PEDIATRICS/ORIGINAL RESEARCH
Emergency Department Analgesia for
Julie C. Brown, MD, MPH
Eileen J. Klein, MD, MPH
Charlotte W. Lewis, MD, MPH
Brian D. Johnston, MD, MPH
Peter Cummings, MD, MPH
From the Department of
Pediatrics, University of
Washington, School of Medicine
(Brown, Klein, Lewis,
Johnston); the Department of
Epidemiology, University of
Washington School of Public
Health and Community
Medicine (Cummings); and the
Children’s Hospital and
Regional Medical Center
(Brown, Klein, Lewis,
Johnston), Seattle, WA.
Copyright © 2003 by the American
College of Emergency Physicians.
0196-0644/2003/$30.00 + 0
doi:10.1067/mem.2003.275
AUGUST 2003 42:2 ANNALS OF EMERGENCY MEDICINE
Fracture Pain
Study objectives: We analyze records of all emergency department (ED) patients
with extremity or clavicular fractures to describe analgesic use, compare analgesia
between adults and children, and compare analgesia between the subset of these
adults and children with documented moderate or severe pain. Among children, we
compare treatment between pediatric and nonpediatric facilities.
Methods: Analysis of the ED component of the National Center for Health Statistics
National Hospital Ambulatory Medical Care Survey for 1997 through 2000 was con-
ducted. The proportion of patients with closed extremity and clavicular fracture that
received any analgesic and narcotic analgesic medications was determined for each
age category. Survey-adjusted regression analyses compared pain and narcotic
medications by age and ED type (pediatric versus other). Analyses were repeated for
the subset of patients with moderate or severe pain severity scores.
Results: Of 2,828 patients with isolated closed fractures of the extremities or clavi-
cle, 64% received any analgesic and 42% received a narcotic analgesic. Pain sever-
ity scores were recorded for 59% of visits overall, 47% of children younger than 4
years, and 34% of children younger than 1 year. Among patients with documented
moderate or severe pain, 73% received an analgesic and 54% received a narcotic
analgesic. Compared with adults, a lower proportion of children (≤15 years) received
either any analgesic or a narcotic analgesic (P<.001). After adjustment for con-
founders and survey design, the proportion of patients aged 0 to 3, 4 to 8, 9 to 15, 16
to 29, 30 to 69, and 70 years and older who received any analgesic was 54% (95%
confidence interval [CI] 41% to 67%), 63% (95% CI 57% to 68%), 60% (95% CI 57% to
64%), 67% (95% CI 62% to 73%), 68% (95% CI 64% to 72%), and 58% (95% CI 52% to
65%), respectively; the proportion who received a narcotic analgesic was 21% (95%
CI 11% to 31%), 30% (95% CI 22% to 37%), 27% (95% CI 23% to 32%), 47% (95% CI 40%
to 54%), 51% (95% CI 46% to 56%), and 41% (95% CI 35% to 48%), respectively. Com-
pared with children treated in other EDs, children treated in pediatric EDs were about
as likely to receive any analgesia (adjusted relative risk [RR] 1.1; 95% CI 0.9 to 1.3) or
narcotic analgesia (adjusted RR 0.9; 95% CI 0.6 to 1.2).
Conclusion: In pediatric and adult patients, pain medications were frequently not
part of ED treatment for fractures, even for visits with documented moderate or
severe pain. Pain severity scores were often not recorded. Pediatric patients were
least likely to receive analgesics, especially narcotics.
[Ann Emerg Med. 2003;42:197-205.]
1 9 7
ANALGESIA FOR FRACTURE PAIN
Brown et al
INTRODUCTION
Several studies have evaluated administration of anal-
gesia in the emergency department (ED) and have
reported that patients frequently do not receive medica-
tion for painful conditions.1-4 Recently, the Joint
Commission on Accreditation of Healthcare Organiza-
tions stated that “unrelieved pain has adverse physical
and psychological effects. The patient’s right to pain
management is respected and supported.” The practice
standard to which health care organizations are now
held states that “pain is assessed in all patients.”5 Given
the current literature on analgesia in ED settings and
the emphasis of the Joint Commission on Accreditation
of Healthcare Organizations on provision of adequate
analgesia to patients in pain, we sought to determine
whether appropriate analgesia was being provided to
pediatric and adult ED patients with potentially painful
fractures during 1997 to 2000.
Using a national survey of ED visits, we evaluated
patients with extremity and clavicular fractures treated
in EDs. We described variations in analgesic use by age
and compared the use of analgesia for adults with that
for children. We repeated these analyses for the subset
of patients with moderate or severe pain on pain sever-
ity scores. In children, we also compared treatment
between pediatric and nonpediatric facilities.
METHODS
We used data from the ED component of the National
Hospital Ambulatory Medical Care Survey, which was
directed by the Centers for Disease Control and
Prevention’s National Center for Health Statistics, for
1997 to 2000. Community and academic hospitals,
including children’s hospitals, were included in this
survey, whereas federal hospitals and freestanding clin-
ics were excluded. EDs were eligible only if they were
open and staffed 24 hours a day. The survey used a 4-
stage sampling design to obtain representative national
estimates and took place during a randomly assigned 4-
week data period. Data collection was performed by
