizationcostswasabletosignificantlyoffsetthecostsofACT.Evenwhencomparedwith

otherservices,ACTresultedinloweroverallcosts(Bondetal.,2001).

Young,Barrett,Engelhardt,andMoore(2014)assessACTasaneffectiveintervention

forimprovingmentalhealthandstablehousingforpeopleexperiencinghomelessness.The

studyexaminesoutcomesformentalhealth,housingstability,andsubstanceusepre-ACTand

sixmonthspost-ACTaswellaspatientsatisfactionandengagement.Theproportionof

participantsthatreportedstablehousingincreasedfrom17.6%to39.2%(Youngetal.,2014).

StreetMedicine 18

Therewereasignificantnumberofreportedgainsinglobalmentalhealthsymptomologyand

lessfrequentepisodesofdepressionandanxiety.However,therewasnosignificantchangedin

reportedsubstanceuse.Thoughamajorityoftheparticipants(78.4%)weresatisfiedwithACT,

theparticipantengagementwasequaltotheCESTnormativeaverage(Youngetal.,2014).

StreetMedicineisbecomingoneoftheleadingmodelsforthedeliveryofprimaryand

mentalhealthcareservicestohomelesspatientsduetoitsstreetoutreach/ACTcomponent.

However,becauseoftheunconventionalsettinginwhichStreetMedicineiscarriedout,itis

oftendifficulttoassessqualitymanagementconcerns.Howe,Buck,andWithers(2009)

conductedaqualitativeanalysisoneightprogramsfromthe2007and2008presentationsfrom

theannualInternationalStreetMedicineSymposia.Thegoaloftheanalysiswastooutlinethe

contextualcomponentsthatimpactqualitymanagementanddefinepresentquality

managementpracticesusedinStreetMedicineprograms(Howeetal.,2009).Thecontextual

componentsidentifiedintheanalysisthatimpactqualitymanagementincludedunconventional

livingarrangementsanddeficiencyoffinancialresourcesforhomelesspatients,inconsistent

contactwithpopulationsofhomelesspatientsthataretransient,andinformalclinicalsettings

(Howeetal.,2009).ThebestpracticesforStreetMedicineprogramsthatdeliverhighqualityof

careincludedtheuseofmobileclinicvans,thedevelopmentofelectronicmedicalrecordsthat

areuniquetoStreetMedicine,collaborationamongcommunityclinicsandproviders,andthe

provisionofsocialsupportservices(Howeetal.,2009).Thoughthisstudyonlypresents

preliminarydata,itmarksanopportunityforfurtherstudiesintoqualitymanagementforStreet

Medicineprograms.

StreetMedicine 19

Previousresearchindicatesthattherearetwochallengesthatmustbeovercomein

ordertoprovidehomelesspatientswithmeaningfulhealthcare.Onechallengeistofindaway

todecreasethebarrierstohealthcarethathomelesspeopleoftenfaceandprovidethemwith

amorepositiveexperiencewiththehealthcaresystem.Thesecondchallengeistodecrease

thenumberofpatientsthatturntothehospitalforavoidableandcostlyhealthcareservices.

Uponreviewingtheliterature,thereisnotalotofresearchaboutStreetMedicine,butitseems

thatthestreetoutreachcomponentofstreetmedicinemaymakeitamethodofhealthcare

deliverythatcanaddressthesechallenges.

StreetMedicine 20

III.ProgramDescription

AboutMercyCare

MercyCareisaFederallyQualifiedHealthCenterthatserves11,392patients,88%of

whomareuninsuredand62%ofwhomarehomeless.ItisalsotheonlyHealthcareforthe

HomelessprograminAtlanta.Thereareeightclinicsandfivemobilesitesalloverthemetro

Atlantaareathatprovideprimarycare,pediatriccare,dentalandvisionservices,health

education,behavioralhealthservices,andHIVtreatmentregardlessofinsurancestatus.

CommunityoutreachisanimportantpartofMercyCare.TheCommunityHomeless

OutreachProgramaimstobuildrapportwithhomelesspeoplewhoarestreetboundand

providesreferralsforhousing,clothing,food,andmedicalandbehavioralhealthservices.The

HIVPreventionOutreachTeamprovidesscreeningsandcommunityeventsthatprovide

informationaboutHIVandprevention.TheFamilyHealthPromotionprogramprovideshealth

informationandencourageshealthylifestylesamongtheimmigrantLatinocommunity.

StreetMedicineProgram

Recognizingthedireneedforhealthcareservicesamongunshelteredindividualsliving

onthestreets,MercyCarestartedtheirStreetMedicineprogramin2013.Thepurposeofthe

StreetMedicineprogramistoengageunshelteredindividualsinthemetroAtlantaareaand

providethemwithphysicalandmentalhealthcareintheplaceswheretheyliveandsleep.This

programstrivestoprovidephysicalandmentalhealthcaredirectlyonthestreetstoreduce

morbidityandmortality,assisthomelessindividualswithobtaininghousing,andeducatefuture

healthcareproviderstoprovideculturallycompetenthealthcaretostreet-boundhomeless

individuals.Bydoingso,theStreetMedicineprogramhopestoimprovetherelationship

StreetMedicine 21

betweenthehomelesscommunityandthehealthcaresystemandreducethebarriersto

healthcareaccess.

TheStreetMedicineteamconsistsoffamilyandinternalmedicinephysicians,a

psychiatrist,aregisterednurse,apeerspecialistandstudentvolunteers.Thisteamgathers6to

8teammembersandcompletesStreetMedicineroundsonceaweekonWednesdaysforfour

hours(6:00pmto10:00pm)travelinginavantostreetcorners,doorwaysofabandoned

buildings,andunderbridges.Theteamintroducesthemselvestonewclientstoengagethemin

thehealthcaresystemandfollowsupwithcurrentclientstoensurecontinuityofcare.Once

thepatientsareestablished,theStreetMedicineprogramusesitspartnershipswithother

organizationsandMercyCareclinicstoprovideadditionalhealthcareservicesandsocial

supportservices.Thismethodofoutreachallowsproviderstomeethomelesspatientswhere

theyareinordertobuildtrustingrelationshipswiththesecommunitiesandprovidebetter

healthcare.Theultimategoaloftheprogramistoprovidehomelesspatientswithaconsistent

sourceofcareaccordingtheprogram’sLogicModel(Figure1).

StreetMedicine 22

IV.Methods

EvaluationDesign

TheStreetMedicineProgramhasbeenoperatingsince2013andhaspredetermined

goalsandobjectivesasreferencedintheprogram’sLogicModelinFigure1.Usingthegoalsand

objectivesfromtheprogramtheoryasstandardsfortheevaluation,anoutcomeevaluation

designwasselectedasthebestmethodforthisevaluation.Thistypeofevaluationwillbeused

toassesshowwelltheprogramachieveditsintendedoutcome,whichisengaginghomeless

patientsinregularfollow-upwithprimaryandbehavioralhealthservicesatMercyCare.

