The emotional, physical, and financial impact of pediatric trauma on
society is enormous. On the basis of the data in the National Health Interview
Survey (1987 to 1994), Danseco et al. estimated the cost of unintentional
childhood injury in the United States in 1994 dollars to be $347 billion
annually1. This
estimate included $17 billion in medical costs, $72 billion in future work
loss, and $257 billion in lost quality of life. Over twenty million children
are injured each year, with an incidence of one of every four
children1,2.
In 1996, approximately 13% of all medical expenditures for children between
the ages of one and nineteen years was for the treatment of
trauma1,3.
Trauma is the leading cause of death and disability in
children4-7,
and, in 1998, trauma accounted for 44% of all childhood
deaths6.
Injury accounts for a substantial number of pediatric emergency-department
visits and inpatient admissions. Using National Hospital Ambulatory Medical
Care Survey data from 1998, Simon et al. documented that children who were
less than nineteen years old accounted for twenty-eight million
emergency-department visits, of which eleven million were
trauma-related5. The
orthopaedic component of this trauma volume was substantial, with 16.2% due to
contusions, 15% due to sprains and strains, and 14.3% due to fractures and
dislocations. Similarly, the orthopaedic injury contribution to inpatient
pediatric admissions is substantial. Galano et al., using data from the
Healthcare Cost and Utilization Project Kids' Inpatient Database (HCUP KID),
estimated that there were 6.7 million pediatric discharges from all hospitals
in the United States during
19978. Pediatric
orthopaedic trauma accounted for 84,000 admissions during this time-period,
generating $932.8 million in hospital charges. Of those admissions, 21.7% was
for femoral fractures; 21.5%, for tibial and/or fibular fractures; 17%, for
humeral fractures; 14.8%, for radial and/or ulnar fractures; and 5.2%, for
vertebral fractures. Importantly, 70.4% of the children seen in that study
because of orthopaedic trauma were treated at nonpediatric hospitals and only
29.6% were treated at pediatric
hospitals8.
It is the impression of many pediatric orthopaedists that trauma care is
becoming a larger component of both office and inpatient pediatric orthopaedic
practice. Although it has not been well documented, many pediatric
orthopaedists have the impression that more pediatric orthopaedic trauma is
being cared for by pediatric orthopaedic specialists practicing in pediatric
hospital settings. Kasser documented this trend of increased pediatric
orthopaedic trauma care by pediatric specialists for supracondylar humeral
fractures9. Using
the trauma registry at Children's Hospital in Boston, in conjunction with HCUP
KID, he demonstrated that, in 1991, 63% of the supracondylar humeral fractures
in New England were treated by general orthopaedists, whereas, in 1999, 68%
were treated by pediatric orthopaedists in a pediatric hospital.
While there are published data describing national inpatient hospital and
emergency-room impact from both general pediatric trauma and, to a lesser
extent, orthopaedic pediatric trauma, we know of no report that has detailed
the impact of trauma on an actual pediatric orthopaedic practice. We used data
from a single urban, academic, pediatric orthopaedic group practice to
document the frequency of pediatric office and operative fracture care
provided by this group. The work effort required for nonoperative and
operative fracture care was measured and is expressed as the physician
component (work relative value units) of the resource-based relative value
scale as defined by the Centers for Medicare and Medicaid Services.
Publication of these data is intended to be of value to others in pediatric
orthopaedic practice.
The data in this analysis were obtained for fiscal year 2005 (July 2004
through June 2005) from a single practice group of fellowship-trained
pediatric orthopaedic surgeons consisting of 3.4 full-time-equivalent
clinicians. This group practices in an academic, university-based, fully
integrated health system environment. The main office of the practice is
located in a freestanding urban pediatric hospital with a level-I trauma
center designation, and secondary offices are located in three suburban
satellite locations. The surgeons performed all surgery at the freestanding
pediatric hospital. The pediatric orthopaedists covered all trauma call at the
hospital, and each participated in an every-fourth-night call rotation.
