Race and Insurance Status as Risk Factorsfor Trauma Mortality
Adil H. Haider, MD, MPH; David C. Chang, MPH, MBA, PhD; David T. Efron, MD;Elliott R. Haut, MD; Marie Crandall, MD, MPH; Edward E. Cornwell III, MD
To determine the effect of race and insur-ance status on trauma mortality.
Review of patients (aged 18-64 years; InjurySeverityScore
9)includedintheNationalTraumaDataBank (2001-2005). African American and Hispanic pa-tients were each compared with white patients and in-sured patients were compared with uninsured patients.Multiple logistic regression analyses determined differ-ences in survival rates after adjusting for demographics,injuryseverity(InjurySeverityScoreandrevisedTraumaScore), severity of head and/or extremity injury, and in- jury mechanism.
A total of 429751 patients met inclusion cri-teria. African American (n=72 249) and Hispanic(n=41770) patients were less likely to be insured andmorelikelytosustainpenetratingtraumathanwhitepa-tients (n=262878). African American and Hispanic pa-tientshadhigherunadjustedmortalityrates(white,5.7%;AfricanAmerican,8.2%;Hispanic,9.1%;
=.05forAfri-can American and Hispanic patients) and an increasedadjusted odds ratio (OR) of death compared with whitepatients(AfricanAmericanOR,1.17;95%confidencein-terval [CI], 1.10-1.23; Hispanic OR,1.47; 95% CI, 1.39-1.57). Insured patients (47%) had lower crude mortal-ityratesthanuninsuredpatients(4.4%vs8.6%;
=.05).InsuredAfricanAmericanandHispanicpatientshadin-creasedmortalityratescomparedwithinsuredwhitepa-tients.Thiseffectworsenedforuninsuredpatientsacrossgroups(insuredAfricanAmericanOR,1.2;95%CI,1.08-1.33;insuredHispanicOR,1.51;95%CI,1.36-1.64;un-insuredwhiteOR,1.55;95%CI,1.46-1.64;uninsuredAfri-can American OR,1.78; 95% CI, 1.65-1.90; uninsuredHispanic OR,2.30; 95% CI, 2.13-2.49). The referencegroupwasinsured white patients.
Race and insurance status each indepen-dently predicts outcome disparities after trauma. Afri-can American, Hispanic, and uninsured patients haveworseoutcomes,butinsurancestatusappearstohavethestronger association with mortality after trauma.
Arch Surg. 2008;143(10):945-949
-care Research and Quality
re-ported that, “Some ethnic mi-noritiesandlow-incomefamiliesof all races tend to be in poorerhealth.” Subsequently, a body of litera-ture has accumulated describing race-basedoutcomedisparitiesinsurgicalpopu-lations.
In trauma care, racial disparitiesinposttraumaticfunctionaloutcomeshavebeen described in both adult
and pediat-ric trauma populations.
Disparitiesbasedonhealthinsurancesta-tushavealsobeendemonstratedinthede-tection, treatment, and outcomes of sev-eraldiseaseconditions,includingcancer.
Adulttraumastudieshavedescribeddiffer-ences in outcomes and discharge disposi-tions based on insurance coverage.
In astudy by Marquez de la Plata and col-leagues,
itwassuggestedthatinsurancesta-tus, rather than race, determined dispari-ties in rehabilitation placement. Anotherstudy of elderly patients hospitalized sec-ondary to a traumatic fall also found thatpaymenttypewasafactorindecidingapa-tient’s discharge location.
Currentliteraturefailstoadequatelyex-plain the interactions between race, in-surance status, and outcomes followingtrauma. It is not clear whether the previ-ouslydescribedracialdisparitiescanbeex-plained solely by insurance or socioeco-nomicstatus.Wethereforeendeavoredtouse a national database to determine theindependent effects of race and insur-ance status on trauma mortality.
STUDY DESIGN ANDPATIENT POPULATION
Departments of Surgery, The Johns Hopkins UniversitySchool of Medicine, Baltimore,Maryland (Drs Haider, Chang,Efron, and Haut); FeinbergSchool of Medicine,Northwestern University,Chicago, Illinois (Dr Crandall)Howard University College of Medicine, Washington, DC(Dr Cornwell).
(REPRINTED) ARCH SURG/VOL 143 (NO. 10), OCT 2008 WWW.ARCHSURG.COM
©2008 American Medical Association. All rights reserved.