OBJECTIVE. Atopic dermatitis is the most common chronic inflammatory skin disease of childhood and is increasing in prevalence throughout the world. Morbidity and resource use for atopic dermatitis are comparable to other chronic diseases. Topical corticosteroids are first-line therapeutic agents for atopic dermatitis; topical calcineurin inhibitors are considered second-line agents for patients who are older than 2 years. The aims of this study were to examine trends in visits for atopic dermatitis in children in the United States between 1997 and 2004, identify factors that were associated with a pediatric visit for atopic dermatitis, and assess changes in the treatment of atopic dermatitis over time.
METHODS. Visits for atopic dermatitis by children (0–18 years) to office-based physicians and hospital outpatient departments using 1997–2004 National Ambulatory Medical Care Survey and National Hospital Ambulatory Care Survey databases were analyzed. Medication prescribing rates during 2 time periods (1997–2000 and 2001–2004) were also analyzed.
RESULTS. There were an estimated 7.4 million visits for atopic dermatitis. Statistically significant differences in patients with atopic dermatitis included age 2 to 5 years, black race, Asian race, and specialist or hospital outpatient clinic evaluation. The increase in atopic dermatitis visits per year was statistically significant. No statistical differences in prescribing rates were identified between the 2 time periods. Between 1997 and 2000, topical corticosteroids were prescribed in 34% of visits, decreasing to 25% between 2001 and 2004. Between 2001 and 2004, topical calcineurin inhibitors were prescribed in 23% of visits. In the same period, topical corticosteroids were prescribed in 24% of visits by children who were younger than 2 years; topical calcineurin inhibitors were prescribed in 22% of visits.
CONCLUSIONS. Visits for atopic dermatitis in children are increasing. A recommended first-line treatment was prescribed in a minority of the visits.
Atopic dermatitis (AD) is the most common chronic inflammatory skin condition seen in the pediatric population. Studies have documented an increase in the prevalence of AD throughout the world.1–3 Although similar increases have been found within the United States, these studies have been limited to small geographic areas.4 Morbidity and resource use, similar to many other chronic illnesses, have been reported in both children and adults with AD.5–12 In addition, families of children with moderate to severe AD have been found to have a higher impact on their quality of life than families of children with insulin-dependent diabetes.5 Compared with looking after a child with chronic asthma, caring for a child with moderate to severe AD has been associated with greater parental sleep disturbances, anxiety levels, and increased maternal depression.11 Costs to third-party payers to care for patients who were younger than 65 years and had AD in the United States were estimated to be between 0.9 billion to 3.8 billion dollars per year in 1997–1998.12
Patients with AD characteristically present with xerosis, pruritus, and relapsing eczematous skin lesions distributed in age-specific patterns.13–15 Frequently, there is a family history of atopy, and the diagnosis of AD often precedes the development of either asthma or allergic rhinitis.16 There is currently no cure for AD; therapy is mainly directed at alleviating the symptoms of the disease.15,17,18 Topical corticosteroids have been the mainstay of treatment for AD for many years. Only recently has a new class of medications, the topical calcineurin inhibitors, been introduced. Topical calcineurin inhibitors have expanded the viable options for treatment of AD. Although estimates have been made regarding the initial use of these newer medications, factors related to their use and changes in treatment use for AD have not yet been measured.19 The aims of this study were to describe the trends in visits for AD in children (0–18 years) in the United States between 1997 and 2004, examine factors that were associated with the diagnosis of AD, and assess changes in the treatment of AD over time.
A retrospective, cross-sectional study of outpatient encounters compiled from the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS) databases from 1997 through 2004 was performed. The NAMCS and the NHAMCS are administered by the Ambulatory Care Statistics Branch of the Centers for Disease Control and Prevention National Center for Health Statistics (NCHS).20–22 The NAMCS collects information on patient visits to non–federally funded, community, office-based physician practices throughout the United States. The NHAMCS collects information on patient visits to hospital outpatient departments (general medicine, surgery, pediatrics, obstetrics and gynecology, and a combined category of other clinics) and to hospital emergency departments (EDs). The data from both databases represent a proportion of visits, not a proportion of patients, because these surveys do not track individual patients.
