research article
Topic: Healthcare Disparities Trauma
Rubic and Peer Review article attached add more if needed
3-5 pages must be able to write part two of this assignment
lable at ScienceDirect
The American Journal of Surgery 218 (2019) 842e846
Contents lists avai
The American Journal of Surgery
journal homepage: www.americanjournalofsurgery.com
Shelby Chun Fat a, Juan P. Herrera-Escobar a, Anupamaa J. Seshadri b, Syeda S. Al Rafai a,
Zain G. Hashmi a, Elzerie de Jager a, Constantine Velmahos c, George Kasotakis d,
George Velmahos c, Ali Salim b, Adil H. Haider a, b, Deepika Nehra b, *
a Center for Surgery and Public Health, Brigham and Women’s Hospital, Harvard Medical School and Harvard T.H Chan School of Public Health, Boston, MA,
USA
b Department of Surgery, Division of Trauma, Burn and Surgical Critical Care, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA
c Department of Surgery, Division of Trauma, Emergency Surgery, and Surgical Critical Care, Massachusetts General Hospital, Harvard Medical School,
Boston, MA, USA
d Department of Surgery, Division of Trauma, Acute Care Surgery & Surgical Critical Care, Boston University School of Medicine, Boston, MA, USA
a r t i c l e i n f o
Article history:
Received 28 February 2019
Received in revised form
18 March 2019
Accepted 23 March 2019
This work was presented at the Academic
Surgical Congress on January 30 – February
1, 2018 in Jacksonville, FL.
Keywords:
Trauma
Racial disparities
Long-term outcomes
Rehabilitation utilization
Post-discharge
Patient-reported
* Corresponding author. 75 Francis St., Boston, MA,
E-mail address: dnehra@bwh.harvard.edu (D. Neh
https://doi.org/10.1016/j.amjsurg.2019.03.024
0002-9610/
© 2019 Published by Elsevier Inc.
a b s t r a c t
Background: Racial disparities in trauma outcomes have been documented, but little is known about
racial differences in post-discharge healthcare utilization. This study compares the utilization of post-
discharge healthcare services by African-American and Caucasian trauma patients.
Methods: Trauma patients with an Injury Severity Score (ISS)�9 from three Level-I trauma centers were
contacted between 6 and 12 months post-injury. Utilization of trauma-related healthcare services was
asked. Coarsened exact matching (CEM) was used to match African-American and Caucasian patients.
Conditional logistic regression then compared matched patients in terms of post-discharge healthcare
utilization.
Results: 182 African-American and 1,117 Caucasian patients were followed. Of these, 141 African-
Americans were matched to 628 Caucasians. After CEM, we found that African-American patients
were less likely to use rehabilitation services [OR:0.64 (95% CI:0.43e0.95)] and had fewer injury-related
outpatient visits [OR:0.59 (95% CI:0.40e0.86)] after discharge.
Conclusions: This study shows the existence of racial disparities in post-discharge healthcare utilization
after trauma for otherwise similarly injured, matched patients.
© 2019 Published by Elsevier Inc.
Background
Trauma remains one of the most common causes of long-term
functional impairment and disability. Traumatically injured pa-
tients commonly suffer from reduced quality of life, poor functional
outcomes, psychologic disturbances, chronic pain, social disinte-
gration and the burden of their oftentimes high medical costs.1,2
The care of these patients does not end on discharge from the
hospital and many of these patients require ongoing rehabilitation
services after discharge. Post-hospitalization care, especially care
provided at a rehabilitation center, has been previously shown to
positively contribute to improving long-term outcomes and func-
tional independence after traumatic injury.3e5
Disparities in healthcare delivery and outcomes have been
02215, USA.
ra).
