Autoimmune disease concomitance among inflammatory bowel disease patients in the United States, 2001-2002.

Abstract

Recent studies suggest that inflammatory bowel disease (IBD) may share an underlying pathogenesis with other autoimmune diseases.
Two United States data sets with patient-level medical and drug claims were used to explore the occurrence of autoimmune diseases in patients with IBD, particularly Crohn's disease (CD) and ulcerative colitis (UC), with that in controls. From 2001 to 2002 IBD patients were identified using International Classification of Diseases, 9th revision, diagnosis codes in the IMS Health Integrated Administration Claims Database and the Market Scan Commercial Claims and Encounters Database. Controls were selected by matching on sex, age, Census Bureau region, and length of previous medical insurance coverage. Odds ratios (ORs) evaluated the risk relationship between IBD patients and controls within an estimated Mantel-Haenszel 95% confidence interval. Sensitivity analysis tested the case identification method used to select IBD patients.
The risk for ankylosing spondylitis (AS) was substantially increased across both data sets: OR (95% confidence interval [CI]) of 7.8 (5.6-10.8) in IMS Health and 5.8 (3.9-8.6) in MarketScan. The risk for rheumatoid arthritis (RA) was 2.7 (2.4-3.0) and 2.1 (1.8-2.3), respectively; for multiple sclerosis (MS); the ORs were 1.5 (1.2-1.9) and 1.6 (1.2-2.1), respectively. There was no increased risk for type 1 diabetes mellitus, and the results for psoriatic arthritis (PsA) were inconsistent. The sensitivity analysis supported these findings.
A much higher risk for RA, AS, PsA, and MS was observed in IBD patients compared with controls. Prospective epidemiologic studies are needed to confirm these findings and explore the pathogenic mechanism of this relationship.

