Geriatric Ground Level Falls: Injury Patterns and Outcome Analysis

This report presents a retrospective review of a Level I trauma center's database, analyzing adult trauma patients admitted after ground level falls (GLFs) between 2003 and 2013. The study examines demographics, injury patterns, and outcomes, with a focus on the geriatric population and the impact of anticoagulation and antiplatelet use. The cohort included 5088 patients, with a significant portion over 60 years old. The research found that mortality increased with age, and the likelihood of home discharge decreased. Age was associated with an increased incidence of certain fractures. Aspirin use was linked to intracranial bleeds, and both aspirin and warfarin were associated with increased mortality. The study concludes that GLF patients are not homogenous and that age and medication use significantly influence injury patterns and outcomes, highlighting implications for patient evaluation and management.

The older theyarethe harderthey fall: Injury patternsand outcomesby
age after ground level falls$
Bishwajit Bhattacharya*, Adrian Maung, Kevin Schuster,Kimberly A. Davis
YaleSchoolof Medicine,Sectionof GeneralSurgery,Traumaand SurgicalCriticalCare,Departmentof Surgery,New Haven,CT 06510,UnitedStates
A R T I C L E I N F O
Articlehistory:
Accepted11 June 2016
Keywords:
Ground level fall
Geriatric
Outcome
Mortality
A B S T R A C T
Background:Trauma centers are seeing an increasing number of geriatric patients that are more
susceptible to injuries even from relatively minor insults such as a ground level fall (GLF). As life
expectancy increases, people are living in the geriatric age bracket for decades and often use
anticoagulationagents for various comorbidities.We hypothesizethat this patient population is not
homogenousand we investigatedthe injury patterns and outcomesafter GLF as a function of age and
anticoagulationuse.We also soughtto identify injury patternsand patientcharacteristicsof GLF patients.
Methods:A retrospectivereview of a LevelI traumacenter’sdatabaseidentifiedall adult (age>18) trauma
patients admitted after GLFs between 1/2003 and 12/2013.Demographics,injury patterns,antiplatelet
use, anticoagulation use (including warfarin, enoxaparin, and rivaroxaban) and outcomes were
abstracted.
Results:The cohort included5088 patients.3990patientswere >60yearsand 38.2%were male.With each
decade,althoughthe mean ISS did not considerablychange(range7.0–8.6),mortality increased(0.9%at
<60yearsvs. 5.5%at >90years),and the likelihood of home dischargedecreaseddramatically(73.7%at
<60years vs. 18.2%at >90years).Abdominal solid organ injuries were rare (0.8%).Age was associated
with an increasedincidenceof cervicalspine (p =0.002),rib (p =0.009) and pelvic fractures(p < 0.001).
Only aspirin use was significantly associatedwith intracranial bleed (p =0.001).Aspirin (p =0.049) or
warfarin (p < 0.001) use was associatedwith increasedoverall mortality.
Conclusion:GLF patients are not homogenousas certain injury patterns change with increasing age.
Aspirin use was associatedwith an increasedincidenceof intracranialbleeds,whereasother antiplatelet
or anticoagulationagentswere not. GLF is also associatedwith significantmorbidity and mortality that
increasesdramaticallywith age.Both aspirin and warfarin are independentlyassociatedwith increased
mortality. These patient differenceshave implications for their evaluationand management.
Levelof evidence:Epidemiological/prognosticstudy level IV.
Published by Elsevier Ltd.
Background
Overthe last six decadeslife expectancyin the US has increased
to 78.7 years from 68.2 years [1,2]. The geriatric population is
expectedto form 25%of the US populationby 2030 [3] including a
significantincreasein thoseover85 [4]. As the populationagesand
the averagelife expectancyincreases,traumacentersare seeingan
increasingnumberof elderlypatients—oftenat the extremesof age
[5]. This segment of the population represents an increasing
demand on the health care systemdue to the unique challenges
related to decreasedphysiologicalreserveand increasedfrailty.