shown for many conditions.6 Trauma has historically been thought
to be immune to such disparities given its emergent nature. Un-
fortunately, this is not the case and racial disparities have been well
established even for traumatically injured patients. When
compared to Caucasian patients, African American trauma patients
have a higher mortality and a higher likelihood of long-term dis-
ability.6e12 Importantly, this relationship between African Amer-
ican race and higher trauma mortality has been shown to be
independent of socioeconomic status.6
African American patients are also more likely to be discharged
home as opposed to rehabilitation centers or skilled nursing facil-
ities compared to Caucasian patients.13e16 There are likely multiple
factors at play other than race, such as insurance and socioeco-
nomic status.13,14 These previous studies are however limited as
they have focused on specific injuries such as spinal cord injury15
and traumatic brain injury,16 or are based on data from the Na-
tional Trauma Data Bank where only information regarding
mailto:dnehra@bwh.harvard.edu
http://crossmark.crossref.org/dialog/?doi=10.1016/j.amjsurg.2019.03.024&domain=pdf
www.sciencedirect.com/science/journal/00029610
www.americanjournalofsurgery.com
https://doi.org/10.1016/j.amjsurg.2019.03.024
https://doi.org/10.1016/j.amjsurg.2019.03.024
https://doi.org/10.1016/j.amjsurg.2019.03.024
S. Chun Fat et al. / The American Journal of Surgery 218 (2019) 842e846 843
discharge disposition is available.13,14 As a result, we to date have
not been able to capture the long-term contact that these trau-
matically injured patients have with other types of post-discharge
medical care, such as that received in clinic or in the emergency
department (ED).
The Functional Outcomes and Recovery after Trauma Emer-
gencies (FORTE) project is a multi-center effort that collects
Patient-Reported Outcomes (PROs) from three Boston level 1
trauma centers. In Massachusetts, insurance is significantly
expanded via Medicaid, with an extremely high proportion of
trauma patients either already insured or able to become insured
during their hospitalization. This data, therefore, is uniquely able to
evaluate post-discharge service utilization in patients based on race
without the confounding effects of insurance status.
The aim of this study was to examine the differences in patient-
reported utilization of post-discharge services, including rehabili-
tation services, outpatient clinics and the ED, between African
American and Caucasian patients who have sustained moderate to
severe traumatic injuries. We hypothesized that African American
patients would utilize rehabilitation services and outpatient clinics
less frequently and the ED more frequently as compared to their
Caucasian counterparts.
Methods
Data sources and patient population
We used the FORTE project data set for the present study. The
FORTE project is a multi-center effort among three Boston level I
trauma centers (Brigham and Women’s Hospital, Massachusetts
General Hospital, and Boston Medical Center), to include long-term
functional and PROs measures into trauma registries. These long-
term functional and PROs measures are collected via a phone
interview conducted at 6 or 12 months after injury for patients who
have sustained moderate or severe trauma (Injury Severity Score
[ISS] >9). This interview consists of an initial screening and verbal
consent followed by a series of survey questions to assess func-
tional and PROs measures, as well as other relevant aspects of the
patient recovery experience. This survey includes the following:
work and insurance status, education, Trauma Quality of Life In-
strument, Short-Form Health Survey Version 2.0, screening for
Posttraumatic Stress Disorder and information regarding post-
discharge contacts with healthcare. All patients who participated
in the FORTE project were either English or Spanish-speaking
adults and all interviews in Spanish were performed by a Spanish
speaking interviewer. Interviewers were only given the patient
name, contact information, date of injury, age, and gender, and
were blinded about injury-related characteristics and some patient
characteristics such as race or ethnicity. Interview data were
collected and managed using REDCap (Research Electronic Data
Capture) hosted at Partners Healthcare. Further details regarding
the patient recruitment and data collection procedures for the
FORTE project have been described previously.17
This dataset was then linked with institutional trauma registry
data to capture patient demographic and injury-related character-
istics. The patients’ race was determined using the institutional
trauma registry. For the present study, we included African Amer-
ican and Caucasian patients from the FORTE data set from
December 2015 through July 2018. Patients whose race was missing
or listed as anything other than African American or Caucasian were
not included. Other clinical variables extracted from the trauma
registry included age, sex, insurance, mechanism of injury, ISS,
Abbreviated Injury Scale (AIS) per body region, intensive care unit
(ICU) admission, ventilator requirement, length of stay, in-hospital
complications and discharge disposition (home, home with health
services, rehabilitation facility, nursing home/skilled nursing facil-
ity, other.)
