The Journal of TRAUMA威 Injury, Infection, and Critical Care

Geriatric Falls: Injury Severity Is High and Disproportionate

to Mechanism
Daniel A. Sterling, MD, Judith A. O’Connor, RN, and John Bonadies, MD

Objective: Falls are a well-known anism in 48% of the older group and 7% group (9.28 vs. 4.64, p < 0.05), whereas
source of morbidity and mortality in the of the younger group (p < 0.05). Falls in there was no difference in mean ISS be-
elderly. Fall-related injury severity in this the older group constituted 65% of pa- tween multilevel and same-level falls
group, however, is less clear, particularly tients with ISS >15, with 32% of all falls within the older group itself (10.12 vs.
as it relates to type of fall. Our purpose is resulting in serious injury (ISS >15). In 9.28, p > 0.05). The fall-related death rate
to explore the relationship between mech- contrast, falls in the younger group con- was higher in the older group (7% vs.
anism of fall and both pattern and severity stituted only 11% of ISS >15 patients, 4%), with falls seven times more likely to
of injury in geriatric patients as compared with falls causing serious injury only 15% be the cause of death compared with the
with a younger cohort. of the time (both p < 0.05). Notably, same- younger group (55% vs. 7.5%) (both p <
Methods: Our trauma registry was level falls resulted in serious injury 30% 0.05). Same-level falls as a cause of death
queried for all patients evaluated by the of the time in the older group versus 4% was 10 times more common in the elderly
trauma service over a 412-year period in the younger group (p < 0.05), and were (25% vs. 2.5%, p < 0.05).
(1994 –1998). Two cohorts were formed on responsible for an ISS >15 30-fold more Conclusion: Falls among the elderly,
the basis of age greater than 65 or less in the older group (31% vs. <1%; p < including same-level falls, are a common
than or equal to 65 years and compared as 0.05). Abbreviated Injury Scale evaluation source of both high injury severity and
to mechanism, Injury Severity Score revealed more frequent head/neck (47% mortality, much more so than in younger
(ISS), Abbreviated Injury Scale score, and vs. 22%), chest (23% vs. 9%), and pelvic/ patients. A different pattern of injury be-
mortality. extremity (27% vs. 15%) injuries in the tween older and younger fall patients also
Results: Over the study period, 1,512 older group for all falls (all p < 0.05). The exists.
patients were evaluated, 333 greater than mean ISS for same-level falls in the older Key Words: Falls, Geriatric trauma,
65 years and 1,179 less than or equal to 65 group was twice that for the younger Injury Severity Score.
years of age. Falls were the injury mech-
J Trauma. 2001;50:116 –119.

A
s the population of the United States ages, the ramifi- Falls constitute the most common mechanism of injury
cations of providing health care for this sector become in the geriatric population, with an annual incidence of 30%
increasingly important. It is expected that by the year in those over 65 rising to 50% in those over 80 years of age.4
2030 the U.S. population over the age of 65 will represent In our own state, recent data from the State of Connecticut
25% of the total population, roughly double the current Committee on Trauma regarding geriatric injury corroborate
proportion.1,2 Trauma affecting this age group thus takes on this finding, with 64% of injuries to patients 70 years old or
similar importance. Trauma remains the seventh leading older occurring as the result of falls.5 Although multilevel
cause of death in patients over 65 years of age. Moreover, it falls do occur, same-level falls predominate. The morbidity of
is well established that the elderly, although injured less these falls is significant; nearly 40% of all nursing home
often, are more likely to die from their injuries than are their admissions are in some way related to falls.6
younger counterparts. With geriatric trauma accounting for a Outcome after trauma in this age group has been studied
disproportionate one third of all trauma-related expenses cur- extensively, both short and long term. The majority of the
rently, it can be expected that the cost of caring for the injured literature supports aggressive care of the injured elderly,
elderly will rise dramatically.3
including those over age 75 and with multiple injuries.7–16
Overall mortality from falls has declined over the last 30
years, primarily as a result of improved trauma
Submitted for publication December 15, 1999. management.17
Accepted for publication September 23, 2000.
Copyright © 2001 by Lippincott Williams & Wilkins, Inc.
Much has been reported regarding geriatric falls includ-
From the Department of Surgery, Hospital of Saint Raphael, New ing patient profile, environmental factors, frequency rates,
Haven, Connecticut. and preventive measures.17–23 Risk factors predisposing this
Poster presentation at the 12th Annual Meeting of the Eastern Associ- sector of the population have been well described previously
ation for the Surgery of Trauma, January 13–16, 1999, Orlando, Florida.
Address for reprints: John Bonadies, MD, Department of Surgery,
in both the community and nursing home settings.5 No pub-
Hospital of Saint Raphael, 1450 Chapel Street, New Haven, CT 06511; lished study, however, addresses the association between falls
email: [email protected]. of all types, injury severity, and pattern of injury, specifically