Ground level fall (GLF), or a fall from standing height, is the
most common mechanismof trauma for geriatric patients in the
US and worldwide [6,7].GLFsincreasein frequencywith increasing
agewith an annualincidenceof 30%in those over 65 rising to 50%
in thoseover 80 yearsof age[8]. This relativelyminor mechanism
of trauma results in greater injury severity in elderly patients
compared to younger patients due to the inherent frailty
associated with aging [9,10]. In addition, the similar energy
transfer that occurs in GLF makes this an ideal mechanism for
investigatinginjury patternsand interactionsof comorbidities.
As more patients live longer, the age distribution for this
populationalso shifts towardsthe more elderly.It is now common
to live in the geriatric age bracket for decades.Since physiologic
$ Presented:Poster Presentation74th Annual Meeting AAST 2015,Las VegasNV
September9–12, 2015.
* Correspondingauthor at: 330 Cedar Street,P.O. Box 208062,New Haven, CT
06520-8062,United States.
E-mail addresses:[e-mail](B. Bhattacharya),
[e-mail](A. Maung), [e-mail](K. Schuster),
[e-mail](K.A. Davis).
http://dx.doi.org/10.1016/j.injury.2016.06.019
0020-1383/Publishedby ElsevierLtd.
Injury, Int. J. Care Injured xxx (2016)xxx–xxx
G Model
JINJ 6778No. of Pages5
Pleasecite this article in press as: B. Bhattacharya,et al., The older they are the harder they fall: Injury patternsand outcomesby age after
ground level falls, Injury (2016),http://dx.doi.org/10.1016/j.injury.2016.06.019
Contents lists available at ScienceDirect
Injury
j o u r n a lhomepa g e :w w w . e l s e vi e r . c o m / l o c a t e/ i n j u r y

reserveprogressivelydecreaseswith age,we hypothesizedthat it
is inaccurateto categorizeall senior patientsinto one homogenous
group and that injury patternsand outcomesafter GLF will vary as
a functionof age.To the bestof our knowledge,this is the first study
to analyzeGLF injury patternsand outcomesstratifiedby decadeof
life. We also investigatedpatient characteristics,anticoagulation
use,injury patternand comorbiditiesassociatedwith the outcome
of this common trauma mechanism.
Methods
Studydesign
This study was approved by the Yale University Human
InvestigationsCommittee.Review of the trauma registry,which
capturesall patientsadmittedwith an ICD-9 code consistentwith
injury, (Traumabase,Denver,CO) identified all ground level fall
patients above the age of eighteen admitted to a single trauma
center, Yale New Haven Hospital between January 2003 and
December2013.YaleNew HavenHospitalis an AmericanCollegeof
Surgeonsverified and ConnecticutDepartmentof Health desig-
natedurban Level One traumacenter,servingthe southernpart of
Connecticutstate averaging2600 trauma admissionsa year.Data
extractedincludedbasicpatient demographicsand outcomesdata
including ICU admission,length of stay, inpatient mortality and
dischargedisposition,which was classifiedas homeor dischargeto
an extended care facility. Injury patterns were gathered for all
fractures of the skull, face, ribs, spine, pelvis and extremities,
intracranial hemorrhage(including subarachnoid,subdural, epi-
dural and all intraventricular bleeds) and intracavitary injury.
Patientswith multiple injuries were countedonce for each injury
pattern. Injury patternswere correlatedwith age and outcomes.
Data was collected on comorbidities and use of anti-coagulant/
anti-plateletmedications.Thesecomorbiditieswere usedto adjust
for the effectsof injury on outcomes.
Statisticalanalysis
Statisticalanalysiswas performedusing IBM SPSSStatistics21
(IBM Corporation,Somers NY) and SAS v9.3 (SAS institute, Cary,
NC) software using an ANOVA test, Chi-square,Yates correction
and exact confidenceintervals as appropriate;statistical signifi-
cance was defined as p < 0.05. Multivariable logistic regression
models were built to assessthe impact of comorbiditieson death
as well as specific injuries.