DataCollectionProcedures

MedicalRecordReview

Aretrospectivemedicalrecordsreviewwasconductedtogatherdatafortheevaluation.

Alistof284patientswasprovidedfordataextractionbasedonaStreetMedicine

administrativereport.Ofthe284patients,26patientshadtobeexcludedfromdataanalysis

becausethepatientshadnotbeenseenduringtheStreetMedicineroundsandbutmayhave

beenpatientsthatwereseenviatelemedicine.Allofthedatathatwasextractedwascollected

inacomputer-baseddatacollectionform.

ThemedicalrecordreviewincludeddatacollectedfromMercyCareandGradyHospital.

ThedatacollectedfromMercyCarecontainedinformationaboutpatientencountersfromthe

datesofJanuary1,2015throughSeptember31,2016.ThoughtheStreetMedicineProgram

beganin2013,MercyCareswitchedtoanewelectronicmedicalrecordsystemin2015,andthe

previousdatawasunabletoberecovered.ThedatacollectedfromGradyHospitalcontained

StreetMedicine 23

informationaboutpatientencountersforalldatespriortothefirstencounterwithMercyCare

throughSeptember31,2016.

ProtectionofPatientHealthInformation

ThemedicalrecordsarestoredandmaintainedatMercyCareusingelectronichealth

recordsoftware.Theinformationcanonlybeaccessedusingaloginnameandpassword

generatedfromMercyCare’sadministrativeofficeuponparticipatinginaHIPPAtraining

programandclinicorientation.Anyremoteaccesstomedicalrecordsmustalsobeapproved

throughMercyCare’sadministrativeofficeandaloginandpasswordmustbeprovidedtothe

user.

Thecomputer-baseddatacollectionformwasencryptedinordertoprotectthe

patient’sinformation.Alloftheprotectedpatientinformationfordataextractionwas

transferredfromtheprogram’sattendingphysiciantothecomputer-baseddatacollectionform

usingaportablestoragedeviceandtookplaceatMercyCare.Thefilesonthestoragedevice

werealsoencryptedfortheprotectionofthepatient’sinformation.

Measures

NumberofPatientswithStreetMedicineEncounters

StreetMedicineencounterswereanimportantsourceofinformationforassessing

medicaltreatmentaccessedbyhomelesspatients.AvisitwasconsideredaStreetMedicine

encounterifthepatient’svisittookplaceonaWednesdaybetweenthehoursof6:00pmand

10:00pm,whichisthedayandtimethattheStreetMedicineTeamconductstheirrounds.The

indicatorstoassessStreetMedicineencountersincluded1)thepatient’stotalnumberofStreet

StreetMedicine 24

Medicinevisits,2)thenumberofvisitsthatwerebehavioralhealthvisits,and3)thenumberof

visitsthatwereprimarycarevisits.

NumberofPatientswithClinicEncounters

Theclinicencounterswerevitalforassessingtreatmentaswell.Theindicatorstoassess

clinicencountersincluded1)thepatient’stotalnumberofclinicvisits,2)thenumberofvisits

thatwerebehavioralhealthvisits,3)thenumberofvisitsthatwereprimarycarevisits,and4)

casemanagementvisits.Casemanagementisoneoftheservicesthatwasprovidedinthe

clinics.CasemanagershelppatientstogetSocialSecuritybenefits,housingassistance,clothing

assistance,foodassistance,transportationservices,etc.Thisindicatorwasaddedforadditional

informationaboutservicesthatwereaccessedintheclinic.

ConsistentMedicalCare

AnoutcomethatwasvitaltoassesstheeffectivenessoftheStreetMedicineprogram

wasthenumberofpatientsthatwereregularlyfollowingupwithahealthcareprofessionalin

theclinicandduringStreetMedicinerounds.Basedonpreviousfollowupintervalstudies

[Schwartz,Woloshin,Wasson,Renfrew,&Welch(1999);Buscheretal.(2013);Welch,Chapko,

James,Schwartz,&Woloshin(1999)],apatientwasconsideredtoberegularlyfollowingupif

theintervalsbetweenthevisitswere6monthsapartorless.Anindicatortoassessconsistency

ofmedicalcarewasthepatient’sconnectionstatustoMercyCare.Patientsthatwerelabeled

“Started”werenewpatientsatthetimeoftheirfirstStreetMedicineencounterandwere

regularlyfollowingupleadingtotheenddateoftherecordreview.Patientsthatwerelabeled

“AlreadyConnected”werealreadybeingseenataMercyCarecliniclocationatthetimeof

theirfirstStreetMedicineencounterandregularlyfollowingupleadingtotheendoftherecord

StreetMedicine 25

review.Patientsthatwerelabeled“Disconnected”wereneworcurrentpatientsthathadnot

seenaMercyCareproviderformorethan6months.Patientsthatwerelabeled“Reconnected”

werepatientsthathaddisconnectedfromMercyCarebutlaterhadatleastonefollowupvisit

duringtherecordreviewperiod.

EDVisitsandHospitalAdmissionDays

EDvisitsandhospitaladmissionsbeforeandafterbecomingapatientatMercyCare

werealsousedtoassesstheeffectivenessoftheprogram.TheStreetMedicineprogram

hypothesizedthatprovidingconsistentmedicalcarecoulddecreasethenumberofEDvisitsand

hospitaladmissionsThemeasurestoassesseffectivenesswere1)thepatient’stotalEDvisits

everatGradyHospital,2)allEDvisitsatGradybeforeandafterthepatient’sfirstMercyCare

encounterleadinguptoSeptember31,2016,3)thetotalnumberofhospitaladmissiondaysat

Gradybeforeandafterthepatient’sfirstMercyCareencounterleadinguptoSeptember31,

2016,4)thehospitaladmissionsthatwerepsychiatricadmissions,and5)thehospital

admissionsthatwerephysicalhealthrelated.

Analysis

TheanalysiswasperformedusingthedatamanagementandanalyticssoftwareSAS9.2.

Frequencytablesandmeanscalculationswereusedtocalculatevisitinformationrelatedto

visitencounters,EDvisitsandhospitaladmissions.T-testswereusedtocalculatesignificant

differencesinthebeforeandafteranalysesofEDencountersandhospitaladmissions.

StreetMedicine 26

V.Results

SampleIntotal,284patientmedicalrecordreviewswereconducted.Ofthe284patients,26

patientshadtobeexcludedbecausetherewerenoStreetMedicinevisitsnotedinthechart.