Post-discharge contacts with healthcare
Information regarding the patient’s post-discharge contact with
healthcare was obtained by patient report. Specifically, participants
were asked whether they received any rehabilitation services after
discharge (i.e. were discharged to a rehabilitation center or skilled
nursing facility or received home or outpatient services like phys-
ical or occupational therapy), whether they received injury-related
outpatient follow-up in the clinic setting and whether they pre-
sented to an ED for an injury-related problem after their discharge
from the hospital.
Statistical analysis
Patient demographics and clinical characteristics were
compared between Caucasian and African American patients. Cat-
egorical data was compared using chi-squared tests and continuous
data was compared using t-tests or Wilcoxon Rank Sum tests as
appropriate. A sensitivity analysis was performed to assess the
presence of response bias by comparing baseline characteristics of
African American and Caucasian patients who participate in the
FORTE study versus those who did not.
The trauma registry discharge disposition was grouped into
those who received any rehabilitation services after discharge
(rehabilitation, nursing home or skilled nursing facility or home
with health services) and those who did not (home). Logistic
regression adjusting for age, sex, education, insurance, injury
mechanism, ISS, head injury, torso injury, extremity injury, ICU
admission, ventilator use, length of stay, and hospital was con-
ducted to compare this between Caucasian and African American
patients.
Coarsened exact matching
The Coarsened Exact Matching (CEM) algorithm18 was used to
match Caucasian patients to African American patients in a 1:many
ratio. CEM creates a balance between patients in both groups with
slightly different variables while maintaining similarity which al-
lows for higher chances of matching than that of an exact matching
algorithm. Matching was used to control for the influence of po-
tential confounding factors: age, gender, injury type (blunt or
penetrating) and ISS. CEM categorizes (coarsens) continuous data
temporarily into bins with predetermined, appropriate widths to
create meaningful groups. The width of these bins allows for con-
trol of the amount of imbalance in the matching process. The bin
widths used in this study were determined based on the distribu-
tion of the data for continuous variables (age and ISS). Conditional
logistic regression (CEM weighted) was then used to compare
matched patients in terms of post-discharge healthcare utilization.
All statistical analyses were performed using Stata Statistical Soft-
ware Analysis (version 14).
Results
There were 3,431 moderate-to-severely injured patients be-
tween December 2015 and July 2018 who were eligible to partici-
pate in the FORTE questionnaire, and of these, 2,894 identified as
Caucasian (74%) or African American (11%). The remaining 15%
corresponded to 8% of patients of other races and 7% of missing
data. Of the eligible Caucasian or African American patients, 1,299
completed the FORTE questionnaire; 1,117 (86.0%) identified as
Caucasian and 182 identified as African American (Fig. 1).
Fig. 1. Study flowchart.
Table 1
Patient demographic, clinical, and injury-specific characteristics by race.a
Caucasian (n ¼ 1,117)
Age (years), mean (SD) 65 (19.7)
Male sex 574 (51)
Education, Greater than high school 635 (58)
Public Insurance 592 (60)
Medicare 520 (88)
Medicaid 42 (7)
Other 30 (5)
Mechanism of Injury, Blunt 1094 (98)
Injury Severity Score (ISS), mean (SD) 14 (7.0)
Head injury, AIS�2 433 (39)
Torso Injury, AIS�2 250 (22)
Extremity Injury, AIS�2 729 (65)
ICU admission 418 (37)
Ventilator 118 (11)
Complications 229 (21)
Length of stay (days), median (IQR) 5 (3e7)
Discharge disposition
Home 270 (24)
Home with health services 192 (17)
Rehabilitation facility 421 (38)
Nursing home/Skilled nursing facility 189 (17)
Other 43 (4)
AIS: Abbreviated Injury Scale.
p-values with significant p-value < 0.05 were highlighted in bold.
a Data presented as n (%) except as otherwise listed.