Results
Patients
Over the eleven-yearperiod 5088 patientswere evaluatedafter
a GLF,3990 of whom were over the age of 60. The overall patient
populationwas 38.2%male.Median injury severityscore(ISS,Scale
0–75) for all patientswas 9 (Table 1).
Injury patterns
Certain injury patterns were commonly seen with this
mechanism.Hip and lower extremity fractures (46.3%)occurred
most frequentlyfollowed by upperextremityfractures(12.5%)and
rib fractures(8.9%)across all age groups.Abdominal solid organ
injuries were rare (N =42, 0.8%)(Table1). Only 19%of patientswith
abdominal solid organ injury required intervention, seven with
spleen embolization and one undergoinga splenectomy.Certain
injuries were seenwith increasingfrequencywith each decadeof
life after the age of 60 including cervical spine, pelvic, upper
extremityand rib fractures.When agewas examinedas continuous
variableit was associatedwith cervicalspine fracture,rib fractures
and pelvic fractures(p < 0.001).
Certain injuries were significantly associated with patient
comorbidities.Dataregardingcomorbiditywas not availablefor 19
patients and were excluded from analysis. Rib fractures were
associatedwith COPD, diabetes and liver cirrhosis. Intracranial
hemorrhagewas significantlyassociatedwith diabetes(Table 2).
The use of aspirin was the only antiplateletor anticoagulantagent
associatedwith intracranialhemorrhage(Table3).
Outcomes
The overallmortality for all patientswith GLF was 3.7%.Patients
youngerthan 60 had a mortalityof 0.1%comparedto 5.6%for those
in the 90 years or older age group (p < 0.001) (Table 4).
Comorbidities associatedwith mortality included a history of
myocardial infarction (4/22, 18.2%,p =0.001), coronary artery
disease (47/879, 5.4%, p =0.004), liver cirrhosis (3/28, 10.7%,
p =0.047)and COPD (30/434,6.9%,p =0.001).Univariateanalysis
demonstrated aspirin or warfarin use was associated with
increasedoverall mortality (Table 3). Women were less likely to
die (85/3134,2.7%)compared to their male counterparts(102/
1935, 5.3%,p < 0.001). Increasing age was also associatedwith
increasedrisk of death.Binary logistic regressionmodel including
Table 1
Patient demographicsand injury pattern.
Age
Bracket
Total
Patients
Mean
Age (Std
Dev)
Median
ISS
IQR Mean
ISS (Std.
Dev)
Male
Gender
%
Intra-
cranial
Bleed%
(n)
Facial
Fx%
(n)
Cervical
Spine Fx
% (n)
TLS
Fx%
(n)
Rib
Fx%
(n)
Hemo/
Pneumo-
thorax%(n)
Solid
Organ
% (n)
Pelvic
Fx%(n)
Hip,
femur LE
fx%(n)
Upper
Extremity fx
%(n)
<60 1085 45.5
(11.4)
5 4–9 7. (5.3) 55.9
(604)
1.38 (15) 2.86
(31)
2.03 (22) 4.4
(48)
7.0
(76)
1.8 (19) 1.0
(11)
1.5
(16)
48.1
(519)
9.5 (103)
60–69 638 64.6
(2.9)
9 4–9 7.9 (5.1) 45.4
(288)
0.94 (6) 1.72
(11)
3.92 (25) 4.2
(27)
9.8
(62)
2.8 (18) 1.3 (8) 3.8
(24)
44.6
(283)
10.7 (68)
70–79 908 75.1
(2.9)
9 4–9 8.7 (5.8) 34.9
(316)
0.41 (21) 2.53
(23)
3.41 (31) 3.6
(33)
7.4
(67)
1.2 (11) 0.8 (7) 5.4
(49)
46.0
(416)
12.5 (113)
80–89 1683 84.6
(2.8)
9 5–10 8.6 (5.1) 31.7
(534)
2.31 (26) 2.37
(40)
4.26 (72) 4.5
(76)
9.5
(160)
2.2 (37) 0.6
(10)
8.1
(137)
45.3
(763)
14.0 (236)
90+ 768 93.0
(2.8)
9 4–9 8.6 (5.4) 25.2
(192)
1.54 (10) 2.37
(26)
5.47 (42) 4.8
(37)
11.3
(86)
2.0 (15) 0.8 (6) 8.1