116 January 2001

 Geriatric Falls: Injury Severity Is High

Statistical analysis was performed using the ␹2 method,

Table 1 Mechanism of Injury
Fisher’s exact test, two-sample t test, and confidence interval.
E-Code Description Survival was defined as discharge from the hospital, and
880.1 Fall from escalator deaths were defined as those occurring during the same ad-
880.2 Fall from sidewalk/curba mission for trauma. ISS was generated from Tricode software
880.9 Fall on or from stairs/steps (Trianalytics, Bel Air, MD) according to the medical record.
884.2 Multilevel fall—chaira
884.3 Multilevel fall—wheelchaira
An ISS ⬎15 was chosen as the accepted standard of severe
884.4 Multilevel fall—beda injury. Patients over the age of 65 are referred to as the older
884.5 Multilevel fall—other furniturea group; those 65 years of age or under constitute the younger
884.6 Multilevel fall—commodea group. Preexisting comorbid disease and length of stay were
884.9 Multilevel fall—unspecified not evaluated, nor was long-term follow-up conducted.
885.0 Same-level fall—slipping/trippinga
886.0 Same-level fall—sports
886.9 Same-level fall—shoving (not sports-related)a RESULTS
887.0 Other and unspecified fall Over the 412-year period, a total of 1,512 patients were
a
Same-level falls. evaluated by the trauma service, 333 older than 65 and 1,179
65 years of age or younger. Falls were the mechanism of
injury in 48% (n ⫽ 159) of the older group compared with
the morbidity associated with the common same-level fall. 7% (n ⫽ 83) in the younger group (p ⬍ 0.05), comparable to
This retrospective analysis was undertaken to explore the previously reported national statistics (Table 2). The balance
relationship between the mechanism of fall and both severity of mechanisms consisted of motor vehicle crashes, penetrat-
and pattern of injury in geriatric patients as compared with a ing trauma, pedestrians struck, and other assault/injury.
younger cohort. Of the patients with an ISS ⬎15, 65% were a result of
falls in the group older than 65, whereas only 11% in the
MATERIALS AND METHODS group younger than 65 occurred secondary to this mechanism
The trauma registry at a Level II trauma center was (p ⬍ 0.05). Falls resulted in serious injury (ISS ⬎15) 32% of
queried for all patients evaluated by the trauma service over the time in the older group and resulted in serious injury only
the 412-period from 1994 to 1998. Two cohorts were formed 15% of the time in the younger group (p ⬍ 0.05). When
on the basis of age greater than 65 years or less than or equal stratified as same-level and multilevel falls, serious injury
to 65 years and compared as to mechanism of injury, Abbre- was further skewed toward the older group. Same-level falls
viated Injury Scale (AIS) score,24 Injury Severity Score resulted in serious injury (ISS ⬎15) 30% of the time in the
(ISS),25 and mortality. Falls were stratified as either same older group versus 4% in the younger group, and were re-
level or multilevel. Same-level falls were defined by the sponsible for an ISS ⬎15 30-fold more in the older group
following E-codes: 880.1, 884.2, 884.3, 884.5, 884.6, 885.0, (31% vs. ⬍1%, both p ⬍ 0.05). The mean ISS for same-level
and 886.9 (Table 1), and included those from a chair, wheel- falls in the older group was 9.28 (SD, 7.22) compared with
chair, commode, sidewalk/curb, slipping/tripping, and shov- 4.64 (SD, 4.96) in the younger group (p ⬍ 0.05). In the older
ing (not related to sports). Multilevel falls included those group, the mean ISS for multilevel falls was 10.12, similar to
from an escalator, stairs/steps, bed, unspecified, and sports- that for same-level falls (9.28, p ⬎ 0.05).
related. Excluded were falls from great heights in either Regarding the pattern of injury for all falls, older patients
cohort. sustained more frequent injuries compared with younger pa-