Table1showsthedemographicsforthe258patientsthatwereincludedinthemedicalrecord

review.Therewere206(79.84%)malepatientsand52(20.16%)femalepatients.Theagerange

ofthesamplewas19-72yearswithanaverageageofapproximately49years.Thesamplewas

mostlycomprisedofAfricanAmerican/Blackhomelesspatients.Ofthe258patients,146

(56.59%)wereAfricanAmerican/Black,28(10.85%)wereWhite,1(<1%)wasHispanic/Latino,

and83(32.17%)wereofunknownraceduetolackofinformationprovidedinthemedical

record.

StreetMedicineEncounters

TotalNumberofStreetMedicineVisits

Table2showsthat132patients(51.16%)hadonlyoneStreetMedicinevisit,and83

patients(32.17%)had2to3StreetMedicinevisits.Therewerefewpatientswith4to9visitsor

10visitsormore,31(12.03%)and12(4.66%)respectively.Theaveragenumberoftotalstreet

medicinevisitsperpatientwas2.47visits.ThehighestnumberofStreetMedicinevisitscame

fromonepatientwhohad18StreetMedicineencounters.

BehavioralHealthStreetMedicineVisits

TheStreetMedicinevisitsweregroupedandanalyzedaccordingtothetypeofservices

thatwereprovidedinthevisit.Forbehavioralhealthvisits,approximatelyhalfofthepatients

didnotseekbehavioralhealthservicesduringtheirStreetMedicinevisit.However,70patients

(27.13%)hadonlyonebehavioralhealthStreetMedicineencounter,and35patients(13.57%)

StreetMedicine 27

had2to3visits.Fewerpatientshad4visitsormore.Eachpatientaveragedabout0.92

behavioralhealthvisits(range=0-8visits).

PrimaryCareStreetMedicineVisits

MorepatientsengagedinprimarycareStreetMedicineservices.Therewere120

patients(46.69%)thathadonlyonevisitand66patients(25.68%)had2to3visits.Similarto

thebehavioralhealthvisits,fewerpatientshad4visitsormore.Eachpatientaveragedabout

1.56primarycarevisits(range=0-14visits).

ClinicEncounters

TotalNumberofClinicVisits

Table3showsthat107patients(41.47%)didnothaveanyclinicvisits,40patients

(15.50%)had2to3visits,and52patients(20.16%)had4to10visits.Only23patients(8.91%)

hadonlyonevisit.Fewerpatientshad11visitsormore.Eachpatientaveraged4.97totalclinic

visits(range=0-65visits).

BehavioralHealthClinicVisits

Theclinicvisitswerealsogroupedaccordingtothetypesofservicesprovidedinthe

visit.Forbehavioralhealthvisits,180patients(69.77%)didnothaveanybehavioralhealthvisits

and28patients(10.85%)hadonlyonevisit.Fewerpatientshad2ormorevisits.Theaverage

numberofbehavioralhealthvisitsperpatientwas1.44visits(range=0-29visits).

PrimaryCareClinicVisits

Forprimarycarevisits,115patients(44.57%)didnothaveanyvisits,and43patients

(16.67%)hadat4to10visits.Only33patients(12.79%)hadonlyonevisit,and37patients

StreetMedicine 28

(14.34%)had2to3visits.Fewerpatientshad11ormorevisitsduringthetimeofthestudy.

Theaveragenumberofprimarycarevisitsperpatientwas3.52visits(range=0-65visits).

CaseManagement

AsnotedinTable4,only83ofthepatients(32.17%)wereaccessingcasemanagement

services.Theother175patients(67.83%)werenotaccessingcasemanagementservices.

StreetMedicineEncountervs.ClinicEncounters

TheresultsinTable5indicatedthat37.21%oftheStreetMedicineencounterswere

behavioralhealthencountersand62.79%wereprimarycareencounters.Oftheclinic

encounters,29.10%werebehavioralhealthencountersand70.83%wereprimarycare

encounters.

ConnectiontoMercyCare

Table6showsthat118(45.74%)ofthepatientsdisconnectedfromMercyCaresince

theinitialStreetMedicineencounterandwerenolongerreceivingcontinuouscarethrough

StreetMedicineroundsorclinicvisits.Ofthepatientsthatdisconnected,83(70.34%)

disconnectedafteraStreetMedicineencounterand35(29.66%)disconnectedafteraclinic

encounter.Therewere49patients(18.99%)thatwerealreadyconnectedtoMercyCareand

werereceivingcontinuousprimarycareand/orbehavioralhealthservices.Eightpatients

(3.10%)reconnectedtoMercyCarethroughStreetMedicineroundsorclinicencountersand

werereceivingcontinuouscare.HalfofthereconnectedpatientsreconnectedthroughStreet

Medicineroundsaswithclinicencounters.Eighty-threepatients(32.17%)werenewpatients

thatwerecurrentlyreceivingcontinuouscare.Atotalof140(54.26%)patientswerereceiving

continuousprimaryand/orbehavioralhealthservices.

StreetMedicine 29

EDVisits

TotalNumberofEDVisits

Table7showsthatamajorityofthepatients(37.29%)had4to10EDvisits,and26

patients(11.82%)hadonlyoneEDvisit.Eighteenpatients(8.18%)didnothaveanyEDvisits.

Fiftypatients(22.73%)had2to3visits,and44patients(20.04%)had11visitsormore.The

averagetotalnumberofEDvisitswas6.14visitsperpatient(range=0-21visits).Approximately

38(15%)patientsdidnothaveEDdatabecausetherewerenoGradyHospitalrecordsthat

correspondedwiththeirinformationatMercyCare.

EDVisitsBeforeMercyCare

InordertoassesswhetherornottheStreetMedicineprogramdecreasedthenumber

ofpatientsseekingEDservices,theEDvisitswereanalyzedaccordingtovisitsbeforethefirst

MercyCareencounterandafterthefirstMercyCareencounter.Beforetheinitialencounter

withMercyCare,75patients(34.09%)didnothaveanyEDvisits.Fifty-twopatients(23.64%)

had4to10EDvisits,45patients(20.45%)had2to3visits,and38patients(17.27%)hadonly

onevisit.Fewerpatientshad11visitsormore.TheaveragenumberofEDvisitsbeforethe

initialMercyCareencounterwas2.82visits(range=0-18visits).

EDVisitsAfterMercyCare

ForEDvisitsaftertheinitialencounterwithMercyCare,70patients(31.82%)didnot

haveanyEDvisits,whichdecreasedfromthenumberofpatientswithoutEDvisitsbefore

MercyCare.Thereare46patients(20.91%)with2to3visits,whichwassimilartotheresults

forpatientswith4to10visits.Thirty-eightpatients(17.24%)hadonlyonevisit,whichwasthe

sameasthenumberofpatientswithatleastoneEDvisitbeforeMercyCare.Therearefew

StreetMedicine 30

patientswith11visitsormore,howeverthenumberofpatientsincreasedfromtheEDvisits

beforeMercyCare.TheaveragenumberofEDvisitsaftertheinitialMercyCareencounterwas

3.33visits(range=0-21),whichishigherthantheaveragenumberforEDvisitsbeforeMercy

Care.

