Clinical Gastroenterology and Hepatology 2015;13:310–315
Prevalence of Chronic Narcotic Use Among Children With Inflammatory Bowel Disease Jessie P. Buckley,*,‡ Suzanne F. Cook,‡ Jeffery K. Allen,‡ and Michael D. Kappelman§ *Department of Epidemiology, University of North Carolina at Chapel Hill, Chapel Hill; ‡Worldwide Epidemiology, GlaxoSmithKline, Research Triangle Park; and §Department of Pediatrics, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina BACKGROUND & AIMS:
Narcotic analgesics are not recommended for long-term management of pain for patients with inflammatory bowel disease (IBD), particularly pediatric patients. We compared chronic use of narcotics among children with IBD and the general population and investigated factors associated with narcotic use in the pediatric IBD population.
METHODS:
This cross-sectional study included children (younger than 18 years old) with continuous enrollment in a large administrative claims database from 2010 through 2011 (n [ 4,911,286). Children with IBD were identified through diagnosis codes and dispensation of IBD medication (n [ 4344); they were matched for age, sex, and region with 5 children without IBD (n [ 21,720). Chronic narcotic use was defined as ‡3 dispensements of narcotics. We estimated prevalence odds ratios (PORs) and 95% confidence intervals (CIs), comparing narcotic use on the basis of IBD status and evaluating variables associated with narcotic use by patients with IBD by using conditional and unconditional logistic regression.
RESULTS:
The prevalence of chronic narcotic use was 5.6% among children with IBD vs 2.3% in the general population (POR, 2.6; 95% CI, 2.2–3.0). Compared with the general population, POR for chronic narcotic use was significantly higher for pediatric IBD patients with psychological impairment (POR, 6.8; 95% CI, 4.3–10.6) than those without (POR, 2.3; 95% CI, 1.9–2.7). Older age, increased healthcare utilization, fracture, and psychological impairment were strongly associated with chronic use of narcotics among children with IBD.
CONCLUSIONS:
Chronic narcotic use is common in pediatric IBD patients, particularly among those with anxiety and depression. Increased awareness of psychological comorbidity, screening, and treatment may reduce symptoms that lead to narcotic use and its complications.
Keywords: Crohn’s Disease; Ulcerative Colitis; UC; CD.
arcotics may be prescribed to patients with inflammatory bowel disease (IBD) for temporary pain relief such as to treat an acute flare or alleviate pain after surgery. However, long-term narcotic analgesia is not recommended because of gastrointestinal (GI) side effects, tolerance, and possibility of addiction.1–4 Furthermore, narcotic use among adult patients with IBD has been associated with increased risk of mortality, serious infection, sleep disorders, pneumonia, and psychiatric disorders.5–8 In children, chronic treatment with narcotics has additional potential adverse clinical sequelae such as the potential risks of growth failure, malnutrition, opportunistic infections, and hepatosplenic lymphoma.1 Although pain control is an important aspect of disease management in pediatric IBD, little is known about chronic narcotic use in children. In a large, primarily adult study of administrative claims data, nearly half of patients with IBD had at least 1 narcotic prescription
N
during a 2-year period, and 20% had 3 narcotic prescriptions.9 Referral center studies of adults with IBD have reported prevalence of narcotic use and dependency of 13% and 3%–4%, respectively.8,10,11 Factors associated with narcotic analgesia in adults include IBD-related surgery, polypharmacy, increased pain, more symptoms, worsened disease activity, and smoking.8–10,12,13 Associations between chronic pain, narcotic use, depression, and anxiety suggest that treating psychiatric comorbidity should be considered as an Abbreviations used in this paper: CI, confidence interval; EphMRA, European Pharmaceutical Market Research Association; GI, gastrointestinal; IBD, inflammatory bowel disease; ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; PBIRG, Pharmaceutical Business Intelligence and Research Group; POR, prevalence odds ratio. © 2015 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2014.07.057
February 2015
alternative to reliance on analgesics in adult patients.14 Although children with Crohn’s disease are at increased risk of anxiety and depression,15 the relationship between narcotic use and psychiatric conditions has not yet been assessed in the pediatric IBD population. Describing the characteristics of children with IBD with long-term narcotic use is important to define the magnitude of this problem in the pediatric population and to identify potential strategies or interventions to reduce chronic narcotic use. The objectives of this study were to compare the prevalence of chronic narcotic use among children with and without IBD and to identify variables associated with narcotic treatment in the pediatric IBD population.