Table 2 Fall Severity and Mortality

p
Older Younger
Value

Fall as mechanism 159/333 (48%) 83/1,179 (7%) ⬍0.05

Fall-related ISS ⬎15 50/159 (31%) 12/83 (15%) ⬍0.05
Fall as % of all ISS ⬎15 50/77 (65%) 12/107 (11%) ⬍0.05
Same-level fall-related ISS ⬎15 24/81 (30%) 1/28 (4%) ⬍0.05
Same-level fall as % of all ISS ⬎15 24/77 (31%) 1/107 (⬍1%) ⬍0.05
Mean ISS same-level falls 9.28 4.64 ⬍0.05
Fall-related deaths/all deaths 11/20 (55%) 3/40 (7.5%) ⬍0.05
Same-level fall deaths/all deaths 5/20 (25%) 1/40 (2.5%) ⬍0.05
Fall-related death ratea 11/159 (7%) 3/83 (4%) ⬍0.05
Same-level fall-related death rateb 5/81 (6.2%) 1/28 (3.6%) NS
a
Fall-related deaths/all falls.
b
Same-level fall-related deaths/all same-level falls.

Volume 50 • Number 1 117

 The Journal of TRAUMA威 Injury, Infection, and Critical Care

Table 3 AIS Pattern for All Falls Table 4 Fall Related Deaths
Older Younger Body
AIS Region p Value Patient No. Cause of Death ISS
(n ⫽ 159) (n ⫽ 83) RegionAIS