trained hospital staff using a standard collection form.
1 9 8
This study was exempt from full institutional review
board review.
We included only ED visits for closed fractures of the
clavicle or extremity as 1 of the first 3 diagnoses (Figure
1). During the 4 study years, there were 3,054 visits
with at least 1 of the following International Classifica-
Figure 1.
Numbers of patient visits included and excluded from the
study. *Including ICD-9-CM fracture diagnoses 812, 813,
810, 814-818, 820-827. †ICD-9-CM diagnostic categories
that specify >1 fracture within the category were not
excluded (eg, ICD-9-CM code 813 for “radius and/or ulna
fracture,” ICD-9-CM code 818 for “arms, multiple or
unspecified,” ICD-9-CM code 823 for “tibia and/or fibula
fracture.”) ‡Excluding ICD-9-CM open fracture diagnoses
812.1-812.19, 812.3, 812.5, 813.1, 813.3, 813.5, 813.9,
810.1-810.13, 814.1, 815.1, 816.1, 817.1, 818.1, 820.1,
820.3, 820.9, 821.1, 821.3, 822.1, 823.1, 823.3, 823.9,
824.1, 824.3, 824.5, 824.7, 824.9, 827.1, 825.1, 825.3,
826.1. §Excluding ICD-9-CM trauma diagnoses 800-809,
811, 819, 850-869, 940-949, 952.
93,109
ED visits surveyed
in 1997-2000
Include all diagnoses of closed
fracture of the extremity or clavicle*
3,054 Visits
Exclude all visits with fractures in
>1 fracture category†
2,989 Visits
Exclude all visits with a diagnosis of
open extremity or clavicle fracture‡
2,888 Visits
Exclude all visits involving
cardiopulmonary resuscitation
intubation, or ICU admission
2,869 Visits
Exclude all visits with additional
trauma diagnoses§
2,842 Visits
Exclude all visits in which
medications are illegible, or
coded as “medications IV”
or “medications continued”
2,828 Visits
ANNALS OF EMERGENCY MEDICINE 42:2 AUGUST 2003
ANALGESIA FOR FRACTURE PAIN
Brown et al
tion of Diseases, Ninth Revision, Clinical Modification
(ICD-9-CM) diagnostic codes for extremity or clavicle
fracture. The 65 visits involving fracture in more than 1
ICD-9-CM category were subsequently excluded,
although it was still possible to have more than 1 frac-
ture coded within a category (eg, a radius and an ulna
fracture). Also excluded were 101 visits involving ICD-
9-CM codes for open fracture; 27 visits involving car-
diopulmonary resuscitation, intubation, ICU admis-
sion, or additional trauma diagnoses; and 14 visits in
which medications were either illegible on the survey
form or the name of the specific medication given was
not listed (eg, listed as “intravenous medications” or
“medications continued”).
The survey medication item allowed the user to
record up to 6 medications that were ordered, supplied,
administered, or continued during the visit, including
prescription and over-the-counter medications, immu-
nizations, allergy shots, and anesthetics. The medica-
tions listed were then coded by unique drug codes and
also categorized by using a drug class variable based on
the National Drug Code Directory of Drug Classes.6 We
found that the drug class variables were missing a few
pain medications otherwise identified by unique drug
codes. To ensure that all pain medications were in-
cluded in the drug class variables, we took the following
steps: to the narcotic medication drug class variable we
added Vicodin (drug code 34110), Lortab (92180),
Lorcet (93089), Tylox (32945), Tylenol with codeine
(32915, 32920, 32925, 32930, 32935), morphine sul-
fate (99123), meperidine (18760), codeine (7180), Hy-
phen (93077) and hydrocodone (14955); to the pain
medication drug class variable we added morphine sul-
fate (99123) and hydrocodone (14955).
For data description, we used 6 categories of age: 3
pediatric categories (0 to 3, 4 to 8, and 9 to 15 years) and
3 adult categories (16 to 29, 30 to 69, and ≥70 years).
For regression analyses, we used 17 age categories (<1
year; 1 year; and 2 to 3, 4 to 5, 6 to 7, 8 to 9, 10 to 12, 13
to 15, 16 to 19, 20 to 29, 30 to 39, 40 to 49, 50 to 59, 60
to 69, 70 to 79, 80 to 89, and ≥90 years). No survey vari-
able identified which hospitals were dedicated pedi-
atric facilities. In keeping with the practice of other
AUGUST 2003 42:2 ANNALS OF EMERGENCY MEDICINE
investigators, we defined a pediatric hospital as one in
which more than 80% of all visits (not just visits for
fractures) were made by patients younger than 18
years.7
We considered the following survey variables as
potential confounders: fracture type (clavicle, humerus,
radius or ulna, hand, femur, tibia or fibula, foot, other),
year of study, sex, race (white, black, other), ethnicity
(Hispanic, non-Hispanic), source of payment (private
insurance, Medicare, Medicaid, workers’ compensa-
tion, self-pay, other), metropolitan area (yes/no), geo-
graphic location (Northeast, Midwest, South, West),
type of provider (staff physician, resident/intern, other
physician, physician assistant, nurse practitioner, reg-
istered nurse, other), type of hospital ownership (vol-
untary, government, proprietary), triage assessment of
the immediacy of care required (<15 minutes, 15 to 60
minutes, >1 to 2 hours, >2 to 24 hours), orthopedic care
provided in the ED (yes or no), shift at patient arrival
(day, evening, night), admission to the hospital (yes or
no), pain severity score on presentation (none, mild,
moderate, or severe), and pediatric versus nonpediatric
hospital. The information on race and ethnicity was
based on the provider’s knowledge of the patient or by
observation because providers were not directed or
