Abstract
The impact of childhood atopic dermatitis (AD) on social and behavioral issues is not well understood. This study sought to determine the prevalence and predictors of social and behavioral symptoms and functional impairment among US children with AD. The 1996-2015 Medical Expenditure Panel Surveys were analyzed, including a representative cross-sectional study of 2,553 US children with AD. Behavioral and functional issues were examined using Columbia Impairment Scale (CIS) scores. Childhood AD was associated with behavioral and functional problems, particularly nervousness (adjusted odds ratio [95% CI]: 1.18 [1.06-1.31]), home behavior (1.18 [1.06-1.32]), staying out of trouble (1.18 [1.06-1.31]), and relationships with other kids (1.17 [1.05-1.31]) and with siblings (1.14 [1.02-1.28]). Higher CIS scores were present in children with AD vs. without AD (adjusted beta [95% CI]: 0.62 [0.22, 1.02]) and with psoriasis (0.86 [0.22, 1.49]). Among children with AD, higher CIS scores were notably associated with male sex, older age, lower household income, public insurance, and comorbid depression and anxiety. In conclusion, AD was associated with behavioral and functional impairment, similar to psoriasis and other common chronic conditions. There are significant socio-demographic differences in CIS scores.
Keywords: atopic dermatitis, eczema, epidemiology, impairment, quality of life, development
Introduction
Atopic dermatitis (AD) is a chronic inflammatory skin condition that affects 13% of children (1). Previous studies found that children with AD have higher rates of mental health diagnoses (2–4) and impaired quality of life (QOL) (5), yet many aspects of the mental health disturbances are not well-understood. There may be mental health symptoms that are not assessed in diagnostic screening. A recent study found that parents report increased mental health disorders with impairment in children with AD, including impairments in conduct, emotions, peer relationships, and attention (6). This study aims to better understand the social and behavioral symptoms of children with AD and how these impair daily functioning. We hypothesized that AD significantly impacts interpersonal relations, behavioral impairment, and psychological symptoms.
Previous studies reported significant racial/ethnic, socio-economic, and healthcare disparities with respect to AD prevalence, severity, and persistence in US children (7, 8). We hypothesized that socio-economic and healthcare disparities lead to greater behavioral and psychosocial symptoms in children with AD. This study thus examined the association of AD with impaired interpersonal relations, behavioral limitations, and psychological distress.
Methods
Data source
We analyzed the 1996-2015 Medical Expenditure Panel Surveys (MEPS), a representative, cross-sectional survey of approximately 15,000 households in the US non-institutionalized population administered annually by the Agency for Healthcare Research and Quality. Respondents were interviewed in-person in English or Spanish using computer-assisted personal interviewing technology. Data included socio-demographic information, diagnostic and procedure codes, and health and functional status for each respondent.
Person-level sample weights provided by MEPS allowed for accurate estimates of health and illness for the US civilian non-institutionalized population(9). This study was approved by the Institutional Review Board at <<deidentified>>. This study complies with all Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines(10).
Identification of AD and controls
The study cohort was limited to children (age 5-17 years) with current AD. Diagnosis of AD was determined using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes. Parents were asked to list their children’s current clinical diagnoses without specific prompt to AD; diagnoses were recorded and converted into ICD-9-CM codes by professional coders. Consistent with previous studies using MEPS data(11), both ICD-9-CM codes (691 and 692) were included owing to under-coding with the 691 code alone. It may be that parents reported AD by the more common lay term, “eczema,” which would then be professionally coded as ICD-9-CM code 692. The control group comprised all MEPS participants (age 5-17 years) without a current diagnosis of AD. We also examined a control group of children with psoriasis (ICD-9-CM 696) and without AD.
We also examined the association of AD treatment and mental health symptoms and functioning. Prescribed medications were identified using unique codes in the Food and Drug Administration National Drug Code directory. In particular, we examined the role of high-potency topical corticosteroids (TCS) according to Hurwitz’s Clinical Pediatric Dermatology text(13) and systemic therapy (azathioprine, cyclosporine, methotrexate, mycophenolate, or systemic steroids) for treatment of AD. Dupilumab did not become available for use in AD until 2017, and was therefore not included in the classification of severe AD. Importantly, MEPS included reason for prescription with each medication; we only included medications if they were prescribed for the treatment of AD.
Global behavioral impairment
The primary study outcomes were domains of behavioral problems in children, assessed using the Columbia Impairment Scale (CIS), a validated tool developed to assess interpersonal relations, functional impairment, and psychological symptoms(14). MEPS administered this questionnaire by interview to describe behavioral problems in children ages 5-17 years. Parents were asked to rate from 0 (“No problem”) to 4 (“A very big problem”) the severity and functional impairment in 13 specific activities (CIS scores summed to range 0-52): relationship with mother, father, adults, siblings, and other children; feeling unhappy; behavior at school; having fun; nervousness; involvement in sports or hobbies; schoolwork; home behavior; and staying out of trouble. For each CIS domain, responses were dichotomized into “No problem” (CIS rating=0) vs “Some/big problem” (rating=1-4).
Other variables
Socio-demographic characteristics evaluated include age (5-10/11-17 years), sex (male/female), race/ethnicity (white/black/multiracial-other), income strata (poor/near poor/low income/middle income/high income), geographic region (Northeast/Midwest/South/West), and type of insurance coverage (private/public/none). Income strata were constructed relative to the federal poverty level (FPL): poor/near poor/low income was <200% of FPL, middle income 200-<400% of FPL, and high income ≥400% of FPL.
Statistical analysis
Statistical analyses were performed using procedures that adjusted for sample weights, strata, and clustering in SAS (version 9.4, SAS Institute, Cary, NC). Weighted frequencies, prevalences, and 95% confidence intervals (CI) were generated, by socio-demographic characteristics for children with vs without AD and with vs without psoriasis. Rao-Scott χ2 tests were used to assess socio-demographic differences.