DifferenceinEDVisitsBeforeandAfterMercyCare

Table8showstheresultsofthepairedt-testanalysistocomparethedifferenceinED

visitsbeforeandafterMercyCare.TherewasnosignificantdifferenceinthenumberofED

visitsbeforeMercyCare(M=2.82,SD=3.61)andafterMercyCare(M=3.33,SD=4.52);

t(219)=1.22,p=0.2235.

HospitalAdmissions

HospitalAdmissionDaysBeforeandAfterMercyCare

Therewere165patients(75.00%)and163patients(74.09%)withnohospitaladmission

daysbeforeandaftertheinitialMercyCareencounter,respectively,asshowninTable9.

Twentypatients(9.09%)hadonlyonehospitaladmissiondaybeforeMercyCare,and11

patients(5.00%)hadonlyonehospitaladmissiondayafterMercyCare.Fifteenpatients(6.82%)

had2to3daysbeforeMercyCare,and14patients(6.36%)had2to3daysafterMercyCare.

Seventeenpatients(7.72%)had4to10daysbeforeMercyCare,and23(10.44%)had4to10

daysafterMercyCare.Fewerpatientshad11daysormorebeforeandafterMercyCare.The

averagenumberofadmissiondaysperpatientwas0.98days(range=0-34)beforetheinitial

MercyCareencounterand1.84days(range=0-40)aftertheinitialMercyCareencounter.

Approximately38(15%)patientsdidnothavehospitaladmissionsdatabecausetherewereno

GradyHospitalrecordsthatcorrespondedwiththeirinformationatMercyCare.

StreetMedicine 31

Table10showstheresultsofthepairedt-testanalysistocomparethedifferencein

totalhospitaladmissiondaysbeforeandaftertheinitialMercyCareencounter.Therewasa

significantdifferenceinthenumberofhospitaladmissiondaysbeforeMercyCare(M=0.98,

SD=3.08)andafterMercyCare(M=1.84,SD=5.27);t(219)=2.03,p=0.0432.

PsychiatricAdmissionsBeforeandAfterMercyCare

Hospitaladmissionvisitswerealsogroupedaccordingtothetypesofservicesprovided

inthatvisit.Forthepsychiatricadmissions,190patients(86.36%)and202patients(91.82%)

hadnoadmissionsbeforeandafterMercyCarerespectivelyasshowninTable9.Twenty

patients(9.09%)hadonlyonepsychiatricadmissionbeforeMercyCare,and12(5.45%)had

onlyonepsychiatricadmissionafterMercyCare.Fewerpatientshad2ormorepsychiatric

admissions.Theaveragenumberofpsychiatricadmissionswas0.23admissions(range=0-4)

beforeMercyCareand0.14admissions(range=0-6)afterMercyCare.

Table11showstheresultsofthepairedt-testanalysistocomparethedifferencein

psychiatricadmissionsbeforeandafterMercyCare.Therewasnosignificantdifferenceinthe

numberofpsychiatricadmissionsbeforeMercyCare(M=0.23,SD=0.70)andafterMercyCare

(M=0.14,SD=0.58);t(219)=-1.53,p=0.1276.

PhysicalHealthAdmissionsBeforeandAfterMercyCare

Forphysicalhealthadmissions,188(85.45%)and174(79.09%)didnothaveany

admissionsbeforeandafterMercyCarerespectivelyasshowninTable9.Twenty-ninepatients

(13.18%)hadonlyoneadmissionbeforeMercyCare,and28(12.73%)hadonlyoneadmission

afterMercyCare.Fewpatientshad2ormoreadmissionsbeforeandafterMercyCare.The

StreetMedicine 32

averagenumberofphysicalhealthadmissionswas0.16admissions(range=0-3)beforeMercy

Careand0.35admissions(range=0-6)afterMercyCare.

Table12showstheresultsofthepairedt-testanalysistocomparethedifferencein

physicalhealthadmissionsbeforeandafterMercyCare.Therewasasignificantdifferencein

thenumberofphysicalhealthadmissionsbeforeMercyCare(M=0.16,SD=0.43)andafter

MercyCare(M=0.35,SD=0.86);t(219)=2.77,p=0.0060.

StreetMedicine 33

VI.DiscussionandConclusion

Discussion

PatientDemographics

Therewere59.69%moremalepatientsinthesamplethanfemalepatients.Thisisnot

surprisingasmanystudies[Berry(2007);Leeetal.(2016);Byrne,Montgomery,&Fargo(2016)]

andGeorgiahomelessnesscensusreportsrevealthatmenaremorelikelytoliveunsheltered

thanwomen.Itmaybethattherearemoresheltersandhousingoptionsavailableforwomen

andfamiliesexperiencinghomelessnessorthatwomenmaybemorewillingtodoubleupwith

friendsorfamiliesforlivingarrangement.Onestudymentionsthathomelesswomenmaybe

morelikelytoengagein“survivalsex”,orexchangingsexforbasicnecessities,toobtainshelter

(Watson,2011).

Mostofthepatientsinthesampleweremiddle-agedAfricanAmerican/Blackmen.This

findingisconsistentwiththeGeorgiahomelessnesscensusreportssince2011.Despitebeinga

minoritypopulation,theAfricanAmerican/Blackcommunityhasconsistentlyhadahigh,andat

timesthehighest,rateofhomelessnessinAmerica(Jones,2016).Therearevarioussocialand

economicfactorsthatcontributetothehighrateofhomelessnesssuchaslowerhousehold

incomes,housingdiscriminationpracticesandlackofeducationtonameafew(NationalLaw

CenteronHomelessness&Poverty,2014).

StreetMedicineEncountersvs.ClinicEncounters

TheaveragenumberoftotalStreetMedicineencountersperpatientwaslowerthanthe

averagenumberofclinicencounters.TheStreetMedicineteamwasonlyabletoconducttheir

roundsonedayaweekfor4hoursandhavelimitedsuppliesavailableforpatienttreatment.At

StreetMedicine 34

theMercyCareclinic,patientshadaccesstomoremedicalcare,dentalcareandbehavioral

healthcareservicesfivedaysaweekfromthemorninguntilevening.Itispossiblethatbecause

theclinicprovidedmoreservicesandflexibletimesfortreatment,patientsaccessedcareata

clinicmorethanduringStreetMedicinerounds.