expected to ask the patient for this information.7
In all analyses, we accounted for the primary sam-
pling units, the sampling strata, and the probability
sampling weights used in the survey by using lineariza-
tion methods in Stata software (version 7.0, Stata
Corporation, College Station, TX).8 Four strata con-
tained a single primary sampling unit; we therefore col-
lapsed these 4 strata into 2 with 2 primary sampling
units each.
We used linear regression to assess whether the
adjusted proportion of patients receiving any analgesic
or a narcotic analgesic varied by category of age.
Although our study outcome, receipt of pain medica-
tion, was binary (and therefore not normally dis-
tributed), in a reasonably large study such as ours, pro-
portions can be thought of as means that are normally
distributed under the central limit theorem.9 Some
results were presented as mean proportions receiving
1 9 9
ANALGESIA FOR FRACTURE PAIN
Brown et al
medication by age group, adjusted for the mean values
of any confounding variables in the regression.
We used Poisson regression10 to estimate risk ratios
for analgesic or narcotic analgesic use according to pain
severity score and whether the hospital was pediatric
(among children aged ≤15 years). The variables frac-
ture type, year, geographic location, admission to the
hospital, and need for orthopedic care were considered
likely confounders and were included in all regression
models. In addition, we adjusted for any other variable
that changed the risk ratios by more than 10%.
All of the statistical analyses assumed that observa-
tions were independent in the sense that no individual
was included more than once in the data. Because the
survey recorded patient visits only and did not include
patient identifiers, it was possible for a patient to be in
our data more than once. However, given the nature of
these injuries and the large number of participating
hospitals (283 to 309 participating hospitals a year,
with no more than 11 patients per hospital a year), it is
unlikely that there were many, if any, patients included
more than once.
R E S U LT S
During the 4 study years, there were 2,828 ED visits for
isolated closed fractures of the extremities or clavicle
that met inclusion and exclusion criteria in the sample
(Figure 1), which represents a national estimate of 12.5
million ED visits a year. The characteristics of these
patients are shown in Table 1. Children younger than 16
years had more arm fractures than adults and fewer
hand and foot fractures. Compared with other age
groups, children aged 9 to 15 years were more likely to
be male, and patients 70 years and older were more
likely to be female. Compared with older children and
adults, a higher proportion of children younger than 4
years were of minority races, from metropolitan areas,
and from the southern United States. Adults aged 30
years or older were more likely to be admitted to the
hospital than patients younger than 30 years. Race and
ethnicity, which have been reported as risk factors for
2 0 0
inadequate analgesia, were not found to be confounders
of the relationship between age and analgesic use.11-16
For 71% of the patients studied, there was documen-
tation that at least 1 medication was ordered, supplied,
administered, or continued during the visit (including
prescription and over-the-counter medications). These
medications included an analgesic in 64% of patients
studied and a narcotic analgesic, either alone or in a
combination product, in 42% of patients studied. Pain
medication use varied only modestly throughout the 4
years of the study, from 62% in 1997 to 66% in 2000;
narcotic use similarly varied from 40% in 1997 to 45%
in 2000 (linear regression tests for trend were not sig-
nificant; P>.05 for both tests).
Our first analysis described the use of analgesics and
narcotic analgesics by age for all patients. The propor-
tions of patients who received any analgesic or a nar-
cotic analgesic were lower in pediatric and elderly
patients, with more prominent differences in narcotic
medication use (Table 2, Figure 2). This trend was most
noticeable in children younger than 4 years; only 54%
of these children received any analgesics, and only 21%
received narcotic analgesics. Our second analysis tested
for differences in analgesic use across age categories.
The differences in any analgesic and narcotic analgesic
use were statistically different across the 17 age cate-
gories (P<.001 for both tests).
Our third analysis evaluated the use of analgesics and
narcotic analgesics by age for patients with documented
moderate or severe pain. To begin, we characterized
patients for whom information on pain severity was
available. Pain severity scores were recorded for 59% of
patients. The proportion of patients who had pain
severity scores recorded was similar throughout the 4
years of the study. For patients with any pain severity
score, the proportion with moderate or severe pain also
did not change throughout the 4 years of study. Young
children were less likely to have a pain severity score
recorded (Table 2). However, patients who had a pain
severity score recorded by ED staff (including a judg-
ment of no pain) were not much more likely to receive
any analgesic than patients with no record of their pain
ANNALS OF EMERGENCY MEDICINE 42:2 AUGUST 2003
ANALGESIA FOR FRACTURE PAIN
Brown et al

