ORIGINAL ARTICLE: GASTROENTEROLOGY

Healthcare Utilization and Spending for Constipation

in Children With Versus Without Complex
Chronic Conditions

John R. Stephens, Michael J. Steiner, Neal DeJong, yJonathan Rodean, yMatt Hall,
y
Troy Richardson, and zJay G. Berry

ABSTRACT
Downloaded from https://journals.lww.com/jpgn by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3Gamkn0m7hy6Mhb8DYgpUeqLgT2Ue6dY4JJBnXyUM1N4= on 08/26/2018

Objectives: The aim of the study was to examine the prevalence of diagnosis
and treatment for constipation among children receiving Medicaid and to
compare healthcare utilization and spending for constipation among children
based on number of complex chronic conditions (CCCs).
Methods: Retrospective cohort study of 4.9 million children ages 1 to 17
years enrolled in Medicaid from 2009 to 2011 in 10 states in the Truven
Marketscan Database. Constipation was identified using International
Classification of Disease, 9th revision codes for constipation (564.0x),
intestinal impaction (560.3x), or encopresis (307.7). Outpatient and
inpatient utilization and spending for constipation were assessed. CCC
status was identified using validated methodology.
Results: A total of 267,188 children (5.4%) were diagnosed with
constipation. Total constipation spending was $79.5 million. Outpatient
constipation spending was $66.8 million (84.1%) during 406,814 visits,
mean spending $120/visit. Among children with constipation, 1363 (0.5%)
received inpatient treatment, accounting for $12.2 million (15.4%) of
constipation spending, mean spending $7815/hospitalization. Of children
hospitalized for constipation, 552 (40.5%) did not have an outpatient visit for
constipation before admission. Approximately 6.8% of children in the study
had 1 CCC; these children accounted for 33.5% of total constipation
spending, 70.3% of inpatient constipation spending, and 19.8% of
emergency department constipation spending. Constipation prevalence
was 11.0% for children with 1 CCC, 16.6% with 2 CCCs, and 27.1%
with 3 CCCs.
Conclusions: Although the majority of pediatric constipation treatment
occurs in the outpatient setting, inpatient care accounts for a sizable
percentage of spending. Children with CCCs have a higher prevalence of
constipation and account for a disproportionate amount of constipation
healthcare utilization and spending.
Key Words: complex chronic conditions, constipation, medical
complexity, pediatrics, utilization
(JPGN 2017;64: 31–36)
What Is Known
 Constipation is a common pediatric condition, affect-
ing children with and without complex chronic
conditions.
 Children with complex chronic conditions have
increased overall healthcare utilization when com-
pared to children without complex chronic con-
ditions.
What Is New
 Constipation prevalence in children ages 1 to 17
years receiving Medicaid was 5.4%.
 Although the large majority of constipation visits
occurred in the outpatient setting, inpatient consti-
pation treatment accounted for a relatively large
percentage of spending.
 Children with complex chronic conditions accounted
for disproportionate amounts of healthcare utiliz-
ation and spending for constipation.
C onstipation is a common pediatric condition world-wide (1).
It is a frequent reason for outpatient visits (2–4), ranking
second among ambulatory digestive disease diagnoses in the United
States (5). Although little is known about the direct healthcare costs
attributable to constipation in children, a case-control study of a
birth cohort of children ages 5 to 18 years found significantly higher
mean outpatient costs in children with versus those without con-
stipation (6). Using a household-reported, nationally representative
survey, Liem et al (7) estimated total annual healthcare costs of
$3430 per child with constipation and $3.9 billion overall.