Full text

Page 1

ORIGINAL ARTICLE
Autoimmune Disease Concomitance Among Inflammatory
Bowel Disease Patients in the United States, 2001-2002
RussellCohen,MD,*DonRobinson,Jr.,MSPH,†ClarkParamore,MSPH,‡KathyFraeman,SM,‡
KevinRenahan,MBA,§andMohanBala,PhD†
Background: Recent studies suggest that inflammatory bowel
disease (IBD) may share an underlying pathogenesis with other
autoimmune diseases.
Methods: Two United States data sets with patient-level medical
and drug claims were used to explore the occurrence of autoimmune
diseases in patients with IBD, particularly Crohn’s disease (CD) and
ulcerative colitis (UC), with that in controls. From 2001 to 2002 IBD
patients were identified using International Classification of Dis-
eases, 9th revision, diagnosis codes in the IMS Health Integrated
Administration Claims Database and the Market Scan Commercial
Claims and Encounters Database. Controls were selected by match-
ing on sex, age, Census Bureau region, and length of previous
medical insurance coverage. Odds ratios (ORs) evaluated the risk
relationship between IBD patients and controls within an estimated
Mantel-Haenszel 95% confidence interval. Sensitivity analysis
tested the case identification method used to select IBD patients.
Results: The risk for ankylosing spondylitis (AS) was substantially
increased across both data sets: OR (95% confidence interval [CI])
of 7.8 (5.6–10.8) in IMS Health and 5.8 (3.9–8.6) in MarketScan.
The risk for rheumatoid arthritis (RA) was 2.7 (2.4–3.0) and 2.1
(1.8–2.3), respectively; for multiple sclerosis (MS); the ORs were
1.5 (1.2–1.9) and 1.6 (1.2–2.1), respectively. There was no increased
risk for type 1 diabetes mellitus, and the results for psoriatic arthritis
(PsA) were inconsistent. The sensitivity analysis supported these
findings.
Conclusions: A much higher risk for RA, AS, PsA, and MS was
observed in IBD patients compared with controls. Prospective epi-
demiologic studies are needed to confirm these findings and explore
the pathogenic mechanism of this relationship.
(Inflamm Bowel Dis 2008;14:738–743)
Key Words: inflammatory bowel disease, autoimmune disease,
prevalence, epidemiology
S
that patients with the inflammatory bowel diseases (IBD)
Crohn’s disease (CD) and ulcerative colitis (UC) are much
more likely to have 1 of the arthritic autoimmune diseases
than are patients without IBD.1–3Patients with IBD appear
more likely to have multiple sclerosis (MS) as well.2,4In
2007, Weng et al reported finding similar results in a northern
California managed care organization.5These data, as well as
others, have led to speculation that IBD may share pathogenic
pathways with some immune-mediated inflammatory dis-
eases, possibly as a member of the larger autoimmune meta-
disease category.6,7
Two administrative health care data sets from the
United States (U.S.) were used to identify IBD and non-IBD
patients and investigate autoimmune disease relationships
using a prevalence based case-control study design. Each data
set had patient-level medical and prescribed drug claims.
Hence, results from 1 database could be confirmed in the
other using similar, relevant medical information.
everal recent population-based studies using data sources
from Canada and the United Kingdom have demonstrated
PATIENTS AND METHODS
Patients
The eligible population consisted of all U.S. adult pa-
tients from 2 national data sets that contained health care
administrative claims from inpatient and ambulatory settings
(outpatient clinics and physician offices). These data sets
represent 2 major forms of health benefit coverage in the
United States. The IMS Health Integrated Administrative
Claims Database (IMS Health) contained a younger popula-
tion with commercial health plans. Approximately 11 million
unique individuals were initially available for analysis over
these 2 years. During the same period, the MarketScan Com-
mercial Claims and Encounters Database (MarketScan) had
approximately 6 million individuals for study. This was an
Received for publication November 7, 2007; Accepted January 3, 2008.
From the*University of Chicago Hospital, Chicago, Illinois;†Johnson &
Johnson Pharmaceutical Services, Malvern, Pennsylvania;
Source Corp. (UBC), Bethesda, Maryland;§Centocor, Inc., Malvern, Penn-
sylvania.
Supported by Centocor, Inc.
Mohan Bala is currently employed by GlaxoSmithKline, Collegeville, PA.
Reprints: Don Robinson, Jr., Johnson & Johnson Pharmaceutical Services,
200 Great Valley Parkway, Malvern, PA 19355 (e-mail: drobins4@
cntus.jnj.com)
Copyright © 2008 Crohn’s & Colitis Foundation of America, Inc.
DOI 10.1002/ibd.20406
Published online 25 February 2008 in Wiley InterScience (www.interscience.
wiley.com).
‡United Bio-
738
Inflamm Bowel Dis ● Volume 14, Number 6, June 2008