status (any analgesic, adjusted relative risk [RR] 1.1;
95% confidence interval [CI] 0.99 to 1.2). Patients who
had any pain severity score recorded were somewhat
more likely to have received a narcotic compared with
patients for whom no information about pain severity
was recorded (adjusted RR 1.3; 95% CI 1.1 to 1.4).
Next, we evaluated patients with documented mod-
erate or severe pain. Compared with a pain severity
score of none or mild, the presence of documented
moderate or severe pain was associated with pain medi-
cation use (adjusted RR 1.3; 95% CI 1.2 to 1.4) and nar-
cotic use (adjusted RR 1.5; 95% CI 1.3 to 1.7). However,
the use of analgesics among patients with documented
moderate to severe pain remained low: only 73% re-
ceived any pain medication and 54% received narcotics
(Table 2, Figure 3). Children younger than 1 year were
unlikely to have documented moderate to severe pain,
but when they did, they were likely to receive anal-
gesics. In patients aged 4 to 15 years, in which the pro-
portions with documented pain severity scores were
more comparable to that of the adult groups, the pro-
portion of children with moderate or severe pain receiv-

Table 1.
Baseline characteristics of patients with closed clavicle or extremity fractures.

Age Category, No. (%)

Characteristic All Ages, No. (%) 0–3 y 4–8 y 9–15 y 16–29 y 30–69 y ≥70

Total visits, No. (%)* 2,828 (100) 102 (3) 232 (9) 487 (19) 514 (18) 1,039 (36) 454 (16)
Closed fracture category
Clavicle 126 (4) 14 (9) 6 (1) 37 (6) 23 (3) 37 (4) 9 (2)
Humerus 242 (9) 12 (15) 33 (18) 33 (8) 25 (4) 71 (7) 68 (16)
Radius/ulna 516 (19) 30 (30) 86 (34) 134 (28) 66 (14) 150 (14) 50 (12)
Hand 756 (27) 17 (16) 52 (20) 163 (34) 223 (43) 250 (24) 51 (10)
Femur 264 (9) 5 (5) 8 (3) 9 (2) 7 (2) 56 (5) 179 (38)
Tibia/fibula 458 (15) 16 (18) 22 (10) 52 (10) 81 (14) 227 (21) 60 (14)
Foot 385 (14) 6 (5) 14 (7) 49 (11) 83 (19) 210 (20) 23 (4)
Other 81 (3) 2 (2) 11 (6) 10 (2) 6 (1) 38 (5) 14 (3)
Male sex 1,466 (52) 54 (52) 129 (55) 346 (70) 340 (65) 496 (47) 101 (22)
Race
White 2,375 (85) 81 (77) 190 (84) 409 (86) 415 (84) 854 (82) 426 (94)
Black 369 (12) 15 (17) 33 (13) 58 (10) 84 (14) 158 (15) 21 (5)
Other 84 (3) 6 (7) 9 (3) 20 (3) 15 (3) 27 (3) 7 (1)
Metropolitan area 2,320 (74) 88 (82) 195 (75) 387 (73) 411 (71) 878 (77) 361 (73)
Geographic location
Northeast 102 (19) 20 (13) 46 (12) 119 (17) 125 (19) 278 (21) 115 (20)
Midwest 232 (29) 22 (24) 74 (35) 129 (28) 153 (33) 255 (27) 115 (29)
South 487 (31) 36 (41) 68 (32) 134 (33) 141 (29) 299 (31) 134 (30)
West 514 (21) 24 (22) 44 (21) 105 (22) 95 (19) 207 (20) 90 (20)
Pediatric hospital† 66 (3) 16 (21) 27 (12) 21 (5) 2 (<1) 0 (0) 0 (0)
Pain on presentation
None or mild 646 (24) 18 (22) 56 (23) 127 (25) 138 (30) 224 (22) 83 (19)
Moderate or severe 977 (35) 28 (25) 85 (39) 150 (32) 149 (28) 380 (38) 185 (42)
No pain score 1,205 (41) 56 (53) 91 (38) 210 (43) 227 (42) 435 (40) 186 (39)
ED care
Extremity radiographs done 2,188 (79) 80 (82) 190 (84) 393 (83) 403 (79) 806 (78) 316 (71)
Other radiographs done 484 (16) 15 (11) 24 (10) 58 (11) 81 (17) 167 (14) 139 (28)
Treated by staff physician 2,544 (90) 94 (92) 205 (90) 428 (88) 459 (89) 934 (90) 424 (93)
Orthopedic care 1,959 (71) 70 (73) 175 (73) 388 (83) 377 (74) 703 (68) 246 (55)
Admitted to hospital 371 (13) 5 (6) 16 (7) 13 (3) 29 (5) 117 (11) 191 (41)
*All proportions are adjusted for survey design.
†A pediatric hospital is defined as one in which >80% of visits (not just visits for fractures) were made by patients aged <18 years.

