Journal of Pediatric Surgery 49 (2014) 123–128

Contents lists available at ScienceDirect

Journal of Pediatric Surgery

journal homepage: www.elsevier.com/locate/jpedsurg

Shifts towards pediatric specialists in the treatment of appendicitis and pyloric

stenosis: Trends and outcomes
Jarod P. McAteer a, b,⁎, Cabrini A. LaRiviere c, Keith T. Oldham d, Adam B. Goldin a, b
a
Pediatric General and Thoracic Surgery, Seattle Children’s Hospital and University of Washington Seattle, WA 98105, USA
b
Department of Surgery, University of Washington School of Medicine, Seattle, WA 98105, USA
c
Department of Surgery, Louisiana State University New Orleans, LA 70112, USA
d
Division of Pediatric Surgery, Children’s Hospital of Wisconsin, Milwaukee, WI 53226, USA

a r t i c l e i n f o a b s t r a c t

Article history: Background: Little data exists on temporal changes in the care of children with common surgical conditions.
Received 20 September 2013 We hypothesized that an increasing proportion of procedures are performed at pediatric hospitals over time,
Accepted 30 September 2013 and that outcomes are superior at these centers.
Methods: We conducted a retrospective cohort study using Washington State discharge records for children
Key words:
0–17 years old undergoing appendectomy (n = 39,472) or pyloromyotomy (n = 3,500). Pediatric hospitals
Outcomes
were defined as centers with full-time pediatric surgeons. Outcomes were examined for two time periods
Specialty
Designation
(1987–2000, 2001–2009).
Regionalization Results: From 1987 to 2009, the proportion of procedures performed at pediatric hospitals steadily increased.
Appendectomy The percentage for appendectomies increased from 17% to 32%, and that for pyloromyotomies increased from
Pyloromyotomy 57% to 99%. For pyloromyotomy, care at a pediatric hospital was associated with decreased risk of
postoperative complications (OR = 0.36, p b 0.001) for both time periods. Appendectomy outcomes did not
differ significantly in the early time period, but in the later time period specialist care was associated with
lower risk of complications in children b5 years (OR = 0.54, p = 0.03).
Conclusion: There has been a shift towards pediatric hospitals for certain procedures, with a widening
disparity in outcomes for younger children. These results suggest that procedures in younger patients may
best be performed by providers familiar with these patient populations.
© 2014 Elsevier Inc. All rights reserved.

Hospital and provider characteristics have been implicated
as key factors that may influence post-operative outcomes in
surgical patients. In adult surgery, operative volume is one such
factor that has been widely studied [1,2]. Although many of the
intricacies of the relationship between provider experience and
outcomes remain undefined, the strength of evidence has led to
changes in policy directed at regionalization of care in adult
surgery, with some indications of improved outcomes following
such measures [3,4].
Recent research efforts have also sought to determine the
influence of surgeon and hospital factors on outcomes in children
with surgical illness. While there is a relatively large body of evidence
supporting a link between provider characteristics (e.g. operative
volume, surgeon specialty training, hospital designation) and out-
comes in pediatric surgery, the quality of these data are highly
variable, and results depend greatly upon the complexity of the
⁎ Corresponding author. Pediatric General and Thoracic Surgery Seattle Children’s
Hospital and University of Washington 4800 Sand Point Way NE Seattle WA 98145.
Tel.: +1 307 259 6262; fax: +1 206 987 3925.
E-mail address: jarodmc@uw.edu (J.P. McAteer).
procedure of interest [5]. In order to simplify comparisons, many
studies in this area have focused on common, low-complexity
procedures (e.g. appendectomy, pyloromyotomy), and many have
shown improved outcomes for children treated at specialty centers
and by fellowship-trained pediatric surgeons [6–12]. Based on such
findings, recommendations have been put forth urging referral of
younger children with surgical needs to pediatric specialists [13]. The
extent to which such changes in practice have occurred, and the
extent to which post-operative outcomes have changed over time,
have not been investigated.
In order to identify the temporal changes in practice with regard to
where children with surgical diseases are treated, and to compare
post-operative outcomes for children treated at pediatric versus non-
pediatric hospitals during two different time periods, we conducted a
population-based retrospective cohort study on children undergoing
non-incidental appendectomy and pyloromyotomy in Washington
State over a 23 year period. We hypothesized that an increasing
proportion of procedures would be performed at hospitals with
pediatric surgeons over time, and that the risk of post-operative
complications at pediatric hospitals relative to non-pediatric hospitals
would change over time.