(62)
47.8
(365)
15.3 (117)
Total 5088 73.3
(17.6)
9 4–9 8.2 (5.4) 38.2
(1936)
1.30 (78) 2.57
(131)
3.77
(192)
4.3
(221)
8.9
(451)
2.0 (100) 0.8
(42)
5.7
(288)
46.3
(2346)
12.6 (639)
P Value <0.01 <0.001 0.233 0.343 0.002 0.716 0.01 0.214 0.541 <0.001 0.461 0.001
Chi
square
246.51 5.56 4.89 16.66 2.11 13.25 5.81 3.10 67.26 3.61 19.25
2 B. Bhattacharyaet al./ Injury, Int. J. CareInjured xxx (2016)xxx–xxx
G Model
JINJ 6778No. of Pages5
Pleasecite this article in press as: B. Bhattacharya,et al., The older they are the harder they fall: Injury patternsand outcomesby age after
ground level falls, Injury (2016),http://dx.doi.org/10.1016/j.injury.2016.06.019

all injury patterns,gender,age as a continuous variable demon-
strateddeathon was associatedwith intracranialhemorrhage(70/
187, 37.4%,p < 0.001), cervical spine fracture (19/187, 10.2%,
p < 0.001),rib fracture(29/187,15.5%,p =0.002)and pelvic fracture
(15/187,8.0%,p =0.052). The overall mean length of stay was
5.6days (median=4 days). In the geriatrics group, the septuage-
narians had the greatestmean length of stay (LOS) of 6.2days
(median=4) and nonagenarianshad the shortest mean LOS of
5.1days (median=4). When LOS was calculatedby excluding in
house mortality cases mean and median LOS were unchanged.
Only 1%of patientsunder age 60 were admittedfrom a long term
facility whereas5% of nonagenarianswere admittedfrom a long
term facility. Overall, only 37.3%patients were dischargedhome.
Patientsunder 60 yearsof agehad a home dischargerate of 73.8%
that incrementallyfell to 18.1%for the >90agebracket(p < 0.001)
(Table 4).
Discussion
GLFis a leadingcauseof traumamorbidityand mortalityamong
the elderly population [6–9]. This relatively minor mechanismof
traumahas a highermorbidity and mortality in the agedcompared
to youngerpatients [10]. Recognizingthe changingprofile of the
geriatricpopulation,we hypothesizedthat outcomemay relate to
decadeof life. With increasingage,thereare progressivechangesin
physiologicalreservethat may result in differencesin outcomeand
injury patterns. This is the first study of the geriatric trauma
population that analyzes injury pattern and outcome by each
decadeof life. We also examinedother patient characteristicsto
define their impact on injury pattern and outcome.
Our resultsdemonstratethat certaininjuries are commonafter
a GLF.Hip, lower extremityfracturesand intracranialhemorrhage
were very common across all age groups. Cervical spine, upper
extremity, rib and pelvic fractures increase with frequency as
patient’s age. These findings may be due to decreasing bone
density associatedwith aging.The changein injury patternsmay
also be associatedwith postural changesthat are seen with age.
The craniopelvicalignmentchangesrapidly with age,particularly
in the eight decadeof life [11].Posturalchangesresult in the center
of gravity being moved forward which may contribute to the
increased incidence of upper extremity, rib and cervical spine
fractures.Another contributingfactor to the increasingfrequency
in upper body injuries with ageis a decreasein reactiontime with
age that may slow down protectivereflexesduring a fall [12].