S. Chun Fat et al. / The American Journal of Surgery 218 (2019) 842e846844
Sensitivity analysis comparing the 1,299 participants to the
1,595 non-participants showed that participants were more likely
to have higher mean ISS (14.1 [SD: 7.2] vs 13.55 [SD: 6.7]; p: 0.022)
and higher ICU admission rate (38% vs 34%; p: 0.045) compared to
non-participants. Participants and non-participants did not differ in
age, sex, insurance, injury type, presence of head injury, extremity
injury, ventilator use, in-hospital complications, length of stay, or
discharge disposition. Additionally, no significant differences in
race were found between participants and non-participants, which
means that the distribution of Caucasians and African Americans in
the study sample reflects the data set in this regard.
Patient demographics are summarized in Table 1. The Caucasian
patients were significantly older (mean age 65 [SD 19.7] years) with
fewer males (51% male) compared to the African American patients
(mean age 45 [SD 20] years; 67% male). There was no significant
difference in insurance status between groups with only 6 total
patients (<1%) being uninsured. Blunt mechanism of injury was the
most common type, with average ISS being slightly higher in the
African American as compared to Caucasian group (15.3 [SD 8.4] vs.
14 [SD 7.0] respectively). There was a significant difference in the
discharge disposition as reported in the institutional trauma reg-
istries, with African American patients being more likely to be
discharged to home and home with services and less likely to be
discharged to rehabilitation facilities and nursing home/skilled
nursing facilities than Caucasians (Table 1). However, after adjust-
ing for potential confounders (age, sex, education, insurance, injury
mechanism, ISS, head injury, torso injury, extremity injury, ICU
admission, ventilator use, length of stay, and hospital), there was no
significant difference in discharge disposition (as reported in the
trauma registry) for African American compared to Caucasian pa-
tients (p ¼ 0.408).
When unadjusted patient-reported post-discharge healthcare
utilization outcomes were examined in the unmatched cohort,
African Americans were significantly less likely to utilize post-
discharge rehabilitation services and more likely to be seen in the
ED for an injury-related issue (Table 2).
After CEM, 628 Caucasian patients were matched to 141 African
American patients. Overall imbalance, which is measured by the
African American (n ¼ 182) p-value
45 (20.0) <0.001
122 (67) <0.001
44 (25) <0.001
96 (57) 0.482
27 (28) <0.001
61 (64)
8 (8)
137 (76) <0.001
15.3 (8.4) 0.021
57 (31) 0.055
71 (39) <0.001
105 (58) 0.048
75 (41) 0.329
32 (18) 0.006
47 (26) 0.104
5 (3e10) 0.288
<0.001 78 (43) 33 (18) 49 (27) 16 (9) 4 (2)
Table 2
Unadjusted analysis of patient-reported post-discharge healthcare utilization by race in the unmatched cohort.a
Total (n ¼ 1,299) Caucasian (n ¼ 1,117) African American (n ¼ 182) p-value
Rehabilitation services 1,008 (77.6) 891 (79.8) 117 (64.3) <0.001
Injury- related outpatient visit 598 (46.0) 525 (47.0) 73 (40.0) 0.084
Injury-related ED visit 147 (11.3) 113 (10.1) 34 (18.7) 0.001
a Data presented as n (%).
S. Chun Fat et al. / The American Journal of Surgery 218 (2019) 842e846 845
multivariable L1 distance, was initially 0.61. After CEM was per-
formed, overall imbalance dropped to L1 ¼0.25. Perfect balance
(exact matching) is indicated by L1 ¼0 with the largest imbalance
being L1 ¼1 (complete separation).
In the CEM matched adjusted analysis, African American pa-
tients were significantly less likely to report receiving rehabilitation
services after discharge and were also less likely to be seen for an
injury-related outpatient visit. There was not a statistical signifi-
cance difference in injury-related ED visits between race groups
(Fig. 2).