0022-3468/$ – see front matter © 2014 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jpedsurg.2013.09.046
124 J.P. McAteer et al. / Journal of Pediatric Surgery 49 (2014) 123–128

1. Methods were quantified using multivariate logistic regression models. Post-

1.1. Study design
We performed a population-based retrospective cohort study
using the Washington State Comprehensive Hospital Abstract
Reporting System (CHARS), a statewide inpatient hospital discharge
database that provides de-identified patient data regarding age,
gender, payer status, diagnoses, procedures, length of stay, and
discharge disposition. The study was approved by the University of
Washington Institutional Review Board (IRB #39105).
1.2. Study subjects
Using hospital discharge records from January 1, 1987 to
December 31, 2009, we identified all patients b 18 years of age who
underwent non-incidental appendectomy or pyloromyotomy. Ap-
pendectomy patients were defined as those with a procedure code for
appendectomy (ICD-9 procedure code 47.0, 47.01, 47.09) in the
absence of a code for incidental appendectomy. Patients with benign
GI neoplasms or malignant neoplasms, and those with inflammatory
bowel disease were excluded. Pyloromyotomy patients were defined
as those with both a procedure code for pyloromyotomy (43.3) and a
diagnostic code for infantile pyloric stenosis (750.5). Pyloromyotomy
patients ≥ 1 year of age were excluded (these data do not record age
in months).
1.3. Covariates of interest
We defined “pediatric hospital” as any hospital employing full-
time, fellowship-trained pediatric surgeons who were available 24/7,
as our goal was to have treatment at a pediatric hospital represent
treatment by a pediatric surgeon. Of 5 hospitals that met these
criteria, 3 were freestanding children’s hospitals employing only
pediatric surgeons. Excluding the other 2 hospitals (designated
children’s units in adult hospitals) from the analysis did not
substantively change our results, so all five were included in the
analysis. All other hospitals were considered “non-pediatric hospi-
tals”. One hospital in our study had part-time pediatric surgeon
coverage from 1988–1996, and thus did not meet criteria as a
pediatric hospital. Neither excluding this hospital nor treating it as a
pediatric hospital for the years 1988–1996 substantively changed our
results, so we left it as a non-pediatric hospital in our analysis.
Demographic factors were extracted including gender, age, Medicaid
insurance status, and perforation status (for appendectomy patients).
Chronic comorbid conditions were identified using a set of ICD-9
codes selected a priori, as these variables are commonly considered as
potential confounding factors in pediatric outcomes research (Ap-
pendix A) [14]. Post-operative complications were defined using ICD-
9 codes indicating complications attributed to medical or procedural
treatment (Appendix B). Duodenal perforation in pyloromyotomy
patients was defined as a code for accidental perforation during
procedure (998.2) [12]. Negative appendectomy was defined as
appendectomy in the absence of a diagnosis of appendicitis [15,16].
There were no missing data for the covariates considered in the study.
operative outcomes were analyzed separately for the time periods
1987–2000 and 2001–2009. We chose this time period breakdown
because in 2001 the American Academy of Pediatrics released a
position statement urging referral of certain childhood surgical cases
to pediatric specialists[13]. Outcomes in appendectomy patients were
analyzed in the full cohort and stratified by age group. P b 0.05 was
considered statistically significant. Regression models were adjusted
for clustering at the hospital level using a random-effects model in
order to account for the non-independence of sampling.
Statistical analysis was performed using Stata 12 (College
Station, TX).
2. Results
2.1. Appendectomy
Over the course of the study period, 39,472 patients underwent
non-incidental appendectomy. The majority of patients were older
than 10 years (62.5%), and 26.8% were perforated (Table 1). The
negative appendectomy proportion was 8.9%. Overall, 22.0% of
patients received treatment at a pediatric hospital.
The proportion of appendectomies performed at pediatric hospi-
tals increased over time, from 17% in 1987 to 32% in 2009. The change
was greatest for children younger than 5 years (Fig. 1). In a
multivariate logistic regression model, the odds of receiving care at
a pediatric hospital increased by 4% each successive year (OR = 1.04,
95% CI 1.03–1.04), with the greatest increase noted in patients
younger than 5 years (OR = 1.05, 95% CI 1.04–1.06). Given that such
results could be related to a relative increase in the population near
pediatric relative to non-pediatric hospitals (rather than from a
conscious treatment decision) we used state census data to look at the
change in population in the three counties with pediatric hospitals.
Rather than increasing, the total proportion of the state population in
these counties decreased over the study (50.8% in 1980, 50.4% in 1990,
48.5% in 2000, and 47.6% in 2010, data not shown).
In comparing children treated in 1987–2000 to those treated in
2001–2009, 19.5% of children in the earlier period received care at a
pediatric hospital, versus 25.3% in the later period. The change was
greatest for children b 5 years (37.5% to 51.5%). Perforation was more
common at pediatric hospitals in both time periods, as were comorbid
conditions and post-operative complications (Table 2). The negative
appendectomy rate decreased from the earlier to the later time period
for both groups, but to a greater degree at pediatric hospitals.
After multivariate adjustment, the odds of post-operative complica-
tions from 1987–2000 was similar at both hospitals types across all age
groups (Table 4). Similar results were observed with regard to the odds
of negative appendectomy. In the later time period, however, the odds of
post-operative complications for children b 5 years was significantly
Table 1
Characteristics of appendectomy and pyloromyotomy patients.
Appendectomy Pyloromyotomy
(n = 39,472) (n = 3,500)
n (%) n (%)