Methods Study Design and Data Source This cross-sectional study used data from the Truven Health MarketScan Commercial Claims and Encounters database, an administrative claims database that has been used widely for health research including studies of IBD.15–18 The database includes approximately 500 million insurance claims on members of the working population younger than age 65 and their dependents who have employer-sponsored health insurance coverage from 1 of approximately 100 employer-sponsored and private health plans. Information includes longitudinal data on inpatient and outpatient diagnoses and procedures linked to demographic characteristics, enrollment details, and prescription claims records.
Sample Population The source population included all patients aged 18 years in the Commercial Claims and Encounters MarketScan database with continuous health plan enrollment and pharmacy benefits from January 1, 2010 to December 31, 2011 (n ¼ 4,911,286). Prevalent patients with IBD were identified by using an adaptation of a validated IBD case definition that has been applied previously.19–23 Children were classified as having IBD if they satisfied either of the following conditions: (1) three or more healthcare contacts for Crohn’s disease or ulcerative colitis (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 555.xx and 556.xx, respectively) on separate days or (2) at least 1 healthcare contact for Crohn’s disease or ulcerative colitis and at least 1 pharmacy claim for any of the following medications: mesalamine, olsalazine, balsalazide, sulfasalazine, 6mercaptopurine, azathioprine, methotrexate, enteral budesonide, or biologics (infliximab, adalimumab, certolizumab, or natalizumab). IBD type was classified as Crohn’s disease or ulcerative colitis on the basis of the majority of the last 9 claims. Patients who could not be classified as Crohn’s disease or ulcerative colitis (n ¼ 40) or who lived
Chronic Narcotic Use Among Children With IBD
311
in an “Unknown” region (n ¼ 34) were excluded. To focus on non-surgery narcotic use, children were also excluded if they had GI surgery during the study period (Current Procedural Terminology 4 codes 43020–43135, 43280–44346, 44500–45190, 45395–46505, 46700– 46999, and 49000–49999; n ¼ 684). At the time each child with IBD entered the study population, he or she was matched on age, gender, and region to 5 randomly sampled members of the study population without IBD.
Chronic Narcotic Use Narcotic analgesic use was ascertained from dispensed prescription drug claims. Medications coded by using the National Drug Code system were mapped to European Pharmaceutical Market Research Association (EphMRA) and the Pharmaceutical Business Intelligence and Research Group (PBIRG) Anatomical Classification System drug classes. Narcotic use was defined as a claim for EphMRA/PBIRG code N02A0, and participants were classified as chronic users if they had at least 3 narcotic drug claims during the 2-year study period.
Covariates of Interest Demographic information included age, gender, and region (Northeast, North Central, South, or West). The remaining covariates were classified by using participant data during the entire 2-year study period. Markers of healthcare utilization included the number of inpatient and outpatient contacts and the number of unique generic medications dispensed (excluding narcotic analgesics). IBD medications included oral steroids, oral and rectal aminosalicylates (including mesalamine, olsalazine, balsalazide, sulfasalazine), rectal steroids (enema, suppository, foam), 6-mercaptopurine or azathioprine, methotrexate, cyclosporine or tacrolimus, and biologics (infliximab, adalimumab, certolizumab, or natalizumab). On the basis of prior associations with narcotic use in the literature15,24 and clinical judgment, the following comorbidities were identified a priori by using ICD-9-CM diagnosis codes: fracture (800.xx–829.xx), trauma (830.xx–959.xx), anxiety (293.84, 300.0x, 313.0x), and depression (296.2x, 296.3x, 298.0x, 300.4x, 309.1x, 311.xx). Because psychiatric conditions may be underdiagnosed in children or under-ascertained by using diagnosis codes in claims data, prescriptions for anxiolytics (EphMRA/PBIRG code N05C0) and antidepressants (EphMRA/PBIRG codes N06A1, N06A4, N06A5, N06A9, N06C0) were also identified. Hydroxyzine was excluded from the anxiolytics category because it is not primarily used for this indication. Clomipramine and alprazolam are prescribed to treat anxiety disorders and were therefore moved from the EphMRA/PBIRG antidepressant category to anxiolytics. A summary variable for any psychological impairment was defined as any diagnosis or prescription for anxiety or depression.