AUGUST 2003 42:2 ANNALS OF EMERGENCY MEDICINE 2 0 1

ANALGESIA FOR FRACTURE PAIN
Brown et al

ing any analgesic and narcotic analgesics remained
lower than that for adults younger than 90 years. The
proportion of elderly patients who were older than 90
years, had moderate or severe pain, and received pain
medications and narcotics was lower than that of
younger adults and similar to that of children.
Our fourth analysis tested for differences in analgesic
use across age categories for patients with moderate or
severe pain. For these patients, the differences in pain
medication use and narcotic use across the 17 age cate-
gories were statistically significant (P<.001 for both
comparisons).
Our final analysis compared analgesic use for chil-
dren treated in pediatric versus general EDs. The 64
Table 2.
Proportion of patients who had closed clavicle or extremity
fractures and received pain and narcotic analgesia (adjusted
for survey design, fracture type, year, hospital admission,
orthopedic care, and geographic location).
children younger than 16 years who were treated in
pediatric EDs had a median age of 7 years compared
with 10 years for the 758 children who were younger
than 16 years and were treated in general EDs. The chil-
dren who were younger than 16 years and treated in
pediatric EDs were less likely to have pain severity
scores recorded than the children who were younger
than 16 years and treated in general EDs (34% versus
64%, respectively; adjusted RR 0.6; 95% CI 0.5 to 0.9).
In pediatric and other EDs, similar proportions of chil-
Figure 2.
A, Point estimates with 95% CIs showing the proportion of
patients with closed clavicle or extremity fractures who
received any analgesia (adjusted for survey design, fracture
type, year, hospital admission, orthopedic care, and geo-
graphic location) (N=2,828). B, Point estimates with 95%
CIs showing the proportion of patients with closed clavicle or
extremity fractures who received narcotic analgesia (adjusted
for survey design, fracture type, year, hospital admission,
orthopedic care, and geographic location) (N= 2,828).

Proportion Proportion A 1
Receiving Any Receiving
receiving any analgesia

Analgesic Narcotic
Adjusted proportion

0.8
No. of Agents (95% (95%
Population/ Patients Confidence Confidence 0.6
Age, y (% Total)* Limits) Limits)
0.4
All patients
0–3 102 0.54 (0.41, 0.67) 0.21 (0.11, 0.31) 0.2
4–8 232 0.63 (0.57, 0.68) 0.30 (0.22, 0.37)
9–15 487 0.60 (0.57, 0.64) 0.27 (0.23, 0.32) 0
16–29 514 0.67 (0.62, 0.73) 0.47 (0.40, 0.54)
30–69 1,039 0.68 (0.64, 0.72) 0.51 (0.46, 0.56) 5 15 25 35 45 55 65 75 85 95
≥70 454 0.58 (0.52, 0.65) 0.41 (0.35, 0.48)
Age, y
Patients with any pain
score recorded
0–3 46 (47) 0.56 (0.40, 0.72) 0.32 (0.17, 0.46) B
4–8 141 (62) 0.63 (0.56, 0.71) 0.28 (0.18, 0.39) 1
receiving narcotic analgesia

9–15 277 (57) 0.61 (0.56, 0.66) 0.32 (0.26, 0.39)

0.8
Adjusted proportion

16–29 287 (58) 0.67 (0.60, 0.74) 0.49 (0.41, 0.57)

30–69 604 (60) 0.73 (0.67, 0.79) 0.56 (0.48, 0.64)
≥70 268 (61) 0.60 (0.52, 0.67) 0.43 (0.35, 0.52) 0.6
Patients with moderate
to severe pain recorded 0.4
0–3 28 (25) 0.62 (0.41, 0.84) 0.45 (0.24, 0.66)
4–8 85 (35) 0.76 (0.63, 0.88) 0.43 (0.26, 0.59) 0.2
9–15 150 (28) 0.68 (0.59, 0.77) 0.41 (0.30, 0.51)
16–29 149 (38) 0.79 (0.71, 0.87) 0.59 (0.48, 0.89) 0
30–69 380 (42) 0.78 (0.71, 0.85) 0.63 (0.55, 0.70)
≥70 185 (35) 0.62 (0.51, 0.72) 0.47 (0.39, 0.55) 5 15 25 35 45 55 65 75 85 95
*Adjusted for survey design only. Age, y