Received November 23, 2015; accepted March 22, 2016.
From the Division of General Pediatrics and Adolescent Medicine,
University of North Carolina, School of Medicine, Chapel Hill, NC, the
yChildren’s Hospital Association, Overland Park, KS, and the
zDepartment of Medicine, Division of General Pediatrics, Boston Chil-
dren’s Hospital, Harvard Medical School, Boston, MA.
Address correspondence and reprint requests to John R. Stephens, MD,
Division of General Medicine and Clinical Epidemiology, UNC Hospi-
tals, 5034 Old Clinic Bldg, CB 7110, Chapel Hill, NC 27599-7110
(e-mail: stephenj@med.unc.edu).
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text, and links to the digital files are provided in the
HTML text of this article on the journal’s Web site (www.jpgn.org).
Dr Berry was supported by the Agency for Healthcare Research and Quality
(R21HS023092). There were no other sources of funding for this work.
The authors report no conflicts of interest.
Copyright # 2016 by European Society for Pediatric Gastroenterology,
Hepatology, and Nutrition and North American Society for Pediatric
Gastroenterology, Hepatology, and Nutrition
DOI: 10.1097/MPG.0000000000001210

JPGN  Volume 64, Number 1, January 2017 31

Copyright © ESPGHAN and NASPGHAN. All rights reserved.

Stephens et al
The prevalence of constipation and spending on related health
services may be different in children with versus without chronic
health problems. Children with complex chronic conditions (CCCs)
may be at particularly increased risk for constipation due to a number
of factors, including side effects from chronic medications; alterations
in physiology, such as neuromuscular weakness and gastrointestinal
dysmotility; and increased likelihood of sedentary behavior. Children
with CCCs have high rates of overall healthcare utilization, and they
account for a large proportion of total healthcare spending (8–11).
One could thus hypothesize that children with CCCs would incur
higher healthcare expenditures due to constipation.
In the present study, we describe the overall prevalence of
constipation diagnosis and treatment along with associated healthcare
utilization and spending in a group of children receiving Medicaid.
We also describe and compare constipation-related healthcare util-
ization and spending among children with versus without CCCs.
METHODS
Study Design and Setting
We conducted a retrospective cohort study with the Truven
MarketScan Medicaid claims database for the years 2009 to 2011.
MarketScan contains enrollment information, retail pharmacy,
inpatient, and outpatient claims data for 4.9 million Medicaid
enrollees ages 0 to 22, from 10 deidentified states during the period
studied. The states are geographically distributed and have demo-
graphic characteristics consistent with the national profile of chil-
dren with Medicaid (Truven correspondence).
Study Population
We analyzed all visits for constipation for children ages 1 to
17 within the study cohort. The upper age of 17 years was chosen as
Medicaid enrollment criteria change at age 18 and we wanted to
have complete data for all patients in the cohort during the study. A
diagnosis of constipation was identified using International Classi-
fication of Disease, 9th revision (ICD-9) diagnosis codes for
constipation, (564.0x), intestinal impaction (560.3x), or encopresis
(307.7). The codes were selected to be as specific as possible for
constipation, similar to methodology employed by Choung et al (6).
Demographic and Clinical Characteristics of
the Study Population
Patient age was assigned at the first enrollment in the study
period. Age at first diagnosis of constipation was also calculated for
descriptive purposes.
We examined all visits for constipation by children with CCC
as a subset of the study cohort. Children with CCCs were identified
using version 2 of the ICD-9 methodology developed by Feudtner
et al (12). CCCs are ‘‘any medical condition that can be reasonably
expected to last at least 12 months (unless death intervenes) and to
involve either several different organ systems or 1 organ system