312
Buckley et al
In an effort to identify additional comorbidities associated with chronic narcotic use, the 10 most common 4-digit ICD-9-CM codes among chronic narcotic users with and without IBD were determined. Comorbidities in this list that were not related to GI symptoms or routine medical care were subsequently included in analyses of predictors of chronic narcotic use among children with IBD.
Statistical Analyses Prevalence odds ratios (PORs) and 95% confidence intervals (CIs) comparing the prevalence of chronic narcotic use among children with IBD with their matched controls were estimated by using conditional logistic regression. Heterogeneity in the relationship between IBD status and chronic narcotic use by selected characteristics was investigated by obtaining stratified estimates from models including an interaction term between narcotic use and each characteristic. The 10 most common 4-digit ICD-9-CM diagnosis codes among chronic narcotic users were determined by summing the number of times a diagnosis code appeared in any diagnostic position (each diagnosis was counted only once per child). Variables associated with chronic narcotic use among children with IBD (overall and stratified by IBD type) were examined in crude and multivariate unconditional logistic regression models. Data analysis was performed in SAS v.9.2 (SAS Institute, Cary, NC).
Ethical Considerations This study used previously existing de-identified data and is therefore not considered human subjects research.
Results Of 4344 children with IBD during the study period, 63% had Crohn’s disease, and 37% had ulcerative colitis. There were more boys with IBD than girls, and the study population was composed predominantly of children aged 15–18 years (Table 1). Children with IBD had significantly increased healthcare utilization compared with age-, gender-, and region-matched members of the general population; 18.8% versus 3.1% had at least 1 inpatient visit, 61.9% versus 17.0% had 20 outpatient visits, and 33.4% versus 7.8% had 10 unique generic prescriptions (P < .0001 for all comparisons). Psychological comorbidity was common, with 18.7% of IBD patients having a psychological impairment compared with 10.8% of the general population (P < .0001). During the 2-year study period, 5.6% of children with IBD were chronic narcotic users. Children with IBD had more than twice the prevalence odds of chronic narcotic use as matched members of the general population (POR, 2.59; 95% CI, 2.21–3.04). In stratified analyses, PORs comparing chronic narcotic use among children with and
Clinical Gastroenterology and Hepatology Vol. 13, No. 2 Table 1. Characteristics of Children With IBD and Matched Members of the General Population, Commercial Claims and Encounters MarketScan Database, 2010 and 2011a General population without IBD
IBD Characteristic
n
Total (N) Age (y) 0–9 10–14 15–18 Gender Boys Girls Region Northeast North Central South West Comorbidities Acute pharyngitis Acute sinusitis/headache Anxiety Depression Fracture Joint pain Trauma Medication use Anxiolytics Antidepressants Any psychological impairmentb Inpatient visits 0 1 Outpatient visits 0–19 20 Unique generic prescriptions 0–9 10 IBD medication use Oral steroid Mesalamine drugs Azathioprine/6-mercaptopurine Rectal steroid Methotrexate Cyclosporine or tacrolimus Biologics
4344
%
n
%
21,720
443 1407 2494
10.2 32.4 57.4
2215 7035 12,470
10.2 32.4 57.4
2413 1931
55.6 44.5
12,065 9655
55.6 44.5
887 1289 1419 749
20.4 29.7 32.7 17.2
4435 6445 7095 3745
20.4 29.7 32.7 17.2
1471 1102 320 312 383 924 1537
33.9 25.4 7.4 7.2 8.8 21.3 35.4
6056 4111 901 1003 1740 3593 7389
27.9 18.9 4.2 4.6 8.0 16.5 34.0
193 527 814
4.4 12.1 18.7
542 1367 2355
2.5 6.3 10.8
3526 818
81.2 18.8
21,047 673
96.9 3.1
1654 2690
38.1 61.9
18,021 3699
83.0 17.0
2892 1452
66.6 33.4
20,036 1684
92.3 7.8
2007 3025 1578 96 316 15 1002
46.2 69.6 36.3 2.2 7.3 0.4 23.1
2591 12 9 0 0 0 0
11.9 0.1 0.0 0.0 0.0 0.0 0.0
a P values for differences between IBD and general population are P < .0001 for all variables except fracture (P ¼ .07), trauma (P ¼ .08), and matching variables (age, gender, and region; P ¼ 1.0). b Anxiety or depression diagnosis or prescription.