Male 22,722 (57.6) 2,885 (82.4)

1.4. Statistical analysis
Age (years)
b5 1,778 (4.5) N/A
Descriptive statistics were used to compare characteristics accord- 5–10 13,012 (33.0) N/A
ing to hospital type for both appendectomy and pyloromyotomy 11–17 24,682 (62.5) N/A
Perforated 10,560 (26.8) N/A
patients. The proportion of all procedures in the state that were
Medicaid 10,079 (25.5) 1,629 (46.5)
performed at a pediatric hospital was determined for each procedure Comorbid conditions 1,664 (4.2) 174 (5.0)
for each year of the study. Multivariate logistic regression was used to Any postop complication 1,713 (4.3) 101 (2.9)
quantify the odds of treatment at a pediatric hospital according to each Negative appendectomy 3,511 (8.9) N/A
successive year in the study. The odds of post-operative outcomes Duodenal perforation N/A 54 (1.5)
Treated at pediatric hospital 8,701 (22.0) 2,383 (68.1)
(complications, negative appendectomy) according to hospital type
 J.P. McAteer et al. / Journal of Pediatric Surgery 49 (2014) 123–128 125

a) b)

Fig. 1. Percentage of a) appendectomies and b) pyloromyotomies performed at pediatric hospitals in Washington State by year, 1987–2009.