The weighted frequency and prevalences of children who experienced problems with each CIS domain were generated. Bivariable logistic regression models were constructed with AD as the independent variable (IV) and CIS domains and severe impairment (CIS ≥16 threshold for mental health diagnosis) as the dependent variables (DV) in the full pediatric population. Crude odds ratios (OR) 95% CI were estimated. We decided a priori to adjust all multivariable models for age, sex, race, income, and asthma diagnosis. We adjusted for asthma because previous studies have shown asthma is associated with increased CIS total(15). Adjusted OR (aOR) and 95% CI were estimated. Diagnosis of psoriasis was similarly analyzed for comparison to AD.
Survey linear regression was used to examine differences in least squares (LS) mean total CIS score (DV) by AD diagnosis and other chronic comorbidities (IVs). In particular, we examined psoriasis, urticaria, asthma, atopic disease, anxiety, depression, immunodeficiency, diabetes, and hypertension, identified using corresponding ICD-9-CM codes. In particular, atopic disease included sinusitis, rhinitis, allergic rhinitis, nasal polyps, conjunctivitis, asthma, chronic bronchitis, allergic alveolitis, anaphylaxis, angioedema, urticaria, and eosinophilic esophagitis. Two-way statistical interactions of AD with depression, anxiety, and asthma were assessed.
Bivariable and multivariable stepwise (α=0.10) linear regression models were used to determine socio-demographic and clinical predictors of overall CIS score (dependent variable) among children with AD. Crude and adjusted beta, each with 95% CI, were estimated.
Complete case analysis was performed due to the low number of missing demographic (<0.01%) and CIS (≤1.2%) data. Benjamini and Hochberg post-hoc adjustment for multiple comparisons (k=118) was used to minimize the false discovery rate. A two-sided P-value <0.05 was considered statistically significant.
Results
Population characteristics
The study cohort comprised a weighted 837,769,147 children (n=98,873) ages 5-17 years between 1996-2015, including 22,716,193 with AD and 1,554,964 with psoriasis. Diagnosis of AD was associated with younger age (Rao-Scott chi-square, P<0.0001), higher household income (P=0.02), black and multiracial/other race/ethnicity (P<0.0001), northeastern, midwestern and southern regions (P<0.0001), and private insurance (P<0.0001) (Table 1). Of the 165 children diagnosed with psoriasis, there were no significant socio-demographic differences compared to children without psoriasis (P≥0.15).
Table 1.
Socio-demographic associations of clinical diagnosis of AD and psoriasis among US children (n=98,873).
| Characteristic | AD Diagnosis |
Psoriasis Diagnosis |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| No |
Yes |
No |
Yes |
|||||||
| Wtd Freq | % Prev [95% CI] | Wtd Freq | % Prev [95% CI] | Rao-Scott P-Value | Wtd Freq | % Prev [95% CI] | Wtd Freq | % Prev [95% CI] | Rao-Scott P-Value | |
| Sex | 0.56 | 0.14 | ||||||||
| Male | 416,538,649 | 51.1 [50.6, 51.6] | 11,835,848 | 52.1 [49.4, 54.8] | 427,728,469 | 51.2 [50.7, 51.6] | 646,028 | 41.5 [31.7, 51.4] | ||
| Female | 398,514,305 | 48.9 [48.4, 49.4] | 10,880,345 | 47.9 [45.2, 50.6] | 408,485,714 | 48.8 [48.4, 49.3] | 908,935 | 58.5 [48.6, 68.3] | ||
| Age (years) | <0.0001 | 0.48 | ||||||||
| 5-11 | 440,341,428 | 54.0 [53.5, 54.5] | 13,724,195 | 60.4 [57.8, 63.0] | 453,296,062 | 54.2 [53.7, 54.7] | 769,561 | 49.5 [39.4, 59.6] | ||
| 12-17 | 374,711,526 | 46.0 [45.5, 46.5] | 8,991,997 | 39.6 [37.0, 42.2] | 382,918,121 | 45.8 [45.3, 46.3] | 785,402 | 50.5 [40.4, 60.6] | ||
| Income | 0.02 | 0.24 | ||||||||
| Poor, near poor, low | 300,101,920 | 36.8 [35.8, 37.8] | 7,520,676 | 33.1 [30.7, 35.6] | 307,168,538 | 36.7 [35.8, 37.7] | 454,085 | 29.2 [20.9, 37.5] | ||
| Middle income | 270,505,226 | 33.2 [32.5, 33.9] | 8,327,717 | 36.7 [34.2, 39.1] | 278,172,566 | 33.3 [32.6, 33.9] | 660,377 | 42.5 [32.9, 52.1] | ||
| High income | 244,445,808 | 30.0 [29.1, 30.9] | 6,867,800 | 30.2 [27.5, 33.0] | 250,873,079 | 30.0 [29.1, 30.9] | 440,529 | 28.3 [19.2, 37.5] | ||
| Race/Ethnicity | <0.0001 | 0.09 | ||||||||
| White | 346,430,453 | 42.5 [41.0, 44.0] | 9,489,744 | 41.8 [38.3, 45.3] | 355,239,734 | 42.5 [41.0, 44.0] | 680,463 | 43.8 [33.3, 54.2] | ||
| Black | 70,082,301 | 8.6 [7.9, 9.3] | 2,774,290 | 12.2 [10.2, 14.2] | 72,745,578 | 8.7 [8.0, 9.4] | 111,013 | 7.1 [3.1, 11.2] | ||
| Hispanic | 143,018,450 | 17.5 [16.3, 18.8] | 3,130,751 | 13.8 [11.7, 15.8] | 146,006,504 | 17.5 [16.2, 18.7] | 142,697 | 9.2 [5.0, 13.4] | ||
| Multiracial/other | 255,494,780 | 31.3 [30.1, 32.6] | 7,321,408 | 32.2 [29.2, 35.3] | 262,195,397 | 31.4 [30.1, 32.6] | 620,791 | 39.9 [30.2, 49.6] | ||
| Region | <0.0001 | 0.24 | ||||||||
| Northeast | 142,786,762 | 17.5 [16.2, 18.9] | 4,706,597 | 20.7 [17.4, 24.0] | 147,096,910 | 17.6 [16.2, 19.0] | 396,449 | 25.5 [15.1, 35.9] | ||
| Midwest | 189,648,007 | 23.3 [21.8, 24.7] | 5,547,507 | 24.4 [21.3, 27.5] | 194,805,162 | 23.3 [21.9, 24.7] | 390,352 | 25.1 [17.0, 33.2] | ||
| South | 288,815,062 | 35.4 [33.8, 37.0] | 8,400,746 | 37.0 [33.5, 40.4] | 296,692,224 | 35.5 [33.9, 37.1] | 523,584 | 33.7 [23.8, 43.6] | ||
| West | 193,803,124 | 23.8 [22.4, 25.2] | 4,061,342 | 17.9 [15.5, 20.3] | 197,619,887 | 23.6 [22.2, 25.0] | 244,579 | 15.7 [8.4, 23.1] | ||
| Insurance coverage | <0.0001 | 0.41 | ||||||||
| Private | 547,056,890 | 67.1 [66.1, 68.2] | 16,209,507 | 71.4 [68.9, 73.8] | 562,123,213 | 67.2 [66.2, 68.3] | 1,143,183 | 73.5 [65.4, 81.7] | ||
| Public | 216,302,835 | 26.5 [25.5, 27.5] | 5,756,778 | 25.3 [23.0, 27.7] | 221,710,819 | 26.5 [25.5, 27.5] | 348,794 | 22.4 [14.6, 30.3] | ||
| None | 51,693,230 | 6.3 [6.0, 6.7] | 749,907 | 3.3 [2.4, 4.2] | 52,380,150 | 6.3 [5.9, 6.6] | 62,987 | 4.1 [0.9, 7.2] | ||
Bold face indicate statistically significant P-values.