MorepatientsaccessedprimarycareservicesduringtheclinicencountersandStreet

Medicineencountersthanbehavioralhealthservices.Thisfindingwasinterestingconsidering

thefactthattherewasahighdemandformentalhealthservices.However,theproblemwith

patientsseekingmental/behavioralhealthatMercyCareismorecomplex.Itispossiblethat

stigmasurroundingmentalillnessanddistrustofmental/behavioralhealthcareprovidersmay

becausingthepatientstodeferfromaccessingtheseservices.MercyCarealsohasmore

primarycareprovidersthanbehavioralhealthproviders,whichmeansthereisapossibilitythat

thepatientscouldbereceivingbehavioralhealthservicesfromprovidersoutsideofMercy

Care,ifatall.Therewerenodatacollectedaboutothersourcesofcareandforwhattypesof

services.

ConnectiontoMercyCare

ThefrequencytableforconnectiontoMercyCarerevealedthat54.26%oftheStreet

MedicinepatientsareconnectedtoMercyCareandengaginginconsistentprimarycareand

behavioralhealthcareservices.Ontheotherhand,45.74%ofthepatientsdisconnectedfrom

MercyCare.Ofthepatientsthatdisconnected,over70%disconnectedafteraStreetMedicine

encounter,andalmost30%disconnectedafteraclinicencounter.Thoughtherewereahigh

numberofpatientsthatweredisconnected,variousfactorsthatcouldhavecontributedtothis

findinghavetobeconsidered.Asmentionedintheliteraturereview,peopleexperiencing

StreetMedicine 35

homelessnessfacemanybarrierstoreceivingcaresuchastransportation,waittimes,stigma,

affordability,lackofphysicalneeds,anddistrustofhealthcareproviders.ThoughStreet

Medicinehasthepotentialtoeliminatemanyofthosebarriers,issueswithstigmaandlackof

attainmentofphysicalneedsmaynotbesolvedwithStreetMedicinealone.Itisalsopossible

thatpatientsmaybeaccessingothersourcesofmedicalandbehavioralhealthcareinthe

metroAtlantaarea.

Becauseoverhalfofthepatientsareregularlyfollowingup,StreetMedicinemaybe

activelyeliminatingsomeofthesebarrierstohealthcareandkeepingpatientsengagedin

services.StreetMedicine’soutreachcomponentdecreasesissueswithtransportationbecause

healthcareprovidersbringhealthcaretothepatientswheretheyliveandsleep.Iteliminates

theneedsforinsuranceandaffordabilitybecausetheservicesareprovidedtopatientsfreeof

charge.StreetMedicineimprovestheavailabilityofresourceslikeprescriptions,referrals,and

informationaboutwheretoseekotherneededhealthservices.Duetothemanypotential

benefitsofandlimitedinformationaboutStreetMedicine,moreresearchneedstobe

conductedtoassessifthereisarelationshipbetweenStreetMedicineprogramsandan

increaseinregularfollow-upsfrompeopleexperiencinghomelessness.

EDVisits

Therewasanaverageof6.14EDvisitsperpatientatGradyHospital.Thoughtheresults

revealedthatpatientshadmoreEDvisitsaftertheirencounterwithMercyCare,therewasno

significantdifferenceinEDvisitsbeforeoraftertheencounter.Theresultsmaybeduetothe

hoursinwhichpatientscanaccesscarethroughMercyCare.StreetMedicineservicesareonly

availableonWednesdaysfrom6pmto10pm,andtheclinicisonlyopenfivedaysaweekfora

StreetMedicine 36

setnumberofhours.Ifapatientneedstreatmentoutsidetheclinic’sortheStreetMedicine

team’shoursofoperation,theclosetfacilityforemergencymedicaltreatmentisGrady

Hospital.

Theresultsmayalsobeduetoanincreaseinthenumberofpatientsthatarebeing

referredtotheEDbyaMercyCareprovider.Ifapatientneedsemergencytreatmentthat

cannotbeaccommodatedatMercyCare,theyareusuallyreferredtoGradyHospitalfor

medicaltreatment.ManyofthepatientsthatcometoMercyCarehavenotbeenseeking

regularhealthcareservicesandareveryill.Forexample,patientshavedangerouslyelevated

bloodpressures,severelyinfectedwoundsthatrequireIVantibiotics,orepisodesofmental

illnessthatmakethepatientadangertohimselforherselforothers.Therefore,thepatients

arereferredtoGradyforlife-savingemergencytreatment.

Patientsexperiencinghomelesscanbedifficulttofollowupwithiftheymovearound

frequentlyorifthepoliceforcethemtomoveoutoftheirencampment,whichcouldleavethe

patientswithoutregularaccesstomedicalcareduringStreetMedicineroundssincetheteam

onlygoesoutonceaweek.Duringtheseperiodsoftimeinwhichpatientswerenotactively

followingupwiththeStreetMedicineteam,itispossiblethattheycouldhavebeenutilizingED

servicesforregularsourcesofcare.

Moredataneedtobecollectedtoassessthedetailsofthepatients’EDvisitsto

determinewhichEDvisitswereavoidable.Theadditionaldatacanbeusedtomoreaccurately

assesswhethertherewereanychangesinavoidableEDvisitsbeforeandafterthepatient’s

initialMercyCareencounter.

HospitalAdmissions

StreetMedicine 37

SimilartoEDvisits,theincreasesintotalhospitaladmissiondaysandnumberof

physicalhealth-relatedadmissionsmaybeduetotheincreasedneedforpatientstohave

emergencycarethatleadstoanadmission,andevensurgicalprocedures,thatMercyCareis

notabletoaccommodate.Asmentionedearlier,manypatientsthatpresenttoMercyCareare

severelyillandindireneedoflife-savingtreatment.Itisalsopossiblethat,duringtheperiodof

timeinwhichthepatients’werenotregularlyfollowingup,orfelloutofcare,withtheStreet

Medicineteam,theywerebeingadmittedtothehospitalforconditionsthatwerepreviously

beingfollowedbyaprovideratMercyCare.Moredatawillneedtobecollectedtoassess

whethertheincreaseinhospitaladmissiondaysisrelatedtothetimesinwhichthepatientsfell

outofcare.

Thoughthedecreaseinpsychiatric-relatedadmissionsisnotstatisticallysignificant,itis

potentiallyasignificantachievementfortheStreetMedicineprogram.Itispossiblethatthat

thisfindingisduetothepatientsexperiencinghomelessnessseekingregularfollow-upsfor

behavioral/mentalhealthcare.Moredatawillneedtobecollectedovertimetofurtherassess

whetherStreetMedicineisabletosignificantlyreducethenumberofpsychiatric-related

hospitaladmissionsforpatientsthatareregularlyfollowingup.