lower at pediatric hospitals (OR = 0.54, 95% CI 0.25–0.94), as was the
odds of negative appendectomy for children b5 years (OR = 0.51, 95%
CI 0.19–0.91) and 5–10 years (OR = 0.46, 95% CI 0.29–0.74).
2.2. Pyloromyotomy
Over the course of the study period, 3,500 infants underwent
pyloromyotomy. The majority of patients were male (82.4%) and
nearly half were on Medicaid (Table 1). Comorbid conditions were
present in 5% of patients. Overall, 68.1% of patients were treated at
pediatric hospitals.
The total proportion of pyloromyotomies performed at pediatric
hospitals increased significantly over the study period, from 57% in
1987 to 99% in 2009 (Fig. 1). After adjustment, the odds of receiving
care at a pediatric hospital increased by 10% for each successive year
in the study (OR = 1.10, 95% CI 1.09–1.11).
During the period 1987–2000, 59.3% of pyloromyotomies were
performed at pediatric hospitals, versus 83.9% from 2001–2009. In
both time periods, comorbid conditions were significantly more
common and post-operative complications significantly less common,
at pediatric hospitals (Table 3). Over the entire study period, the
adjusted odds of post-operative complications was significantly lower
at pediatric hospitals (OR = 0.36, 95% CI 0.20–0.64), as was the odds
of duodenal perforation specifically (OR = 0.27, 95% CI 0.13–0.58).
These effects did not differ by time period (Table 4).
3. Discussion
This study represents, to our knowledge, the first analysis of
population-level temporal changes in the practice environment and
outcomes of treatment for children with appendicitis and pyloric
stenosis. Using a large, population-based cohort, we found that an
increasing proportion of appendectomies and pyloromyotomies were
performed at hospitals with pediatric surgeons over time, and that
among appendectomy patients this pattern was most apparent in the
youngest patients. We also found that post-operative outcomes were
improved at pediatric hospitals, and that this effect only became
apparent for appendectomy patients in recent years.
The volume of literature investigating the association between
surgeon and hospital characteristics and outcomes in children has
grown in recent years, with several studies focusing on appendectomy
and pyloromyotomy [5]. Appendectomy studies have analyzed
numerous post-operative outcomes (e.g. complications, misdiagnosis,
length of stay), with most finding a positive association between
measures of provider experience (operative volume, hospital desig-
nation, surgeon specialty) and improved outcomes [6,7,9,15–17].
Other studies, however, have found more mixed results and a less
clear association [8,18,19]. Studies focusing on pyloromyotomy have
generally been more consistent than appendectomy studies in
documenting positive results for the association between surgeon
experience and outcomes [10–12,20,21]. A recent systematic review
concluded that while pyloromyotomy patients might benefit from
more centralized care, current evidence does not support differential
outcomes for appendectomy, though no studies analyzed outcomes
by age group or according to different time periods [22].
While studies in adults have shown that centralization of care has
occurred for some procedures over time, no such data exist in children
[23]. Our results show that such changes have occurred, most notably
for younger children. These changes may be due to a number of
underlying trends. It is possible that, over time, there has been a
decrease in the willingness of adult general surgeons to operate on
younger children, thus leading to increased rates of referral to

Table 2
Appendectomy patient characteristics according to hospital type (by time period).

1987–2000 2001–2009

Pediatric Non-pediatric P value Pediatric Non-pediatric P value

(n = 4,533) (n = 18,708) (n = 4,168) (n = 12,063)

Male 2,563 (56.5) 10,705 (57.2) 0.41 2,394 (57.4) 7,060 (58.5) 0.22
Age (years)
b5 385 (8.5) 641 (3.4) b0.001 387 (9.3) 365 (3.0) b0.001
5 to 10 1,878 (41.4) 5,778 (30.9) 1,741 (41.8) 3,615 (30.0)
11 to 17 2,270 (50.1) 12,289 (65.7) 2,040 (48.9) 8,083 (67.0)
Perforated 1,385 (30.6) 4,842 (25.9) b0.001 1,364 (32.7) 2,969 (24.6) b0.001
Medicaid 867 (19.1) 3,875 (20.7) 0.02 1,468 (35.2) 3,869 (32.1) b0.001
Comorbid conditions 242 (5.3) 425 (2.3) b0.001 324 (7.8) 673 (5.6) b0.001
Any postop complication 250 (5.5) 829 (4.4) 0.002 197 (4.7) 437 (3.6) 0.002
Negative appendectomy 493 (10.9) 2,109 (11.3) 0.45 155 (3.7) 753 (6.3) b0.001
126 J.P. McAteer et al. / Journal of Pediatric Surgery 49 (2014) 123–128

Table 3
Pyloromyotomy patient characteristics according to hospital type (by time period).