The clinical practice was supported by cast technicians and/or medical
assistants in the office and by orthopaedic residents in the emergency
department. The practice did not employ nurse practitioners or physician
assistants during fiscal year 2005 and did not share in any physical therapy
or radiology revenue support. Fracture reductions were not initially performed
by the pediatric orthopaedists in either the offices or emergency department
but rather were routinely done in the emergency department by orthopaedic
residents without the direct physical presence of the pediatric orthopaedists.
Orthopaedic attending staff support was available to the residents at all
times, either physically or by telephone, if needed. Fracture reduction
techniques were routinely discussed with senior staff, and no fracture was
manipulated more than twice in the emergency department. In addition,
computerized digital radiography was available in real time to senior staff at
home and was frequently used to monitor fracture care by the residents when
necessary. The cases of all patients were reviewed by senior staff with either
the treating orthopaedic resident, or all residents on rotation at the
institution, either the next day or within a few days of the emergency
department visit.
No professional orthopaedic fee or associated work relative value unit
(RVU) was generated in the emergency department for the work performed by
orthopaedic residents unless an attending pediatric orthopaedist was
physically present for the fracture reduction. A substantial majority of the
patients seen by the orthopaedic residents in the emergency department
returned to the study practice for follow-up care and provide part of the data
for this study. The data on the operative treatment of fractures do not
include information on a small number of patients (less than ten) with a
finger phalanx or metacarpal fracture in fiscal year 2005, as a separate adult
hand team performed this work. Essentially, all data on the nonoperative
treatment provided in the office for finger phalanx or metacarpal fractures
are included. A professional orthopaedic charge and calculation of the
associated work RVU were first generated in the practice office at the time of
the child's initial office visit or at the time of surgery. All operative
procedures were physically staffed by one of the full-time pediatric
orthopaedists and were performed in a hospital setting.
Computerized billing data for the study group were used to compile the
information presented in this study. Every outpatient office encounter and
operative intervention with a Current Procedural Terminology
(CPT)10 code that
was performed in the hospital by the group in fiscal year 2005 was included in
the billing database. This database included the particular evaluation and
management code generated for every office visit and the CPT code for every
operative procedure. More than one CPT code was included for a single patient
if multiple procedures were done at the same anesthetic setting. CPT codes for
irrigation and débridement (10180 to 11044) were compiled independent
of the associated fracture CPT code. Treatment of a fracture in the office was
coded with use of either evaluation and management codes or, more commonly,
nonmanipulation global fracture CPT codes at the orthopaedist's discretion,
depending on when the child first presented to the office. Assignments of the
work RVU were listed for every outpatient encounter or operative procedure
with use of the 2005 values provided by the Centers for Medicare and Medicaid
Services11. For
example, an established, focused evaluation and management office visit
(99212) was assigned 0.45 work RVU, a detailed consultative office visit
(99243) was assigned 1.72 work RVU, a percutaneous pinning of a displaced
supracondylar humeral fracture (24538) was assigned 9.42 work RVU, and an
instrumented scoliosis fusion with use of instrumentation and autogenous bone
graft (22802, 22843, and 20937) was assigned 47.06 work RVU.
The billing database was specifically analyzed to determine the
contribution of trauma to the pediatric orthopaedic practice of the group.
Descriptive statistics, including the actual numbers and percentages of office
fracture visits and CPT-coded trauma-related operations and the actual numbers
and percentages of work RVU production for fiscal year 2005, are
presented.
Work RVU for Fracture Treatment in the Office and Operating Room
This pediatric orthopaedic group generated a total of 36,771 work RVU, with
18,693 units (51%) generated from treatment in the operating room and 18,078
(49%) from treatment in the office (Fig.
1). A total of 1903 new patients with a fracture were seen in the
office, and their care accounted for 5286 work RVU or 29% of all
office-generated work RVU. An additional 412 work RVU (2% of all work RVU for
treatment provided in the office) were generated from 520 cast applications or
removals required to care for these 1903 fractures. Combining the work RVU
generated from general fracture care in the office with the work RVU generated
from the associated cast applications results in 5698 work RVU (32% of all
office-related work RVU) generated as a direct result of fracture care in the
office. Of the 19,389 office visits in fiscal year 2005, 6661 (34%) were for
nonoperative fracture care.