Association of AD with CIS domains of impairment
AD was associated with variable impact on different CIS domains (Table 2). In multivariable logistic regression models, childhood AD was associated with problems with nervousness (adjusted OR [95% CI]: 1.18 [1.06, 1.31]), behavior at home (1.18 [1.06, 1.32]), staying out of trouble (1.18 [1.06, 1.31]), and relationships with other kids (1.17 [1.05, 1.31]) and with siblings (1.14 [1.02-1.28]). In contrast, history of psoriasis was not associated with impairment in any of the CIS domains (Table 2).
Table 2.
Associations of AD or psoriasis diagnosis and behavioral domains using the Columbia Impairment Scale.
| Variable | AD |
Psoriasis |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No |
Yes |
No |
Yes |
|||||||||
| Wtd Freq (% Prev [95% CI]) | Wtd Freq (% Prev [95% CI]) | Crude OR [95% CI] | P | aOR* [95% CI] | P | Wtd Freq (% Prev [95% CI]) | Wtd Freq (% Prev [95% CI]) | Crude OR [95% CI] | P | aOR* [95% CI] | P | |
| Social relationships | ||||||||||||
|
| ||||||||||||
| Getting along with mother | ||||||||||||
| No problem | 506,573,630 (64.3 [63.7, 65.0]) | 13,839,439 (62.7 [60.1, 65.4]) | 1.00 [ref] | – | 1.00 [ref] | – | 519,502,820 (64.3 [63.7, 64.9]) | 910,250 (59.0 [49.6, 68.5]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 280,680,300 (35.7 [35.1, 36.3]) | 8,223,864 (37.3 [34.6, 39.9]) | 1.07 [0.96, 1.20] | 0.34 | 1.10 [0.99, 1.24] | 0.17 | 288,272,058 (35.7 [35.1, 36.3]) | 632,106 (41.0 [31.5, 50.4]) | 1.25 [0.85, 1.85] | 0.39 | 1.20 [0.80, 1.79] | 0.50 |
|
| ||||||||||||
| Getting along with father | ||||||||||||
| No problem | 478,207,801 (63.6 [62.9, 64.2]) | 13,416,619 (63.1 [60.5, 65.7]) | 1.00 [ref] | – | 1.00 [ref] | – | 490,815,546 (63.6 [62.9, 64.2]) | 808,874 (56.0 [46.0, 66.0]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 273,912,858 (36.4 [35.8, 37.1]) | 7,835,047 (36.9 [34.3, 39.5]) | 1.02 [0.91, 1.14] | 0.76 | 1.06 [0.94, 1.18] | 0.48 | 281,111,585 (36.4 [35.8, 37.1]) | 636,321 (44.0 [34.0, 54.0]) | 1.37 [0.92, 2.06] | 0.24 | 1.33 [0.87, 2.01] | 0.33 |
|
| ||||||||||||
| Getting along with brothers and sisters | ||||||||||||
| No problem | 354,354,307 (47.9 [47.1, 48.6]) | 9,118,219 (44.6 [41.9, 47.4]) | 1.00 [ref] | – | 1.00 [ref] | – | 362,917,430 (47.8 [47.0, 48.6]) | 555,096 (39.9 [29.5, 50.2]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 9,118,219 (44.6 [41.9, 47.4]) | 11,309,026 (55.4 [52.6, 58.1]) | 1.14 [1.02, 1.27] | 0.06 | 1.14 [1.02, 1.28] | 0.05 | 396,514,599 (52.2 [51.4, 53.0]) | 837,307 (60.1 [49.8, 70.5]) | 1.38 [0.90, 2.12] | 0.26 | 1.36 [0.89, 2.08] | 0.29 |
|
| ||||||||||||
| Getting along with adults | ||||||||||||
| No problem | 658,612,463 (82.9 [82.5, 83.4]) | 18,174,004 (82.0 [80.3, 83.8]) | 1.00 [ref] | – | 1.00 [ref] | – | 675,470,274 (82.9 [82.5, 83.3]) | 1,316,193 (85.3 [79.2, 91.5]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 135,523,605 (17.1 [16.6, 17.5]) | 3,980,330 (18.0 [16.2, 19.7]) | 1.06 [0.94, 1.20] | 0.43 | 1.08 [0.95, 1.22] | 0.33 | 139,277,772 (17.1 [16.7, 17.5]) | 226,163 (14.7 [8.5, 20.8]) | 0.83 [0.51, 1.36] | 0.56 | 0.85 [0.52, 1.39] | 0.60 |
|
| ||||||||||||
| Getting along with other kids | ||||||||||||
| No problem | 582,882,532 (73.4 [72.9, 73.9]) | 15,528,584 (70.2 [67.8, 72.5]) | 1.00 [ref] | – | 1.00 [ref] | – | 597,312,342 (73.3 [72.8, 73.8]) | 1,098,775 (71.2 [62.4, 80.1]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 211,289,602 (26.6 [26.1, 27.1]) | 6,606,098 (29.8 [27.5, 32.2]) | 1.17 [1.05, 1.31] | 0.02 | 1.17 [1.05, 1.31] | 0.01 | 217,452,119 (26.7 [26.2, 27.2]) | 443,581 (28.8 [19.9, 37.6]) | 1.11 [0.72, 1.71] | 0.68 | 1.12 [0.73, 1.73] | 0.66 |
|
| ||||||||||||