1987–2000 2001–2009

Pediatric Non-Pediatric P value Pediatric Non-Pediatric P value

(n = 1,332) (n = 915) (n = 1,051) (n = 202)

Male 1,095 (82.2) 754 (82.4) 0.9 865 (82.3) 171 (84.7) 0.42
Medicaid 483 (36.3) 409 (44.7) b0.001 614 (58.4) 123 (60.9) 0.51
Comorbid conditions 78 (5.9) 17 (1.9) b0.001 76 (7.2) 3 (1.5) 0.002
Any postop complication 25 (1.9) 48 (5.3) b0.001 19 (1.8) 9 (4.5) 0.02
Duodenal Perforation 10 (0.8) 26 (2.8) b0.001 9 (0.9) 9 (4.5) b0.001

pediatric surgeons. Alternatively, as an increased focus has been
placed on safety of anesthesia care in children, adult anesthesiologists
may be less willing to offer general anesthesia to infants and young
children, thus prompting transfer to centers with pediatric anesthesia
capabilities [24]. The patterns may also reflect increasing parental
preference for specialty care.
Whatever the mechanism for the observed change in practice, our
results indicate that post-operative outcomes are generally improved
at pediatric hospitals. For pyloromyotomy, the risk of post-operative
complications was 65% less for children treated at a pediatric hospital
versus a non-pediatric hospital. In our study, essentially all patients at
pediatric hospitals would have been treated by pediatric surgeons. If
surgeon training is indeed the key exposure captured by our hospital
definition, then improved outcomes at pediatric centers may reflect
greater comfort and familiarity on the part of pediatric surgeons in
treating infants with pyloric stenosis. The fact that the incidence of
duodenal perforation, a purely technical complication, is so much
lower at pediatric hospitals suggests that surgeon-level characteristics
may be the key factors underlying our observations.
Appendectomy is an even more intriguing procedure to consider,
because it is frequently performed by both pediatric and adult general
surgeons. An important consideration in such studies is the analysis of
outcomes stratified according to age group, as the association
between surgeon training and outcomes may be modified by the
age of the patient. Other studies have taken this into consideration
with regard to length of stay and hospital charges, but not with regard
to post-operative complications [6,19]. Our results indicate that as
fewer appendectomies were performed at non-pediatric hospitals
over time, the outcomes at pediatric relative to non-pediatric
hospitals became better for the youngest children. It is also notable
that while negative appendectomy was similarly common at both
hospital types in the early time period (1987–2000), it was
significantly less common at pediatric relative to non-pediatric
hospitals in the later time period (2001–2009). These findings may
be due to a decrease in experience over time for adult surgeons
operating on younger children with appendicitis (and thus a
divergence in outcomes as the number of younger children they
operate on decreases), or may reflect greater standardization of care
or adoption of clinical treatment pathways at pediatric hospitals
relative to non-pediatric hospitals. One might presume that treatment
of older children with appendicitis is similar to treatment of adults,
thus explaining the similar outcomes for both hospital types in this
age group.
This study has several limitations. These are observational data,
and associations cannot be assumed to be causal in nature.
Additionally, the CHARS database, similar to other administrative
data sources, is subject to potential misclassification, which we have
attempted to minimize by employing stringent inclusion/exclusion
criteria for our populations. Given the generally healthy nature of
children with appendicitis and pyloric stenosis, it can be difficult to
adequately risk stratify these patients. We utilized several measures
of increased risk, including perforation status and chronic comorbid-
ities, but some residual confounding due to differences in disease
severity across hospitals may be present. Lastly, although surgeon
training may be the key factor responsible for our results, other
features of non-pediatric hospitals, including lack of multidisciplinary
specialists, fewer ancillary services, rural location and more delayed
presentation for treated patients, may be important drivers of these
findings as well.
Our findings clearly show that the treatment of certain surgical
conditions has shifted to more specialized environments. There is
good evidence to suggest that optimal outcomes after pyloromyotomy
can be achieved in the hands of pediatric surgeons. Similarly,
appendicitis in children b 5 years of age may warrant treatment by
pediatric specialists, especially as trends in care have shifted. For
appendicitis especially, more data are needed to determine what
aspects of surgical care (e.g. surgeon experience, clinical care
pathways) are most important in driving improved outcomes at
pediatric hospitals. This will afford us a greater understanding of these
associations and may facilitate the development of best-practice
standards for certain common surgical conditions in children.
Adoption of such standards may help to further improve outcomes
at non-pediatric hospitals in cases in which patient transfer might not
be feasible.