Of the 18,693 work RVU generated in the operating room, 5975 (32%) were
from fracture treatment. This represented the largest single category of work
done in the operating room, followed by other operations on the spine (31%),
foot and/or ankle (11%), hip (9%), knee (8%), and other procedures (9%)
(Fig. 2). Combining the
fracture care provided in the office and the operating room resulted in 11,673
work RVU, or 32% of all work RVU generated by the group in that year.
Most Common Fractures Treated in the Office
The fractures that were most commonly seen in the office during fiscal year
2005, sorted by frequency and production of work RVU, are listed in a table in
the Appendix. No fracture reductions were performed in the office setting. The
most common locations of the fractures were the distal part of the radius
(23%), forearm (14%), tibia (13%), and elbow (10%). In contrast, the four
fracture locations seen in the office in terms of the production of work RVU
were the distal part of the radius (22%), tibia (15%), forearm (12%), and
elbow (11%). Femoral fractures accounted for 5% of the fractures treated in
the office but generated 10% of work RVU generated by fracture care in the
office. Of all fractures treated in the office, 62% occurred in the upper
extremity, while work RVU production for these same fractures equaled 57% of
the total work RVU production for fracture care in the office.
Most Common Operatively Treated Fractures
A table listing the most common trauma-related CPT-coded operations sorted
by frequency and work RVU production, and one with a detailed breakdown of all
such trauma-related CPT-coded operative events, are provided in the Appendix.
Of the surgically treated fractures, the most common locations were the elbow
(25.3%), tibia (12%), femur (9.8%), forearm (5.5%), distal aspect of the
radius (5%), and humerus (2.1%). One hundred and five separate irrigation and
débridement procedures for open injuries accounted for 13.2% of all
trauma-related CPT-coded operative events but for only 8.5% of all work RVU
for trauma-related operations. One hundred and forty-nine procedures to remove
a metal implant accounted for 18.8% of all trauma-related operative events but
for only 8.2% of the work RVU produced for such operations.
Elbow Fractures
Elbow fractures accounted for 201 (25.3%) of all trauma-related CPT-coded
procedures. Supracondylar humeral fractures alone accounted for 141 elbow
fractures (70%), representing 17.8% of all trauma-related CPT-coded procedures
and 22.9% of all work RVU produced for operative procedures for trauma-related
fractures. One hundred and twenty-four (88%) of the 141 supracondylar humeral
fractures were treated with closed reduction and percutaneous pin fixation,
while only ten fractures (7%) required an open reduction.
Tibial Fractures
Ninety-five tibial fractures were treated operatively. Forty-nine (52%) of
them were diaphyseal fractures; the proximal end of the tibia or the knee
joint was involved in eighteen fractures (19%) and the distal end of the tibia
or the ankle involved in twenty-eight fractures (29%). Tibial fractures
accounted for 12% of all trauma-related CPT-coded operative events and 14.2%
of all work RVU produced for operative treatment of trauma-related fractures.
Thirty-one tibial fractures (33%) were treated by closed reduction; forty-one
(43%), by open reduction and plate fixation, mostly for proximal and distal
fractures; nine (9%), by intramedullary nailing; thirteen (14%), by external
fixation; and, one (1%), by percutaneous pinning.
Femoral Fractures
Seventy-eight femoral fractures had surgical treatment. Fifty-nine (76%) of
them were diaphyseal fractures, the neck or subtrochanteric region was
involved in eleven (14%), and the supracondylar or condylar region was
involved in eight (10%). While femoral fractures accounted for 9.8% of all
trauma-related CPT-coded operative events, these fractures accounted for 19.7%
of all work RVU generated for operative treatment. Forty-eight femoral
fractures (62%) were treated with intramedullary nailing, performed
predominantly with flexible nails, and accounted for 837 work RVU or 14% of
all work RVU generated for trauma-related operative treatment. Ten femoral
fractures (13%) were treated with closed reduction; eight (10%), with open
reduction and plate fixation; and three (4%), with external fixation,
accounting for 1.8%, 2.1%, and 0.2%, respectively, of all work RVU generated
for trauma-related operative treatment.