Limitations

Collectingandanalyzingdataonhomelesspatientshadmanylimitations.Therewere

issueswithrandomerror.Forexample,itwasdifficulttocollecthospitalencounterdatafor

15%ofthepatientsbecausepeopleprovideddifferentnames,datesofbirth,orsocialsecurity

numbers.Therefore,theMercyCarechartscouldnotbeaccuratelymatchedwiththeGrady

Hospitalcharts.

StreetMedicine 38

Itwasalsodifficulttocollectaccuratedemographicinformationaboutthepatients

basedoninconsistenciesinthemedicalrecords.About32.17%oftherecordsdidnotindicate

theraceofthepatients.Therefore,thestatisticsforracecouldhavebeenmoreaccurately

reportediftheracewasconsistentlyindicatedinthemedicalrecord.

Alloftheencountershadtobemanuallycountedandseparatedintogroupsdepending

onthetypesofservicesthatwereaccessed.Somevisitsmayhavebeenmiscountedor

miscategorizedintheprocess.Tomitigatethiserror,allvisitswererecountedtomakesurethat

thenumberofvisitsinthechartmatchedthenumberofvisitsrecorded.

AliteraturereviewrevealedthatthereisnotalotofliteratureaboutStreetMedicine

programs.Further,thereisnotalotofliteratureaboutassessingtheeffectivenessofStreet

Medicineprograms.Thelackofresearchrelatedtoassessingeffectivenessmadeitdifficultto

determinewhatdatashouldbecollected.MorestudiesshouldfocusonaddressingStreet

Medicineprogramsandhowtoassesstheeffectivenessofsuchprograms.

Recommendations

FutureResearch

Asmentionedpreviously,moredataneedstobecollectedregardingthedetailsofthe

EDvisitsandhospitaladmissionstodetermineiftheStreetMedicineprogramwaseffectively

reducingEDvisitsandhospitalizations.Oneareathatneedsmoreanalysisistherelationship

betweenconnectionstatusandEDvisitsandhospitalizations.Bycomparingtheperiodsoftime

inwhichthepatientsfelloutofcarewithMercyCareandGradyHospitalrecords,theresulting

informationcouldbeusedtodetermineifthereisarelationshipbetweenEDvisitsand

hospitalizationsandconnectionstatustoMercyCare.IfthenumberofEDvisitsand

StreetMedicine 39

hospitalizationsincreasedduringtheperiodsoftimeinwhichthepatientwasnotregularly

followingup,thenthiscouldindicatethatStreetMedicinemayhavebeenusefulindeterring

someEDvisitsandhospitalizations.

AnotherareathatneedsmoreanalysisisthenumberofavoidableEDvisits.Thedetails

oftheEDvisitsatGradyneedtobere-examinedtodetermineifthevisitwasnecessaryor

couldhavebeenavoided.UponseparatingandtallyingthenecessaryEDvisitsfromthe

avoidableEDvisits,wecanmoreaccuratelyassesswhethertheStreetMedicineprogramhas

beeneffectiveinreducingavoidableEDvisits.

OutcomeEvaluation

OneofthemaingoalsoftheStreetMedicineprogramistoreducemortalityand

morbidity.However,thedatathathasbeencollectedcannotbeusedtoassesswhetherthe

StreetMedicineprogramhasaccomplishedthisgoal,andtheStreetMedicineteamiscurrently

notcollectinganydatarelatedtomorbidityormortality.Therefore,itmaybeusefultoconduct

anotheroutcomeevaluation.Theoutcomeevaluationwouldspecificallybeusedtodetermine

iftheprogramhasbeeneffectivelydecreasingtheratesofmortalityandmorbidityforpeople

experiencinghomelesslivingonthestreet.Uponcompletingtheevaluation,theprogramwill

haveaprocessforcollectingandassessingmortalityandmorbiditydatafortheStreetMedicine

patients.

Conclusion

Inconclusion,theStreetMedicineprogrammaybeapromisingsolutionforidentifying

andretainingpeopleexperiencinghomelessnessengagedinhealthcareanddecreasingthe

numberofpatientsthatturntothehospitalforavoidableandcostlyhealthcareservices.Due

StreetMedicine 40

toitsstreetoutreachcomponent,theStreetMedicineprogrameliminatesmanyofthebarriers

thathomelesspeoplefacesuchastransportation,affordability,andaccesstoresources(e.g.

prescriptions,referrals,informationaboutwheretoaccessservices).Byeliminatingthese

barriers,theStreetMedicineteamhopestodecreasetheratesofmorbidityandmortalityin

thisvulnerablecommunityandprovidethemwithastablemedicalhome.

StreetMedicine 41

VII.References

Baggett,T.P.,Hwang,S.W.,O’Connell,J.J.,Porneala,B.C.,Stringfellow,E.J.,Orav,E.J.,…

Rigotti,N.A.(2013).MortalityamonghomelessadultsinBoston:Shiftsincausesof

deathovera15-yearperiod.JAMAInternalMedicine,173(3),189–195.

https://doi.org/10.1001/jamainternmed.2013.1604

Berry,B.(2007).Arepeatedobservationapproachforestimatingthestreethomeless

population.EvaluationReview,31(2),166–199.

https://doi.org/10.1177/0193841X06296947

Bond,G.R.,Drake,R.E.,Mueser,K.T.,&Latimer,E.(2001).Assertivecommunitytreatmentfor

peoplewithseverementalillness:Criticalingredientsandimpactonpatients.Disease

Management&HealthOutcomes,9(3),141–159.

Buscher,A.,Mugavero,M.,Westfall,A.O.,Keruly,J.,Moore,R.,Drainoni,M.-L.,…Giordano,T.

P.(2013).Theassociationofclinicalfollow-UpintervalsinHIV-infectedpersonswith

viralsuppressiononsubsequentviralsuppression.AIDSPatientCareandSTDs,27(8),

459–466.https://doi.org/10.1089/apc.2013.0105

Byrne,T.,Montgomery,A.E.,&Fargo,J.D.(2016).Unshelteredhomelessnessamongveterans:

Correlatesandprofiles.CommunityMentalHealthJournal,52(2),148–157.

https://doi.org/10.1007/s10597-015-9922-0

GeorgiaDepartmentofCommunityAffairs.(2015,September).2015reportonhomelessness:

Georgia’s14,000.RetrievedAugust9,2016,from

https://www.dca.ga.gov/housing/specialneeds/programs/documents/HomelessnessRep

ort2015.pdf

StreetMedicine 42

HealthQualityOntario.(2016).Interventionstoimproveaccesstoprimarycareforpeoplewho

arehomeless:Asystematicreview.OntarioHealthTechnologyAssessmentSeries,16(9),

1–50.