The surveys have multistage probability designs. The NAMCS has a 3-stage sampling design, with sampling based on geographic location, physician practices within a geographic location (stratified by physician specialty), and visits within individual physician practices. The NHAMCS has a 4-stage sampling design, with sampling based on geographic area, hospitals within a geographic area, clinics or EDs within hospitals, and patient visits within clinics or EDs. Visit weights accounting for selection probability, adjustment for no response, and other adjustments to reflect the universe of ambulatory visits in the United States allow extrapolation to national estimates for all aspects of the surveys.23,24
Demographics such as age, gender, race, and insurance status are included. Physician and hospital information includes self-identified specialty (in the NAMCS only), clinic type (in the NHAMCS only) such as hospital outpatient department or ED, geographic region, and whether the practice is in a rural area. Clinical characteristics include up to 3 diagnoses (coded using the International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM]), up to 8 medications, and the corresponding National Drug Code Directory class number for each medicine.25
All visits by children who were aged newborn to 18 years and had a diagnosis of AD (ICD-9-CM code of 691.8) in any of the primary or secondary diagnoses fields in the NAMCS and hospital outpatient departments (excluding hospital EDs) of the NHAMCS were included. A total of 510 sample records for AD met inclusion criteria. An additional category was created for analysis of medication use for patients who were younger than 2 years at the time of their visit. There were 238 sample records in this category.
Topical corticosteroids were identified using the National Drug Code Directory class prefix 1268, oral corticosteroids by the prefix 1032, antihistamines by the prefix 1944, and topical calcineurin inhibitors by the drug mention code 01087 for topical tacrolimus (Protopic; Astellas Pharma US, Inc, Deerfield, IL) and code 02031 for topical pimecrolimus (Elidel; Novartis Pharmaceuticals Corp, East Hanover, NJ). Only prescription medications were included in the data analysis. No over-the-counter medications, such as topical 1% hydrocortisone or moisturizers/emollients, were included. It was possible for >1 drug to be prescribed in a single visit. A total of 237 records had at least 1 of the selected drugs prescribed, and in 52 (22%) of those cases, ≥2 of the drugs were prescribed. Cases with multiple prescriptions were included in all individual drug analyses.
Race/ethnicity variables in the NHAMCS and NAMCS were combined into single variables with the following values: white not of Hispanic origin, Hispanic, black not of Hispanic origin, Asian not of Hispanic origin, and other/unknown. When the value of Hispanic was blank, the variables were treated as though they were not of Hispanic origin. The pay type variable was combined into 4 categories: commercial insurance (private insurance and workers’ compensation), government (Medicare, Medicaid), self-pay, and other/unknown (blank, no charge, other and unknown). Age categories were <1 year, 1 year, 2 to 5 years, 6 to 10 years, and 11 to 18 years. Any patient who was older than 18 years was excluded from the analysis. We defined specialist as allergy, allergy and immunology, dermatology, and pediatric allergy. All other physicians were considered generalists.
SEs were calculated using Stata software (Stata Corp, College Station, TX) as recommended by the NCHS, which accounts for the complex sampling design of the NAMCS/NHAMCS.23,24,26 All statistical tests were based on estimates that had a <30% relative SE (ie, the SE divided by the estimate expressed as a percentage of the estimate) and were based on ≥30 cases in the sample data. According to the NCHS, estimates with a >30% relative SE or based on <30 sample cases may be unreliable.
Ninety-five percent confidence intervals (CIs) for all counts and proportions were estimated. We used the svy:logistic procedure in Stata to compare various demographic categories and to evaluate interactions among these variables. Trends over time were tracked using year of visit as a continuous predictor variable. For events in which the sample size was small (ie, medication use in children younger than 2 years), the time variable was dichotomized into two 4-year intervals (1997–2000 and 2001–2004).
All analyses were performed with Stata 9 for Windows. All tests were conducted using a 2-sided α level of .01.
Demographic Characteristics of All Pediatric Visits
There were an estimated 1.5 billion (95% CI: 1.4–1.7) visits by patients who were aged ≤18 years in the United States during the time interval 1997 to 2004. Of those visits, 51% were by boys and 69% were by white patients (Table 1). For these visits, 65% were by children with commercial insurance coverage. Only 14% of the visits were in a nonurban setting. Most children were seen in a physician's office (88%) rather than in a hospital outpatient clinic (12%), and 69% of the visits were to a generalist. Of the pediatric outpatient visits, 16% were by children who were younger than 1 year, 10% by children who were 1 year of age, 22% by children who were aged 2 to 5 years, 20% by children who were aged 6 to 10 years, and 33% by children who were aged 11 to 18 years.
Factors Associated With AD Visits and Visit Rates
There were an estimated 7.4 million (95% CI: 5.9–8.8) visits in the United States by children who were aged newborn to 18 years and had AD to offices of pediatricians, other primary care physicians, specialists, and hospital outpatient clinics between 1997 and 2004. Of the pediatric AD visits, the demographic characteristics were similar to all pediatric visits with the exception of race (only 51% white), specialty (only 43% generalists), and age (only 21% 11–18 years, but >31% between 2 and 5 years; Table 1). These trends were tested with a logistic regression model. All of the disparities mentioned were statistically significant at an α level of .01 (Table 2).
Annual AD visits started in 1997 at 620000 and increased to a peak of 1.7 million visits in 2003. In 2004, however, the number of visits declined to 850000. During this period, the total number of pediatric visits was very stable. The increase in AD visits per year was statistically significant (odds ratio [OR]: 1.10; 95% CI: 1.02–1.18) as shown in Fig 1.
We examined interactions between time periods (1997–2000 vs 2001–2004) and all of the specified demographic groups to determine whether the increasing rate of AD was preferentially associated with these groups. No interactions were significant at a .01 level, indicating that the increase was more or less uniform across these groups.
During 1997 to 2000, there were 2.8 million AD visits, and 0.9 million (34%) involved the prescription of topical corticosteroids. From 2001 to 2004, there were 4.6 million AD visits, and 1.1 million (25%) involved the prescription of topical corticosteroids. This decline was not statistically significant (OR: 0.6; 95% CI: 0.3–1.2). A total of 400000 (14%) of the AD visits in 1997 to 2000 and 1.0 million (21%) in 2001 to 2004 involved the prescription of antihistamines. This increase was also not statistically significant (OR: 1.6; 95% CI: 0.5–4.8).
Topical tacrolimus (Protopic) was prescribed in 0.5 million (10%) and topical pimecrolimus (Elidel) was prescribed in 0.6 million (13%) of the 4.6 million AD visits in 2001 to 2004. Figure 2 shows the trends in proportions of prescriptions for these drugs over time.
Among children who were younger than 2 years, there were 0.7 million AD visits in 1997 to 2000 and 1.3 million in 2001 to 2004. Of these visits, 150000 (21%) and 310000 (24%), respectively, involved the prescription of topical corticosteroids. This trend was not statistically significant (OR: 1.2; 95% CI: 0.4–3.9). For children who were younger than 2 years in 1997–2000, antihistamines were prescribed in 50000 (7%) visits for AD and in 220000 (18%) of the visits in 2001–2004. For children who were younger than 2 years in 2001–2004, topical tacrolimus was prescribed in 97000 (8%) visits for AD, whereas topical pimecrolimus was prescribed in 180000 (14%) visits.
Between 1997 and 2004, oral corticosteroids were prescribed in 1.1 million (17%) of the 6.7 million AD visits (asthma cases were excluded from this analysis). There was no significant time trend in the prescription of oral corticosteroids in this group (OR: 1.3; 95% CI: 0.6–3.1). There were 130000 (6%) prescriptions for oral corticosteroids among the 2.0 million AD visits in children who were younger than 2 years and did not have asthma between 1997 and 2004.
The estimated number of pediatric visits for AD has been increasing annually in the United States. The increase in AD visits per year in our cohort was statistically significant. This finding is consistent with the increasing prevalence of AD seen throughout the world, although it must be noted that our data represent a proportion of visits, not a proportion of children.1–3 Therefore, visit trends can serve only as a surrogate measure of the prevalence of AD in children over time throughout the United States.
Reasons for this national increase in the number of visits for AD are unknown. Experts have developed numerous hypotheses to account for the reported worldwide increased prevalence of atopic diseases; however, consensus agreement on 1 specific cause remains elusive.16,27,28 Possible explanations for this increase over time include the “hygiene hypothesis,” exposure to environmental risk factors, lifestyle changes, and westernization.29–32 Another possibility is that physicians may simply be better at diagnosing children with AD than in the past and therefore may be diagnosing more mild cases.3
Although the total number of annual pediatric patient visits remained stable over time, the number of estimated annual AD visits peaked at 1.7 million visits in 2003. Although this peak in 2003 may be an anomaly, a potential explanation for this finding may be related to the introduction of the topical calcineurin inhibitors in 2000 and 2001, which were initially marketed to primary care physicians, specialists, and the public.19,33 With increased public knowledge of topical calcineurin inhibitors, parents who had previously not seen a physician for their child's AD may have potentially sought this novel treatment.34 Media influences on demand for testing and treatment have also been noted to occur in other medical specialties.35,36 Marketing of the topical calcineurin inhibitors may have also increased primary care physician knowledge about AD, which could have led to an increase in their comfort with diagnosing and treating AD. Future evaluation of visit trends for AD during 2005 and after the January 2006 Food and Drug Administration (FDA) approval of a topical calcineurin inhibitor boxed warning will be important to follow.
The demographic characteristics of the pediatric patients seen with AD were similar to all other pediatric visits that occurred between the same time interval except for age, race, and physician specialty. Patients who were 2 to 5 years of age were seen more often for a diagnosis of AD. Although AD often develops before 1 year of age, a delay in making the diagnosis of AD or potential parental hesitation to obtain treatment for AD during infancy may explain why children 2 to 5 years of age were more commonly seen for AD than any other pediatric age group.
Black and Asian children were also seen more frequently for the diagnosis of AD in our cohort. These results mirror those reported by Janumpally et al,37 who found that black and Asian/Pacific Islander individuals were more likely than white individuals to seek medical care for the diagnosis of AD. A British study found that London-born black Caribbean children were at a potentially higher risk for developing AD when compared with their white counterparts, suggesting that certain racial groups may be at an increased risk for developing AD.38
Specialists and hospital outpatient clinic physicians were the providers who saw the majority of patients with a diagnosis of AD. These results suggest that children with AD may require or families may seek subspecialty care for the treatment of AD. Therefore, there may be an increased need in the future for more subspecialists who are trained to care for AD. Hospital outpatient clinics (staffed by generalists), most commonly located in urban settings, also reported more pediatric visits for AD. Urban patients with AD may not have access to subspecialists or private generalist practices and therefore may seek their care for AD from hospital outpatient clinic physicians.
Topical corticosteroids are accepted first-line anti-inflammatory agents to treat flares of AD.15,17,18 When selected and used appropriately, topical corticosteroids are usually not associated with significant adverse effects.14 Surprising, fewer than one third of all pediatric visits and fewer than 1 in 4 visits for AD of patients who were younger than 2 years were treated with topical corticosteroids in our cohort.
It is unclear why topical corticosteroids were not prescribed in a majority of the pediatric visits for AD in our cohort. Unfortunately, steroid phobia of the physician or the parent may limit the prescribing and use of this first-line therapy for patients with AD.39 In a questionnaire-based study performed in the United Kingdom in 2000, 72.5% of patients reported that they worried about using topical corticosteroids either on their own or on their child's skin, and 24% were noncompliant with their treatment regimen because of these concerns.40 In another study that surveyed patients and physicians about pediatric AD, 94% of the physicians had concerns, such as the long-term consequences of topical corticosteroids, and admitted to taking precautions when treating children for AD.41 Misunderstandings about the use and adverse effects of topical corticosteroids by all types of medical professionals and conflicting messages to patients/parents about topical corticosteroid safety may also contribute significantly to the undertreatment of pediatric AD.42 This limited use of evidence-based therapy is not dissimilar to other studies that have shown low usage rates of other effective therapies in children.43
Although it seemed that the introduction of topical calcineurin inhibitors was chronologically coincident with both an increase in the number of visits for AD and a decrease in the number of topical corticosteroid prescriptions, there was an insufficient number of records in the database to allow a statistical analysis of this trend.
Oral corticosteroids were prescribed in 1 in 6 visits for AD in this study, although oral corticosteroids are not a recommended first-line therapy for AD because of their possible adverse effects and the potential for rebound flare of symptoms.14,17,18 Topical calcineurin inhibitors are also not indicated as first-line therapies for the treatment of AD but are recommended for patients who are unresponsive to or intolerant of other first-line therapies.44,45 Topical calcineurin inhibitors are approved for the short-term or intermittent treatment of mild to moderate AD (topical pimecrolimus) or moderate to severe AD (topical tacrolimus) in patients who are older than 2 years.33 In this study, we found similar rates of use of topical corticosteroids and topical calcineurin inhibitors in patients who were younger than 2 years (24% vs 22%). It is unknown whether these children who received off-label-use topical calcineurin inhibitors had previously had topical corticosteroid therapy that failed. The FDA public health advisory, which initially raised the concern of a potential risk for malignancy associated with the use of topical calcineurin inhibitors, and the subsequent FDA-approved boxed warning occurred after our study period.46–48 Evaluation of prescribing pattern trends of topical corticosteroids and topical calcineurin inhibitors over the next 5 years will be important to follow, specifically evaluating for any changes that may have occurred after the announcement of the public health advisory and boxed warning.
This is a secondary data analysis; as such, there are limitations associated with its use. We acknowledge that we cannot confirm the accuracy of the diagnosis of AD reported by the participating providers to the national databases. We restricted our study to include the ICD-9-CM code for AD to obtain our data; however, providers were not instructed to follow any defined diagnostic criteria before making a diagnosis of AD. This may have affected our results. We may have overestimated the number of visits and prescribing patterns for AD if the data included other skin conditions that were incorrectly diagnosed and treated as AD.
Because this study looks at proportion of visits, not proportion of patients, we cannot determine whether the results are potentially inflated by individual patients who had severe AD and had multiple visits. Furthermore, because the NHAMCS provides data on patient visits only to outpatient hospital departments and hospital EDs, hospital-based subspecialty clinic data are not separately available for analysis in our cohort.
Although the data provided much of the key information for determining the use of the medications, the NAMCS and NHAMCS lack indicators of severity of disease and information about previous use or failure of medications. We therefore could not determine whether medications prescribed during visits for AD were given preferentially or after failure of other medications. Also, the sample size did not allow for evaluation of differential prescribing patterns of specific providers, such as between specialists and generalists or among type of specialists (eg, dermatologists or allergists). Furthermore, because emollients and over-the-counter topical steroids are not prescribed “medications,” we were not able to evaluate whether either were recommended therapeutic options during visits for AD in our cohort. Nevertheless, the NAMCS and NHAMCS are well suited to study practice variations from a public health or policy perspective. The availability of large numbers of patients, detailed clinical information, and sophisticated sampling techniques provides nationally representative data that are unavailable elsewhere.
We found that the estimated number of visits for the diagnosis of AD in children in the United States has increased over time. Reasons to account for this national increase in the number of visits for AD have yet to be fully elucidated. We also found that black and Asian children were more likely than white children to be seen for the diagnosis of AD and that specialists and hospital outpatient clinic physicians were seeing the majority of pediatric patients with AD in our cohort. We also determined that recommended first-line treatment options for AD in children seem to have been prescribed in a minority of pediatric patient visits for AD. Continued evaluation of visit trends and prescribing patterns for AD in children over the next several years will be important to follow.
- Accepted March 7, 2007.
- Address correspondence to Kimberly A. Horii, MD, Section of Dermatology, Children's Mercy Hospitals and Clinics, 2401 Gillham Rd, Kansas City, MO 64108. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Su JC, Kemp AS, Varigos GA, Nolan TM. Atopic eczema: its impact on the family and financial cost. Arch Dis Child.1997;76 :159– 162
- Chamlin SL, Frieden IJ, Williams ML, Chren MM. Effects of atopic dermatitis on young American children and their families. Pediatrics.2004;114 :607– 611
- Emerson RM, Williams HC, Allen BR. What is the cost of atopic dermatitis in preschool children? Br J Dermatol.2001;143 :514– 522
- ↵Hanifin JM, Rajka G. Diagnostic features of atopic dermatitis. Acta Derm Venereol.1980;92 :44– 47
- ↵Eichenfield LF, Hanifin JM, Beck LA, et al. Atopic dermatitis and asthma: parallels in the evolution of treatment. Pediatrics.2003;111 :608– 616
- ↵Department of Health and Human Services, Public Health Service, Food and Drug Administration, Center for Drug Evaluation and Research. Available at: www.fda.gov/ohrms/dockets/ac/05/briefing/2005-4089b2_01_05_Cleared%20version%20Elidel-Protopic%20Drug%20Use%20Review%20D040389%207–2004.pdf. Accessed January 19, 2007
- ↵Woodwell DA, Cherry DK. National Ambulatory Medical Care Survey: 2002 Summary—Advance Data From Vital and Health Statistics: No. 346. Hyattsville, MD: National Center for Health Statistics; 2004
- Hing E, Middleton K. National Hospital Ambulatory Medical Care Survey: 2002 Outpatient Department Summary—Advance Data From Vital and Health Statistics: No. 338. Hyattsville, MD: National Center for Health Statistics; 2004
- ↵McCaig LF, Burt CW. National Hospital Ambulatory Medical Care Survey: 2002 Emergency Department Summary—Advance Data From Vital and Health Statistics: No. 335. Hyattsville, MD: National Center for Health Statistics; 2004
- ↵National Center for Health Statistics. Public Use Microdata File Documentation, National Hospital Ambulatory Medical Care Survey, 2003. Hyattsville, MD: National Technical Information Service; 2005
- ↵National Center for Health Statistics. Public Use Microdata File Documentation, National Ambulatory Medical Care Survey, 2003. Hyattsville, MD: National Technical Information Service; 2005
- ↵US Food and Drug Administration. National Drug Code Directory, 1995 Edition. Washington, DC: Public Health Service; 1995
- ↵Hing E, Gousen S, Shimizu I, Burt C. Guide to using masked design variables to estimate standard errors in public use files for the National Ambulatory Medical Care Survey and the National Hospital Ambulatory Care Survey. Inquiry.2003/2004;40 :416– 415
- Strachan DP. Hay fever, hygiene, and household size. BMJ.1989;299 :1259– 1260
- ↵Food and Drug Administration Pediatric Advisory Committee. Briefing Information. Available at: www.fda.gov/ohrms/dockets/ac/05/briefing/2005-4089b2.htm. Accessed January 19, 2007
- ↵Paller AS, McAlister RO, Doyle JJ, Jackson A. Perceptions of physicians and pediatric patients about atopic dermatitis, its impact, and its treatment. Clin Pediatr (Phila).2002;41 :323– 332
- ↵Marshall BC, Henshaw C, Evans DA, Bleyl K, Alder S, Liou TG. Influenza vaccination coverage level at a cystic fibrosis center. Pediatrics.2002;109(5) . Available at: www.pediatrics.org/cgi/content/full/109/5/e80
- ↵Fonacier L, Spergel J, Charlesworth EN, et al. Report of the topical calcineurin inhibitor task force of the American college of allergy, asthma, and immunology and the American academy of allergy, asthma, and immunology. J Allergy Clin Immunol.2005;115 :1249– 1253
- ↵US Food and Drug Administrion, Center for Drug Evaluation and Research. Alert for Healthcare Professionals: Tacrolimus (Marketed as Protopic). Available at: www.fda.gov/cder/drug/InfoSheets/HCP/ProtopicHCP.htm. Accessed January 19, 2007
- US Food and Drug Administrion, Center for Drug Evaluation and Research. Alert for Healthcare Professionals: Pimecrolimus (Marketed as Elidel). Available at: www.fda.gov/cder/drug/InfoSheets/HCP/elidelHCP.htm. Accessed January 19, 2007
- ↵US Food and Drug Administrion. FDA News: FDA Approves Updated Labeling With Boxed Warning and Medication Guide for Two Eczema Drugs, Elidel and Protopic. Available at: www.fda.gov/bbs/topics/news/2006/NEW01299.html. Accessed January 19, 2007
- Copyright © 2007 by the American Academy of Pediatrics