severely enough to require specialty pediatric care and probably some
period of hospitalization in a tertiary care center (13).’’ The meth-
odology uses specific ICD-9 codes to identify CCCs, divided into 10
categories (cardiovascular, respiratory, neuromuscular, renal, gastro-
intestinal, hematologic or immunologic, metabolic, other congenital
or genetic, malignancy, and premature/neonatal) along with domains
of complexity from dependence on medical technology and having
received bone marrow or organ transplantation. We examined all
claims with ICD-9 codes during the 3-year study period to determine
the presence of CCCs within the cohort. Chronic constipation alone
would not meet the above CCC definition.
32
Copyright © ESPGHAN and NASPGHAN. All rights reserved.
JPGN  Volume 64, Number 1, January 2017
Outcomes
The main outcome measures were the number of children in
Medicaid identified with diagnosis and treatment for constipation
and the associated healthcare spending for constipation. Outpatient
treatment was defined as a visit with one of the selected ICD-9
codes for constipation and a prescription for a therapeutic laxative
filled within 2 days of the visit. The 2-day time period was chosen to
closely tie the prescription to a specific constipation visit. Prescrip-
tions and related spending were identified in the retail pharmacy
data within the therapeutic category ‘‘gastrointestinal agent’’ and
therapeutic group ‘‘laxatives’’ (eg, docusate, senna, bisacodyl).
Outpatient constipation visits were subcategorized into emer-
gency department (ED) and clinic (non-ED) visits. Inpatients
treated for constipation were identified by hospital visits with
any of the study constipation ICD-9 codes as a primary diagnosis
or as a secondary diagnosis code, with another nonspecific abdomi-
nal diagnosis code (eg, abdominal pain) as a primary code.
(Appendix 1, Supplemental Digital Content, http://links.lww.com/
MPG/A660).
We also assessed the outpatient use of abdominal radiographs
for constipation-related visits. Abdominal radiology use was ident-
ified via search of diagnosis and procedural codes linked to
identified outpatient encounters, using ICD-9 codes for ‘‘other x-
ray of the urinary system/kidneys, ureters, and bladder x-ray’’
(87.79, commonly termed ‘‘kidneys, ureters, and bladder’’) and
‘‘x-ray of the abdomen’’ (88.19) along with current procedural
technology codes for single, multiple, and complete abdominal
series (74000, 74010, and 74020, respectively).
The outcome measures were compared for children with
versus without CCCs.
Statistical Analysis
Characteristics of patients with and without an outpatient
constipation diagnosis were compared using x2 tests for categorical
variables and Wilcoxon rank-sum tests for continuous variables that
were not normally distributed. The outcome measures were treated
as dependent variables in logistic regression models to assess the
relation of age, sex, and the number of CCCs with constipation-
related outcomes. All analyses were performed with SAS 9.3 (SAS
Institute, Cary, NC). P values <0.05 were considered statistically
significant.
RESULTS
Study Cohort
There were 267,188 (5.4%) children in Medicaid diagnosed
with constipation (Table 1). The median age of children with
constipation was 5 years (interquartile range [IQR]: 2–9), and
54.6% were girls. The median age at diagnosis was 7 years
(IQR: 2–12). Constipation prevalence decreased with age, with
children ages 1 to 2 years most frequently diagnosed with consti-
pation (7.5%) and children ages 13 to 17 years least frequently
diagnosed (2.9%, P < .001). Among race/ethnic groups, non-His-
panic white children had the highest prevalence of constipation
(5.8%), followed by non-Hispanic black and Hispanic children
(5.4% and 5.0%, respectively; P < .001).
Prescription Treatment of Constipation
Of the children diagnosed with constipation, 12,927 (4.8%)
filled a prescription for a therapeutic laxative within 2 days of an
outpatient visit. Docusate was the most frequently prescribed
www.jpgn.org
JPGN  Volume 64, Number 1, January 2017 Healthcare Utilization and Spending for Constipation in Children