Forearm Fractures
Forty-four forearm fractures accounted for 5.5% of the CPT-coded
trauma-related operative events and 5.5% of the work RVU produced by those
operations. Of the forearm fractures, twenty-four (55%) were treated with open
reduction, with use of either plates or flexible nails, and twenty (45%) were
treated with closed reduction alone. Of the operatively treated forearm
fractures, 82% involved both the radius and the ulna.
Distal Radial Fractures
Forty distal radial fractures accounted for 5% of all trauma-related
CPT-coded operative events and 4.3% of the work RVU produced by those
operations. Of the distal radial fractures, thirty (75%) were treated by
closed reduction and ten (25%) by either open reduction or percutaneous pin
fixation.
Fractures of the Humeral Neck or Shaft
Seventeen humeral neck or shaft fractures accounted for 2.1% of all
CPT-coded trauma-related operative procedures and 2% of all work RVU produced
by such operations. Eleven of these fractures were treated with closed
reduction alone, and only six required open reduction and internal
fixation.
This study documents the large contribution of trauma care to an urban
pediatric orthopaedic practice in both the outpatient and inpatient setting.
Trauma care accounted for roughly one-third of the work RVU produced in both
settings. There was a roughly equal split in total production of work RVU
between the office and the operating room: 18,078 work RVU were generated from
treatment in the office and 18,693 units were generated from treatment in the
operating room. The pediatric fracture operations have become increasingly
similar to the operations used to treat fractures in adults.
In 1997, 70.4% of the pediatric patients seen because of orthopaedic trauma
were treated at general hospitals and only 29.6% were treated at pediatric
hospitals8. As an
increasing number of pediatric trauma patients are being directed to pediatric
facilities, an even larger demand for operative treatment of trauma-related
fractures will be placed on pediatric orthopaedic surgeons. According to a
2005 survey on manpower by the Pediatric Orthopaedic Society of North America,
there were only twenty-five pediatric orthopaedic fellows in training with
plans to practice in North
America12. The
limited number of pediatric orthopaedic surgeons being trained may not be
sufficient to address the quantity of operative procedures needed for
pediatric orthopaedic trauma-related injuries, particularly if children with
fractures continue to be selectively referred to pediatric facilities. The
question remains as to how this potential demand for pediatric fracture care
can best be addressed. Options include the recruitment of greater numbers of
orthopaedic residents into full-time pediatric orthopaedic practice, sharing
of trauma responsibilities at pediatric hospitals by orthopaedic surgeons who
do not have training in pediatrics, redirection of pediatric orthopaedic
trauma to adult trauma hospitals, and increased public education in preventing
childhood injury. Each of these strategies has its inherent strengths and
weaknesses, but a discussion of them is beyond the scope of this report.
The finding that 49% of the total production of work RVU was generated from
outpatient treatment in the office was surprising. This large contribution of
outpatient treatment in the office to the work RVU produced by an urban
pediatric practice is in contrast to the work RVU produced by our partners in
adult orthopaedic practice, where the outpatient treatment component of work
RVU was only 14% for a spine surgeon, 11% for a hip surgeon, 22% for a sports
medicine surgeon, 27% for a hand surgeon, and 10% for a trauma surgeon
(unpublished data derived from the same billing database). The large component
generated by treatment in the office, particularly the care of uncomplicated
common fractures seen in pediatric orthopaedic practice, has implications for
the allocation of pediatric orthopaedic manpower and resources. Depending on
the local circumstances, certain practices may find it more appropriate to
address outpatient pediatric orthopaedic manpower needs with nurse
practitioners, physician assistants, or pediatricians possessing specialized
nonoperative pediatric orthopaedic skills. Additional time, education, and
expense will be required for the physician extender to acquire the pediatric
orthopaedic skill sets needed to provide this service. This additional
education will either need to be incorporated into the current formal
education of physician extenders, possibly similar to the model of physician
fellowship training, or alternatively provided to the physician extenders in
an apprentice model by the pediatric orthopaedic surgeons whose practices will
be expanded to include such individuals. Pediatric orthopaedic leadership and
physician extender specialty boards should develop guidelines to address the
content of educational material to be included in the required skill sets. If
an on-the-job pediatric orthopaedic surgeon apprentice model is adopted for
training, the practice will need to accommodate for a training period before
its new physician extender employees will be able to function in an autonomous
manner. The duration of training for these extenders will depend on the
complexity of office-based treatment that they will be required to handle and
may range from months to as long as one year.