2 0 2 ANNALS OF EMERGENCY MEDICINE 42:2 AUGUST 2003

ANALGESIA FOR FRACTURE PAIN
Brown et al

dren younger than 16 years had moderate to severe pain DISCUSSION

when pain severity scores were used (adjusted RR 1.1;
95% CI 0.6 to 1.9), and similar proportions of children
younger than 16 years had documented treatment with
any analgesic (adjusted RR 1.1; 95% CI 0.9 to 1.3) or a
narcotic analgesic (adjusted RR 0.9; 95% CI 0.6 to 1.2).
Results were similar when children younger than 5
years were analyzed separately.
Figure 3.
A, Point estimates with 95% CIs showing the proportion of
patients with closed clavicle or extremity fractures and docu-
mented pain severity scores who received any analgesia
(adjusted for survey design, fracture type, year, hospital
admission, orthopedic care, and geographic location)
(N=977). B, Point estimates with 95% CIs showing the pro-
portion of patients with closed clavicle or extremity fractures
and documented pain severity scores who received narcotic
analgesia (adjusted for survey design, fracture type, year,
hospital admission, orthopedic care, and geographic location)
(N=977).
A 1
receiving any analgesia
A third of the fracture patients in our study did not re-
ceive any pain medications. The proportion of patients
receiving pain medications was even lower in children.
Pain severity scores appear to be underused in the ED
setting. Pain scores were recorded in only 59% of patients
overall, and in only 47% of children younger than 4
years. Even when pain scores were recorded as moderate
or severe, analgesics were not routinely used. For
patients with documented moderate to severe pain,
73% overall and 62% of patients younger than 4 years
received any analgesia.
Our findings are in keeping with previous North
American studies of pain management in the ED. In pre-
vious reports, the proportion of adults receiving analge-
sia for painful conditions ranged between 31% and
74%.1,4,14,17-20 This range was the same for studies lim-
ited to fractures. The proportion of children receiving
analgesia for painful conditions ranged between 25%
and 96%.3,4,17,18,21,22
As in this study, previous studies comparing adult
and pediatric patients have reported that children re-

ceive less analgesia than adults.4,17,23,24 Schechter et

Adjusted proportion

0.8

0.6
al23 retrospectively compared narcotic use for hospital-
ized patients with 4 diagnoses: hernias, appendec-
0.4
tomies, burns, and fractured femurs. They found that
0.2 adults received significantly more doses of narcotics
0
than children (2.2 versus 1.1 doses; P=.0001). Infants
and young children were less likely than older children
5 15 25 35 45 55 65 75 85 95
Age, y to have narcotics prescribed. Selbst and Clark17 studied
268 patients with painful conditions (ie, sickle cell
B 1 crises, lower-extremity fractures, second- or third-
receiving narcotic analgesia

degree burns) and reported that 28% of children and

0.8
Adjusted proportion

60% of adults received analgesia. In an analysis of the

0.6
National Hospital Ambulatory Medical Care Survey for
0.4 1992 to 1997, 5.6% of children versus 7.8% of adults
0.2 received parenteral sedation or analgesia for orthopedic
0 injuries (P=.05).24 This analysis was not adjusted for
fracture type, pain severity score, or survey design.
5 15 25 35 45 55 65 75 85 95 Petrack et al4 retrospectively compared use of anal-
Age, y gesia in 120 children and 120 adults treated for long
bone fractures between 1993 and 1994 in 1 of 3 settings:

AUGUST 2003 42:2 ANNALS OF EMERGENCY MEDICINE 2 0 3

ANALGESIA FOR FRACTURE PAIN
Brown et al
academic centers with combined adult and pediatric
EDs, community EDs, and academic centers with sepa-
rate adult and pediatric EDs. Overall, pediatric patients
were significantly less likely to receive analgesia while
in the ED than adults (53% versus 73%; P=.05). In sepa-
rate analyses of each center, the differences between
children and adults were significant at the academic
centers with combined pediatric and adult EDs (58%
versus 88%, respectively; P<.003) and the community
EDs (38% versus 65%; P<.02) but not at the academic
centers with separate adult and pediatric EDs (65% ver-
sus 68%).4 These findings are contrary to our results of
no difference in analgesia provided to children in pri-
mary pediatric EDs compared with general EDs.
Although it is possible that the study by Petrack et al
was detecting regional differences that are not apparent
nationally, it seems more likely that this difference in
results is caused by variation in hospital case mix be-
tween the 2 studies. In our study, we were unable to sep-
arate academic from community hospitals. Although it
is likely that EDs treating more than 80% pediatric
patients are connected with pediatric academic centers,
we do not have detailed information about the remain-
ing EDs, and we cannot analyze the use of analgesia by
specific general hospital type.
In our study, pain medication use was lowest at the 2
extremes of age, suggesting that providers may be less
comfortable assessing pain and providing pain relief for
the very old and the very young. Some providers may be
concerned about their ability to assess pain in nonver-
bal patients, but simple pain scales can be used for non-
verbal or verbal but illiterate patients. These scales,
such as the Modified Infant Pain Scale; Face, Legs,
Activity, Cry, Consolability pain rating scale; and the
Wong-Baker Faces pain rating scale have been validated
in infants and young children.25-28 Alternatively,
providers may be concerned about the safety of anal-
gesics, particularly narcotic medications. However,
narcotic medications have been shown to be safe even
in young children. In 1 study of fentanyl-induced respi-
ratory depression, the authors did not detect a differ-
ence in respiratory depression between children older
than 3 months and adults.29 In a large prospective case
2 0 4
series, adverse events were noted in only 2.3% of 1,180
pediatric patients who underwent procedural sedation
and analgesia in the ED, with no serious complications
noted.30
In the analysis of patients with moderate to severe
pain, a higher than average proportion of infants (chil-
dren <1 year) received analgesia compared with all
other age categories, which most likely reflects a differ-
ence in the use of pain severity scores. Compared with
all other age groups, a lower proportion of infants had
pain severity scores recorded. It is not surprising that
the small subset of infants who received pain severity
scores also received analgesia. This does not mean we
are doing well caring for infants, because most infants
did not have pain scores and did not receive analgesia. It
probably means that the small subset of infants who had
someone paying enough attention to their pain to give
them a pain score also paid enough attention to give
them analgesia.
We believe this is the largest detailed study to date
comparing pediatric and adult analgesia for fracture
pain. The study has several limitations. The survey
medication item allowed the user to record medications
that were ordered, supplied, administered, or contin-
ued during the visit. It is possible that some patients
received medications before arrival that were not con-
tinued during the visit and thus not recorded. Although
directed to list all medications, including over-the-
counter medications, some hospital staff may have
neglected to list all analgesics given, especially if over-
the-counter pain medications were routinely adminis-
tered by nursing staff in triage. It is also possible that
providers evaluated their patient’s pain severity without
recording this information on the survey. Finally, there
was no way to distinguish visits for acute fracture man-
agement from visits for follow-up care or fracture com-
plications. However, 97% of the patients with pain
scores reported pain, and analgesics were given to a sim-
ilar proportion of patients with and without pain scores.
Thus, it is likely that analgesia was indicated for the
majority of patients, regardless of the timing of injury.
The results of this survey indicate that ED analgesic
use for extremity and clavicular fractures remained low
ANNALS OF EMERGENCY MEDICINE 42:2 AUGUST 2003
ANALGESIA FOR FRACTURE PAIN
Brown et al
and the use of pain severity scores inadequate, from
1997 through 2000. Children and elderly patients were
least likely to receive analgesics. It is widely known that
assessment and management of pain should be part of
routine care of all patients in the ED, especially those
with disorders likely to be painful. We conclude that
additional effort and resources are needed to address
the issue of undertreatment of pain in children and
adults with fractures in the ED setting. Special attention
should be given to analgesia in the very old and very
young. Educating providers on nonverbal options for
measuring pain, especially in young children, may