without IBD were higher among younger children and those with psychiatric diagnoses or medication use (Table 2). For example, the association between IBD status and chronic narcotic use was significantly stronger among children with IBD who also had psychological impairment (POR, 6.75; 95% CI, 4.29–10.62) than among those without (POR, 2.27; 95% CI, 1.91–2.70).
February 2015
Chronic Narcotic Use Among Children With IBD
313
Table 2. Chronic Narcotic Use Among Children With and Without IBD Stratified by Selected Characteristics, Commercial Claims and Encounters MarketScan Database, 2010 and 2011 Group
IBD, n (%)
General population, n (%)
PORa (95% CI)
Interaction P valueb
Overall Age (y) 0–9 10–14 15–18 Gender Boys Girls IBD type Crohn’s disease Ulcerative colitis Anxiety diagnosis Yes No Depression diagnosis Yes No Anxiolytics Yes No Antidepressants Yes No Any psychological impairmentc Yes No
241 (5.6)
489 (2.3)
2.59 (2.21–3.04)
11 (2.5) 28 (2.0) 202 (8.1)
7 (0.3) 49 (0.7) 433 (3.5)
8.18 (3.14–21.32) 2.91 (1.82–4.65) 2.46 (2.07–2.93)
.06
114 (4.7) 127 (6.6)
253 (2.1) 236 (2.4)
2.35 (1.87–2.95) 2.87 (2.29–3.59)
.2
153 (5.6) 88 (5.5)
287 (2.1) 202 (2.5)
2.82 (2.30–3.45) 2.28 (1.76–2.95)
.2
40 (12.5) 201 (5.0)
40 (2.5) 449 (2.2)
5.81 (3.65–9.24) 2.34 (1.97–2.78)
Prevalence of Chronic Narcotic Use Among Children With Inflammatory Bowel Disease Jessie P. Buckley,*,‡ Suzanne F. Cook,‡ Jeffery K. Allen,‡ and Michael D. Kappelman§ *Department of Epidemiology, University of North Carolina at Chapel Hill, Chapel Hill; ‡Worldwide Epidemiology, GlaxoSmithKline, Research Triangle Park; and §Department of Pediatrics, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina BACKGROUND & AIMS:
Narcotic analgesics are not recommended for long-term management of pain for patients with inflammatory bowel disease (IBD), particularly pediatric patients. We compared chronic use of narcotics among children with IBD and the general population and investigated factors associated with narcotic use in the pediatric IBD population.
METHODS:
This cross-sectional study included children (younger than 18 years old) with continuous enrollment in a large administrative claims database from 2010 through 2011 (n [ 4,911,286). Children with IBD were identified through diagnosis codes and dispensation of IBD medication (n [ 4344); they were matched for age, sex, and region with 5 children without IBD (n [ 21,720). Chronic narcotic use was defined as ‡3 dispensements of narcotics. We estimated prevalence odds ratios (PORs) and 95% confidence intervals (CIs), comparing narcotic use on the basis of IBD status and evaluating variables associated with narcotic use by patients with IBD by using conditional and unconditional logistic regression.
RESULTS:
The prevalence of chronic narcotic use was 5.6% among children with IBD vs 2.3% in the general population (POR, 2.6; 95% CI, 2.2–3.0). Compared with the general population, POR for chronic narcotic use was significantly higher for pediatric IBD patients with psychological impairment (POR, 6.8; 95% CI, 4.3–10.6) than those without (POR, 2.3; 95% CI, 1.9–2.7). Older age, increased healthcare utilization, fracture, and psychological impairment were strongly associated with chronic use of narcotics among children with IBD.
CONCLUSIONS:
Chronic narcotic use is common in pediatric IBD patients, particularly among those with anxiety and depression. Increased awareness of psychological comorbidity, screening, and treatment may reduce symptoms that lead to narcotic use and its complications.
Keywords: Crohn’s Disease; Ulcerative Colitis; UC; CD.
arcotics may be prescribed to patients with inflammatory bowel disease (IBD) for temporary pain relief such as to treat an acute flare or alleviate pain after surgery. However, long-term narcotic analgesia is not recommended because of gastrointestinal (GI) side effects, tolerance, and possibility of addiction.1–4 Furthermore, narcotic use among adult patients with IBD has been associated with increased risk of mortality, serious infection, sleep disorders, pneumonia, and psychiatric disorders.5–8 In children, chronic treatment with narcotics has additional potential adverse clinical sequelae such as the potential risks of growth failure, malnutrition, opportunistic infections, and hepatosplenic lymphoma.1 Although pain control is an important aspect of disease management in pediatric IBD, little is known about chronic narcotic use in children. In a large, primarily adult study of administrative claims data, nearly half of patients with IBD had at least 1 narcotic prescription
N
during a 2-year period, and 20% had 3 narcotic prescriptions.9 Referral center studies of adults with IBD have reported prevalence of narcotic use and dependency of 13% and 3%–4%, respectively.8,10,11 Factors associated with narcotic analgesia in adults include IBD-related surgery, polypharmacy, increased pain, more symptoms, worsened disease activity, and smoking.8–10,12,13 Associations between chronic pain, narcotic use, depression, and anxiety suggest that treating psychiatric comorbidity should be considered as an Abbreviations used in this paper: CI, confidence interval; EphMRA, European Pharmaceutical Market Research Association; GI, gastrointestinal; IBD, inflammatory bowel disease; ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; PBIRG, Pharmaceutical Business Intelligence and Research Group; POR, prevalence odds ratio. © 2015 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2014.07.057
February 2015
alternative to reliance on analgesics in adult patients.14 Although children with Crohn’s disease are at increased risk of anxiety and depression,15 the relationship between narcotic use and psychiatric conditions has not yet been assessed in the pediatric IBD population. Describing the characteristics of children with IBD with long-term narcotic use is important to define the magnitude of this problem in the pediatric population and to identify potential strategies or interventions to reduce chronic narcotic use. The objectives of this study were to compare the prevalence of chronic narcotic use among children with and without IBD and to identify variables associated with narcotic treatment in the pediatric IBD population.
Methods Study Design and Data Source This cross-sectional study used data from the Truven Health MarketScan Commercial Claims and Encounters database, an administrative claims database that has been used widely for health research including studies of IBD.15–18 The database includes approximately 500 million insurance claims on members of the working population younger than age 65 and their dependents who have employer-sponsored health insurance coverage from 1 of approximately 100 employer-sponsored and private health plans. Information includes longitudinal data on inpatient and outpatient diagnoses and procedures linked to demographic characteristics, enrollment details, and prescription claims records.
Sample Population The source population included all patients aged 18 years in the Commercial Claims and Encounters MarketScan database with continuous health plan enrollment and pharmacy benefits from January 1, 2010 to December 31, 2011 (n ¼ 4,911,286). Prevalent patients with IBD were identified by using an adaptation of a validated IBD case definition that has been applied previously.19–23 Children were classified as having IBD if they satisfied either of the following conditions: (1) three or more healthcare contacts for Crohn’s disease or ulcerative colitis (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 555.xx and 556.xx, respectively) on separate days or (2) at least 1 healthcare contact for Crohn’s disease or ulcerative colitis and at least 1 pharmacy claim for any of the following medications: mesalamine, olsalazine, balsalazide, sulfasalazine, 6mercaptopurine, azathioprine, methotrexate, enteral budesonide, or biologics (infliximab, adalimumab, certolizumab, or natalizumab). IBD type was classified as Crohn’s disease or ulcerative colitis on the basis of the majority of the last 9 claims. Patients who could not be classified as Crohn’s disease or ulcerative colitis (n ¼ 40) or who lived
Chronic Narcotic Use Among Children With IBD
311
in an “Unknown” region (n ¼ 34) were excluded. To focus on non-surgery narcotic use, children were also excluded if they had GI surgery during the study period (Current Procedural Terminology 4 codes 43020–43135, 43280–44346, 44500–45190, 45395–46505, 46700– 46999, and 49000–49999; n ¼ 684). At the time each child with IBD entered the study population, he or she was matched on age, gender, and region to 5 randomly sampled members of the study population without IBD.
Chronic Narcotic Use Narcotic analgesic use was ascertained from dispensed prescription drug claims. Medications coded by using the National Drug Code system were mapped to European Pharmaceutical Market Research Association (EphMRA) and the Pharmaceutical Business Intelligence and Research Group (PBIRG) Anatomical Classification System drug classes. Narcotic use was defined as a claim for EphMRA/PBIRG code N02A0, and participants were classified as chronic users if they had at least 3 narcotic drug claims during the 2-year study period.
Covariates of Interest Demographic information included age, gender, and region (Northeast, North Central, South, or West). The remaining covariates were classified by using participant data during the entire 2-year study period. Markers of healthcare utilization included the number of inpatient and outpatient contacts and the number of unique generic medications dispensed (excluding narcotic analgesics). IBD medications included oral steroids, oral and rectal aminosalicylates (including mesalamine, olsalazine, balsalazide, sulfasalazine), rectal steroids (enema, suppository, foam), 6-mercaptopurine or azathioprine, methotrexate, cyclosporine or tacrolimus, and biologics (infliximab, adalimumab, certolizumab, or natalizumab). On the basis of prior associations with narcotic use in the literature15,24 and clinical judgment, the following comorbidities were identified a priori by using ICD-9-CM diagnosis codes: fracture (800.xx–829.xx), trauma (830.xx–959.xx), anxiety (293.84, 300.0x, 313.0x), and depression (296.2x, 296.3x, 298.0x, 300.4x, 309.1x, 311.xx). Because psychiatric conditions may be underdiagnosed in children or under-ascertained by using diagnosis codes in claims data, prescriptions for anxiolytics (EphMRA/PBIRG code N05C0) and antidepressants (EphMRA/PBIRG codes N06A1, N06A4, N06A5, N06A9, N06C0) were also identified. Hydroxyzine was excluded from the anxiolytics category because it is not primarily used for this indication. Clomipramine and alprazolam are prescribed to treat anxiety disorders and were therefore moved from the EphMRA/PBIRG antidepressant category to anxiolytics. A summary variable for any psychological impairment was defined as any diagnosis or prescription for anxiety or depression.
312
Buckley et al
In an effort to identify additional comorbidities associated with chronic narcotic use, the 10 most common 4-digit ICD-9-CM codes among chronic narcotic users with and without IBD were determined. Comorbidities in this list that were not related to GI symptoms or routine medical care were subsequently included in analyses of predictors of chronic narcotic use among children with IBD.
Statistical Analyses Prevalence odds ratios (PORs) and 95% confidence intervals (CIs) comparing the prevalence of chronic narcotic use among children with IBD with their matched controls were estimated by using conditional logistic regression. Heterogeneity in the relationship between IBD status and chronic narcotic use by selected characteristics was investigated by obtaining stratified estimates from models including an interaction term between narcotic use and each characteristic. The 10 most common 4-digit ICD-9-CM diagnosis codes among chronic narcotic users were determined by summing the number of times a diagnosis code appeared in any diagnostic position (each diagnosis was counted only once per child). Variables associated with chronic narcotic use among children with IBD (overall and stratified by IBD type) were examined in crude and multivariate unconditional logistic regression models. Data analysis was performed in SAS v.9.2 (SAS Institute, Cary, NC).
Ethical Considerations This study used previously existing de-identified data and is therefore not considered human subjects research.
Results Of 4344 children with IBD during the study period, 63% had Crohn’s disease, and 37% had ulcerative colitis. There were more boys with IBD than girls, and the study population was composed predominantly of children aged 15–18 years (Table 1). Children with IBD had significantly increased healthcare utilization compared with age-, gender-, and region-matched members of the general population; 18.8% versus 3.1% had at least 1 inpatient visit, 61.9% versus 17.0% had 20 outpatient visits, and 33.4% versus 7.8% had 10 unique generic prescriptions (P < .0001 for all comparisons). Psychological comorbidity was common, with 18.7% of IBD patients having a psychological impairment compared with 10.8% of the general population (P < .0001). During the 2-year study period, 5.6% of children with IBD were chronic narcotic users. Children with IBD had more than twice the prevalence odds of chronic narcotic use as matched members of the general population (POR, 2.59; 95% CI, 2.21–3.04). In stratified analyses, PORs comparing chronic narcotic use among children with and
Clinical Gastroenterology and Hepatology Vol. 13, No. 2 Table 1. Characteristics of Children With IBD and Matched Members of the General Population, Commercial Claims and Encounters MarketScan Database, 2010 and 2011a General population without IBD
IBD Characteristic
n
Total (N) Age (y) 0–9 10–14 15–18 Gender Boys Girls Region Northeast North Central South West Comorbidities Acute pharyngitis Acute sinusitis/headache Anxiety Depression Fracture Joint pain Trauma Medication use Anxiolytics Antidepressants Any psychological impairmentb Inpatient visits 0 1 Outpatient visits 0–19 20 Unique generic prescriptions 0–9 10 IBD medication use Oral steroid Mesalamine drugs Azathioprine/6-mercaptopurine Rectal steroid Methotrexate Cyclosporine or tacrolimus Biologics
4344
%
n
%
21,720
443 1407 2494
10.2 32.4 57.4
2215 7035 12,470
10.2 32.4 57.4
2413 1931
55.6 44.5
12,065 9655
55.6 44.5
887 1289 1419 749
20.4 29.7 32.7 17.2
4435 6445 7095 3745
20.4 29.7 32.7 17.2
1471 1102 320 312 383 924 1537
33.9 25.4 7.4 7.2 8.8 21.3 35.4
6056 4111 901 1003 1740 3593 7389
27.9 18.9 4.2 4.6 8.0 16.5 34.0
193 527 814
4.4 12.1 18.7
542 1367 2355
2.5 6.3 10.8
3526 818
81.2 18.8
21,047 673
96.9 3.1
1654 2690
38.1 61.9
18,021 3699
83.0 17.0
2892 1452
66.6 33.4
20,036 1684
92.3 7.8
2007 3025 1578 96 316 15 1002
46.2 69.6 36.3 2.2 7.3 0.4 23.1
2591 12 9 0 0 0 0
11.9 0.1 0.0 0.0 0.0 0.0 0.0
a P values for differences between IBD and general population are P < .0001 for all variables except fracture (P ¼ .07), trauma (P ¼ .08), and matching variables (age, gender, and region; P ¼ 1.0). b Anxiety or depression diagnosis or prescription.
without IBD were higher among younger children and those with psychiatric diagnoses or medication use (Table 2). For example, the association between IBD status and chronic narcotic use was significantly stronger among children with IBD who also had psychological impairment (POR, 6.75; 95% CI, 4.29–10.62) than among those without (POR, 2.27; 95% CI, 1.91–2.70).
February 2015
Chronic Narcotic Use Among Children With IBD
313
Table 2. Chronic Narcotic Use Among Children With and Without IBD Stratified by Selected Characteristics, Commercial Claims and Encounters MarketScan Database, 2010 and 2011 Group
IBD, n (%)
General population, n (%)
PORa (95% CI)
Interaction P valueb
Overall Age (y) 0–9 10–14 15–18 Gender Boys Girls IBD type Crohn’s disease Ulcerative colitis Anxiety diagnosis Yes No Depression diagnosis Yes No Anxiolytics Yes No Antidepressants Yes No Any psychological impairmentc Yes No
241 (5.6)
489 (2.3)
2.59 (2.21–3.04)
11 (2.5) 28 (2.0) 202 (8.1)
7 (0.3) 49 (0.7) 433 (3.5)
8.18 (3.14–21.32) 2.91 (1.82–4.65) 2.46 (2.07–2.93)
.06
114 (4.7) 127 (6.6)
253 (2.1) 236 (2.4)
2.35 (1.87–2.95) 2.87 (2.29–3.59)
.2
153 (5.6) 88 (5.5)
287 (2.1) 202 (2.5)
2.82 (2.30–3.45) 2.28 (1.76–2.95)
.2
40 (12.5) 201 (5.0)
40 (2.5) 449 (2.2)
5.81 (3.65–9.24) 2.34 (1.97–2.78)