TABLE 1. Demographics and number of complex chronic conditions for children receiving Medicaid, comparing those with versus without
constipation

All children receiving Children without Children with

Characteristic Medicaid (column %) constipation (column %) constipation (column %)

Patients 4,929,967 4,668,779 (94.6) 267,188 (5.4)

Median age (Q1, Q3) 7 (3, 12) 7 (3, 12) 5 (2, 9)
Age groups
1–2 y 1,188,268 (24.1) 1,099,467 (23.6) 88,801 (33.2)
3–5 y 922,179 (18.7) 865,715 (18.6) 56,464 (21.1)
6–12 y 1,770,379 (35.9) 1,678,500 (36.0) 91,879 (34.4)
13–17 y 1,049,141 (21.3) 1,019,097 (21.9) 30,044 (11.2)
Sex
Female 2,412,716 (48.9) 2,266,846 (48.6) 145,870 (54.6)
Male 2,517,251 (51.1) 2,395,933 (51.4) 121,318 (45.4)
Race/ethnicity
Non-Hispanic black 1,422,087 (28.8) 1,345,501 (28.9) 76,586 (28.7)
Non-Hispanic white 2,289,343 (46.4) 2,157,089 (46.3) 132,254 (49.5)
Hispanic 514,588 (10.4) 488,694 (10.5) 25,894 (9.7)
Other 703,949 (14.3) 671,495 (14.4) 32,454 (12.1)
Number of complex chronic conditions
No CCCs 4,593,867 (93.2) 4,369,780 (93.7) 224,087 (83.9)
1 CCC 270,200 (5.5) 240,527 (5.2) 29,673 (11.1)
2 CCC 42,119 (0.9) 35,141 (0.8) 6978 (2.6)
3 or more CCC 23,781 (0.5) 17,331 (0.4) 6450 (2.4)
Complex chronic conditions
Other congenital or genetic defect 80,087 (1.6) 69,378 (1.5) 10,709 (4.0)
Cardiovascular 71,919 (1.5) 62,814 (1.3) 9105 (3.4)
Neurologic and neuromuscular 70,506 (1.4) 58,300 (1.3) 12,206 (4.6)
Metabolic 66,695 (1.4) 5,7517 (1.2) 9178 (3.4)
Malignancy 39,723 (0.8) 35,390 (0.8) 4333 (1.6)
Hematologic or immunologic 32,329 (0.7) 27,944 (0.6) 4385 (1.6)
Technology dependence 24,087 (0.5) 18,458 (0.4) 5629 (2.1)
Gastrointestinal 20,332 (0.4) 14,191 (0.3) 6141 (2.3)
Renal and urologic 17,068 (0.3) 13,390 (0.3) 3678 (1.4)
Respiratory 12,894 (0.3) 10,470 (0.2) 2424 (0.9)
Premature and neonatal 8421 (0.2) 6606 (0.1) 1815 (0.7)
Transplant 2501 (0.1) 2216 (0.0) 285 (0.1)

All P values comparing characteristics between children with and without constipation were <0.001.
CCC ¼ complex chronic condition.

laxative (56.5%). The majority of children (88.6%) had a prescrip-
tion filled for only 1 type of laxative, whereas 9.5% received
prescriptions for 2 and 1.8% of children received prescriptions
for 3 or more laxatives.
Healthcare Utilization and Spending for
Constipation
Total spending on constipation-related care during the study
period was $79.4 million (Table 2). Spending on constipation
represented 0.4% of total healthcare spending for all children in
the database during the 3 years studied.
Outpatient constipation spending, including both outpatient
clinic and ED utilization, was $66.8 million, and accounted for the
majority (84.1%) of constipation spending. There were 406,814
outpatient clinic visits for constipation among the 267,188 patients
who were diagnosed, with mean spending per visit of $120, median
$75 (IQR: 54–133). There were 13,878 outpatient clinic visits for
constipation treatment (diagnosis plus a filled prescription for thera-
peutic laxative) with mean spending per visit of $117, median $69
(IQR: 54–109). ED spending was $17.8 million, which represents
22.4% of all constipation spending and 36.4% of all outpatient
constipation spending. There were 71,935 ED visits for constipation,
with mean spending per visit of $248, median $167 (IQR: 81–279).
Inpatient constipation spending was $12.2 million, represent-
ing 15.4% of constipation spending. Of the 267,188 patients
diagnosed with constipation, there were 1363 patients (0.5%) with
1562 inpatient encounters, with mean spending per encounter of
$7815, median $5167 (IQR: 2756–8531), and median length of stay
of 2 days (IQR: 1–3). The majority of inpatient encounters, 1399
(89.6%) had one of the study ICD-9 codes as the primary diagnosis,
with 163 (10.4%) having a study code as a secondary diagnosis
(Appendix 2, Supplemental Digital Content, http://links.lww.com/
MPG/A661). For patients with a study ICD-9 code as the primary
diagnosis, constipation codes (564.0x) were most common, 81.8%,
followed by intestinal impaction codes (560.3x), 17.9%. Among the
hospitalized patients, 552 (40.5%) did not have a previous out-
patient visit for constipation in the study period. The hospitalized
patients without a prior outpatient visit were less likely than those
with a prior visit to have a study ICD-9 code as the primary
diagnosis code (86.1% vs 92.4%, P < 0.001).

www.jpgn.org 33

Copyright © ESPGHAN and NASPGHAN. All rights reserved.