| Psychological symptoms | ||||||||||||
|
| ||||||||||||
| Feeling unhappy or sad | ||||||||||||
| No problem | 487,629,478 (61.4 [60.8, 62.1]) | 13,237,851 (59.8 [57.2, 62.4]) | 1.00 [ref] | – | 1.00 [ref] | – | 500,037,818 (61.4 [60.8, 62.0]) | 829,512 (54.1 [44.3, 63.9]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 305,973,904 (38.6 [37.9, 39.2]) | 8,903,737 (40.2 [37.6, 42.8]) | 1.07 [0.96, 1.19] | 0.33 | 1.10 [0.99, 1.22] | 0.15 | 314,173,568 (38.6 [38.0, 39.2]) | 704,073 (45.9 [36.1, 55.7]) | 1.35 [0.91, 2.01] | 0.26 | 1.29 [0.86, 1.93] | 0.35 |
|
| ||||||||||||
| Feeling nervous or afraid | ||||||||||||
| No problem | 459,047,314 (57.9 [57.3, 58.4]) | 11,839,161 (53.5 [50.9, 56.1]) | 1.00 [ref] | – | 1.00 [ref] | – | 470,056,352 (57.7 [57.2, 58.3]) | 830,123 (54.1 [44.4, 63.8]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 334,411,587 (42.1 [41.6, 42.7]) | 10,297,410 (46.5 [43.9, 49.1]) | 1.19 [1.07, 1.33] | 0.004 | 1.18 [1.06, 1.31] | 0.009 | 344,005,535 (42.3 [41.7, 42.8]) | 703,462 (45.9 [36.2, 55.6]) | 1.16 [0.79, 1.71] | 0.56 | 1.14 [0.77, 1.68] | 0.61 |
|
| ||||||||||||
| Behavioral impairment | ||||||||||||
|
| ||||||||||||
| Behavior at home | ||||||||||||
| No problem | 460,940,676 (58.0 [57.4, 58.7]) | 11,923,087 (53.9 [51.1, 56.6]) | 1.00 [ref] | – | 1.00 [ref] | – | 472,020,832 (57.9 [57.3, 58.6]) | 842,931 (54.7 [44.8, 64.5]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 333,240,716 (42.0 [41.3, 42.6]) | 10,211,595 (46.1 [43.4, 48.9]) | 1.19 [1.06, 1.32] | 0.01 | 1.18 [1.06, 1.32] | 0.009 | 342,752,886 (42.1 [41.4, 42.7]) | 699,425 (45.3 [35.5, 55.2]) | 1.14 [0.77, 1.70] | 0.58 | 1.15 [0.77, 1.73] | 0.58 |
|
| ||||||||||||
| Staying out of trouble | ||||||||||||
| No problem | 579,103,087 (72.9 [72.4, 73.5]) | 15,300,604 (69.1 [66.8, 71.4]) | 1.00 [ref] | – | 1.00 [ref] | – | 593,231,920 (72.8 [72.3, 73.3]) | 1,171,770 (76.2 [67.5, 85.0]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 579,103,087 (72.9 [72.4, 73.5]) | 6,834,078 (30.9 [28.6, 33.2]) | 1.20 [1.08, 1.34] | 0.003 | 1.18 [1.06, 1.31] | 0.007 | 221,517,957 (27.2 [26.7, 27.7]) | 365,367 (23.8 [15.0, 32.5]) | 0.84 [0.52, 1.35] | 0.56 | 0.88 [0.54, 0.62] | 0.68 |
|
| ||||||||||||
| Behavior at school | ||||||||||||
| No problem | 595,662,793 (75.6 [75.2, 76.1]) | 16,610,316 (75.6 [73.3, 77.9]) | 1.00 [ref] | – | 1.00 [ref] | – | 611,052,336 (75.6 [75.1, 76.1]) | 1,220,774 (80.8 [74.2, 87.4]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 191,913,496 (24.4 [23.9, 24.8]) | 5,367,529 (24.4 [22.1, 26.7]) | 1.00 [0.89, 1.13] | 0.97 | 1.00 [0.89, 1.14] | 0.90 | 196,990,532 (24.4 [23.9, 24.9]) | 290,493 (19.2 [12.6, 25.8]) | 0.74 [0.48, 1.13] | 0.29 | 0.78 [0.50, 1.20] | 0.39 |
|
| ||||||||||||
| Schoolwork | ||||||||||||
| No problem | 532,129,962 (67.7 [67.2, 68.2]) | 14,794,988 (67.1 [64.7, 69.5]) | 1.00 [ref] | – | 1.00 [ref] | – | 545,836,245 (67.7 [67.2, 68.2]) | 1,088,705 (70.6 [62.7, 78.4]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 254,060,153 (32.3 [31.8, 32.8]) | 7,253,586 (32.9 [30.5, 35.3]) | 1.03 [0.92, 1.15] | 0.68 | 1.06 [0.95, 1.19] | 0.42 | 560,860,089 (32.3 [31.8, 32.8]) | 453,651 (29.4 [21.6, 37.3]) | 0.88 [0.60, 1.28] | 0.57 | 0.89 [0.59, 1.34] | 0.62 |
|
| ||||||||||||
| Enjoyment and hobbies | ||||||||||||
|
| ||||||||||||
| Having fun | ||||||||||||
| No problem | 693,220,258 (87.3 [86.9, 87.7]) | 19,033,303 (86.0 [84.3, 87.7]) | 1.00 [ref] | – | 1.00 [ref] | – | 710,952,712 (87.3 [86.9, 87.6]) | 1,300,849 (84.3 [77.4, 91.3]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 100,919,029 (12.7 [12.3, 13.1]) | 3,101,379 (14.0 [12.3, 15.7]) | 1.12 [0.97, 1.29] | 0.24 | 1.16 [1.01, 1.34] | 0.09 | 103,778,901 (12.7 [12.4, 13.1]) | 241,507 (15.7 [8.7, 22.6]) | 1.27 [0.75, 2.15] | 0.48 | 1.24 [0.72, 2.12] | 0.55 |
|
| ||||||||||||
| Getting involved in sports or hobbies | ||||||||||||
| No problem | 602,326,942 (76.2 [75.8, 76.7]) | 16,405,255 (74.6 [72.3, 76.8]) | 1.00 [ref] | – | 1.00 [ref] | – | 617,562,755 (76.2 [75.7, 76.7]) | 1,169,443 (76.2 [68.7, 83.7]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Some/big problem | 187,636,359 (23.8 [23.3, 24.2]) | 5,599,515 (25.4 [23.2, 27.7]) | 1.10 [0.97, 1.24] | 0.26 | 1.12 [0.99, 1.26] | 0.15 | 192,870,348 (23.8 [23.3, 24.3]) | 365,526 (23.8 [16.3, 31.3]) | 1.00 [0.66, 1.52] | 1.00 | 0.98 [0.64, 1.49] | 0.92 |
|
| ||||||||||||
| Severe impairment: CIS ≥ 16 | ||||||||||||
| No | 731,685,560 (89.8 [89.4, 90.1]) | 20,164,944 (88.8 [87.2, 90.3]) | 1.00 [ref] | – | 1.00 [ref] | – | 750,412,879 (89.7 [89.4, 90.1]) | 1,437,625 (92.5 [88.5, 96.4]) | 1.00 [ref] | – | 1.00 [ref] | – |
| Yes | 83,367,394 (10.2 [9.9, 10.6]) | 2,551,249 (11.2 [9.7, 12.8]) | 1.11 [0.95, 1.30] | 0.33 | 1.16 [0.99, 1.36] | 0.15 | 85,801,304 (10.3 [9.9, 10.6]) | 117,339 (7.5 [3.6, 11.5]) | 0.71 [0.40, 1.27] | 0.38 | 0.70 [0.39, 1.26] | 0.35 |
Adjusted for age (continuous), sex, race, income, and asthma.
Bold face indicate statistically significant P-values.
Association of AD with total CIS scores
AD was associated with similar or higher total CIS score as compared to many other chronic diseases affecting children and adolescents, such as asthma; as expected, children with depression and/or anxiety had the highest CIS scores (Figure 1). In multivariable linear regression models, children with AD had higher total CIS scores compared to children without AD (least squares [LS] mean: 7.5 vs 6.9; adjusted beta [95% CI]: 0.62 [0.22, 1.02], P=0.009), and children with psoriasis (7.8 vs 7.0; 0.86 [0.22, 1.49], P=0.02).
Figure 1.
Columbia Impairment Scale scores for children with AD compared to other common diseases and no chronic health disorders. Total CIS scores are expressed as least squares means. Adjusted beta values with 95% CI of and corresponding P-values are reported.
Compared to children without AD, total CIS scores were numerically increased in children prescribed super potent TCS (10.2 vs 6.9; 3.27 [−1.78, 8.32], P=0.33) or oral systemic therapy (7.5 vs 6.9, 0.64 [−0.71, 1.99], P=0.48) for AD. There were no significant two-way interactions of AD with asthma (interaction P=0.29), depression (P=0.55), or anxiety (P=0.71) as predictors of CIS scores.
Neither AD nor psoriasis were associated with severe global impairment as judged by total CIS scores ≥16 (Table 2). In multivariable models, children with vs without AD had similar rates of depression (proportion [95% CI]: 2.6% [1.8-3.5%] vs 2.2% [2.0-2.4%], P=0.15) and anxiety (2.8% [1.6-3.9%] vs 2.3% [2.1-2.5%], P=0.33). Psoriasis was likewise not associated with higher rates of depression (3.7% [0.0-7.7%] vs 2.2% [2.1-2.4%], P=0.54) or anxiety (5.8% [0.0-12.3%] vs 2.3% [2.1-2.5%], P=0.24).
Predictors of increased CIS score in children with AD
In multivariable linear regression invoking stepwise variable selection, total CIS score among AD children was associated with male sex (adjusted beta [95% CI]: 1.10 [0.76, 1.45]), older age (12-17 years: 1.23 [0.84, 1.61]), lower income (poor/near poor/low: 1.59 [0.90, 2.28]; middle income: 0.81 [0.34, 1.28]), insurance coverage (public: 1.51 [0.92, 2.10]; none: −1.10 [−1.96, −0.24]), and comorbid depression (8.34 [6.58, 10.11]), anxiety (6.48 [4.82, 8.14]), and asthma (1.29 [0.85, 1.74]), and inversely associated with non-white race/ethnicity (black: −0.86 [−1.22, −0.51]; Hispanic: −2.21 [−2.82, −1.59]) (Table 3).
Table 3.