Henwood,B.F.,Cabassa,L.J.,Craig,C.M.,&Padgett,D.K.(2013).Permanentsupportive

housing:Addressinghomelessnessandhealthdisparities?AmericanJournalOfPublic

Health,103Suppl2,S188–S192.https://doi.org/10.2105/AJPH.2013.301490

Howe,E.C.,Buck,D.S.,&Withers,J.(2009).Deliveringhealthcareonthestreets:Challenges

andopportunitiesforqualitymanagement.QualityManagementInHealthCare,18(4),

239–246.https://doi.org/10.1097/QMH.0b013e3181bee2d9

Hwang,S.W.,&Burns,T.(2014).Healthinterventionsforpeoplewhoarehomeless.The

Lancet,384(9953),1541–7.https://doi.org/10.1016/S0140-6736(14)61133-8

Jones,M.M.(2016).Doesracematterinaddressinghomelessness?Areviewoftheliterature.

WorldMedical&HealthPolicy,8(2),139–156.https://doi.org/10.1002/wmh3.189

Ku,B.S.,Fields,J.M.,Santana,A.,Wasserman,D.,Borman,L.,&Scott,K.C.(2014).Theurban

homeless:Super-usersoftheemergencydepartment.PopulationHealthManagement,

17(6),366–371.https://doi.org/10.1089/pop.2013.0118

Laswell,A.(2015,October).Merycare2015needsassessemnt:Reportoffindings.Retrieved

October18,2016,from

https://www.mercyatlanta.org/2015_HCH_Needs_Assessment.pdf

Lebrun-Harris,L.A.,Baggett,T.P.,Jenkins,D.M.,Sripipatana,A.,Sharma,R.,Hayashi,A.S.,…

Ngo-Metzger,Q.(2013).Healthstatusandhealthcareexperiencesamonghomeless

StreetMedicine 43

patientsinfederallysupportedhealthcenters:Findingsfromthe2009PatientSurvey.

HealthServicesResearch,48(3),992–1017.https://doi.org/10.1111/1475-6773.12009

Lee,C.T.,Guzman,D.,Ponath,C.,Tieu,L.,Riley,E.,&Kushel,M.(2016).Residentialpatternsin

olderhomelessadults:Resultsofaclusteranalysis.SocialScience&Medicine,153,131–

140.https://doi.org/10.1016/j.socscimed.2016.02.004

Martens,W.(2009).VulnerablecategoriesofhomelesspatientsinWesternsocieties:

Experienceseriousbarrierstohealthcareaccess.Medicine&Law,28(2),221–239.

NationalLawCenteronHomelessness&Poverty.(2014,July3).Racialdiscriminationinhousing

andhomelessnessintheunitedstates:AreporttotheUNcommitteeontheelimination

ofracialdiscrimination.RetrievedNovember7,2016,from

https://www.nlchp.org/CERD_Housing_Report_2014.pdf

Nickasch,B.,&Marnocha,S.K.(2009).Healthcareexperiencesofthehomeless.Journalofthe

AmericanAcademyofNursePractitioners,21(1),39–46.

O’Connell,J.,Oppenheimer,S.,Judge,D.,Taube,R.,Blanchfield,B.,Swain,S.,&Koh,H.(2010).

Thebostonhealthcareforthehomelessprogram:Apublichealthframework.American

JournalofPublicHealth,100(8),1400–1408.https://doi.org/10.2105/AJPH.2009.173609

Oppenheimer,S.C.,Nurius,P.S.,&Green,S.(2016).Homelessnesshistoryimpactsonhealth

outcomesandeconomicandriskbehaviorintermediaries:Newinsightsfrompopulation

data.FamiliesinSociety:TheJournalofContemporarySocialServices,97(3),230–242.

https://doi.org/10.1606/1044-3894.2016.97.21

StreetMedicine 44

Parker,R.,&Dykema,S.(2013).Therealityofhomelessmobilityandimplicationsforimproving

care.JournalofCommunityHealth,38(4),685–689.https://doi.org/10.1007/s10900-

013-9664-2

Schwartz,L.M.,Woloshin,S.,Wasson,J.H.,Renfrew,R.A.,&Welch,H.G.(1999).Settingthe

revisitintervalinprimarycare.JournalofGeneralInternalMedicine,14(4),230–235.

https://doi.org/10.1046/j.1525-1497.1999.00322.x

Watson,J.(2011).Understandingsurvivalsex:youngwomen,homelessnessandintimate

relationships.JournalofYouthStudies,14(6),639–655.

https://doi.org/10.1080/13676261.2011.588945

Welch,H.G.,Chapko,M.K.,James,K.E.,Schwartz,L.M.,&Woloshin,S.(1999).Theroleof

patientsandprovidersinthetimingoffollow-upvisits.Telephonecarestudygroup.

JournalofGeneralInternalMedicine,14(4),223–229.

Wen-Chieh,L.,Bharel,M.,Jianying,Z.,O’Connell,E.,&Clark,R.E.(2015).Frequentemergency

departmentvisitsandhospitalizationsamonghomelesspeoplewithmedicaid:

Implicationsformedicaidexpansion.AmericanJournalofPublicHealth,105,S716–

S722.https://doi.org/10.2105/AJPH.2015.302693

Young,M.,Barrett,B.,Engelhardt,M.,&Moore,K.(2014).Six-monthoutcomesofan

integratedassertivecommunitytreatmentteamservingadultswithcomplexbehavioral

healthandhousingneeds.CommunityMentalHealthJournal,50(4),474–479.

https://doi.org/10.1007/s10597-013-9692-5

StreetMedicine 45

VIII.AppendicesFigure1.LogicModel

Goal:UseMercyCare’sStreetMedicineprogramtoassisthomelesspatientsinthedowntownAtlantaareawithgettingconsistentprimarycareandmentalhealthservices

Input

Activities Participation

Short-Term Outcomes

Intermediate Outcomes

Long-Term Outcomes

Grant money

Physicians, nurses, physician assistants, medical

students

Volunteers

Provide medical care during street medicine

rounds

Refer to a Mercy Care Clinic for additional

services and programs

# of patients seen during street medicine rounds

# of patients seen at a Mercy

Care clinics

# of patients that receive primary care services

# of patients that receive behavioral health services

Improvement of medical and psychiatric conditions

of homeless patients

Consistent follow-up in clinics or during Street

Medicine rounds and stable housing

Maintain stable housing and reduce morbidity and

mortality

Output

- Homeless patients do not want medical or psychiatric care - These patients do not get medical care anywhere or mostly from Grady Hospital

Homeless patients

Assumptions

StreetMedicine 46

TableSociodemographicsoftheStudyPopulation

VariableN(%)(N=258)

Gender

Female 52(20.16%)Male 206(79.84%)