Page 2

older population from mostly self-insured employers. Both
data sources were mainly Midwestern and Southern in U.S.
regional representation, although some data from the Eastern
region were present. Continuous enrollment with medical and
drug benefits was required from January 1, 2001, to Decem-
ber 31, 2002. An enrollment gap of up to 60 days was
allowed, and all patients included in the analysis had to have
at least 1 medical encounter during the observation period.
The elderly had limited representation in these data sets,
although retiree and Medicare plans were included. Addi-
tional details on these data and methods have been previously
reported.8
Patients were defined as either IBD cases or controls
based on the presence or absence of an International Classi-
fication of Diseases, 9th revision (ICD-9), diagnosis code
indicating IBD in the patient-level medical claims record over
the 2-year period. Similarly, the presence of another autoim-
mune disease, such as rheumatoid arthritis (RA) or MS,
among IBD cases or controls was indicated by the presence of
at least 1 ICD-9 diagnosis code for that disease in the treat-
ment record. Only those autoimmune diseases with estab-
lished clinical criteria and contrasting clinical features were
included to avoid the effects of misdiagnosis or “rule-out”
diagnoses on the evaluation (Table 1). Control patients were
randomly selected and frequency matched at a 3:1 ratio to
each IBD patient based on the criteria of age (within 2 years),
sex (exact), Census Bureau region (exact), and duration of
previous medical insurance coverage (within the same
month). Two data sets were used for validation purposes,
where consistency of results across the data sets was consid-
ered a reliable, substantiated finding.
Sensitivity Analysis
The possible misidentification of IBD and other auto-
immune disease patients through misdiagnosis or coding er-
rors was addressed by comparing 2 methods of case identi-
fication using ICD-9 criteria. The standard model of
identifying IBD and other autoimmune disease patients re-
quired at least 1 medical service claim carrying an appropri-
ate ICD-9 code within any diagnostic field (Table 1). The
sensitivity analysis required at least 1 inpatient event or at
least 2 ambulatory face-to-face encounters carrying the rele-
vant ICD-9 code when identifying IBD and other autoim-
mune disease patients.2,9,10The IMS Health data set was used
for the sensitivity analysis because it had sufficient numbers
of controls to support the 3:1 matching ratio.
Data Analysis
Co-occurrence rates for autoimmune diseases were
compared to controls using the odds ratio (OR) statistic.11
The 95% confidence intervals (CIs) for the ORs were esti-
mated using the Mantel-Haenszel method. Standard summary
descriptive statistics were also applied, as appropriate. In the
standard population model, there was approximately 80%
power at the 0.05 level to detect the concomitant risk of
autoimmune diseases with ORs as low as 1.12 for the IMS
Health data and 1.13 for the MarketScan data (data not
shown). In the sensitivity analysis, the same statistical power
existed for an OR of 1.18 in IMS Health. These estimates
assume an autoimmune disease prevalence of 4.0%.8A single
SAS programmer who was familiar with both data sets (K.F.)
constructed the standardized analytic programming across the
data sets, thus reducing the confounding factor of program-
ming variation in the study. The SAS system (version 8.2,
SAS Institute, Cary, NC) was used for all study analyses.
RESULTS
The IBD populations across the data sets were similar
in sex distribution and prevalence rate, but varied by age. IBD
prevalence ranged from 0.65 to 0.62 in the IMS Health and
MarketScan data sets, respectively, and mean age was 47 and
54 years, respectively, with both populations being 56%
female (Table 2). The prestudy length of medical insurance
coverage was similar for both IBD patients and controls (data
not shown). The mean age of patients with IBD was approx-
imately 4 years older than the mean age of the overall
population in each data set.8In both data sets IBD patients
and their matched controls were predominantly from the
Midwest and South, as were the parent populations.8
Autoimmune Disease Co-occurrence
Across both data sets, patients with IBD were at in-
creased risk of an arthritic autoimmune disease, especially
TABLE 1. International Classification of Diseases, 9th Revision,
Diagnosis Codes for Identifying Inflammatory Bowel Disease
and Concomitant Autoimmune Diseases
Autoimmune diseases*
ICD-9 codes
Atopic dermatitis
Ankylosing spondylitis
Diabetes mellitus, type 1
Inflammatory bowel disease
Crohn’s disease
Ulcerative colitis
Multiple sclerosis
Psoriasis
Psoriatic arthritis (PsA)
Psoriasis, nonarthritic (PsO)
Rheumatoid arthritis
Musculoskeletal†
691.x
720.0
250.x1
555.x, 556.x
555.x
556.x
340.x
696.0-1
696.0
696.1
714.x
696.0, 714.x, 720.0
ICD-9, International Classification of Diseases, 9th revision.*Restricted to
autoimmune diseases considered clinically distinct from inflammatory bowel
disease. Autoimmune diseases with overlapping clinical manifestations or
diagnostic ambiguity (sarcoidosis, Sjogren’s syndrome, systemic lupus ery-
thematosus) were removed from analysis.†Musculoskeletal conditions are
ankylosing spondylitis, psoriatic arthritis, and rheumatoid arthritis.
Inflamm Bowel Dis ● Volume 14, Number 6, June 2008 Autoimmune Disease Concomitance in IBD Patients
739