improve measurement and documentation of pain sta-
tus and facilitate recognition and treatment of pain in
these vulnerable populations.
Author contributions: JCB and EJK conceived and designed the
study, drafted and revised the manuscript, and performed the anal-
ysis and interpretation. JCB, CWL, and BDJ managed the dataset.
CWL, BDJ, and PC reviewed the manuscript and assisted with the
analysis. PC provided statistical consultation regarding accuracy
of the analysis and presentation of the results. JCB takes responsi-
bility for the paper as a whole.
Received for publication January 31, 2003. Revision received March
26, 2003. Accepted for publication April 1, 2003.
The authors report this study did not receive any outside funding or
support.
Address for reprints: Julie C. Brown, MD, MPH, Children’s Hospital
and Regional Medical Center, 4800 Sand Point Way NE, Mailstop
5D-1, Seattle, WA 98105; 206-987-2599, fax 206-729-2070; E-mail
julbrown@u.washington.edu.
REFERENCES
1. Wilson JE. Oligoanalgesia in the emergency department. Am J Emerg Med.
1989;7:620-623.
2. Reichl M, Bodiwala GG. Use of analgesia in severe pain in the accident and emer-
gency department. Arch Emerg Med. 1987;4:25-31.
3. Friedland LR, Kulick RM. Emergency department analgesic use in pediatric trauma
victims with fractures. Ann Emerg Med. 1994;23:203-207.
4. Petrack EM, Christopher NC, Kriwinsky J. Pain management in the emergency
department: patterns of analgesic utilization. Pediatrics. 1997;99:711-714.
5. Comprehensive Accreditation Handbook for Hospitals: The Official Handbook. Oak-
brook Terrace, IL: Joint Commission on Accreditation of Healthcare Organizations; 2001.
6. US Food and Drug Administration. The national drug code directory. Available at:
http://www.fda.gov/cder/ndc/index.htm. Accessed June 11, 2003.
7. Stone S, Gonzales R, Maselli J, et al. Antibiotic prescribing for patients with colds,
upper respiratory tract infections, and bronchitis: a national study of hospital-based
emergency departments. Ann Emerg Med. 2000;36:320-327.
AUGUST 2003 42:2 ANNALS OF EMERGENCY MEDICINE
8. Levy PS, Lemeshow S. Sampling of Populations: Methods and Applications. New
York, NY: John Wiley; 1999:365-370.
9. Lumley T, Diehr P, Emerson S, et al. The importance of the normality assumption in
large public health data sets. Ann Rev Public Health. 2002;23:151-169.
10. Long JS. Regression Models for Categorical and Limited Dependent Variables.
Thousand Oaks, CA: SAGE Publications; 1997:217-250.
11. Todd KH, Samaroo N, Hoffman JR. Ethnicity as a risk factor for inadequate emer-
gency department analgesia. JAMA. 1993;269:1537-1539.
12. Todd KH, Lee T, Hoffman JR. The effect of ethnicity on physician estimates of pain
severity in patients with isolated extremity trauma. JAMA. 1994;271:925-928.
13. Todd KH. Pain assessment and ethnicity. Ann Emerg Med. 1996;27:421-423.
14. Todd KH, Deaton C, D’Adamo AP, et al. Ethnicity and analgesic practice. Ann
Emerg Med. 2000;35:11-15.
15. Todd KH. Influence of ethnicity on emergency department pain management. Emerg
Med. 2001;13:274-278.
16. Yen K, Kim M, Stremski ES, et al. The effect of ethnicity and race on the use of pain
medications on children with long bone fractures in the emergency department: use of
a national database. Acad Emerg Med. 2001;8:447-448.
17. Selbst SM, Clark M. Analgesic use in the emergency department. Ann Emerg Med.
1990;19:1010-1013.
18. Lewis LM, Lasater LC, Brooks CB. Are emergency physicians too stingy with anal-
gesics? South Med J. 1994;87:7-9.
19. Jones JS, Johnson K, McNinch M. Age as a risk factor for inadequate emergency
department analgesia. Am J Emerg Med. 1996;14:157-160.
20. Ngai B, Ducharme J. Documented use of analgesics in the emergency department
and upon release of patients with extremity fractures. Acad Emerg Med. 1997;4:1176-1178.
21. Friedland LR, Pancioli AM, Duncan KM. Pediatric emergency department analgesic
practice. Pediatr Emerg Care. 1997;13:103-106.
22. Chan L, Russell TJ, Robak N. Parental perception of the adequacy of pain control in
their child after discharge from the emergency department. Pediatr Emerg Care.
1998;14:251-253.
23. Schechter NL, Allen DA, Hanson K. Status of pediatric pain control: a comparison
of hospital analgesic usage in children and adults. Pediatrics. 1986;77:11-15.
24. Hostetler MA, Auinger P, Szilagyi PG. Parenteral analgesic and sedative use among
ED patients in the United States: combined results from the National Hospital Ambula-
tory Medical Care Survey (NHAMCS) 1992-1997. Am J Emerg Med. 2002;20:139-143.
25. Buchholz M, Karl HW, Pomietto M, et al. Pain scores in infants: a modified infant
pain scale versus visual analogue. J Pain Symptom Manage. 1998;15:117-124.
26. Merkel SI, Voepel-Lewis T, Shayevitz JR, et al. The FLACC: a behavioral scale for
scoring postoperative pain in young children. Pediatr Nurs. 1997;23:293-297.
27. Wong DL. Assessing pain at face value [letter]. Am J Nurs. 1999;99:16.
28. Bulloch B, Tenenbein M. Validation of 2 pain scales for use in the pediatric emer-
gency department. Pediatrics [serial online]. 2002;110:e33.
29. Hertzka RE, Gauntlett IS, Fisher DM, et al. Fentanyl-induced ventilatory depression:
effects of age. Anesthesiology. 1989;70:213-218.
30. Pena BM, Krauss B. Adverse events of procedural sedation and analgesia in a
pediatric emergency department. Ann Emerg Med. 1999;34(4, pt 1):483-491.
2 0 5