Socio-demographic and clinical associations of increased CIS score among children with AD.
| Characteristic | AD |
||||
|---|---|---|---|---|---|
| CIS score |
Bivariable Model |
Multivariable Model |
|||
| Mean [95% CI] | Crude Beta [95% CI] | P | Adj. Beta [95% CI] | P | |
| Sex | |||||
| Male | 7.6 [7.1, 8.2] | 1.08 [0.71, 1.44] | <0.0001 | 1.10 [0.76, 1.45] | <0.0001 |
| Female | 6.5 [6.0, 7.0] | 1.00 [ref] | - | 1.00 [ref] | - |
| Age (years) | |||||
| 5-11 | 6.4 [5.9, 6.9] | 1.00 [ref] | - | 1.00 [ref] | - |
| 12-17 | 8.1 [7.4, 8.8] | 1.69 [1.29, 2.10] | <0.0001 | 1.23 [0.84-1.61] | <0.0001 |
| Income | |||||
| Poor, near poor, low | 8.1 [7.5, 8.8] | 1.96 [1.40, 2.53] | <0.0001 | 1.59 [0.90, 2.28] | <0.0001 |
| Middle income | 7.0 [6.3, 7.6] | 0.78 [0.29, 1.26] | 0.007 | 0.81 [0.34, 1.28] | 0.003 |
| High income | 6.2 [5.4, 6.9] | 1.00 [ref] | - | 1.00 [ref] | - |
| Race/Ethnicity | |||||
| White | 7.8 [7.2, 8.5] | 1.00 [ref] | - | 1.00 [ref] | - |
| Black | 6.7 [5.9, 7.6] | −1.09 [−1.42, −0.77] | <0.0001 | −1.54 [−1.94, −1.14] | <0.0001 |
| Hispanic | 6.1 [5.0-7.1] | −1.74 [−2.24, −1.24] | <0.0001 | -2.21 [−2.82, −1.59] | <0.0001 |
| Other or multiple | 6.8 [6.1, 7.5] | −1.03 [−1.58, −0.49] | 0.0009 | −0.60 [−1.14, −0.06] | 0.08 |
| Region | |||||
| Northeast | 7.3 [6.4, 8.1] | 1.00 [ref] | - | 1.00 [ref] | - |
| Midwest | 7.6 [6.8, 8.4] | 0.32 [−0.37, 1.00] | 0.48 | 0.42 [−0.31, 1.14] | 0.38 |
| South | 6.5 [5.8, 7.2] | −0.75 [−1.30, −0.20] | 0.02 | −0.55 [−1.14, 0.03] | 0.15 |
| West | 7.5 [6.6, 8.3] | 0.19 [−0.43, 0.81] | 0.61 | 0.29 [−0.38, 0.97] | 0.50 |
| Insurance coverage | |||||
| Private | 6.8 [6.3, 7.2] | 1.00 [ref] | - | 1.00 [ref] | - |
| Public | 8.5 [7.7, 9.4] | 1.77 [1.28, 2.25] | <0.0001 | 1.51 [0.92, 2.10] | <0.0001 |
| None | 4.9 [3.5, 6.3] | −1.85 [−2.83, −0.87] | 0.0009 | −1.10 [−1.96, −0.24] | 0.03 |
| Depression | |||||
| No | 6.9 [6.5, 7.3] | 1.00 [ref] | - | 1.00 [ref] | - |
| Yes | 16.2 [12.8, 19.6] | 9.35 [7.63, 11.07] | <0.0001 | 8.34 [6.58, 10.11] | <0.0001 |
| Anxiety | |||||
| No | 6.9 [6.5, 7.3] | 1.00 [ref] | - | 1.00 [ref] | - |
| Yes | 14.6 [11.0, 18.2] | 7.65 [6.05, 9.26] | <0.0001 | 6.48 [4.82, 8.14] | <0.0001 |
| Asthma | |||||
| No | 6.9 [6.5, 7.3] | 1.00 [ref] | - | 1.00 [ref] | - |
| Yes | 8.4 [7.4, 9.5] | 1.54 [1.05, 2.02] | <0.0001 | 1.29 [0.85, 1.74] | <0.0001 |
Bold face indicate statistically significant P-values.
Discussion
This study showed that AD, but not psoriasis, is associated with mental health symptoms that may impair daily functioning and behavior in the pediatric population. Though children with vs without AD had similar odds of severe global impairment (CIS ≥16), mental health symptoms in AD occurred in specific domains. In particular, children with AD had higher prevalence of issues with nervousness, home behavior, staying out of trouble, and getting along with other kids and siblings, after adjusting for socio-demographics and comorbid asthma. Taken together, these findings suggest AD is associated with clinically significant symptoms that may not be captured by global assessments or diagnostic screening criteria. Childhood AD was associated with total CIS scores that were comparable to or even higher than other chronic disorders. CIS scores were even higher among adolescent AD male children that had lower income and public insurance.
Previous studies showed poor QOL and increased prevalence of comorbid mental health diagnoses among children with AD (16). The results of the present study indicate that many mental health symptoms and behavioral issues may not be reflected in diagnostic assessments of depression and/or anxiety, highlighting the need for clinicians to screen broadly for mental health symptoms. That is, behavioral, functional, and relationship issues are a part of a different outcome domain (quality of life) than depression or anxiety. Similarly, a recent study examined the measurement properties of Hospital Anxiety and Depression Scale for assessing mental health symptoms in adult AD and found overall good validity(17). Additional research is necessary to determine and validate the optimal approach to assess the full mental and behavioral health burden of pediatric AD.
Importantly, mean CIS scores for AD children on super potent TCS and/or oral systemic therapy were numerically higher than those with low or mid-potency TCS or no systemic therapy. This is likely attributable to insufficient statistical power to determine the true-effect of treatment on CIS scores. Previous studies used super potent TCS and systemic therapy to denote more severe AD disease (12). Taken together, there may be a dose-response effect, such that more severe AD is associated with higher functional CIS burden. Conversely, stronger treatments may have a greater treatment burden on CIS scores than less potent therapies. Future large-scale studies are needed to examine these points.
Previous studies found that children with AD are at increased risk of being teased or bullied (18). The present study suggests that children with AD also have problems with behavior at home, staying out of trouble, and getting along with siblings and other kids. These result build on studies that found lower family QOL associated with childhood AD(19, 20). Taken together, these findings suggest the household detriment of childhood AD may be more specifically described as concomitant behavioral problems. Unlike psoriasis, patients with AD are significantly pruritic, tend to co-sleep with parents (21), have more sleep disturbance and have visible lesions which are associated with significant bullying and stigma. Poor sleep in pediatric patients is associated with inattention and behavioral disruption (22). Families may benefit for a holistic, family-based approach to care (23). Moreover, behavioral and mental health symptoms may impact school productivity. A recent study found that 3.3% of children with AD met the criteria for chronic school absenteeism, with significant socio-economic disparities(24). Lower school productivity may have long-term consequences on future career and success. Further research will guide interventions to address these longitudinal effects of AD.
The exact mechanism of association between AD and nervousness is not fully understood. Chronic itch, pain, sleep disturbance, and self-consciousness may contribute to nervousness and the overall mental health burden(25–28). It may also be that dysregulation of the hypothalamic-pituitary-adrenal axis results in both AD and mental health symptoms(29). Previous studies also showed adults with AD have high rates of anxiety symptoms(27). We found that children who have AD with comorbid asthma, depression, and/or anxiety had even greater impairment than AD alone. Thus, children (and adults) may benefit from multidisciplinary assessment and management of AD, its comorbidities, and functional or mood disturbances in order to reduce these negative impacts. Of note, we did not find any significant associations of pediatric psoriasis with these mental health disturbances. We compared AD with psoriasis, because both disorders have similarities with respect to skin lesions impacting physical appearance and associated symptoms of itch and skin pain. Yet, AD was associated with mental health impairments and psoriasis was not. Thus, the increased odds of mental health disturbances observed in AD may be due to its unique pathophysiology and/or burden.
Association of childhood AD with impaired QOL has been shown in many studies(30, 31). However, the CIS offers novel insight into the social impairment and impacts on relationships, particularly difficulty having fun and getting along with other children. Social difficulties during youth may lead to impaired development of social skills that eventually manifest in long-term problems during adulthood. In addition, CIS is not disease-specific or dermatology-specific, allowing for comparison of impairment between diseases, and showed similar impairment between AD and other chronic diseases.
AD children from lower-income households and on public insurance had higher total CIS scores. These findings may be due to broader socio-economic and healthcare disparities that occur in AD. Previous studies found that low income and public insurance were associated with difficulty accessing and paying for care, increased emergency department utilization, hospital admissions and readmissions in children and/or adults (32–35). These disparities may result in inadequate control of AD, ultimately leading to greater impairment.
Strengths of this study include examination of a large-scale, US population-based cohort, and incorporation of sample weights that allowed for nationally representative estimates of functional impairment in children. Limitations include the inability to assess the role of AD severity, trajectory, and phenotype in impairment. Second, AD diagnosis was determined by caregiver report, and underreporting due to recall bias may partially explain the lower than expected prevalence. (36)Additionally, the CIS may perform differently by race/ethnicity and vary by cultural differences(14). The measurement properties of CIS across racial/ethnic groups, and of AD children in particular, merit further study. Many of the mental health disturbances in AD may be related to sleep disturbances. Unfortunately, sleep disturbances were not assessed in MEPS, and should be examined in future studies. Finally, the cross-sectional nature of the study precluded from making conclusions about causality. Future longitudinal studies are warranted to elucidate the true effect of AD on child functional impairment.
In conclusion, childhood AD is associated with impairment across multiple psychosocial and behavioral domains. These mental health symptoms may not be fully reflected in commonly used diagnostic questionnaires. AD children with comorbid atopic and mental health disorders, and/or lower socio-economic backgrounds are even more likely to have functional limitations. These findings indicate the need for clinicians to screen broadly and address psychosocial and functional impairment in children with AD and their family members.
Acknowledgments
Funding Support: This publication was made possible with support from the Dermatology Foundation.
Abbreviations used:
- ICD-9-CM
International Classification of Disease 9th edition Clinical Modification
- MEPS
Medical Expenditure Panel Survey
- FPL
federal poverty level
- OR
odds ratio
- CI
confidence interval
- AD
atopic dermatitis/eczema
Footnotes
Financial disclosures: None
Conflicts of interest: None
References
- 1.Silverberg JI, Gelfand JM, Margolis DJ, Boguniewicz M, Fonacier L, Grayson MH, et al. Patient burden and quality of life in atopic dermatitis in US adults: A population-based cross-sectional study. Annals of Allergy, Asthma & Immunology. 2018;121(3):340–7. [DOI] [PubMed] [Google Scholar]
- 2.Yaghmaie P, Koudelka CW, Simpson EL. Mental health comorbidity in patients with atopic dermatitis. The Journal of allergy and clinical immunology. 2013;131(2):428–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Strom MA, Fishbein AB, Paller AS, Silverberg JI. Association between atopic dermatitis and attention deficit hyperactivity disorder in U.S. children and adults. The British journal of dermatology. 2016;175(5):920–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Patel KR, Immaneni S, Singam V, Rastogi S, Silverberg JI. Association between atopic dermatitis, depression, and suicidal ideation: A systematic review and meta-analysis. Journal of the American Academy of Dermatology. 2019;80(2):402–10. [DOI] [PubMed] [Google Scholar]
- 5.Hon K, Leung T, Wong K, Chow C, Chuh A, Ng P. Does age or gender influence quality of life in children with atopic dermatitis? Clinical and Experimental Dermatology: Clinical dermatology. 2008;33(6):705–9. [DOI] [PubMed] [Google Scholar]
- 6.Wan J, Takeshita J, Shin DB, Gelfand JM. Mental health impairment among children with atopic dermatitis: A United States population-based cross-sectional study of the 2013–2017 National Health Interview Survey. Journal of the American Academy of Dermatology. 2020;82(6):1368–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Silverberg JI, Simpson EL. Associations of childhood eczema severity: a US population-based study. Dermatitis : contact, atopic, occupational, drug : official journal of the American Contact Dermatitis Society, North American Contact Dermatitis Group. 2014;25(3):107–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.McKenzie C, Silverberg JI. The prevalence and persistence of atopic dermatitis in urban United States children. Annals of allergy, asthma & immunology : official publication of the American College of Allergy, Asthma, & Immunology. 2019;123(2):173–8 e1. [DOI] [PubMed] [Google Scholar]