Page 3

ankylosing spondylitis (AS). The ORs for AS among IBD
patients ranged from 5.8 to 7.8 (Fig. 1). A markedly higher
prevalence of 2 dermatologic autoimmune diseases, atopic
dermatitis and psoriasis, was also found. The ORs for atopic
dermatitis and psoriasis were approximately 1.5 in both data
sets. Patients with IBD were also at higher risk for MS, with
ORs of 1.5 (IMS Health) and 1.6 (MarketScan). These pat-
terns were consistent across the 2 IBD subtypes, CD and UC,
although the CD cohort had a higher risk of several muscu-
loskeletal autoimmune diseases (Figs. 2 and 3). Of the IBD
patients in the IMS Health and MarketScan data sets, 10%
and 8%, respectively, had ICD-9 diagnosis codes, for both
CD and UC, at least once in their 2-year medical record (data
not shown).
Sensitivity
IBD prevalence decreased by about 50% to 0.28% and
0.32% in the MarketScan and IMS Health data sets, respec-
tively, when using the sensitivity analysis rule for case iden-
tification (data not shown). The age and sex distributions
remained the same as those previously reported, but chronic
disease burden scores generally increased (data not shown).8
The risks for co-occurrence with other autoimmune diseases
under the sensitivity analysis were comparable to those found
with the standard population model, with an increased risk for
AS, MS, and RA being the most pronounced (Fig. 4). Inter-
estingly, in the IMS Health data set, the risk for MS increased
while the risk for atopic dermatitis and psoriasis decreased.
The lowest values of the CIs for atopic dermatitis and psori-
asis were also below 1.0.
DISCUSSION
In the largest database study to date in a geographically
diverse U.S. population, patients with IBD had an increased
risk of other autoimmune diseases, particularly AS, RA, and
MS. These results were confirmed by a case-recognition
sensitivity analysis. Contrary to the finding of Weng et al,5a
higher likelihood for psoriasis was not found in the current
study when tested by the sensitivity analysis. The occurrence
of asthma was more frequent among IBD patients in the
standard population model, with an OR of 1.5 in both data
sets. However, the risk for asthma among IBD patients drops
to 1.3 under sensitivity analysis in the IMS Health data set
(data not shown). These findings are consistent with previous
studies.2,5
This report adds to the current evidence of an elevated
risk of autoimmune disease in IBD, a disease pattern that
repeatedly appears in these data sources despite differences in
study design and country.2–5An underlying pathogenic mech-
anism based on a dysfunctional immune system is clearly
suggested by these data. The concordance of these results
across the 2 distinct data sets, IMS Health and MarketScan,
TABLE 2. Inflammatory Bowel Disease, Control, and Immune-Mediated Inflammatory Disease Population Characteristics*
IMID
MarketScanIMS Health
N Age
Female
(%)
Prevalence
(%)N Age
Female
(%)
Prevalence
(%)
Atopic dermatitis
Ankylosing spondylitis
Diabetes mellitus, type 1
Inflammatory bowel disease
Inflammatory bowel disease
controls
Crohn’s disease
Crohn’s disease controls
Ulcerative colitis
Ulcerative colitis
controls
Multiple sclerosis
Psoriasis
Psoriatic arthritis
Psoriasis, nonarthritic
Rheumatoid arthritis
18,720
1370
5339
16,139
50.6 (17.4)
54.4 (13.8)
56.7 (17.5)
53.5 (15.6)
63.3
37.9
51.5
55.5
0.71
0.05
0.20
0.62
24,039
1843
4907
18,603
43.0 (13.9)
47.3 (12.0)
46.9 (16.2)
46.7 (13.0)
65.7
40.4
49.4
55.7
0.84
0.06
0.17
0.65
64,556
7401
29,604
10,104
53.5 (15.6)
52.3 (15.6)
52.3 (15.6
54.2 (15.6)
55.5
57.3
57.3
54.3
N/A
0.29
N/A
0.39
66,969
9267
37,068
11,220
46.6 (13.1)
45.6 (13.1)
45.6 (13.1
47.2 (13.0)
55.8
57.0
57.0
54.0
N/A
0.32
N/A
0.39
40,416
6826
20,614
2366
18,248
28,255
54.2 (15.6)
51.4 (12.6)
54.9 (15.5)
54.8 (12.8)
55.0 (15.8)
59.6 (14.1)
54.3
75.3
52.7
52.2
52.7
72.5
N/A
0.26
0.79
0.09
0.70
1.08
44,880
9131
25,556
3066
22,490
28,208
47.2 (13.0)
46.7 (10.3)
47.1 (12.6)
49.7 (11.2)
46.8 (12.8)
51.9 (12.5)
54.0
76.1
51.2
50.9
51.3
72.6
N/A
0.32
0.89
0.11
0.78
0.98
*Reported as means, (standard deviation) and percentages, as indicated. Two-year prevalence (%) based on a population of 2,619,719 adults continuously
enrolled from January 1, 2001, through December 31, 2002, with medical benefits in MarketScan; 2,872,333 in IMS Health. Because of rounding, psoriasis
clinical subtype (arthritic and nonarthritic psoriasis) prevalence does not equal that of the total psoriasis population (N/A, not applicable; MarketScan,
MarketScan Commercial Claims and Encounters Database; IMS Health, IMS Health Integrated Adminstrative Claims Database).
Cohen et al
Inflamm Bowel Dis ● Volume 14, Number 6, June 2008
740

Page 4

validates these findings, as these data represent the 2 main
sources of health care insurance coverage in the U.S. The
findings of this study support the evaluation of arthritic im-
pairment and neurologic characteristics among IBD patients
in standard care.
Health care claims data sets have been previously used
to explore IBD epidemiology.1,3–5,12The 2-year observation
period used in this study has been found to raise sensitivity in
other conditions, albeit in a Medicare population.10Matching
cases and controls for prebaseline medical insurance cover-
age standardized the likelihood for a chronic condition such
as the autoimmune diseases explored in this study.9
However, an observational bias may exist because pa-
tients with IBD have more medical service visits than the
control group, providing greater opportunity for physicians to
identify another autoimmune disease among the IBD popu-
lation during this period.13,14Similarly, IBD could be de-
tected in patients with an existing autoimmune disease who
have high resource utilization rates. Consequently, the higher
occurrence of AS, RA, and MS among IBD patients observed
in this study design could have been a result of this observa-
tion bias. Yet, the elevated occurrence of AS, RA, and MS
continued at almost the same magnitude after a higher level
of utilization was standardized across IBD patients and con-
trols in the sensitivity analysis.
Health care claims data have many confounders be-
cause of the relationship between access to care, health status,
utilization, and reimbursement inherent in their naturalistic
character. The number of IBD cases not recognized or coded
as such is unknown. Changes in diagnostic criteria and case
recognition rates have historically challenged IBD epidemi-
ologic research.15Several aspects of this study helped to
minimize the role of these factors. The data were recent and
confined to a single country and reduced the impact of prac-
tice pattern variation by having a similar regional distribution
across both data sets used for comparison. Additionally,
given the difficulty of an IBD diagnosis and its low preva-
lence as a chronic medical condition, it seems unlikely that
IBD ICD-9 diagnosis codes would appear in the medical
FIGURE 2. Concomitant risk of another autoimmune disease
among patients with ulcerative colitis. Only autoimmune dis-
eases with clinically distinct manifestations and organ systems
are listed (UC, ulcerative colitis; OR, odds ratio; CI, confidence
interval; AD, atopic dermatitis; AS, ankylosing spondylitis;
DMT1, diabetes mellitus type 1; MS, multiple sclerosis; Ps, pso-
riasis; PsO, psoriasis, nonarthritic; PsA, psoriatic arthritis; RA,
rheumatoid arthritis; MSK, musculoskeletal disorders including
AS, PsA, and RA).
FIGURE 1. Concomitant risk of another autoimmune disease
among patients with inflammatory bowel disease. Only auto-
immune diseases with clinically distinct manifestations and or-
gan systems are listed (OR, odds ratio; CI, confidence interval;
AD, atopic dermatitis; AS, ankylosing spondylitis; DMT1, diabe-
tesmellitustype1;MS,multiplesclerosis;Ps,psoriasis;PsO,pso-
riasis, nonarthritic; PsA, psoriatic arthritis; RA, rheumatoid ar-
thritis; MSK, musculoskeletal disorders including AS, PsA, and
RA; P ? 0.05 across both data sets for AD, AS, MS, Ps, PsO, PsA,
and RA).
Inflamm Bowel Dis ● Volume 14, Number 6, June 2008 Autoimmune Disease Concomitance in IBD Patients
741

Page 5

claims record as an initial, tentative rule-out diagnosis pend-
ing further medical evaluation.
The prevalence-based study design included new and
existing cases over the 2-year period of analysis. Conse-
quently, disease duration may have been a confounding in-
fluence on the OR results. Also, no measure of disease
severity was available in these data, so the association of this
factor with the occurrence of other autoimmune diseases
could not be explored. Data were also not available for health
behavior such as smoking, social factors such as ethnicity,
and physical characteristics such as body mass index. These
multiregional U.S. data sets lack a strong Western represen-
tation and are therefore not fully national in geographic
character. There is a likely “healthy worker” bias in these
patient populations, which would underestimate IBD preva-
lence. Additionally, laboratory data were not collected fre-
quently enough to merit inclusion into the analysis plan.
Confirmation of these results in other Western U.S. data
sources as well as those with Medicaid and older populations
would be useful. Countries have various health care delivery
systems and benefits, as well as practice patterns and popu-
lation demographics, all of which limit the relevance of these
results in other regions. Because of the proprietary nature of
these data sets, the accuracy of ICD-9 coding could not be
confirmed through comparison with patient medical charts,
although both data suppliers regularly conduct quality checks
on their data sources.
Large longitudinal studies that evaluate IBD onset,
occurrence of other diseases, family history, genetics, and
treatment would clarify the relationships among these health
factors.16The incorporation of biomarkers and patient qual-
ity-of-life questionnaires in such studies would deepen the
knowledge base regarding the natural history of IBD and
patient treatment planning.17,18
FIGURE 4. Concomitant risk of another autoimmune disease
among inflammatory bowel disease patients: sensitivity analy-
sis. Data are from the IMS Health data set and include only im-
mune-mediated inflammatory diseases with clinically distinct
manifestations and organ systems and require at least 1 inpa-
tient or at least 2 ambulatory medical claims (outpatient clinic
or physician office). OR, odds ratio; CI, confidence interval; AD,
atopic dermatitis; AS, ankylosing spondylitis; DMT1, diabetes
mellitustype1;MS,multiplesclerosis;Ps,psoriasis;PsO,psoria-
sis,nonarthritic;PsA,psoriaticarthritis;RA,rheumatoidarthritis;
MSK, musculoskeletal disorders including AS, PsA, and RA. In-
flammatory bowel disease, n ? 9160; control, n ? 26,844 (P
? 0.05 for AS, MS, and RA).
FIGURE 3. Concomitant risk of another autoimmune disease
among patients with Crohn’s disease. Only autoimmune dis-
eases with clinically distinct manifestations and organ systems
are listed (CD, Crohn’s disease; OR, odds ratio; CI, confidence
interval; AD, atopic dermatitis; AS, ankylosing spondylitis;
DMT1, diabetes mellitus type 1; MS, multiple sclerosis; Ps, pso-
riasis; PsO, psoriasis, nonarthritic; PsA, psoriatic arthritis; RA,
rheumatoid arthritis; MSK, musculoskeletal disorders including
AS, PsA, and RA).
Cohen et al
Inflamm Bowel Dis ● Volume 14, Number 6, June 2008
742