Discussion
The goal of trauma care is not only to reduce injury-related
morbidity and mortality, but also to improve long-term outcomes
in trauma patients. The impact of traumatic injuries persists far
beyond discharge from the hospital. As we better understand the
long-term consequences of trauma, we strive to improve both in-
hospital and post-hospitalization care and services to mitigate the
negative long-term consequences of traumatic injury. Our results
suggest that racial disparities exist in the post-discharge utilization
of healthcare services, which we know affect long term functional
outcomes after injury. While African American patients were found
to have no significant difference in discharge disposition as
compared to Caucasian patients as reported in the trauma registry,
African American patients were less likely to actually receive
rehabilitation services after discharge and were less likely to be
seen for injury-related outpatient visits. These racial discrepancies
Fig. 2. CEM-weighted analysis comparing the utilization of post-discharge health
in post-discharge health services utilization may contribute to
worse long-term trauma outcomes.
This study is unique in that we utilized patient-reported data to
assess post-discharge utilization of injury-related outpatient ser-
vices. Patient-reported information are a way of identifying areas
for improvement from a patient’s perspective to emphasize effi-
ciency, safety, and high-quality care without bias or interpreta-
tion.19 Symptom severity, treatment impact, outcomes, and
identification of areas important to patients can be assessed to
further facilitate the patient-provider relationship.20,21 This makes
it especially useful in measuring physical, mental, and social
health.22 Furthermore, it enhances patient engagement in shared
decision-making by prompting the patient to assess their experi-
ences, values, preferences, and goals about their healthcare.20
Prior studies investigating discharge disposition or post-
discharge healthcare utilization have used institutional databases
for assessment which do not capture care outside the primary
trauma provider site. In contrast to our study results, these previous
studies have found that Caucasian patients were less likely to
follow-up in outpatient trauma clinics as compared to African
American and Hispanic patients.23,24 Patients may not return to the
same trauma center for follow-up care due to insurance re-
strictions, geographical distance, or personal preference.23 Using
patient-reported information in our study allowed for consider-
ation of the post-discharge healthcare utilization not limited to the
primary trauma care provider facility.
An important strength of this study is the fact that less than one
percent of patients included were uninsured. There was no
care services between African American and Caucasian (reference) patients.
S. Chun Fat et al. / The American Journal of Surgery 218 (2019) 842e846846
significant difference in insurance status in our study population,
though numerous studies have attributed racial disparities in out-
comes to insurance status.8,23,25 This is likely due to the unique,
expanded health insurance coverage via Medicaid in Massachu-
setts. Having the majority of our patients insured mitigates the
effect of insurance in our study. Therefore, our findings of decreased
utilization of post-discharge rehabilitation and outpatient services
is potentially more attributable to the variable of race, as compared
to other studies whose cohorts are not as uniformly insured.
Limitations of this study include a selection bias in that only
patients who answered the phone calls and were willing to
participate were included. We assessed the potential of selection
bias by comparing baseline characteristics of patients who partic-
ipated to those who did not, and the only significant differences
were that responders had a higher ISS and ICU admission rate. The
patient-reported nature of our outcome measures also potentially
introduces some room for error based upon the patient’s recall or
understanding of services provided, but is likely still the most ac-
curate way of capturing a patient’s actual service utilization. So-
cioeconomic status has been associated with post-discharge
healthcare utilization13,26e29 and there was a significant difference
in the education level between the Caucasian and African American
cohort. However, education was not a variable included in the CEM
analysis because we do not feel this variable should independently
influence the discharge disposition or services provided to a pa-
tient. Lastly, the generalizability of this study is uncertain as it was
conducted in a single city. However, the study does include data
from three level I trauma centers with somewhat different patient
populations.
Conclusions
In this multi-institutional study using long-term patient-re-
ported data we demonstrate racial disparities in post-discharge
healthcare utilization after trauma for similarly injured, cohort
matched patients. African American patients were less likely to use
post-discharge rehabilitation services and less likely to be seen in
the outpatient setting for injury-related services as compared to
Caucasian counterparts. These differences in post-discharge
resource utilization may contribute to differences in long-term
outcomes. There are likely many factors that contribute to these
differences, such as an unconscious provider bias, patient under-
standing, miscommunication, access to care, and evidence of
distrust toward medical providers.26,28e30 Better understanding the
reasons for these differences in post-discharge resource utilization
may provide insight into avenues for equalizing long-term out-
comes for traumatically injured patients.