The use of physician extenders to manage the less complex demands of
pediatric orthopaedic treatment in the office setting has implications on
resource allocation for operating-room demands. On the one hand, the
increasing complexity and volume of operations for trauma-related injuries
increases the need for pediatric orthopaedic surgeons to be present in the
operating room; on the other hand, unless this need is substantial, the volume
of elective procedures for non-trauma-related conditions in the pediatric
orthopaedic practice may not be large enough to utilize all of the time that
becomes available to the pediatric orthopaedic surgeon as a result of
off-loading the less complex office-based treatment to physician extenders.
Simply freeing pediatric orthopaedic surgeons from spending their time in the
office setting may not necessarily translate into increased operative
production.
The type and magnitude of the resources and manpower required for a
pediatric orthopaedic practice depends on the specific practice environment.
Factors to be considered in order to implement a correct allocation will vary
but should include the following: (1) the percentage of treatment provided in
the office setting appears to be much higher in a pediatric orthopaedic
practice compared with an adult practice, (2) in an urban setting, a
substantial component of treatment provided in the office is trauma-related
and of a complexity that can be handled by physician extenders, (3) pediatric
orthopaedic surgery in an urban setting has a substantial trauma-related
component frequently requiring technical skills used to treat factures in
adults, and (4) the amount of elective operative procedures performed in the
practice for the treatment of pediatric orthopaedic conditions that are not
related to trauma is an important element. The data in the present study
provide important background information required to answer some of these
work-force issues.
Orthopaedic residents contemplating a career in pediatric orthopaedics
should understand that potentially one-half of their work production may be
performed in an office setting if the data reported here are representative.
This should not necessarily be seen as a negative since a well-managed,
efficient office practice (with appropriate support from a nurse practitioner
or physician assistant) can be a very positive experience. Surgeons with the
appropriate skill set (a good personality and/or enjoyment of the
parent-child-doctor interface) may even prefer this balance to one in which
most of one's time is spent in the operating room. Most of the fractures seen
and treated in the office setting were uncomplicated, although many required
the judgment of a specialized pediatric orthopaedic surgeon when determining
acceptable fracture position. Designated fracture clinics that include ready
access to efficient radiography, including mini c-arm units and efficient
casting systems, may increase the efficiency in managing this high volume of
fracture work in the office. For instance, we instituted evening fracture
clinics during the study year, and they were greatly appreciated by the
parents of school-aged children.
Lastly, the demographic data on pediatric orthopaedic trauma care described
in this study do not apply to all physicians practicing pediatric
orthopaedics. The study practice analyzed in this paper was an urban,
academic, full-time, pediatric orthopaedic practice centered within a busy,
freestanding pediatric hospital with a level-I trauma center. These
demographic data need to be considered when attempting to apply these findings
to other pediatric orthopaedic environments. Alternate pediatric orthopaedic
care settings and pediatric orthopaedic practices not affiliated with a
freestanding level-I trauma center, or private practices where pediatric
orthopaedics does not comprise the bulk of the work, may not follow the
patterns noted in this report.
Tables listing all fractures sorted by treatment site (office or hospital)
and detailed lists of all operative events are available with the electronic
versions of this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM).