Treatment patterns of pediatric patients with atopic dermatitis: A claims data analysis

Journal of the American Academy of Dermatology, March 2020, Volume 82, Issue 3, Pages 651-660

Abstract

Background

Real-world evidence on treatment patterns of pediatric patients with atopic dermatitis (AD) is sparse.

Objective

To assess current treatment patterns in pediatric AD patients.

Methods

Retrospective observational analysis of commercial insurance and Medicaid administrative claims data (January 2011-December 2016) for pediatric AD patients, stratified by age and provider type.

Results

The analytic sample comprised 607,258 pediatric AD patients. Median observation period was 30.3 months. Overall, 78.6% were prescribed ≥1 AD medication; 86.7% were prescribed topical corticosteroids, and 5.4% were prescribed a calcineurin inhibitor. Systemic corticosteroids (SCSs) were prescribed for 24.4% of patients, 51.8% of whom did not have asthma or allergic comorbidities. Of the 46.6% prescribed an antihistamine and 16.2% prescribed montelukast, 62.0% and 41.3%, respectively, did not have asthma or allergic comorbidities. Systemic immunosuppressants were rarely prescribed (<0.5%). Higher potency topical corticosteroid and SCS use increased with age. Treatment patterns varied by provider type; specialists were more likely to prescribe higher potency topicals and/or systemics, regardless of patient age. A minority of patients were treated by or referred to a specialist.

Limitations

Identification of AD patients relied on billing diagnoses; the disease severity was proxied by the treatment prescribed.

Conclusion

Results indicate that SCSs, despite known risks, and other medications with disproven efficacy in AD are frequently prescribed, suggesting a need for safer and more effective alternatives.

Key words: adolescents, atopic dermatitis, children, eczema, infants, systemic treatment, topical calcineurin inhibitors, topical corticosteroids, topical treatment.

Abbreviations used: AD-atopic dermatitis, FDA-Food and Drug Administration, SCS-systemic corticosteroid, TCI-topical calcineurin inhibitor, TCS-topical corticosteroid.

Capsule Summary

  • Pediatric atopic dermatitis treatments varied by provider type, with only a minority of patients ever being treated by a specialist.

  • Pediatric patients with atopic dermatitis commonly used drugs without proven efficacy, as well as agents with an unfavorable safety profile, suggesting the need for safe and effective alternatives for long-term disease control.

Atopic dermatitis (AD) is the most common chronic pediatric inflammatory skin disease. 1 2 This disease presents within the first year of life in most patients and before 5 years of age in 85%-95% of cases. 3 4 Given the early onset, primary care physicians, especially pediatricians, play a critical role in initial management and subsequent subspecialist referrals. 5

Optimal AD management focuses on maintenance skin care and topical medications to minimize flares. 6 First-line topical prescription options include corticosteroids (TCSs), calcineurin inhibitors (TCIs), and a phosphodiesterase 4 inhibitor (crisaborole, Food and Drug Administration [FDA]–approved in December 2016). Skin care regimens are time-consuming and can be confusing, requiring education and monitoring. 7

Off-label treatment, especially with systemic immunosuppressants and systemic (oral and injectable) corticosteroids (SCSs), are frequently prescribed for patients unresponsive to topical therapy. 8 Until recently, the only FDA-approved systemic therapy for pediatric patients with AD was oral corticosteroids, despite potential adverse effects and the known tendency to rebound. 9 In March 2019, the FDA approved dupilumab, an interleukin-4 receptor α antagonist, to treat patients aged ≥12 years with moderate-to-severe AD inadequately controlled by topical prescription treatments or when those therapies are not advisable.

Published literature is scant on current pediatric-specific practice patterns. 10 The objective of this study was to assess current treatment patterns in a large population of insured pediatric patients with AD to identify opportunities for improved management.

Methods

Study design, data, and patient population

This study used a retrospective observational design. Combined administrative medical and pharmacy claims from the IBM MarketScan Commercial Database (Armonk, NY) and the multistate Medicaid database covering January 01, 2011-December 31, 2016, were analyzed. Both databases include de-identified patient-level claims that are compliant with the Health Insurance Portability and Accountability Act.

Patients with ≥1 claim with a diagnosis of AD (International Classification of Diseases, 9th and 10th revision, codes 691.8 and L20.x, respectively) were included. The index date was defined as the first observed medical claim for AD. Other inclusion criteria were age <18 years on the index date and continuous health plan eligibility 6 months before the index date (baseline period, up to 6 months for infants [ie, children <1 year of age]) and ≥12 months after the index date. Patients who received a diagnosis of an immune-mediated inflammatory disorder during the baseline period or on the index date were excluded to avoid ambiguity for medications used for these conditions rather than AD. The observation period spanned the patients' index date to the first of 2 potential endpoints, the end of health plan continuous eligibility or the data cut-off date.

Outcomes

The following medications were considered to be prescribed most likely to treat AD: TCSs, TCIs, antihistamines (topical and oral), montelukast sodium, SCSs, immunosuppressants (azathioprine, cyclosporine A, methotrexate, mycophenolate mofetil, interferon γ), intravenous immunoglobulin, and phototherapy. Potential use for related conditions was explored in a subset analysis excluding patients with diagnostic codes for asthma and allergies. Although topical and oral antibiotics are often prescribed for infected AD, these medications are also used for many unrelated, common childhood infections, so antibiotics were not included in the analysis. Crisaborole and dupilumab were not included because their approval did not coincide with the period of data availability. Class of AD medications received, total number of prescriptions filled, and combination therapies (overlap ≥3 months between ≥2 distinct AD treatments) were assessed.

A proxy for AD severity was developed by using AD treatment regimens. TCSs were classified by potency class (highest 1 to lowest 7). 11 The proxy for severity level 1 included a class 5-7 TCS or TCI used alone. The proxy for severity level 2 included a class 1-4 TCS or TCI used with other AD therapies. Finally, the proxy for severity level 3 included an SCS, another immunosuppressant, an intravenous immunoglobulin, and phototherapy. Patients were categorized on the basis of treatment with the highest potency received. Untreated patients were categorized in the proxy for severity level 1.

Statistical analyses

AD treatment patterns were reported among treated patients. Results were stratified by age group (<2, 2-5, 6-11, and 12-17 years) and by type of health care provider on the index date (dermatologist, allergist/immunologist, pediatrician, and other provider types [mainly primary care providers, family practices, acute care providers, and nurse practitioners]). Analyses were performed by using SAS version 9.4 (SAS Institute, Cary, NC).

Results

Patient characteristics

A total of 607,258 children were identified. Patients with Medicaid coverage represented 44.2% of the study population. Overall, 29.2% of patients had ≥1 atopic comorbidity, including allergic rhinitis (17.4%), asthma (12.1%), and allergic conjunctivitis (7.0%). Atopic comorbidities were more commonly observed among patients of severity level 3 (48.4%) than severity level 1 (23.4%) or 2 (25.6%) (results not shown). Patients were observed for a median of 30.3 months ( Table I ).

Table I
Demographic and clinical characteristics of patients with AD
Characteristic Value, n = 607,258
Demographics
Age, y, mean ± SD; median 5.4 ± 5.0; 4.0
Sex, male 307,992 (50.7)
Commercial insurance plan type 338,678 (55.8)
Medicaid coverage 268,580 (44.2)
Comorbidities during the 6-month baseline period
Combined Charlson-Elixhauser comorbidity index, mean ± SD; median 0.2 ± 0.4; 0
Patients with ≥1 comorbidity 289,104 (47.6)
Patients with ≥1 atopic comorbidity 177,212 (29.2)
Allergic conjunctivitis 42,405 (7.0)
Allergic rhinitis 105,537 (17.4)
Allergic urticaria 4798 (0.8)
Asthma 73,351 (12.1)
Chronic rhinosinusitis 9693 (1.6)
Eosinophilic esophagitis 471 (0.1)
Food allergy 9572 (1.6)
Nasal polyps 159 (0)
Patients with ≥1 other comorbidity 157,872 (26.0)
Autoimmune disorders 1920 (0.3)
Bacterial infections 20,007 (3.3)
Fungal infections 80,410 (13.2)
Obesity 10,257 (1.7)
Psychiatric comorbidities 29,511 (4.9)
Anxiety 6614 (1.1)
Attention deficit hyperactivity disorder 19,978 (3.3)
Depression 6804 (1.1)
Sleep disorders 1582 (0.3)
Viral infections and disorders 36,856 (6.1)
Patients who received ≥1 treatment for AD during the observation period 477,484 (78.6)
Proxy for AD severity during the observation period
Severity level 1 261,335 (43.0)
Severity level 2 224,864 (37.0)
Severity level 3 121,059 (19.9)
Duration of observation period, mon, mean ± SD; median 33.6 ± 16.1; 30.3
Values are n (%) unless stated otherwise.
AD , Atopic dermatitis; SD , standard deviation; TCI , topical calcineurin inhibitor; TCS , topical corticosteroids.

The baseline period includes the index date. For infants, the baseline period includes up to 6 months of follow-up.

Patients considered in the analysis of treatment patterns.

Patients were categorized into 3 levels on the basis of the highest potency treatment received. Level 1 was defined as no treatment for AD, TCS class 7 (least potent), TCS class 6 (mild), TCS class 5 (lower midstrength), or TCI used alone. Level 2 was defined as TCS class 4 (midstrength), TCS class 3 (upper midstrength), TCS class 2 (potent), TCS class 1 (super potent), and TCI (when used with any other treatment for AD). Level 3 was defined as systemic corticosteroids, azathioprine, cyclosporine A, methotrexate, mycophenolate mofetil, interferon γ, intravenous immunoglobulin, and phototherapy.

Treatment patterns

Most pediatric patients with AD diagnoses received ≥1 medication for AD during their observation period (78.6%; Table I ). Patients filled a mean (standard deviation) of 2.7 (3.7) AD prescriptions annually; just under one third (30.1%) received combination AD therapy ( Table II ). Most patients (86.7%) received ≥1 prescription topical medication, mainly low- (class 6-7, 36.5%) and medium- (class 3-5, 63.8%) potency TCSs. Antihistamines were the most commonly used systemic treatments (46.6%) and 16.2% received montelukast. SCSs were prescribed in a quarter (24.4%) of the pediatric patients. Use of immunosuppressants (0.2%) and phototherapy (0.2%) was marginal. Comorbid asthma or allergies were not present in most patients receiving antihistamines (62.0%) or SCSs (51.8%) and were present in 41.3% of those prescribed montelukast sodium.

Table II
Treatment patterns of patients with AD for entire observation period
Treatment category Value, n = 477,484
Total filled prescriptions for AD treatment per year, mean ± SD; median 2.7 ± 3.7; 1.3
Combination therapy with ≥2 distinct AD treatments 143,614 (30.1)
Topical treatments 413,965 (86.7)
Antihistamines 24 (0)
Corticosteroids
Any potency 410,134 (85.9)
Low 174,444 (36.5)
Medium 304,462 (63.8)
High 70,083 (14.7)
Calcineurin inhibitors 25,577 (5.4)
Systemic treatments 297,313 (62.3)
Antihistamines 222,328 (46.6)
Corticosteroids 116,635 (24.4)
Any immunosuppressants 1106 (0.2)
Azathioprine 147 (0)
Cyclosporine A 252 (0.1)
Interferon γ 1 (0)
Methotrexate 600 (0.1)
Mycophenolate mofetil 276 (0.1)
Intravenous immunoglobulin 5123 (1.1)
Montelukast sodium 77,529 (16.2)
Phototherapy 938 (0.2)
Values are n (%) unless stated otherwise.
AD , Atopic dermatitis; SD , standard deviation.

The percentage of patients without comorbid asthma or allergies was 62.0% among patients prescribed systemic antihistamines, 51.8% among patients prescribed systemic corticosteroids, and 41.3% among patients prescribed montelukast sodium.

Stratifications

Patients across age groups had similar comorbid profiles ( Table III ). Overall, younger patients were the least likely to have seen a dermatologist on the index date ( Table III ). Proportions of patients with severity level 3 increased with age (infants 10.0%, adolescents [ie, children aged 12-17 years] 34.5%). The frequency of dispensed AD medications was consistent across all age groups (median 1.3-1.4 annually), although specific treatments varied ( Table IV ). Notable differences were a positive relationship between TCS potency and age (infants 50.9% low and 10.1% high potency, adolescents 22.6% low and 24.3% high potency). Use of SCSs also increased with age (infants 12.3%, adolescents 40.7%). Regardless of age, immunosuppressants and phototherapy were very rarely used.

Table III
Demographic and clinical characteristics of patients with AD, stratified by age group and provider type
Characteristic Age group, y Provider type
0-1, n = 195,591 (32.2% of patients) 2-5, n = 152,270 (25.1% of patients) 6-11, n = 165,113 (27.2% of patients) 12-17, n = 94,284 (15.5% of patients) Dermatologist, n = 71,899 (11.8% of providers) Allergist/immunologist, n = 39,238 (6.5% of providers) Pediatrician, n = 233,342 (38.4% of providers) Other provider, n = 262,779 (43.3% of providers)
Demographics
Age, mean ± SD; median 0.3 ± 0.4; 0 3.4 ± 1.1; 3.0 8.3 ± 1.7; 8.0 14.2 ± 1.7; 14.0 7.8 ± 5.3; 8.0 6.0 ± 4.5; 5.0 4.3 ± 4.7; 3.0 5.6 ± 5.1; 5.0
Sex, male 109,240 (55.9) 77,052 (50.6) 81,051 (49.1) 40,649 (43.1) 34,035 (47.3) 21,910 (55.8) 120,382 (51.6) 131,665 (50.1)
Commercial insurance plan type 106,255 (54.3) 82,290 (54.0) 91,400 (55.4) 58,733 (62.3) 63.354 (88.1) 31,626 (80.6) 166,009 (71.1) 77,689 (29.6)
Medicaid coverage 89,336 (45.7) 69,980 (46.0) 73,713 (44.6) 35,551 (37.7) 8545 (11.9) 7612 (19.4) 67,333 (28.9) 185,090 (70.4)
Comorbidities during 6-month baseline period
Combined Charlson-Elixhauser comorbidity index, mean ± SD; median 0.1 ± 0.3; 0 0.2 ± 0.4; 0 0.2 ± 0.4; 0 0.2 ± 0.5; 0 0.1 ± 0.3; 0 0.4 ± 0.5; 0 0.1 ± 0.4; 0 0.2 ± 0.5; 0
Patients with ≥1 comorbidity 76,992 (39.4) 74,496 (48.9) 91,206 (55.2) 46,410 (49.2) 33,484 (46.6) 30,615 (78.0) 93,171 (39.9) 131,834 (50.2)
Patients with ≥1 atopic comorbidity 31,357 (16.0) 54,410 (35.7) 62,762 (38.0) 28,683 (30.4) 13,678 (19.0) 28,992 (73.9) 53,340 (22.9) 81,202 (30.9)
Allergic conjunctivitis 13,716 (7.0) 13,225 (8.7) 11,147 (6.8) 4317 (4.6) 3417 (4.8) 5547 (14.1) 15,372 (6.6) 18,069 (6.9)
Allergic rhinitis 10,589 (5.4) 32,263 (21.2) 43,273 (26.2) 19,412 (20.6) 7078 (9.8) 22,930 (58.4) 27,494 (11.8) 48,035 (18.3)
Allergic urticaria 1269 (0.6) 1589 (1.0) 1301 (0.8) 639 (0.7) 312 (0.4) 1716 (4.4) 965 (0.4) 1805 (0.7)
Asthma 7308 (3.7) 21,965 (14.4) 30,296 (18.3) 13,782 (14.6) 5315 (7.4) 12,324 (31.4) 19,525 (8.4) 36,187 (13.8)
Chronic rhinosinusitis 1458 (0.7) 3301 (2.2) 3088 (1.9) 1846 (2.0) 1067 (1.5) 1572 (4.0) 2582 (1.1) 4472 (1.7)
Eosinophilic esophagitis 46 (0) 139 (0.1) 183 (0.1) 103 (0.1) 46 (0.1) 160 (0.4) 80 (0) 185 (0.1)
Food allergy 3035 (1.6) 3093 (2.0) 2586 (1.6) 858 (0.9) 463 (0.6) 3899 (9.9) 1988 (0.9) 3222 (1.2)
Nasal polyps 6 (0) 21 (0) 59 (0) 73 (0.1) 18 (0) 36 (0.1) 39 (0) 66 (0)
Patients with ≥1 other comorbidity 55,916 (28.6) 31,079 (20.4) 44,858 (27.2) 26,019 (27.6) 25,080 (34.9) 6237 (15.9) 52,693 (22.6) 73,862 (28.1)
Autoimmune disorders 194 (0.1) 412 (0.3) 734 (0.4) 580 (0.6) 652 (0.9) 90 (0.2) 376 (0.2) 802 (0.3)
Bacterial infections 5446 (2.8) 6298 (4.1) 5735 (3.5) 2528 (2.7) 3439 (4.8) 984 (2.5) 7286 (3.1) 8298 (3.2)
Fungal infections 49,291 (25.2) 13,130 (8.6) 10,808 (6.5) 7181 (7.6) 9617 (13.4) 2223 (5.7) 29,985 (12.9) 38,585 (14.7)
Psychiatric comorbidities 819 (0.4) 2663 (1.7) 15,258 (9.2) 10,771 (11.4) 2892 (4.0) 1658 (4.2) 7802 (3.3) 17,159 (6.5)
Anxiety 80 (0) 818 (0.5) 2771 (1.7) 2945 (3.1) 1016 (1.4) 505 (1.3) 1972 (0.8) 3121 (1.2)
Attention deficit hyperactivity disorder 35 (0) 1469 (1.0) 12,370 (7.5) 6104 (6.5) 1641 (2.3) 1126 (2.9) 5586 (2.4) 11,625 (4.4)
Depression 705 (0.4) 492 (0.3) 1730 (1.0) 3877 (4.1) 687 (1.0) 264 (0.7) 1112 (0.5) 4741 (1.8)
Obesity 312 (0.2) 1532 (1.0) 4737 (2.9) 3676 (3.9) 658 (0.9) 445 (1.1) 2901 (1.2) 6253 (2.4)
Sleep disorders 163 (0.1) 346 (0.2) 629 (0.4) 444 (0.5) 98 (0.1) 98 (0.2) 403 (0.2) 983 (0.4)
Viral infections and disorders 3987 (2.0) 12,452 (8.2) 15,086 (9.1) 5331 (5.7) 11,918 (16.6) 1510 (3.8) 9755 (4.2) 13,673 (5.2)
Provider type on index date
Dermatologist 11,390 (5.8) 16,184 (10.6) 23,570 (14.3) 20,755 (22.0) - - - -
Allergist/immunologist 7296 (3.7) 12,957 (8.5) 13,513 (8.2) 5472 (5.8) - - - -
Pediatrician 95,717 (48.9) 58,820 (38.6) 54,461 (33.0) 24,344 (25.8) - - - -
Other 81,188 (41.5) 64,309 (42.2) 73,569 (44.6) 43,713 (46.4) - - - -
Patients with ≥1 visit with dermatologist or allergist/immunologist after index visit 33,046 (46.0) 20,061 (51.1) 31,516 (13.5) 13,144 (5.0)
Patient age group, y
0-1 - - - - 6719 (59.0) 4139 (56.7) 14,230 (14.9) 3874 (4.8)
2-5 - - - - 7405 (45.8) 6382 (49.3) 7288 (12.4) 3119 (4.9)
6-11 - - - - 9819 (41.7) 6860 (50.8) 6620 (12.2) 3415 (4.6)
12-17 - - - - 9103 (43.9) 2680 (49.0) 3378 (13.9) 2736 (6.3)
Patients who received ≥1 treatment for AD during observation period 145,683 (74.5) 120,070 (78.9) 133,472 (80.8) 78,259 (83.0) 58,758 (81.7) 29,994 (76.4) 169,455 (72.6) 219,277 (83.4)
Proxy for AD severity during the observation period §
Severity level 1 104,428 (53.4) 67,066 (44.0) 61,754 (37.4) 28,087 (29.8) 26,062 (36.2) 15,730 (40.1) 115,465 (49.5) 104,078 (39.6)
Severity level 2 71,613 (36.6) 59,726 (39.2) 59,828 (36.2) 33,697 (35.7) 29,087 (40.5) 10,061 (25.6) 80,698 (34.6) 105,018 (40.0)
Severity level 3 19,550 (10.0) 25,478 (16.7) 43,531 (26.4) 32,500 (34.5) 16,750 (23.3) 13,447 (34.3) 37,179 (15.9) 53,683 (20.4)
Duration of observation period, mon, mean ± SD; median 32.9 ± 16.6; 28.7 34.9 ± 16.2; 32.5 33.9 ± 15.8; 31.2 32.1 ± 15.1; 28.9 34.1 ± 16.1; 31.3 32.1 ± 15.6; 28.4 33.1 ± 16.0; 29.5 34.1 ± 16.2; 31.3
Values are n (%) unless stated otherwise.
AD , Atopic dermatitis; SD , standard deviation; TCI , topical calcineurin inhibitor; TCS , topical corticosteroids.

The baseline period includes the index date. For infants (ie, children ≤1 year of age), the baseline period includes up to 6 months of follow-up.

Percentages for the age groups are calculated out of the total number of patients in the corresponding age group.

Patients considered in the analysis of treatment patterns.

§ Patients are categorized into 3 levels on the basis of highest potency treatment received. Level 1 was defined as no treatment for AD, TCS class 7 (least potent), TCS class 6 (mild), TCS class 5 (lower midstrength), or TCI used alone. Level 2 was defined as TCS class 4 (midstrength), TCS class 3 (upper midstrength), TCS class 2 (potent), TCS class 1 (super potent), and TCI (when used with any other treatment for AD). Level 3 was defined as systemic corticosteroids, azathioprine, cyclosporine A, methotrexate, mycophenolate mofetil, interferon γ, intravenous immunoglobulin, and phototherapy.

Table IV
Treatment patterns of patients with AD for entire observation period, stratified by age group and provider type
Treatment category Age group, y Provider type
0-1, n = 145,683 2-5, n = 120,070 6-11, n = 133,472 12-17, n = 78,259 Dermatologist, n = 58,758 Allergist/immunologist,
n = 29,994
Pediatrician, n = 169,455 Other provider, n = 219,277
Total no. filled prescriptions for AD treatment per year, mean ± SD; median 2.4 ± 3.1; 1.3 2.7 ± 3.8; 1.3 2.9 ± 4.1; 1.4 2.6 ± 3.7; 1.3 2.4 ± 3.4; 1.2 4.2 ± 5.2; 2.2 2.2 ± 3.0; 1.1 2.9 ± 3.9; 1.5
Combination therapy with ≥2 distinct AD treatments 35,611 (24.4) 38,042 (31.7) 45,092 (33.8) 24,869 (31.8) 16,881 (28.7) 11,273 (37.6) 40,085 (23.7) 75,375 (34.4)
Topical treatments 129,920 (89.2) 103,829 (86.5) 113,325 (84.9) 66,891 (85.5) 53,780 (91.5) 21,757 (72.5) 149,268 (88.1) 189,160 (86.3)
Topical antihistamines 8 (0) 5 (0) 7 (0) 4 (0) 0 (0) 0 (0) 2 (0) 22 (0)
Topical corticosteroids
Any 129,515 (88.9) 102,953 (85.7) 111,826 (83.8) 65,840 (84.1) 52,428 (89.2) 21,430 (71.4) 148,309 (87.5) 187,967 (85.7)
Low potency 74,122 (50.9) 43,820 (36.5) 38,837 (29.1) 17,665 (22.6) 20,139 (34.3) 7913 (26.4) 63,210 (37.3) 83,182 (37.9)
Medium potency 90,466 (62.1) 77,545 (64.6) 86,126 (64.5) 50,325 (64.3) 38,597 (65.7) 17,001 (56.7) 109,289 (64.5) 139,575 (63.7)
High potency 14,771 (10.1) 15,052 (12.5) 21,249 (15.9) 19,011 (24.3) 16,539 (28.1) 4347 (14.5) 19,538 (11.5) 29,659 (13.5)
TCIs 5157 (3.5) 6171 (5.1) 8217 (6.2) 6032 (7.7) 7843 (13.3) 2344 (7.8) 6809 (4.0) 8581 (3.9)
Systemic treatments 83,566 (57.4) 77,036 (64.2) 86,829 (65.1) 49,882 (63.7) 30,123 (51.3) 23,283 (77.6) 91,502 (54.0) 152,405 (69.5)
Systemic antihistamines 69,659 (47.8) 60,430 (50.3) 61,226 (45.9) 31,013 (39.6) 17,195 (29.3) 13,146 (43.8) 63,904 (37.7) 128,083 (58.4)
Systemic corticosteroids 17,914 (12.3) 24,230 (20.2) 42,616 (31.9) 31,875 (40.7) 15,998 (27.2) 13,207 (44.0) 35,522 (21.0) 51,908 (23.7)
Any systemic immunosuppressants 124 (0.1) 218 (0.2) 399 (0.3) 365 (0.5) 289 (0.5) 75 (0.3) 263 (0.2) 479 (0.2)
IVIG 1851 (1.3) 1379 (1.1) 1096 (0.8) 797 (1.0) 688 (1.2) 351 (1.2) 1936 (1.1) 2148 (1.0)
Montelukast sodium § 16,895 (11.6) 23,125 (19.3) 26,608 (19.9) 10,901 (13.9) 8374 (14.3) 11,049 (36.8) 23,756 (14.0) 34,350 (15.7)
Phototherapy 62 (0) 184 (0.2) 346 (0.3) 346 (0.4) 363 (0.6) 80 (0.3) 173 (0.1) 322 (0.1)
Values are n (%) unless stated otherwise.
AD , Atopic dermatitis; IVIG , intravenous immunoglobulin; SD , standard deviation; TCI , topical calcineurin inhibitor.

The percentage of patients without comorbid asthma or allergies was 80.9% among 0–1-year-olds, 55.2% among 2–5-year-olds, 49.3% among 6–11-year-olds, 57.7% among 12–17-year-olds, 74.9% among dermatologists, 23.6% among allergists/immunologists, 67.9% among pediatricians, and 61.2% among other provider types.

The percentage of patients without comorbid asthma or allergies was 77.0% among 0–1-year-olds, 41.4% among 2–5-year-olds, 42.0% among 6–11-year-olds, 58.9% among 12–17-year-olds, 69.8% among dermatologists, 15.2% among allergists/immunologists, 58.7% among pediatricians, and 51.0% among other provider types.

Immunosuppressants included azathioprine, cyclosporine A, interferon γ, methotrexate, and mycophenolate mofetil.

§ The percentage of patients without comorbid asthma or allergies was 72.1% among 0–1-year-olds, 36.0% among 2–5-year-olds, 29.0% among 6–11-year-olds, 34.7% among 12–17-year-olds, 59.0% among dermatologists, 13.5% among allergists/immunologists, 52.1% among pediatricians, and 38.4% among other provider types.

Referral to a dermatologist or allergist/immunologist by pediatricians and other providers was infrequent ( Table III ). Only 13.5% of patients had a visit with a dermatologist or allergist/immunologist after seeing a pediatrician on the index date and 5.0% after seeing a provider other than a pediatrician.

Use of high-potency TCSs was greatest among patients initially seen by a dermatologist (28.1%; 11.5%-14.5% among other provider categories), as was use of TCIs (13.3%; 3.9%-7.8% among other provider categories) ( Table IV ). Conversely, use of oral antihistamines was lowest among those initially seen by a dermatologist (29.3%; 37.7%-58.4% among other provider categories). Use of SCSs (44.0%; 21.0%-27.2% among other provider categories) and montelukast sodium (36.8%; 14.0%-15.7% among other provider categories) were highest among patients initially seen by an allergist/immunologist. Immunosuppressants and phototherapy were rarely prescribed, regardless of provider type.

Among patients treated by allergists/immunologists, comorbid asthma or allergies were present in most patients receiving antihistamines (76.4%), SCSs (84.8%), and montelukast sodium (86.5%); in contrast, these comorbidities were present in the minority of patients treated by dermatologists, pediatricians, and other providers receiving these treatments.

Discussion

This study provides a comprehensive overview of medications prescribed for more than half a million pediatric patients with AD and characterizes their clinical management. Key differences in how pediatric patients with AD are treated across provider types were observed. Furthermore, only 1 in 4 patients were initially evaluated or subsequently referred to a specialist (dermatologist or allergist/immunologist) by a pediatrician or other provider type. In addition to insurance-restricted access, shortages of pediatric dermatologists in the United States, particularly outside of metropolitan areas, might be a contributing factor in this low observed rate of dermatology clinical visits. 12 13 This database included 44.2% pediatric patients insured by Medicaid, a group most severely affected by access to specialist care, with dermatology among the most restricted. 14 15 16 17 18

With an observed utilization >85%, topical treatments were the mainstay for the management of AD, consistent with current guideline recommendations. 6 TCSs were most widely used and were dispensed to 85.9% of patients. These percentages are higher than those recently reported for female patients (by 72.0%) and male patients (by 56.7%) aged 0-19 years who saw a dermatologist for a primary diagnosis of AD. 19 Likewise, a US National Ambulatory Medical Care Survey (2003-2012)–based analysis of visits for AD indicated that ∼61% of patients aged 0-15 years received a TCS. 20 The probable explanation for this higher reported utilization of TCSs is the longer patient follow-up.

Although low-potency and medium-potency TCSs were equally prescribed by all types of providers, use of high-potency TCSs were more common among patients initially seen by dermatologists. This finding supports greater comfort and familiarity with prescribing higher potency agents among dermatologists. 20 Infants and young children were also least likely to receive high-potency TCSs and to see a dermatologist. Primary care physicians are often hesitant to prescribe stronger than low-potency TCSs, even for flaring AD.

The observed utilization of TCIs by 13.3% of patients seen by a dermatologist is consistent with that reported in a recent analysis of the US National Ambulatory Medical Care Survey data for patients <15 years old. 20 Of note, this proportion was 3 times higher than that observed among patients cared for by providers other than allergists/immunologists or pediatricians. As for TCSs, dermatologists might have more comfort with the risk-benefit profile of this class of treatments when used as prescribed. 10

There was widespread use of medications not recommended by published AD guidelines, consistent with a previously published observational study. 21 Oral antihistamines were the most commonly used agents after TCSs, despite concerns over adverse effects, such as undesired sedation, dry mouth, and blurred vision, which might be particularly detrimental for school-aged children. 21 22 23 Furthermore, a potential association between sedating antihistamine use in AD patients and attention deficit hyperactivity disorder was recently identified in a single report. 23 There is a lack of evidence supporting their benefit in AD, although physicians might be prescribing them for the soporific effect. 22 24 25 Moreover, this widespread use could be partially explained by the documented misperception of parents that antihistamines are useful in reducing AD-related itch. 19 Similar results were observed with montelukast, despite the absence of evidence supporting its efficacy in patients with AD 26 27 28 and potential risk of neuropsychiatric adverse effects in children, including depression, aggression, and nightmares. 29

A limitation of our study design was the inherent inability to assess the widespread use of oral antibiotics to treat AD, rather than unrelated infections. Furthermore, over-the-counter AD treatments, such as emollients and topical 1% hydrocortisone, typically used for maintenance therapy for milder forms of AD, were not captured in the database. The use of these treatments could help explain why 21.4% of patients did not have any filled prescriptions for AD medication. These data are in line with a recently published study, in which >30% of patients did not report use of any topical medications within the previous week. 30

Until the approval of dupilumab in March 2019 to treat patients aged ≥12 years with moderate-to-severe AD inadequately controlled by topical prescription medications and those for whom these topical medications are not advisable, corticosteroids were the only FDA-approved systemic medication for AD in children. However, guidelines and a consensus from the International Eczema Council recommended against this approach, with perhaps the exception of short-term transitional administration for acute, severe exacerbations. 22 24 25 31 Long-term utilization of SCSs has been associated with rebound worsening after treatment discontinuation and side effects, including growth suppression in children, osteoporosis, osteonecrosis, adrenal insufficiency, Cushing syndrome, hypertension, glucose intolerance, and diabetes, to name a few. 9 Yet, in the present study 24.4% of patients had ≥1 filled prescription for a SCS during the observation period (12.6% when excluding patients with concomitant asthma or allergies).

The prevalence of SCS use was highest for adolescent patients (40.7%) and those treated by allergists/immunologists (44.0%) and dermatologists (27.2%). A retrospective study involving national survey data acquired during 1997-2004 reported SCS rates of 17% for all AD-related visits, excluding those for patients with concomitant asthma or allergies. However, this visit-level rate is difficult to compare with patient-level results from the current analysis. 32 The previously reported utilization rate among US clinical practices was lower, 14 28 likely because of the longer average observation period in the present study. The high prevalence of SCS use in pediatric patients with AD emphasizes the need for safe and effective therapies to treat the disease.

A survey among North American pediatric dermatologists cited cyclosporine, methotrexate, and mycophenolate mofetil as the systemic first-line treatments of choice for AD, not systemic steroids. 33 Yet, filled prescriptions for systemic immunosuppressants other than corticosteroids were marginal across all age groups and provider types. At <1%, the utilization of immunosuppressants in the present study is less than expected for patients in the most severe group, previously documented at around 7%. 34 Despite the safety concerns limiting their acceptance, for many patients with severe AD, the benefits of immunosuppressants often outweigh the risks. Still, a recent survey of adult patients with moderate-to-severe AD treated with systemic agents, including immunosuppressants, indicated that the lack of disease control was a concern, with patients reporting a substantial number of, prolonged duration of, and recurrent flares.

Reference: Silverberg JI, Ghorayeb E, Chen Z. A real-world study evaluating adequacy of existing systemic treatments for patients with moderate-to-severe atopic dermatitis (AD-QUEST): 6-month survey data on flares. Presented at the International Investigative Dermatology/Society for Investigative Dermatology Annual Meeting (IID/SID), May 16–19, 2018, Orlando, Florida.

One key challenge with the identification of patients with AD in administrative health care claims stems from the fact that a variety of International Classification of Diseases diagnostic codes have been assigned to AD, eczema, or dermatitis and a clinician's choice of code might vary for a patient seen for AD. Thus, restricting the sample to those with a specific diagnostic code for AD might inadvertently exclude some true AD patients who were misdiagnosed. However, the objective here was not to measure the prevalence of pediatric AD, but rather to characterize treatment patterns in patients with AD. Thus, a conservative approach in selecting patients with a specific diagnosis of AD was favored. A sensitivity analysis of a broader set of codes for AD and AD-related conditions (dyshidrosis, other erythematosquamous dermatosis, infective dermatitis, other specified dermatitis, dermatitis unspecified, contact dermatitis and other eczema, unspecified cause, and nummular dermatitis) yielded similar results, supporting our findings.

Additional limitations include the fact that diagnostic codes used for billing might reflect suspected rather than confirmed clinical diagnoses. AD severity was defined on the basis of treatments received rather than on clinical data. As such, the proportion of patients with higher severity might be underestimated due to reluctance of health care providers to treat with more potent therapies. Last, claims data only provide information on filled prescriptions, not medications actually used.

Conclusion

Despite high prevalence and disease burden of pediatric AD in the United States, there is no consensus on a uniform approach to disease management. Results from this claims data analysis indicate that most patients were not seen by a specialist; moreover, treatment varied greatly by provider type, with specialists more likely to treat with higher potency topicals and/or systemics. Oral antihistamines and montelukast were frequently prescribed for patients with AD without asthma and allergic comorbidities, especially in younger children, despite their lack of demonstrated efficacy. Systemic immunosuppressants were rarely prescribed, despite the potential benefit of these drugs for severe AD patients. Instead, SCSs were far more commonly prescribed, especially for adolescents, despite their known risks. These results emphasize the need for effective therapies that are well tolerated by pediatric patients for long-term control of the disease.

Writing and editorial assistance in the preparation of this article was provided by Éric Gravel from StatLog Inc and was payed for by Regeneron and Sanofi.

Dr Paller and Dr Siegfried contributed equally to the data analysis, interpretation, drafting, and revising of this article.

Funding sources: Supported by Regeneron Pharmaceuticals Inc and Sanofi .

Conflicts of interest: Dr Paller has been a consultant for and received honorarium from Regeneron Pharmaceuticals and Sanofi and has served as an investigator for Regeneron Pharmaceuticals. Dr Siegfried has been a consultant for, received honorarium from, and served as an investigator for Regeneron Pharmaceuticals and Sanofi. Mr Vekeman and Ms Héroux are employees of StatLog Inc, which received research funding for the current study. Dr Gadkari and Dr Mina-Osorio, are employees of and stockholders in Regeneron Pharmaceuticals Inc. Dr Kaur, Dr Mallya, and Mr Miao are employees of and stockholders in Sanofi.

References

  • 1 C. Avena-Woods. Overview of atopic dermatitis. Am J Manag Care. 2017;23(8 Suppl):S115-S123
  • 2 S. Nutten. Atopic dermatitis: global epidemiology and risk factors. Ann Nutr Metab. 2015;66(1):8-16
  • 3 H.C. Williams. Clinical practice. Atopic dermatitis. N Engl J Med. 2005;352(22):2314-2324
  • 4 S. Illi, E. von Mutius, S. Lau, et al. The natural course of atopic dermatitis from birth to age 7 years and the association with asthma. J Allergy Clin Immunol. 2004;113(5):925-931
  • 5 J.M. Saavedra, M. Boguniewicz, S. Chamlin, et al. Patterns of clinical management of atopic dermatitis in infants and toddlers: a survey of three physician specialties in the United States. J Pediatr. 2013;163(6):1747-1753
  • 6 L.F. Eichenfield, W.L. Tom, T.G. Berger, et al. Guidelines of care for the management of atopic dermatitis: part 2. Management and treatment of atopic dermatitis with topical therapies. J Am Acad Dermatol. 2014;71(1):116-132
  • 7 L.F. Eichenfield, M. Boguniewicz, E.L. Simpson, et al. Translating atopic dermatitis management guidelines into practice for primary care providers. Pediatrics. 2015;136(3):554-565
  • 8 J. Krejci-Manwaring, M.G. Tusa, C. Carroll, et al. Stealth monitoring of adherence to topical medication: adherence is very poor in children with atopic dermatitis. J Am Acad Dermatol. 2007;56(2):211-216
  • 9 S.H. Yu, A.M. Drucker, M. Lebwohl, J.I. Silverberg. A systematic review of the safety and efficacy of systemic corticosteroids in atopic dermatitis. J Am Acad Dermatol. 2017;78(4):733-740.e11
  • 10 E.C. Siegfried, J.C. Jaworski, P. Mina-Osorio. A systematic scoping literature review of publications supporting treatment guidelines for pediatric atopic dermatitis in contrast to clinical practice patterns. Dermatol Ther (Heidelb). 2018;8(3):349-377
  • 11 WHO model prescribing information. drugs used in skin diseases: annex: classification of topical corticosteroids. http://apps.who.int/medicinedocs/en/d/Jh2918e/32.html
  • 12 B. Prindaville, K.A. Horii, E.C. Siegfried, H. Brandling-Bennett. Pediatric dermatology workforce in the United States. Pediatr Dermatol. 2019;36(1):166-168
  • 13 B. Prindaville, R.J. Antaya, E.C. Siegfried. Pediatric dermatology: past, present, and future. Pediatr Dermatol. 2015;32(1):1-12
  • 14 Group TMAS. Access of Medicaid recipients to outpatient care. N Engl J Med. 1994;330(20):1426-1430
  • 15 J. Bisgaier, K.V. Rhodes. Auditing access to specialty care for children with public insurance. N Engl J Med. 2011;364(24):2324-2333
  • 16 A.C. Skinner, M.L. Mayer. Effects of insurance status on children's access to specialty care: a systematic review of the literature. BMC Health Serv Res. 2007;7:194
  • 17 J. Resneck, M.J. Pletcher, N. Lozano. Medicare, Medicaid, and access to dermatologists: the effect of patient insurance on appointment access and wait times. J Am Acad Dermatol. 2004;50(1):85-92
  • 18 Siegfried EC, Paller AS, Vekeman F, et al. Differences in treatment patterns and healthcare resource utilization between Medicaid and commercially insured children with atopic dermatitis. In development.
  • 19 T. Alexander, E. Maxim, L.A. Cardwell, A. Chawla, S.R. Feldman. Prescriptions for atopic dermatitis: oral corticosteroids remain commonplace. J Dermatolog Treat. 2018;29(3):238-240
  • 20 A. He, S.R. Feldman, A.B. Fleischer. Trends in atopic dermatitis management: comparison of 1990-1997 to 2003-2012. J Drugs Dermatol. 2018;17(2):135-140
  • 21 A. He, S.R. Feldman, A.B. Fleischer. An assessment of the use of antihistamines in the management of atopic dermatitis. J Am Acad Dermatol. 2018;79(1):92-96
  • 22 R. Sidbury, D.M. Davis, D.E. Cohen, et al. Guidelines of care for the management of atopic dermatitis: section 3. Management and treatment with phototherapy and systemic agents. J Am Acad Dermatol. 2014;71(2):327-349
  • 23 J. Schmitt, A. Buske-Kirschbaum, F. Tesch, et al. Increased attention-deficit/hyperactivity symptoms in atopic dermatitis are associated with history of antihistamine use. Allergy. 2018;73(3):615-626
  • 24 J. Ring, A. Alomar, T. Bieber, et al. Guidelines for treatment of atopic eczema (atopic dermatitis) Part I. J Eur Acad Dermatol Venereol. 2012;26(8):1045-1060
  • 25 J. Ring, A. Alomar, T. Bieber, et al. Guidelines for treatment of atopic eczema (atopic dermatitis) Part II. J Eur Acad Dermatol Venereol. 2012;26(9):1176-1193
  • 26 E. Roekevisch, P.I. Spuls, D. Kuester, J. Limpens, J. Schmitt. Efficacy and safety of systemic treatments for moderate-to-severe atopic dermatitis: a systematic review. J Allergy Clin Immunol. 2014;133(2):429-438
  • 27 A.Y. Lee. Is montelukast benefical in children with atopic dermatitis?. Allergy Asthma Immunol Res. 2016;8(4):279-281
  • 28 W.K. Chin. Leukotriene receptor antagonism may not be effective in atopic dermatitis treatment after all. J Clin Pharm Ther. 2018;43(1):159-162
  • 29 M.G. Haarman, F. van Hunsel, T.W. de Vries. Adverse drug reactions of montelukast in children and adults. Pharmacol Res Perspect. 2017;5(5):e00341
  • 30 E.L. Simpson, E. Guttman-Yassky, D.J. Margolis, et al. Association of inadequately controlled disease and disease severity with patient-reported disease burden in adults with atopic dermatitis. JAMA Dermatol. 2018;154(8):903-912
  • 31 A.M. Drucker, K. Eyerich, M.S. de Bruin-Weller, et al. Use of systemic corticosteroids for atopic dermatitis: International Eczema Council consensus statement. Br J Dermatol. 2018;178(3):768-775
  • 32 K.A. Horii, S.D. Simon, D.Y. Liu, V. Sharma. Atopic dermatitis in children in the United States, 1997 2004: visit trends, patient and provider characteristics, and prescribing patterns. Pediatrics. 2007;120(3):e527-e534
  • 33 C.R. Totri, L.F. Eichenfield, K. Logan, et al. Prescribing practices for systemic agents in the treatment of severe pediatric atopic dermatitis in the US and Canada: The PeDRA TREAT survey. J Am Acad Dermatol. 2017;76(2):281-285
  • 34 J.I. Silverberg, E.L. Simpson. Associations of childhood eczema severity: a US population-based study. Dermatitis. 2014;25(3):107-114

This e-Reprint is made available with support of

Boehringer Ingelheim International GmbH