Conflicts of interest
The authors of this manuscript have no conflicts of interest.
Funding
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
Acknowledgements
We acknowledge the entire FORTE project team for their
contribution and dedication with data collection as well as their
continued support.
References
1. Sleet DA, Dahlberg LL, Basavaraju SV, et al. Injury prevention, violence pre-
vention, and trauma care: building the scientific base. CDC MMWR. 2011;60:
78e85.
2. Haider AH, Herrera-Escobar JP, Al Rafai SS, et al. Factors associated with long-
term outcomes after injury: results of the functional outcomes and recovery
after trauma emergencies (FORTE) multi-center cohort study. Ann Surg. 2018.
Article in press.
3. Nusselder WJ, Looman CWN, Franco OH, et al. The relation between non-
occupational physical activity and years lived with and without disability.
J Epidemiol Community Health. 2008;62:823e828.
4. Jette DU, Warren RL, Wirtalla C. The relation between therapy intensity and
outcomes of rehabilitation in skilled nursing facilities. Arch Phys Med Rehabil.
2005;86:373e379.
5. Hansen L, Shaheen A, Crandall M. Outpatient follow-up after traumatic injury:
challenges and opportunities. J Emerg Trauma Shock. 2014;7(4):256e260.
6. Haider AH, Weygandt PL, Bentley JM, et al. Disparities in trauma care and
outcomes in the United States: a systematic review and meta-analysis.
J Trauma Acute Care Surg. 2013;74(5):1195e1205.
7. Glance LG, Osler TM, Mukamel DB, et al. Trends in racial disparities for injured
patients admitted to trauma centers. Health Serv Res. 2013;48(5):1684e1703.
8. Haider AH, Chang DC, Efron DT, et al. Race and insurance status as risk factors
for trauma mortality. Arch Surg. 2008;143(10):945e949.
9. Hicks CW, Hashmi ZG, Velopulos C, et al. Association between race and age in
survival after trauma. JAMA Surg. 2014;149(7):642e647.
10. Strong BL, Torain JM, Greene CR, Smith GS. Outcomes of trauma admission for
falls: influence of race and age on in-hospital and post-discharge mortality. Am
J Surg. 2016;212(4):638e644.
11. Lin S-F, Beck AN, Finch BK. Blackewhite disparity in disability among U.S. Older
adults: age, period, and cohort trends. J Gerontol B Psychol Sci Soc Sci.
2014;69(5):784e797.
12. Ng JH, Bierman AS, Elliott MN, et al. Beyond black and white: race/ethnicity and
health status among older adults. Am J Manag Care. 2014;20(3):239e248.
13. Englum BR, Villegas C, Bolorunduro O, et al. Racial, ethnic, and insurance status
disparities in use of posthospitalization care after trauma. J Am Coll Surg.
2011;213(6):699e708.
14. Sacks GD, Hill C, Rogers SO. Insurance status and hospital discharge disposition
after trauma: inequities in access to postacute care. J Trauma. 2011;71:
1011e1015.
15. Arango-Lasprilla JC, Ketchum JM, Stevens LF, et al. Ethnicity/racial differences
in employment outcomes following spinal cord injury. NeuroRehabilitation.
2009;24(1):37e46.
16. Marquez de la Plata C, Hewlitt M, de Oliveira A, et al. Ethnic differences in
rehabilitation placement and outcome after TBI. J Head Trauma Rehabil.
2007;22(2):113e121.
17. Rios-Diaz AJ, Herrera-Escobar JP, Lilley EJ, et al. Routine inclusion of long-term
functional and patient-reported outcomes into trauma registries: the FORTE
project. J Trauma Acute Care Surg. 2017;83(1):97e104.
18. Blackwell M, Iacus S, King G, Porro G. Cem: coarsened exact matching in Stata.
STATA J. 2009;9(4):524e546.
19. Botero JP, Thanarajasingam G, Warsame R. Capturing and incorporating
patient-reported outcomes into clinical trials: practical considerations for cli-
nicians. Curr Oncol Rep. 2016;18:61.
20. Lavallee DC, Chenok KE, Love RM, et al. Incorporating patient-reported out-
comes into health care to engage patients and enhance care. Health Aff.
2016;35(4):575e582.
21. Jenkinson C, Morley D. Patient reported outcomes. Eur J Cardiovasc Nurs.
2015;15(2):112e113.
22. Driban JB, Morgan N, Price LL, et al. Patient-Reported Outcomes Measurement
Information System (PROMIS) instruments among individuals with symp-
tomatic knee osteoarthritis: a cross-sectional study of floor/ceiling effects and
construct validity. BMC Musculoskelet Disord. 2015;16:253.
23. Stone ME, Marsh J, Cucuzzo J, et al. Factors associated with trauma clinic
follow-up compliance after discharge: experience at an urban Level I trauma
center. J Trauma Acute Care Surg. 2014;76(1):185e190.
24. Stewart MG, Chen AY. Factor predictive of poor compliance with follow-up
after facial trauma: a prospective study. Otolaryngol Head Neck Surg.
1997;117:72e75.
25. Greene WR, Oyetunji TA, Bowers U, et al. Insurance status is a potent predictor
of outcomes in both blunt and penetrating trauma. Am J Surg. 2010;199:
544e557.
26. Brown LE, Burton R, Hixon B, et al. Factors influencing emergency department
preference for access to healthcare. West J Emerg Med. 2012;13(5):410e415.
27. Leukhardt WH, Golob JF, McCoy AM, et al. Follow-up disparities after trauma: a
real problem for outcomes research. Am J Surg. 2010;199(3):348e353.
28. Baker DW, Stevens C, Brook R. Determinants of emergency department use: are
race and ethnicity important? Ann Emerg Med. 1996;28:677e682.
29. Lee JS, Tamayo-Sarver J, Kinneer P, Hobgood C. Association between patient
race/ethnicity and perceived interpersonal aspects of care in the emergency
department. J Natl Med Assoc. 2008;100:79e85.
30. Burgess DJ, Fu SS, van Ryn M. Why do providers contribute to disparities and
what can Be done about it? J Gen Intern Med. 2004;19(11):1154e1159.
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref1
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref1
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref1
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref1
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref2
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref2
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref2
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref2
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref3
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref3
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref3
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref3
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref4
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref4
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref4
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref4
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref5
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref5
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref5
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref6
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref6
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref6
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref6
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref7
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref7
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref7
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref8
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref8
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref8
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref9
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref9
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref9
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref10
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref10
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref10
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref10
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref11
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref11
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref11
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref11
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref11
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref12
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref12
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref12
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref13
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref13
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref13
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref13
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref14
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref14
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref14
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref14
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref15
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref15
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref15
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref15
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref16
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref16
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref16
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref16
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref17
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref17
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref17
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref17
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref18
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref18
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref18
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref19
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref19
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref19
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref20
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref20
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref20
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref20
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref21
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref21
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref21
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref22
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref22
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref22
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref22
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref23
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref23
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref23
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref23
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref24
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref24
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref24
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref24
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref25
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref25
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref25
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref25
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref26
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref26
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref26
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref27
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref27
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref27
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref28
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref28
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref28
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref29
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref29
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref29
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref29
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref30
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref30
http://refhub.elsevier.com/S0002-9610(19)30306-X/sref30
Reproduced with permission of copyright owner. Further reproduction
prohibited without permission.
- Racial disparities in post-discharge healthcare utilization after trauma
Background
Methods
Data sources and patient population
Post-discharge contacts with healthcare
Statistical analysis
Coarsened exact matching
Results
Discussion
Conclusions
Conflicts of interest
Funding
Acknowledgements
References