Intracavitaryinjuries including abdominalsolid organ injuries
were uncommon.The relativelow kinetic energyassociatedwith
this mechanism likely explains this low incidence despite the
theoretically more at risk organs with weaker capsules.These
injury patternssuggestthat the energyfrom the fall is absorbedby
the skeletal system and dissipated thereby protecting the
intrathoracicand intraabdominalorgans.
The survivaloutcomesof GLF patientswere stronglycorrelated
with age. There is a stepwise increment of in-hospital mortality.
This finding is in line with previous studies that have shown a
worse outcome of geriatric patients from falls compared to
younger patients.However what is significant is the doubling of
mortality from 2.5%to 5.4%from the 7th to tenth decadeof life
with no change in ISS. This sharp fall in survival is probably
reflectiveof the decline in physiologicalreservewith age.Parallel
to this finding is the sharp fall in home dischargewith age.This
trend is demonstrativeof the increasingmorbidity of GLF injury
with agewithout an increasein ISS likely due to increasingfrailty.
Interestinglythe 90+agebrackethad the lowest ICU admissionrate
and shortestlength of stay despite having the highest mortality.
Table 2
Injury pattern correlationto comorbidity.
Rib Fractures(n =451) IntracranialBleeds(n =78)
Co-morbidity (n =Total
number)
Rib Fractures+Comorbidity
(n =patients)
Odds ratio (95%
CI)
P value Intracranialbleed+comorbidity
(n =patients)
Odds ratio (95%
CI)
P value
COPD (n =434) 59 1.68 (1.25–2.26) <0.001 6 0.84 (0.360–1.96) 0.688
Diabetes(n =1010) 76 0.75 (0.58–0.98) 0.034 25 1.76(1.07–2.89) 0.026
Liver Cirrhosis (n =28) 6 2.82 (0.93–7.21) 0.033 2 5.03 (0.57–20.67) 0.068
Hypertension(n =2857) 264 1.11(0.91–1.36) 0.310 52 1.38 (0.84–2.26) 0.199
Table 3
Drug agentand intracranialbleed and death by all injuries.
Agent (N =number of patients) Number of intracranialbleed
patients
Odds Ratio (95%CI) P value No of deathsby all causeon agent Odds Ratio (95%CI) P Value
Aspirin (N =872) 25 2.32 (1.43–3.75) 0.001 42 1.42 (1.0–2.03) 0.049
Warfarin (N =571) 13 1.60 (0.87–2.91) 0.125 40 2.24 (1.56–3.21) <0.001
Clopidogrel(N =224) 3 0.87 (0.27–2.77) 0.809 13 1.66 (0.93–2.97) 0.083
Table 4
Patient disposition and outcomes.
Age Bracket Total
Patients
Admit from facility%
(n)
Mean Lengthof Stay(Std.Dev.) Median/IQR (Length of
Stay)
ICU Admit%
(n)
Home Discharge%
(n)
Death%(n)
<60 1085 1.0%(11) 5.4 (8.3) 3 (2–6) 23.5 (255) 75.6 (800) 0.1 (10)
60–69 638 1.1%(7) 6.1 (10.9) 4 (2–6) 23.3 (149) 47.3(302) 2.5 (16)
70–79 908 1.5%(14) 6.2 (11.5) 4 (3–6) 25.5 (232) 31.2(283) 4.2 (38)
80–89 1689 3.6%(60) 5.3 (5.4) 4 (3–6) 21.7(368) 22.2 (375) 4.8 (81)
90+ 768 5.1%(39) 5.1 (4.4) 5 (3–6) 18.9 (145) 18.2 (140) 5.4 (42)
Total 5088 2.6%(130) 5.0 (8.1) 3–6 (4) 22.6 (1149) 37.3(1900) 3.7 (187)
p value 0.008 0.018 <0.001 <0.001
B. Bhattacharyaet al. / Injury, Int. J. CareInjured xxx (2016)xxx–xxx 3
G Model
JINJ 6778No. of Pages5
Pleasecite this article in press as: B. Bhattacharya,et al., The older they are the harder they fall: Injury patternsand outcomesby age after
ground level falls, Injury (2016),http://dx.doi.org/10.1016/j.injury.2016.06.019

This may be reflectiveof predeterminedgoalsof carethat resulted
in limitations in care after injury. Due to changes in medical
records modalities at our institution we were not able to extract
this data. The high morbidity and mortality associatedwith this
mechanismof injury should warrant early discussionsof goals of
care if not established.In addition certain comorbiditiessuch as
history of myocardialinfarction,coronaryarterydiseaseand COPD
were statisticallyassociatedwith an increasedrisk for mortality
following GLF.