Stephens et al JPGN  Volume 64, Number 1, January 2017

TABLE 2. Healthcare spending on constipation for children in Medicaid by the presence or absence of a complex chronic condition

Complex chronic condition

All children receiving Medicaid Present (row %) Absent (row %)

Health service N ¼ 4,789,967 N ¼ 336,100 (6.8%) N ¼ 4,453,867 (93.2%)

Total constipation spending $79,452,475 $26,600,062 (33.5) $52,852,413 (66.5)

Outpatient $48,986,526 $14,278,791 (29.1) $34,707,735 (70.9)
Emergency department $17,815,115 $3,520,277 (19.8) $14,294,838 (80.2)
Inpatient $12,207,633 $8,582,300 (70.3) $3,625,333 (29.7)
Pharmacy $443,200 $218,693 (49.3) $224,506 (50.7)

Complex chronic condition defined according to Feudtner methodology, ‘‘any medical condition that can be reasonably expected to last at least 12 months
(unless death intervenes) and to involve either several different organ systems or 1 organ system severely enough to require specialty pediatric care and probably
some period of hospitalization in a tertiary care center’’ (13).

Retail pharmacy spending for constipation during the study
period was $443,200, representing 0.6% of constipation spending.
A minority of patients (15.7%) had abdominal radiography
during their outpatient constipation visits. Radiography use was
higher in the ED (33.3%) versus the non-ED (15.0%) outpatient
setting (P < 0.001).
Patients With Complex Chronic Conditions
Within the study cohort, 336,100 (6.8%) children had one or
more CCCs (Table 1); 5.5% had 1 CCC and the remaining 1.4% of
the cohort had 2 or more CCCs. The most prevalent CCC was other
congenital or genetic defect (1.6%) and the least prevalent was
organ transplant (0.1%). Gastrointestinal CCCs were present in
0.4% of children in the study.
Constipation diagnosis was more common in children with
1 versus no CCCs: 12.8% versus 4.9%, P < 0.001. A diagnosis of
constipation was also more common as the number of CCCs
increased: 11% with 1 CCC; 16.6% with 2 CCCs; and 27.1% with
3 CCCs (P < 0.001).
Constipation-related spending was proportionally higher in
patients with versus without CCCs (Table 2). Children with CCCs
80 All children ($79.4 million)
Children with complex chronic conditions ($26.6 million)
Children w/o complex chronic conditions ($52.8 million)
% of total constipation spending
accounted for $26.6 million (33.5%) of total spending for children
with constipation. Children with CCCs also incurred 70.3% of
inpatient constipation spending, 29.1% of outpatient clinic consti-
pation spending, and 19.8% of ED constipation spending. The
relative proportions of categories of spending in children with
and without CCC are represented in Figure 1.
Healthcare utilization for constipation was also proportion-
ally higher in patients with versus without CCCs. Although 6.8% of
the total cohort had one or more CCCs, those with any CCC
accounted for 19.3% of patients with acute outpatient treatment
and 55.3% of patients hospitalized for constipation. Inpatient
encounters for children with CCCs were less likely than those
without CCCs to have a study ICD-9 code as the primary diagnosis
code (87.9% vs 91.8%, P ¼ 0.01). The likelihood for all categories
of healthcare utilization increased with increasing number of CCCs
(Table 3).
For children diagnosed with constipation, those with CCCs
were more likely to fill a prescription for laxative (5.8% vs 4.7%,
P < 0.001). Of those who filled a prescription, patients with versus
without a CCC were also more likely to fill prescriptions for
multiple constipation medications: 11.3% versus 9.1% for 2 con-
stipation medications and 2.7% versus 1.6% for 3 medications
(P < 0.001 for all).
Children with CCCs were slightly less likely to have radi-
ography during outpatient constipation visits (15.0% vs 15.8%,
P < 0.001).