Head/neck 75 (47%) 18 (22%) 0.00 ⬎65 Years

Face 14 (9%) 10 (12%) 0.264 1 SDH/IVH/SAH/ICH 25 15
Chest 37 (23%) 7 (8%) 0.003 2 Pneumonia/C-spine fracture 5 12 31
Abdomen 3 (2%) 10 (12%) 0.001 3 SDH 16 14
Pelvis/extremity 43 (27%) 12 (14%) 0.021 4 SDH/SAH/arrhythmia 16 14
Skin/soft tissue 75 (47%) 50 (60%) 0.025 5 SDH 16 14
6 SDH/cerebral contusion/MI 25 14 53
7 Respiratory failure/pneumonia 9 33
8 C-spine fracture/ischemic stroke 9 13
tients in the head/neck region (47% vs. 22%), chest region 9 SDH 24 4 2 2 1
1 3 5 6
(23% vs. 9%), and pelvic/extremity region (27% vs. 15%) (all 10 Cerebral contusions/SAH 9 13
p ⬍ 0.05). Younger patients, on the other hand, sustained 11 Thoracic spinal cord injury 20 4 2
1 3
abdominal and skin/soft tissue injuries more frequently than Mean ISS 15.8
their older counterparts (12% vs. 2%, and 62% vs. 47%, ⱕ65 Years
respectively, all p ⬍ 0.05) (Table 3). Moreover, for same- 1 Pelvic fracture/postoperative 4 52
level falls determined on the basis of AIS score, head and hemorrhage
2 SDH 25 15
neck injuries were more severe on average in the older group
3 SDH 17 14 61
(AIS score of 3.5 vs. 2.86, p ⬍ 0.05) as were pelvic/extremity
injuries (2.48 vs. 2, p ⬍ 0.05). No regional severity differ- Mean ISS 15.3
ences between the two groups were identified for all falls. MI, myocardial infarction; ICH, intracerebral hemorrhage; IVH,
Regarding fall-related deaths, there was a significant intraventricular hemorrhage; SDH, subdural hematoma; SAH, sub-
arachnoid hemorrhage; ISS, injury severity scale; AIS, Abbreviated
difference between the older and younger cohorts; 11 of 159
Injury Scale.
(7%) and 3 of 83 (4%), respectively, died from injuries Body Region: 1, Head/Neck; 2, Face; 3, Chest; 4, Abdomen/
related to a fall (p ⬍ 0.05). Furthermore, 11 of the 20 deaths pelvic contents; 5, Extremity or pelvic girdle; 6, Skin/subcutaneous
that occurred in the older group were the result of a fall tissue.
(55%), whereas only 3 of 40 deaths in the younger group AIS: 1, Minor; 2, Moderate; 3, Serious; 4, Severe; 5, Critical; 6,
were fall related (7.5%) (p ⬍ 0.05). Deaths from same-level Maximum.
falls, as a proportion of all deaths, were significantly higher
in the older group as well (5 of 20 [25%] vs. 1 of 40 [2.5%], group. Interestingly, for the older group the ISS in the same-
p ⬍ 0.05). The mean ISS for fall-related deaths in the older level and multilevel fall groups was similar. In fact, nearly
group was 16; 7 had an ISS ⬎15. The cause in 7 of these 11 one third of the older group sustained severe injuries regard-
deaths was the result of closed head injury (CHI); two pa- less of whether the fall was from a commode, for example, or
tients died from pulmonary complications (Table 4). Of the down a flight of stairs.
11 deaths from falls in the older group, 5 resulted from In this retrospective study, falls were also found to ac-
same-level falls. One of the three deaths from falls in the count for a significantly different injury profile and severity
younger group also was a result of a same-level fall, render- between the older and younger cohorts. The analysis of AIS
ing same-level fall death rates not statistically significant data pointed out that patients in the older group sustained
between older and younger cohorts. This single same-level more frequent head/neck injuries, more frequent chest inju-
fall death in the younger group occurred in a patient with a ries, and more frequent pelvis/extremity injuries when com-
congenital defect (osteogenesis imperfecta) resulting in re- pared with younger fall patients. Both abdominal and skin/
current long-bone fractures. His death followed operative soft tissue injuries, conversely, were more common in the
fixation of a pathologic acetabular fracture as a result of younger cohort. Geriatric same-level fall patients, moreover,
coagulopathy and hemorrhage. The balance of deaths in the sustained more severe head/neck and pelvic/extremity inju-
younger group resulted from CHI. ries than their younger counterparts.
When analyzing fall-related deaths, proportionately more
DISCUSSION occurred in the older group. Similarly, a disproportionately
Falls represent a significant source of morbidity and high number of deaths from same-level falls occurred in the
mortality among the elderly, accounting for nearly half of all older group. This further supports the premise that these
geriatric trauma. Indeed, in our study not only were falls the simple falls, despite their seemingly innocuous mechanism,
principal mechanism of injury among the elderly but the more can result in disproportionately severe injury and even death.
severely injured elderly patients were predominantly those The most common cause of death was CHI. The single
involved in falls. The ISS was worse in the older group for all same-level fall-related death in the younger group probably
falls and same-level falls as compared with the younger prevented statistically significant conclusions regarding the