Table 4
Adjusted odds of complications at pediatric vs. non-pediatric hospitals for appendectomy and pyloromyotomy patients, broken down by time period (Models adjusted for all
covariates, age treated as a continuous variable in appendectomy models).

1987–2000 2001–2009

Appendectomy OR 95% C.I. P value OR 95% C.I. P value

Any Complication
All ages 1.02 0.65–1.61 0.93 1.05 0.69–1.60 0.82
b5 years 1.11 0.47–2.58 0.82 0.54 0.25–0.94 0.03
5–10 years 1.03 0.57–1.85 0.93 1.09 0.70–1.69 0.70
11–17 years 1.12 0.69–1.80 0.65 1.15 0.71–1.88 0.57
Negative Appendectomy
All ages 1.04 0.78–1.38 0.78 0.70 0.42–1.18 0.18
b5 years 1.20 0.79–1.83 0.40 0.51 0.19–0.91 0.02
5–10 years 0.88 0.61–1.27 0.48 0.46 0.29–0.74 0.001
11–17 years 1.10 0.80–1.51 0.57 0.90 0.50–1.59 0.71
Pyloromyotomy
Any complication 0.36 0.18–0.72 0.004 0.36 0.16–0.82 0.02
Duodenal perforation 0.26 0.10–0.71 0.008 0.20 0.08–0.50 0.001
 J.P. McAteer et al. / Journal of Pediatric Surgery 49 (2014) 123–128 127