DISCUSSION
60 In our sample of 4.9 million children insured by Medicaid,
constipation was identified in health claims in 5.4%, or about 1 in
20 children, during a 3-year period. Children ages 1 to 2 years had
the highest rate of constipation diagnosis. The large majority of
40 constipation-related care in our study occurred in the outpatient
setting. These results are similar to a recent constipation study in
the adult US population (14). The mean spending per visit for
outpatient care ($120) in our cohort was much smaller than the
20
mean for inpatient treatment ($7815). This large relative differ-
ence in spending is also similar to findings in adult patients
(14,15) and accounts for the disproportionate effect on overall
0 spending in our study, whereby inpatients represented only 0.5%
Outpatient Emergency Inpatient Pharmacy of patients diagnosed with constipation, yet accounted for 15.4%
of total spending. The outsized effect on expenditures from
Health service
inpatient care when the majority of patients might be effectively
FIGURE 1. Allocation of healthcare spending for children with con- managed as outpatients argues for judicious use of hospitalization
stipation in Medicaid by the presence or absence of a complex chronic in the treatment of constipation. As 40.5% of inpatients in our
condition. study did not have a preceding diagnosis of constipation, a

34 www.jpgn.org

Copyright © ESPGHAN and NASPGHAN. All rights reserved.

JPGN  Volume 64, Number 1, January 2017
TABLE 3. Multivariable analysis of the likelihood of constipation diagnosis and treatment in children with Medicaid by number of complex chronic
conditions
Likelihood of constipation diagnosis Likelihood of outpatient constipation treatment
OR (95% CI)
CCC count
0 Reference
1 2.6 (2.6, 2.6)
2 4.1 (4.0, 4.2)
3 or more 7.3 (7.1, 7.5)
Age
1–2 2.6 (2.8, 2.9)
3–5 2.4 (2.3, 2.4)
6–12 1.9 (1.9, 2.0)
13–17 Reference
Sex
Female 1.3 (1.3, 1.3)
Male Reference
CCC ¼ complex chronic conditions; CI ¼ confidence interval; OR ¼ odds ratio.
substantial percentage of admissions appear to have occurred
without a prior attempt at outpatient treatment.
Over a third of the outpatient constipation-related spending
in our study occurred in the emergency setting. This occurred, in
part, because the mean spending on an ED visit for constipation was
twice that for an outpatient clinic visit. The preventability of ED
visits for constipation has not been assessed, and it is likely that
some children presented with a nonspecific symptom such as
abdominal pain but were discharged with a constipation diagnosis.
There may, however, be an opportunity to reduce unnecessary ED
visits for constipation through improved outpatient and community
care management for children in Medicaid.
The overall utilization rate of radiography for outpatient
constipation visits in our study was low at about 15%. This may
represent adoption of current practice guidelines, which recom-
mend against routine radiography in diagnosing constipation (16).
Radiography use was significantly higher in the ED setting, with
one third of these patients receiving abdominal x-rays. The ED
radiography rate in our study is actually lower than the 46% rate
found by Freedman et al (17) in a recent pediatric ED study of x-
rays during constipation visits. Although the higher ED rate of
radiography likely reflects diagnostic uncertainty around a non-
specific complaint, such as abdominal pain, it may also represent
another area of potential practice improvement.
The rate of constipation treatment, which we defined in our
study as a prescription for therapeutic laxative filled within 2 days
of an outpatient constipation visit, was quite low, occurring in only
4.8% of those diagnosed with constipation. It is also interesting that
docusate was the most commonly prescribed agent, as this is not
recommended as first-line therapy for constipation per current
practice guidelines (16). Although both of these findings could
be due to regional practice variation, we believe our results more
likely underestimate those treated, as some commonly used agents
to treat constipation, including polyethylene glycol 3350, which is
first-line therapy per current practice guidelines, are available
over–the-counter in some states, and thus would not generate a
Medicaid claim. As the states included in the Truven database are
deidentified, we are therefore unable to fully account for all laxative
use. It is also possible that nonpharmaceutical treatment or family