Page 6

ACKNOWLEDGMENTS
This study was funded by Centocor, Inc. The authors
appreciate the medical writing assistance of Mary Whitman,
PhD, and Rebecca E. Clemente, PhD, of Centocor, Inc., as
well as the statistical support of Paul Nietert, PhD, of the
Medical University of South Carolina.
REFERENCES
1. Bernstein CN, Blanchard JF, Rawsthorne P, et al. The prevalence of
extraintestinal diseases in inflammatory bowel disease: a population-
based study. Am J Gastroenterol. 2001;96:1116–1122.
2. Bernstein CN, Wajda A, Blanchard JF. The clustering of other chronic
inflammatory diseases in inflammatory bowel disease: a population-
based study. Gastroenterology. 2005;129:827–836.
3. Cucino C, Sonnenberg A. The comorbid occurrence of other diagnoses
in patients with ulcerative colitis and Crohn’s disease. Am J Gastroen-
terol. 2001;96:2107–2112.
4. Gupta G, Gelfand JM, Lewis JD. Increased risk for demyelinating
diseases in patients with inflammatory bowel disease. Gastroenterology.
2005;129:819–826.
5. Weng X, Liu L, Barcellos LF, et al. Clustering of inflammatory bowel
disease with immune mediated diseases among members of a northern
California-managed care organization. Am J Gastroenterol. 2007;102:
1429–1435.
6. Davidson A, Diamond B. Autoimmune diseases. N Engl J Med. 2001;
345:340–350.
7. Williams JP, Meyers JA. Immune-mediated inflammatory disorders
(I.M.I.D.s): the economic and clinical costs. Am J Manag Care. 2002;
8:S664–S681; quiz S82–S85.
8. Robinson D Jr, Hackett M, Wong J, et al. Co-occurrence and comor-
bidities in patients with immune-mediated inflammatory disorders: an
exploration using US healthcare claims data, 2001–2002. Curr Med Res
Opin. 2006;22:989–1000.
9. Peterson E, Van Vorst K, Wickstom S, et al. Exploring issues in
managed care: six illustrative case studies. In: McGlynn EA, ed. RAND
Draft Report DRU-1560-HGI; 1997;11:11–20.
10. Rector TS, Wickstrom SL, Shah M, et al. Specificity and sensitivity of
claims-based algorithms for identifying members of Medicare?Choice
health plans that have chronic medical conditions. Health Serv Res.
2004;39:1839–1857.
11. Woodward M. Epidemiology: Study Design and Data Analysis. 1st ed.
Boca Raton, Fl.: Chapman & Hall; 1999:113–117.
12. Bernstein CN, Wajda A, Svenson LW, et al. The epidemiology of
inflammatory bowel disease in Canada: a population-based study. Am J
Gastroenterol. 2006;101:1559–1568.
13. Longobardi T, Bernstein CN. Health care resource utilization in inflam-
matory bowel disease. Clin Gastroenterol Hepatol. 2006;4:731–743.
14. Longobardi T, Jacobs P, Bernstein CN. Utilization of health care re-
sources by individuals with inflammatory bowel disease in the United
States: a profile of time since diagnosis. Am J Gastroenterol. 2004;99:
650–655.
15. Loftus EV Jr, Sandborn WJ. Epidemiology of inflammatory bowel
disease. Gastroenterol Clin North Am. 2002;31:1–20.
16. Sloka S. Observations on recent studies showing increased co-occur-
rence of autoimmune diseases. J Autoimmun. 2002;18:251–257.
17. Bernklev T, Jahnsen J, Aadland E, et al. Health-related quality of life in
patients with inflammatory bowel disease five years after the initial
diagnosis. Scand J Gastroenterol. 2004;39:365–373.
18. Loftus EV Jr. Clinical epidemiology of inflammatory bowel disease:
incidence, prevalence, and environmental influences. Gastroenterology.
2004;126:1504–1517.
Inflamm Bowel Dis ● Volume 14, Number 6, June 2008Autoimmune Disease Concomitance in IBD Patients
743