Appendix A [4] MacKenzie EJ, Rivara FP, Jurkovich GJ, et al. A national evaluation of the effect of
trauma-center care on mortality. N Engl J Med 2006;354:366–78.
ICD-9 codes for chronic medical conditions of childhood. [5] McAteer JP, LaRiviere CA, Drugas GT, et al. Influence of surgeon experience,
hospital volume, and specialty designation on outcomes in pediatric surgery: a
Diagnosis ICD-9 code(s) systematic review. JAMA Pediatr 2013:E1–8.
[6] Kokoska ER, Minkes RK, Silen ML, et al. Effect of pediatric surgical practice on the
Asthma 493
treatment of children with appendicitis. Pediatrics 2001;107:1298–301.
Diabetes mellitus 250.0–250.9 [7] Emil SG, Taylor MB. Appendicitis in children treated by pediatric versus general
Neonatal 775.1 surgeons. J Am Coll Surg 2007;204:34–9.
Brain and spinal cord malformations 740.0–742.9 [8] Somme S, To T, Langer JC. Effect of subspecialty training on outcome after pediatric
Mental retardation 318.0–318.2 appendectomy. J Pediatr Surg 2007;42:221–6.
Central nervous system degeneration and disease 330.0–330.9 [9] Collins HL, Almond SL, Thompson B, et al. Comparison of childhood appendicitis
334.0–334.2 management in the regional paediatric surgery unit and the district general
335.0–335.9 hospital. J Pediatr Surg 2010;45:300–2.
Infantile cerebral palsy 343.0–343.9 [10] Brain AJ, Roberts DS. Who should treat pyloric stenosis: the general or specialist
Muscular dystrophies and myopathies 359.0–359.3 pediatric surgeon? J Pediatr Surg 1996;31:1535–7.
Heart and great vessel malformations 745.0–747.4 [11] Pranikoff T, Campbell BT, Travis J, et al. Differences in outcome with subspecialty
care: pyloromyotomy in North Carolina. J Pediatr Surg 2002;37:352–6.
Cardiomyopathies 425.0–425.4
[12] Langer JC, To T. Does pediatric surgical specialty training affect outcome after
429.1
Ramstedt pyloromyotomy? A population-based study. Pediatrics 2004;113:
Conduction disorders 426.0–427.4
1342–7.
Dysrhythmias 427.6–427.9 [13] Pediatrics SAPAAo. Guidelines for referral to pediatric surgical specialists.
Respiratory malformations 748.0–748.9 Pediatrics 2002;110:187–91.
Chronic respiratory disease 770.7 [14] Feudtner C, Christakis DA, Connell FA. Pediatric deaths attributable to complex
Cystic fibrosis 277 chronic conditions: a population-based study of Washington State, 1980–1997.
renal congenital anomalies 753.0–753.9 Pediatrics 2000;106:205–9.
Chronic renal failure 585 [15] Smink DS, Finkelstein JA, Kleinman K, et al. The effect of hospital volume of
Gastrointestinal congenital anomalies 750.3 pediatric appendectomies on the misdiagnosis of appendicitis in children.
751.6–751.9 Pediatrics 2004;113:18–23.
Chronic liver disease and cirrhosis 571.4–571.9 [16] Ponsky TA, Huang ZJ, Kittle K, et al. Hospital- and patient-level characteristics and
Inflammatory bowel disease 555.0–556.9 the risk of appendiceal rupture and negative appendectomy in children. JAMA
Sickle Cell disease 282.5–282.6 2004;292:1977–82.
[17] Alexander F, Magnuson D, DiFiore J, et al. Specialty versus generalist care of
Hereditary anemia 282.0–282.4
children with appendicitis: an outcome comparison. J Pediatr Surg 2001;36:
Hereditary immunodeficiency 279.00–279.9
1510–3.
288.1–288.2 [18] Cosper GH, Hamann MS, Stiles A, et al. Hospital characteristics affect outcomes for
446.1 common pediatric surgical conditions. Am Surg 2006;72:739–45.
Acquired immunodeficiency 0420–0421 [19] Whisker L, Luke D, Hendrickse C, et al. Appendicitis in children: a comparative
Amino acid metabolism 270.0–270.9 study between a specialist paediatric centre and a district general hospital. J
Carbohydrate metabolism 271.0–271.9 Pediatr Surg 2009;44:362–7.
Lipid metabolism 272.0–272.9 [20] Ly DP, Liao JG, Burd RS. Effect of surgeon and hospital characteristics on outcome
Storage disorders 277.3, 277.5 after pyloromyotomy. Arch Surg 2005;140:1191–7.
Other metabolic disorders 275.0–275.3, 277.2, 277.4, [21] Safford SD, Pietrobon R, Safford KM, et al. A study of 11,003 patients with
277.6, 277.8–277.9 hypertrophic pyloric stenosis and the association between surgeon and hospital
Chromosomal anomalies 758.0–758.9 volume and outcomes. J Pediatr Surg 2005;40:967–72 [discussion 972-963].
Bone and joint anomalies 2 59.4, 737.3, 756.0–756.5 [22] Evans C, van Woerden HC. The effect of surgical training and hospital
characteristics on patient outcomes after pediatric surgery: a systematic review.
Diaphragm and abdominal wall 553.3, 756.6–756.7
J Pediatr Surg 2011;46:2119–27.
Other congenital anomalies 759.7–759.9
[23] Finks JF, Osborne NH, Birkmeyer JD. Trends in hospital volume and operative
mortality for high-risk surgery. N Engl J Med 2011;364:2128–37.
[24] Hackel A, Badgwell JM, Binding RR, et al. Guidelines for the pediatric perioperative
anesthesia environment. American Academy of Pediatrics. Section on Anesthesi-
ology. Pediatrics 1999;103:512–5.

Appendix B Discussion
ICD-9 codes for post-operative complications.