nonadherence contributed to the low prescription medication usage.
But given incomplete information on some important therapeutic
www.jpgn.org
Copyright © ESPGHAN and NASPGHAN. All rights reserved.
Healthcare Utilization and Spending for Constipation in Children
Likelihood of inpatient constipation treatment
OR (95% CI) OR (95% CI)
Reference Reference
2.8 (2.7, 3.0) 8.5 (7.4, 9.8)
4.9 (4.4, 5.4) 31.2 (26.2, 37.0)
6.1 (5.5, 6.9) 96.1 (83.5, 110.7)
1.2 (1.1, 1.2) 0.9 (0.8, 1.1)
1.1 (1.0, 1.2) 1.5 (1.3, 1.8)
1.1 (1.1, 1.2) 1.6 (1.4, 1.9)
Reference Reference
1.3 (1.3, 1.4) 0.9 (0.8, 1.0)
Reference Reference
agents that may not be captured in prescription claims, our data on
laxative use may not generalize to all children.
Comparison of the findings from the present study with the
pediatric literature is difficult given the small number of studies on the
prevalence of childhood constipation and related health services.
Choung and colleagues (6) used a case-control design to assess the
effect of constipation on healthcare cost in children ages 5 and 18 years.
This cohort excluded most of the age range that we found were most
frequently diagnosed with constipation (ie, age 1–2 years). Their
findings, significantly increased outpatient and inpatient costs in the
constipation cases compared with controls, certainly support that there
are increased costs associated with constipation diagnosis and treat-
ment, but the direct costs cannot be calculated by their methodology.
Liem and colleagues (7) reported a constipation prevalence of 1.1% in
children using a nationally representative household survey. This
prevalence is lower than that reported in the present study. One reason
for this difference may be that the prior study had a shorter follow-
up period.
The present study complements the existing literature by
describing the prevalence of constipation and related health services
in children with CCCs. Rates of diagnosis, outpatient treatment, and
inpatient treatment all increased progressively with an increasing
number of CCCs. The large majority of CCCs in our patients were
not gastrointestinal. Thus, the effect of CCCs may be due to other
factors, such as medication side effects, impaired mobility, or the
physiologic effects of another organ system (eg, neurologic) on the
digestive system. Children with CCCs accounted for substantial
percentages of constipation spending across categories, despite
representing a small minority of the study cohort. The outsized
effect on constipation utilization from a small group of patients with
CCCs mirrors similar effect found on overall healthcare utilization
by this population. For example, a recent Canadian study of a cohort
of 15,771 children with medical complexity, representing 0.67% of
the eligible study population, reported healthcare utilization of
32.7% of total spending for children in that province (8).
Our study has several limitations. As mentioned above, we
may have underestimated the number of visits for outpatient treat-
ment for constipation due to our not being able to measure use of
laxatives not covered by Medicaid in some states, particularly
polyethylene glycol 3350. In addition, our data are retrospective
and may have missed some cases of constipation if other diagnosis
35
Stephens et al
codes than the ones we studied were used. We, however, did attempt
to use a comprehensive list of ICD-9 codes for constipation, similar
to Choung et al (6), who also hand-reviewed charts to confirm the
codes’ accuracy. All spending for visits including a diagnosis for
constipation were attributed to constipation-related care, so the esti-
mates for spending could be overestimated if these visits included
treatments for conditions unrelated to constipation. The large majority
of inpatient visits, however, which accounted for a large proportion of
spending, had one of our study ICD-9 codes as the primary diagnosis,
for patients with and without CCC, suggesting that this was the major
issue addressed. It is possible that some diagnoses of constipation were
made incorrectly, although the overall low rate of radiography in
patients with and without CCCs argues against the diagnoses being
made based on x-rays alone. Our study is drawn from a group of
children receiving Medicaid, who may differ from children with other
insurance in ways relevant to this analysis. Furthermore, we can only
observe these children for as long as they are enrolled in Medicaid in the
same state, so instances of constipation that occurred outside those