Age 19-29 17(6.59%)30-39 38(14.72%)40-49 68(26.37%)50-59 98(37.98%)

60orolder 37(14.34%)Race

AfricanAmerican/Black 146(56.59%)Hispanic/Latino 1(<1%)

White 28(10.85%)Unknown 83(32.17%)

StreetMedicine 47

Table2.StreetMedicineEncounters

TotalNumberofPatientswithStreetMedicineVisits

N(%)(N=258)

TotalNumberofPatientswithBehavioralHealth

VisitsN(%)

(N=258)

TotalNumberofPatientswithPrimary

CareVisitsN(%)

(N=257)0visits 0 *137(53.10%) *51(19.84%)1visit 132(51.16%) 70(27.13%) 120(46.69%)2-3visits 83(32.17%) 35(13.57%) 66(25.68%)4visitsormore 43(16.69%) 16(6.20%) 20(7.79%)*Note:Thepatientswithnobehavioralhealthvisitsonlyhadprimarycarevisits,andthepatientswithnoprimarycarevisitsonlyhadbehavioralhealthvisits.

StreetMedicine 48

Table3.ClinicEncounters TotalNumberof

PatientswithClinicVisitsN(%)

(N=258)

TotalNumberofPatientswithBehavioralHealth

VisitsN(%)

(N=258)

TotalNumberofPatientswithPrimary

CareVisitsN(%)

(N=258)0visits 107(41.47%) *180(69.77%) *115(44.57%)1visit 23(8.91%) 28(10.85%) 33(12.79%)2–3visits 40(15.50%) 19(7.37%) 37(14.34%)4–10visits 52(20.16%) 20(7.76%) 43(16.67%)11visitsormore 36(14.02%) 11(4.28%) 30(11.65%)*Note:Thepatientswithnobehavioralhealthvisitsonlyhadprimarycarevisits,andthepatientswithnoprimarycarevisitsonlyhadbehavioralhealthvisits.

StreetMedicine 49

Table4.CaseManagement

N(%)(N=258)

Yes 83(32.17%)No 175(67.83%)

StreetMedicine 50

Table5.StreetMedicineEncountersvs.ClinicEncounters

BehavioralHealthVisits PrimaryCareVisits

StreetMedicineEncountersN(%)

(N=637)237(37.21%) 400(62.79%)

ClinicEncountersN(%)

(N=1282)373(29.10%) 908(70.83%)

StreetMedicine 51

Table6.PatientConnectionStatusN(%)

N=258*AlreadyConnected 49(18.99%)

**Started 83(32.17%)

***Disconnected 118(45.74%)

****Reconnected 8(3.10%)

N(%)

N=118DisconnectedafterStreetMedicine 83(70.34%)

DisconnectedafterClinicVisit 35(29.66%)

N(%)N=8

ReconnectedafterStreetMedicine 4(50%)

ReconnectedafterClinicVisit 4(50%)Notes:*Patientsthatwerelabeled“Started”werenewpatientsatthetimeoftheirfirstStreetMedicineencounterandwereregularlyfollowingup.**Patientsthatwerelabeled“AlreadyConnected”werealreadybeingseenataMercyCarecliniclocationatthetimeofthetheirfirstStreetMedicineencounterandregularlyfollowingup.***Patientsthatwerelabeled“Disconnected”wereneworcurrentpatientsthatdiscontinuedseeingaMercyCareproviderformorethan6months.****Patientsthatwerelabeled“Reconnected”werepatientsthathaddisconnectedfromMercyCarebutlaterre-engagedinregularfollowupvisits.

StreetMedicine 52

Table7.EDVisits TotalNumberof

PatientswithEDVisitsN(%)

(N=220)

TotalNumberofPatientswithEDVisitsBeforeMercy

CareN(%)

(N=220)

TotalNumberofPatientswithEDVisits

AfterMercyCareN(%)

(N=220)0visits 18(8.18%) 75(34.09%) 70(31.82%)1visit 26(11.82%) 38(17.27%) 38(17.24%)2–3visits 50(22.73%) 45(20.45%) 46(20.91%)4–10visits 82(37.29%) 52(23.64%) 46(20.91%)11visitsormore 44(20.04%) 10(4.54%) 20(9.07%)

StreetMedicine 53

Table8.PairedSampleTTestforEDVisitsBeforeandAfterMercyCare PairedDifferences

t df

Sig.(2-

tailed)N MeanStd.Dev.

Std.Err.

95%CLLower

95%CLUpper

PairAfter-Before

220 0.5091 6.1855 0.4170 -0.3128 1.3310 1.22 219 0.2235

StreetMedicine 54

Table9.HospitalAdmissionsBeforeandAfterMercyCare BeforeMercyCare

N(%)(N=220)

AfterMercyCareN(%)

(N=220)TotalNumberofPatientswithHospitalAdmissionDays

Nodays 165(75.00%) 163(74.09%)1day 20(9.09%) 11(5.00%)

2-3days 15(6.82%) 14(6.36%)4–10days 17(7.72%) 23(10.44%)

11daysormore 3(1.35%) 9(14.91%) TotalNumberofPatientswithPsychiatricAdmissionVisits

Novisits 190(86.36%) 202(91.82%)1visit 20(9.09%) 12(5.45%)2visits 4(1.82%) 3(1.36%)

3visitsormore 6(2.73%) 3(1.36%) TotalNumberofPatientswithPhysicalHealthAdmissionVisits

Novisits 188(85.45%) 174(79.09%)1visit 29(13.18%) 28(12.73%)2visits 2(<1%) 12(5.45%)

3visitsormore 1(<1%) 6(2.72%)

StreetMedicine 55

Table10.PairedSampleTTestforTotalHospitalAdmissionDaysBeforeandAfterMercyCare PairedDifferences

t df

Sig.(2-

tailed)N MeanStd.Dev.

Std.Err.

95%CLLower

95%CLUpper

PairAfter-Before

220 0.8591 6.2660 0.4225 0.0265 1.6917 2.03 219 0.0432

StreetMedicine 56

Table11.PairedSampleTTestforPsychiatricAdmissionsBeforeandAfterMercyCare PairedDifferences

t df

Sig.(2-

tailed)N MeanStd.Dev.

Std.Err.

95%CLLower

95%CLUpper

PairAfter-Before

220 -0.0909 0.8815 0.0594 -0.2080 0.0262 -1.53 219 0.1276

StreetMedicine 57

Table12.PairedSampleTTestforPhysicalHealthAdmissionsBeforeandAfterMercyCare PairedDifferences

t df

Sig.(2-

tailed)N MeanStd.Dev.

Std.Err.

95%CLLower

95%CLUpper

PairAfter-Before

220 0.1864 0.9962 0.0672 0.0540 0.3187 2.77 219 0.0060