These results have implications for the trauma evaluation of
geriatricspatients.Although elderly patientswith GLF experience
relatively low kinetic energy, they are still prone to significant
injury. In our cohort the incidence of intraabdominal and
intrathoraciccavitary injury were extremelyrare. Our study did
not intend to study the optimal imaging approachto this patient
population. However our data suggeststhe routine imaging of
chest, abdomen and pelvis by CT scan maybe unwarranted.
Numerousstudiessupporta more conscientiousCT scanutilization
in trauma evaluations[13–15].Our data would suggestthat only
patients with positive ultrasound imaging and or concerning
physical exam should have further cross-sectional imaging.
Further prospectivestudies are warranted.
The elderly commonly use anti-platelet or anticoagulation
treatment for their underlying comorbidities. In recent years
newer agentshave been introduced in the market that are more
convenientfor patientsas they do not requirefrequentmonitoring.
These drugs have the disadvantagefrom a trauma perspectiveof
not beingeasilyreversible.In our studyonly aspirinwas associated
with intracranialbleeds.However,we did not segregatehigh and
low dose aspirin in the study and cannot determineif this was a
dosedependentrisk or if the presenceof aspirin by itself was a risk
factor.Aspirin and warfarin were the only drugs associatedwith
mortality in our experience.The role of these agents in trauma
brain injury mortality has been disputed [16,17].Recent studies
haveidentifiedwarfarin use as beingassociatedwith mortality but
not the use of aspirin [18].Our databarelydemonstratedstatistical
significance between aspirin use and brain trauma mortality.
Studiesexaminingthis issue,including ours, haverelativelysmall
number of patientson aspirin that may accountfor the statistical
differencesbetweenstudies.Our study in particularwas confined
to one sub-populationof all traumatic brain injuries by a single
mechanismthat may also account for the difference.The small
number of patientstaking FactorXa inhibitors (rivaroxabann =3,
enoxaparinn =15) limited our ability to assessmortality risk in
elderly patients, and the failure to reach statistical significance
may representa Type II error. The mortality associatedwith these
agentsmaybean indirect marker of individuals with more severe
cardiovasculardisease.The widespread use of anti-platelet and
anti-coagulationdrugsraisesa more generalquestion.At what age
do the risks of bleeding while on these drugs outweigh their
cardiovascularbenefits?
This data confirms previous studies examiningthe morbidity
and mortality associatedwith this mechanismin the elderly[9,10].
Previousstudieshavedemonstratedgeriatricpatientsmay benefit
from specialized trauma care [19–22]. Our data identifies this
segmentof the populationas distinct subsetwith higher mortality
despitelow ISSscore.The increasedmorbidityand mortalityof this
mechanism may warrant the addition of age-basedcriteria to
standardtraumatriagecriteria[23].The majorityof nonagenarians
survived the initial hospital stay but the vast majority required
admission to an extended care facility. Previous studies have
demonstratedan increasedone year mortality for nonagenarians
and centenarian surviving the initial hospitalization [5]. The
results of our study combined with previous studies should be
consideredin guiding healthcareand family discussions.As is the
casefor all traumas,the bestcure is prevention.Fall risk awareness
and prevention along with osteoporosisprevention will have to
play an increasinglyimportantrole as our populationcontinuesto
age [24–26].
Our study is limited by the retrospectivenature of the study.
Due to limitations of medicalrecordswe were not able to decipher
how many spine fractures that were recorded were acute vs.
chronic.The study period included a small number of patientson
newer generationanticoagulationagentsand maybe underpow-
eredto detectthe impactof thesedrugs.This studywas basedupon
a single institution experienceat a Level 1 trauma center. Our
experience may be biased towards a higher acuity patient
population.
Conclusions
In summary,our results show that the geriatric population is
not a homogenousgroup.Morbidity, mortality and injury pattern
resulting from GLF are a function of increasingage.This relatively
minor mechanismof injury is a leading cause of morbidity and
mortality that increaseswith each decadeof life. As we see more
patientsliving to be nonagenariansand beyond,our resourcesand
goals of care will have to be tailored to this segment of the
population.
Conflict of interest
The authors of the manuscripthave no conflicts of intereststo
disclose.
References
[1] CDC/NCHS,National Vital StatisticsSystem,Grove RD, Hetzel AM. Vital
statisticsrates in the United States,1940–1960.Washington,DC: U.S.
GovernmentPrinting Office; 1968.
[2] Murphy SL, Xu J, KochanekKD. Deaths:final data for 2010.Natl Vital StatRep
2013 May 8;61(4):1–117.
[3] MandaviaD, Newton K. Geriatrictrauma.EmergMed Clin N Am 1998;6:257–
74.
[4] Vincent GK, Velkoff VA. The next four decades:the older population in the
United States:2010to 2050.Washington,DC: US CensusBureau;2010.p. 25–
1138CurrentPopulationReporthttp://www.aoa.gov/aoaroot/aging_statistics/
future_growth/DOCS/p25-1138.pdf[accessedMarch 2013].
[5] HwabejireJO, KaafaraniHM, LeeJ, Yeh DD, FagenholzP, King DR, et al. Patterns
of injury, outcomes,and predictorsof in-hospital and 1-year mortality in
nonagenarianand centenariantrauma patients.JAMA Surg 2014;149
(10):1054–9.
[6] National TraumaData Bank. Annual Report: American Collegeof Surgeons;
2012.
[7] PedenM, McGee K, SharmaG. The injury chart book: a graphicaloverviewof
the global burden of injuries. Geneva:World Health Organization;2002.
[8] StevensJA, Ryan G, Kresnow M. Centersfor DiseaseControl and Prevention
(CDC) Fatalitiesand injuries from falls among older adults-UnitedStates,
1993Y2003and 2001Y2005.MMWR Morb Mortal Wkly Rep
2006;55:1221Y1224.
[9] SpaniolasK, Cheng JD, GestringML, SangosanyaA, StassenNA, BankeyPE.
Groundlevel falls areassociatedwith significantmortalityin elderlypatients.J
Trauma2010 Oct;69(4):821–5.
[10] SterlingDA, O’ConnorJA, BonadiesJ. Geriatricfalls: injury severityis high and
disproportionateto mechanism.J Trauma2001;50(1):116–9.
[11] YoshidaG, YasudaT, TogawaD, HasegawaT, YamatoY, KobayashiS, et al.
Craniopelvicalignmentin elderly asymptomaticindividuals: analysisof 671
cranial centersof gravity.Spine (Phila Pa 1976)2014;39(14):1121–7.
[12] GottsdankerR. Age and simple reactiontime. J Gerontol 1982;37(3):42–348.
[13] RodriguezRM, Anglin D, LangdorfMI, BaumannBM, HendeyGW, BradleyRN,
et al. NEXUS chest: validationof a decision instrumentfor selectivechest
imaging in blunt trauma.JAMA Surg 2013;148:940–6.
[14] Healy DA, HegartyA, FeeleyI, Clarke-MoloneyM, Grace PA, Walsh SR.
Systematicreview and meta-analysisof routine total body CT comparedwith
selectiveCT in trauma patients.Emerg Med J 2014;31:101–8.
[15] SierinkJC, SaltzherrTP,EdwardsMJ, BeukerBJ, PatkaP, GoslingsJC, et al. Direct
total body CT scanin multi-traumapatients.Ned TijdschrGeneeskd2012;156:
A4897.
[16] Ahmed N, BialowasC, Kuo YH, ZawodniakL. Impact of preinjury
anticoagulationin patients with traumaticbrain injury. South Med J 2009
May;102(5):476–80.
4 B. Bhattacharyaet al./ Injury, Int. J. CareInjured xxx (2016)xxx–xxx
G Model
JINJ 6778No. of Pages5
Pleasecite this article in press as: B. Bhattacharya,et al., The older they are the harder they fall: Injury patternsand outcomesby age after
ground level falls, Injury (2016),http://dx.doi.org/10.1016/j.injury.2016.06.019

[17] Collins CE, Witkowski ER, Flahive JM, AndersonJr. FA, Santry HP. Effect of
preinjury warfarin use on outcomesafter head trauma in Medicare
beneficiaries.Am J Surg 2014;208(4):544–9.
[18] Grandhi R, Harrison G, Voronovich Z, Bauer J, Chen SH, Nicholas D, et al.
Preinjury warfarin, but not antiplateletmedications,increasesmortality in
elderly traumaticbrain injury patients.J TraumaAcute Care Surg 2015;78
(3):614–21.
[19] Zafar SN, Obirieze A, SchneiderEB, Hashmi ZG, Scott VK, GreeneWR, et al.
Outcomesof trauma care at centerstreatinga higher proportion of older
patients: the casefor geriatric trauma centers.J TraumaAcute Care Surg
2015;78(4):852–9.
[20] Garwe T, RobertsZV, Albrecht RM, Morgan AE, Johnson JJ, Neas BR. Direct
transportof geriatrictraumapatientswith pelvic fracturesto a LevelI trauma
centerwithin an organizedtraumasystem:impact on two-week incidenceof
in-hospital complications.Am J Surg 2012;204(6)921–5discussion925–6.
[21] BradburnE, RogersFB, KrasneM, RogersA, Horst MA, BeelenMJ, et al. High-
risk geriatricprotocol:improvingmortalityin the elderly.J TraumaAcuteCare
Surg 2012;73(2):435–40.
[22] Mangram AJ, Mitchell CD, ShiffletteVK, Lorenzo M, Truitt MS, Goel A, et al.
Geriatric trauma service:a one-yearexperience.J TraumaAcute Care Surg
2012;72(1):119–22.
[23] Ichwan B, Darbha S, Shah MN, ThompsonL, EvansDC, Boulger CT, et al.
Geriatric-specifictriagecriteriaare moresensitivethan standardadult criteria
in identifyingneed for traumacentercare in injured older adults.Ann Emerg
Med 2015;65(1):92–100.
[24] VlaeyenE, CoussementJ, LeysensG, Van der Elst E, DelbaereK, CambierD, et al.
Characteristicsand effectivenessof fall preventionprogramsin nursing
homes: a systematicreview and meta-analysisof randomizedcontrolled
trials. J Am Geriatr Soc 2015;63(2):211–21.
[25] SchoeneD, ValenzuelaT, Lord SR, de Bruin ED. The effect of interactive
cognitive-motortraining in reducing fall risk in older people: a systematic
review. BMC Geriatr 2014;14:107.
[26] Shaw JA, Connelly DM, McWilliam CL. Enactingfall preventionin community
outreachcare.Qual Health Res 2014;24(7):901–12.
B. Bhattacharyaet al. / Injury, Int. J. CareInjured xxx (2016)xxx–xxx 5
G Model
JINJ 6778No. of Pages5
Pleasecite this article in press as: B. Bhattacharya,et al., The older they are the harder they fall: Injury patternsand outcomesby age after
ground level falls, Injury (2016),http://dx.doi.org/10.1016/j.injury.2016.06.019