Diagnosis ICD-9 code(s) Dr Oliver Muensterer (New York, NY) (Discussant): I think this is a really
Cellulitis/abscess 682.2, 682.9, 683
important paper particularly for those that go from one hospital to
Complications secondary to procedure 996 the another to operate in outside hospitals. It would be interesting
Complications of specific body systems 997.1–997.9 to see if it’s the actual hospital, the environment, or the physician
Post-operative shock 998.0 operating, the pediatric surgeon versus an adult surgeon. My
Hemorrhage or hematoma 998.1–998.13
feeling is that it is not only us going somewhere else but it’s the
Accidental puncture/laceration 998.2
Wound disruption 998.3–998.33 residents you work with, the hospital environment, and the team
Retained foreign body 998.4, 998.7 that you are working with. I’m very much for centralization of
Post-operative fistula 998.6 these specialized procedures.
Subcutaneous emphysema 998.81
Non-healing surgical wound 998.83
Other/unspecified complications 998.89, 998.9
Dr Jarod Mcateer (Response): Thank you for your comment. Absolutely,
it is very difficult in any study that looks at physician character-
istics to separate out the characteristics of the physician from the
hospital or center at which they operate and so it is important for
References us to try to go forward and try to define the resources that are most
important and optimizing those outcomes.
[1] Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical
mortality in the United States. N Engl J Med 2002;346:1128–37.
[2] Birkmeyer JD, Stukel TA, Siewers AE, et al. Surgeon volume and operative Dr Dennis Lund (Phoenix, AZ) (Discussant): I also rise to applaud this
mortality in the United States. N Engl J Med 2003;349:2117–27. study. I think it’s very important and I think it’s important for
[3] Birkmeyer JD, Finlayson EV, Birkmeyer CM. Volume standards for high-risk
surgical procedures: potential benefits of the Leapfrog initiative. Surgery another reason. One of the things that we have to understand is the
2001;130:415–22. changing demographic of the healthcare spend in the United
128 J.P. McAteer et al. / Journal of Pediatric Surgery 49 (2014) 123–128
States. Children are becoming a smaller and smaller portion of
health care spending in the United States as the population ages,
and we’re now down to about nine percent of the total health care
spending in the country, so it really behooves children’s hospital to
begin to show the value added for why they exist. I think one of the
things that we as pediatric surgeons can do is drive the bus on this.
We can disagree about what kind of resources we ought to have in
a children’s surgical center but at the end of the day if pediatric
surgeons are the ones who are setting the bar here, I think we’ll
ultimately continue to do the right thing for children. I applaud
your paper.
Dr Jarod Mcateer (Response): Thank you for your comments. I
appreciate it.
Dr Helene Flageole (Hamilton, ON) (Discussant): We certainly have
seen a very similar phenomenon in Canada. I really enjoyed your
paper. I was just wondering in terms of the negative appendectomy
rates being higher in non-children’s hospitals if those hospitals
have the same access to imaging that the children’s hospital had.
From our perspective in Ontario we will not infrequently get a
referral of a child and the reason for the transfer is the inability of
the community hospital to perform imaging after 5 p.m.
Dr Jarod Mcateer (Response): Yes, thank you for the question.
Differences in imaging availability are an issue in Washington
State. When we look at rural community hospitals especially
compared to our experience at Seattle Children’s, we have a very
well integrated radiology program for diagnosis of appendicitis by
ultrasound. Many of these children come from outside hospitals
without imaging or many of them have gotten CT scans, some of
which are not sent to us, and so in general our ability to
radiographically diagnose appendicitis at the children’s hospital
is significantly more advanced than a lot of the rural hospitals from
which these children are referred.
Dr Charles Stolar (Santa Barbara, Ca) (Discussant): Do you think it’s
possible that the non-children’s hospitals have limited abilities to
do a history and physical as opposed to imaging?
Dr Jarod Mcateer (Response): It would be difficult for me to comment
on that.
Dr Michael Skinner (Dallas, Tx) (Discussant): Very nice study. I thought
it was a rather unorthodox definition of a children’s hospital. If I
were to move to George, Washington, and work in the community
hospital there, would that then become a children’s hospital?
Dr Jarod Mcateer (Response): That’s an interesting question. The way
we defined it – certainly there were hospitals in our data that
had both pediatric surgeons and adult surgeons working at them.
Of the five pediatric hospitals in our data, three were free-
standing children’s hospitals that had only pediatric surgeons
working at them over the entire study period. Two, however,
were children’s units within adult hospitals, and to try and
separate that out, we excluded those two hospitals that had both
types of surgeons working there. These are the results we
achieved with that, with exclusion of that to try and get
specifically at surgeon training, but, yes, you are correct. If you
were at that hospital full-time and available 24/7, if there is
always a pediatric surgeon available, then that would have been
coded as a pediatric hospital in our data.
Dr Michael Skinner: So it is a little misleading for us to generalize your
conclusions to children’s hospitals. Really it’s about pediatric
surgical care. Is that not true?
Dr Jarod Mcateer (Response): Correct, absolutely. Maybe I did not make
that clear. That is the exposure we were trying to capture in terms
of how we defined our exposure based on the training of the
surgeons who were available on-call at that specific hospital.