118 January 2001

 Geriatric Falls: Injury Severity Is High

same-level fall death rate. Interestingly, the condition that 11. Horst HM, Obeid FN, Sorensen VJ, Bivins BA. Factors influencing
rendered him susceptible to severe injury equated his physi- survival of elderly trauma patients. Crit Care Med. 1986;14:681–
684.
ologic status to that of an elderly person.
12. Knudson MM, Lieberman J, Morris JA, Cushing BM, Stubbs HA.
Clearly, the “graying of America” has huge implications Mortality factors in geriatric blunt trauma patients. Arch Surg. 1994;
on U.S. health care in general and for those institutions caring 129:448 – 453.
for injured patients in particular. This article highlights how 13. Morris JA, Mackenzie EJ, Edelstein SL. The effect of preexisting
simple and commonplace mechanisms often result in devas- conditions on mortality in trauma patients. JAMA. 1990;263:1942–1946.
14. Oreskovich MR, Howard JD, Copass MK, Carrico CJ. Geriatric
tating injury and death in the geriatric patient. A high index
trauma: injury patterns and outcome. J Trauma. 1984;24:565–569.
of suspicion for potential serious injury even after simple 15. Osler T, Hales K, Baack B, et al. Trauma in the elderly. Am J Surg.
same-level falls is necessary to diagnose and treat geriatric 1988;156:537–543.
patients in a timely fashion. Efforts at preventing falls clearly 16. Van der Sluis CK, Klasen HJ, Eisma WH, ten Duis HJ. Major
need to include strategies to decrease both same-level and trauma in young and old: what is the difference? J Trauma. 1996;
40:78 – 82.
multilevel falls.
17. Riggs JE. Mortality from accidental falls among the elderly in the
Unite States, 1962–1988: demonstrating the impact of improved
REFERENCES trauma management. J Trauma. 1993;35:212–219.
1. Mandavia D, Newton K. Geriatric trauma. Emerg Med Clin North 18. Sattin RW, Rodriguez JG, Devito CA, Wingo PA. Home
Am. 1998;6:257–274. environmental hazards and the risk of fall injury events among
2. Santora TA, Schinco MA, Trooskin SZ. Management of trauma in community-dwelling older persons. Study to assess falls among the
the elderly patient. Surg Clin North Am. 1994;74:163–186. elderly (SAFE) group. J Am Geriatr Soc. 1998;46:669 – 676.
3. Schwab CW, Kauder DR. Trauma in the geriatric patient. Arch Surg. 19. Province MA, Hadley EC, Wolf SL, et al. The effects of exercise on
1992;127:701–706. falls in elderly patients. A preplanned meta-analysis of the FICSIT
4. Tinetti ME. Falls. In: Cassel CK, Riesenberg DE, Sorensen LB, et trials. Frailty and injuries: cooperative studies of intervention
al., eds. Geriatric Medicine. 2nd ed. New York: Springer-Verlag, techniques. JAMA. 1995;273:1341–1347.
1990:528 –534. 20. Koski K, Luukinen H, Laippala P, Kivela SL. Physiologic factors
5. State of Connecticut Committee on Trauma. State of Connecticut and medications as predictors of injurious falls by elderly people: a
Injury Data, Fiscal Year 1996. Wallingford, CT: The Connecticut prospective population-based study. Age Ageing. 1996;25:29 –38.
Healthcare Research and Education Foundation; 1997. 21. Koski K, Luukinen H, Laippala P, Kivela SL. Risk factors for major
6. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among injurious falls among the home-dwelling elderly by functional
elderly persons living in the community. N Engl J Med. 1988; abilities. A prospective population-based study. Gerontology. 1998;
319:1701–1707. 44:232–238.
7. Battistella FD, Din AM, Perez L. Trauma patients 75 years and 22. Myers AH, Young Y, Langlois JA. Prevention of falls in the elderly.
older: long-term follow-up results justify aggressive management. Bone. 1996;18(suppl 1):87S-101S.
J Trauma. 1998;44:618 – 624. 23. Tinetti ME, Baker DI, McAvay G, et al. A multifactorial
8. Demaria EJ, Kenney PR, Merriam MA, et al. Survival after trauma intervention to reduce the risk of falling among elderly people living
in geriatric patients. Ann Surg. 1987;206:738 –743. in the community. N Engl J Med. 1994;331:821– 827.
9. Finelli FC, Jonsson J, Champion HR, Morelli S, Fouty WJ. A case 24. Association for the Advancement of Automotive Medicine. The
control study for major trauma in geriatric patients. J Trauma. 1989; Abbreviated Injury Scale, 1990 Revision. Des Plaines, IL:
29:541–548. Association for the Advancement of Automotive Medicine; 1990.
10. Gubler KD, Davis R, Koepsell T, Soderberg R, Maier RV, Rivara 25. Baker SP, O’Neil B, Haddon W, Long WB, The Injury Severity
FP. Long-term survival of elderly trauma patients. Arch Surg. 1997; Score. A method for describing patients with multiple injuries and
132:1010 –1014. evaluating emergency care. J Trauma. 1974;14:187–196.

Volume 50 • Number 1 119