parameters would be missed. Lastly, our data were drawn from 10
deidentified states. Although the states are reported to represent large
and small states from varied geographic regions, we cannot exclude
that institutional or regional practice variation contributed to the study
results, potentially limiting generalizability to all children.
There are important implications to consider from the study
findings. First, the large majority of constipation cases were man-
aged in the outpatient setting. The large differential expense of
inpatient care for constipation should generate further study to
understand which patients necessitate admission for constipation
treatment. In addition, the large proportion of inpatients that did not
have a prior outpatient visit for constipation suggests a potential role
for better constipation screening and treatment in the primary care
setting. Second, children with CCCs merit special attention to
constipation given their significantly increased risks of diagnosis
and high acuity healthcare utilization. The proportionally high
utilization for constipation in children with CCCs suggests that
common—and seemingly straightforward pediatric problems—
may have significant effect on health for patients with medical
complexity. Potential future areas of study include prospective
examination of risk factors for admission for constipation and
hospital-level variation in practice on treatment of constipation.
REFERENCES
1. Van den Berg MM, Benninga MA, Di Lorenzo C. Epidemiology of
childhood constipation: a systematic review. Am J Gastroenterol
2006;101:2401–9.
36
Copyright © ESPGHAN and NASPGHAN. All rights reserved.
JPGN  Volume 64, Number 1, January 2017
2. Sonnenberg A, Koch TR. Physician visits in the United States for
constipation: 1958 to 1986. Dig Dis Sci 1989;34:606–11.
3. Chitkara DK, Talley NJ, Locke GR 3rd et al. Medical presentation of
constipation from childhood to early adulthood: a population-based
cohort study. Clin Gastroenterol Hepatol 2007;5:1059–64.
4. Chitkara DK, Talley NJ, Weaver AL, et al. Incidence of presentation
of common functional gastrointestinal disorders in children from
birth to 5 years: a cohort study. Clin Gastroenterol Hepatol
2007;5:186–91.
5. Everhart JE, Ruhl CE. Burden of digestive diseases in the United States
part II: lower gastrointestinal diseases. Gastroenterology 2009;136:
741–54.
6. Choung RS, Shah ND, Chitkara D, et al. Direct medical costs of
constipation from childhood to early adulthood: a population-based
birth cohort study. J Pediatr Gastroenterol Nutr 2011;52:47–54.
7. Liem O, Harman J, Benninga M, et al. Health utilization and cost impact
of childhood constipation in the United States. J Pediatr 2009;154:258–
62.
8. Cohen E, Berry JG, Camacho X, et al. Patterns and costs of health care
use of children with medical complexity. Pediatrics 2012;130:e1463–
70.
9. Simon TD, Berry J, Feudtner C, et al. Children with complex chronic
conditions in inpatient hospital settings in the United States. Pediatrics
2010;126:647–55.
10. Burns KH, Casey PH, Lyle RE, et al. Increasing prevalence of
medically complex children in US hospitals. Pediatrics 2010;126:
638–46.
11. Berry JG, Hall DE, Kuo DZ, et al. Hospital utilization and character-
istics of patients experiencing recurrent readmissions within children’s
hospitals. JAMA 2011;305:682–90.
12. Feudtner C, Feinstein JA, Zhong W, et al. Pediatric complex chronic
conditions classification system version 2: updated for ICD-10 and
complex medical technology dependence and transplantation. BMC
Pediatr 2014;14:199.
13. Feudtner C, Christakis DA, Connell FA. Pediatric deaths attribut-
able to complex chronic conditions: a population-based study of
Washington State, 1980–1997. Pediatrics 2000;106 (1 Pt 2):
205–9.
14. Martin BC, Barghout V, Cerulli A. Direct medical costs of constipation
in the United States. Manag Care Interface 2006;19:43–9.
15. Sethi S, Mikami S, Leclair J, et al. Inpatient burden of constipation in the
United States: an analysis of national trends in the United States from
1997 to 2010. Am J Gastroenterol 2014;109:250–6.
16. Tabbers MM, DiLorenzo C, Berger MY, et al. Evaluation and treatment
of functional constipation in infants and children: evidence-based
recommendations from ESPGHAN and NASPGHAN. J Pediatr Gas-
troenterol Nutr 2014;58:258–74.
17. Freedman SB, Thull-Freedman J, Manson D, et al. Pediatric abdominal
radiograph use, constipation, and significant misdiagnoses. J Pediatr
2014;164:83–8e82.
www.jpgn.org