- 9.Medical Expenditure Panel Survey, MEPS survey questionnaires, household questionnaire sections. [Internet]. Agency for Healthcare Research and Quality. Available from: http://meps.ahrq.gov/mepsweb/survey_comp/survey.jsp. [Google Scholar]
- 10.von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453–7. [DOI] [PubMed] [Google Scholar]
- 11.Fischer AH, Shin DB, Margolis DJ, Takeshita J. Racial and ethnic differences in health care utilization for childhood eczema: An analysis of the 2001–2013 Medical Expenditure Panel Surveys. Journal of the American Academy of Dermatology. 2017;77(6):1060–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Thyssen JP, Skov L, Hamann CR, Gislason GH, Egeberg A. Assessment of major comorbidities in adults with atopic dermatitis using the Charlson comorbidity index. Journal of the American Academy of Dermatology. 2017;76(6):1088–92.e1. [DOI] [PubMed] [Google Scholar]
- 13.Hurwitz S Hurwitz’s clinical pediatric dermatology : a textbook of skin disorders of childhood and adolescence. Fifth edition.. ed. Paller A, Mancini AJ, editors. Edinburgh: Elsevier; 2016. [Google Scholar]
- 14.Bird HR, Andrews H, Schwab-Stone M, Goodman S, Dulcan M, Richters J, et al. Global measures of impairment for epidemiologic and clinical use with children and adolescents. International Journal of Methods in Psychiatric Research. 1996;6(4):295–307. [Google Scholar]
- 15.Sullivan PW, Ghushchyan V, Navaratnam P, Friedman HS, Kavati A, Ortiz B, et al. Indicators of poorly controlled asthma and health-related quality of life among school-age children in the United States. Allergy and asthma proceedings. 2017;38(6):431–9. [DOI] [PubMed] [Google Scholar]
- 16.Silverberg JI, Simpson EL. Association between severe eczema in children and multiple comorbid conditions and increased healthcare utilization. Pediatric Allergy and Immunology. 2013;24(5):476–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Silverberg JI, Gelfand JM, Margolis DJ, Boguniewicz M, Fonacier L, Grayson MH, et al. Measurement Properties of the Hospital Anxiety and Depression Scale Used in Atopic Dermatitis in Adults. The Journal of investigative dermatology. 2019;139(6):1388–91. [DOI] [PubMed] [Google Scholar]
- 18.Lewis-Jones S Quality of life and childhood atopic dermatitis: the misery of living with childhood eczema. Int J Clin Pract. 2006;60(8):984–92. [DOI] [PubMed] [Google Scholar]
- 19.Jang HJ, Hwang S, Ahn Y, Lim DH, Sohn M, Kim JH. Family quality of life among families of children with atopic dermatitis. Asia Pacific allergy. 2016;6(4):213–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Marciniak J, Reich A, Szepietowski JC. Quality of Life of Parents of Children with Atopic Dermatitis. Acta dermato-venereologica. 2017;97(6):711–4. [DOI] [PubMed] [Google Scholar]
- 21.Chamlin SL, Mattson CL, Frieden IJ, Williams ML, Mancini AJ, Cella D, et al. The price of pruritus: sleep disturbance and cosleeping in atopic dermatitis. Archives of pediatrics & adolescent medicine. 2005;159(8):745–50. [DOI] [PubMed] [Google Scholar]
- 22.L LK, Cices A, Fishbein AB, Paller AS. Neurocognitive function in moderate-severe pediatric atopic dermatitis: A case-control study. Pediatric dermatology. 2019;36(1):110–4. [DOI] [PubMed] [Google Scholar]
- 23.Goudie A, Havercamp S, Jamieson B, Sahr T. Assessing functional impairment in siblings living with children with disability. Pediatrics. 2013;132(2):e476–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wan J, Margolis DJ, Mitra N, Hoffstad OJ, Takeshita J. Racial and Ethnic Differences in Atopic Dermatitis–Related School Absences Among US ChildrenRacial and Ethnic Differences in Atopic Dermatitis–Related School AbsencesLetters. JAMA dermatology. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Chiesa Fuxench ZC, Block JK, Boguniewicz M, Boyle J, Fonacier L, Gelfand JM, et al. Atopic Dermatitis in America Study: A Cross-Sectional Study Examining the Prevalence and Disease Burden of Atopic Dermatitis in the US Adult Population. The Journal of investigative dermatology. 2019;139(3):583–90. [DOI] [PubMed] [Google Scholar]
- 26.Kantor R, Dalal P, Cella D, Silverberg JI. Research letter: Impact of pruritus on quality of life-A systematic review. Journal of the American Academy of Dermatology. 2016;75(5):885–6 e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Silverberg JI, Kantor RW, Dalal P, Hickey C, Shaunfield S, Kaiser K, et al. A Comprehensive Conceptual Model of the Experience of Chronic Itch in Adults. American journal of clinical dermatology. 2018;19(5):759–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Vakharia PP, Chopra R, Sacotte R, Patel KR, Singam V, Patel N, et al. Burden of skin pain in atopic dermatitis. Annals of allergy, asthma & immunology : official publication of the American College of Allergy, Asthma, & Immunology. 2017;119(6):548–52 e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Suárez AL, Feramisco JD, Koo J, Steinhoff M. Psychoneuroimmunology of psychological stress and atopic dermatitis: pathophysiologic and therapeutic updates. Acta dermato-venereologica. 2012;92(1):7–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bridgman AC, Eshtiaghi P, Cresswell-Melville A, Ramien M, Drucker AM. The Burden of Moderate to Severe Atopic Dermatitis in Canadian Children: A Cross-Sectional Survey. Journal of cutaneous medicine and surgery. 2018;22(4):443–4. [DOI] [PubMed] [Google Scholar]
- 31.Na CH, Chung J, Simpson EL. Quality of Life and Disease Impact of Atopic Dermatitis and Psoriasis on Children and Their Families. Children (Basel, Switzerland). 2019;6(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kwa L, Silverberg JI. Financial burden of emergency department visits for atopic dermatitis in the United States. Journal of the American Academy of Dermatology. 2018;79(3):443–7. [DOI] [PubMed] [Google Scholar]
- 33.Narla S, Hsu DY, Thyssen JP, Silverberg JI. Predictors of Hospitalization, Length of Stay, and Costs of Care Among Adult and Pediatric Inpatients With Atopic Dermatitis in the United States. Dermatitis : contact, atopic, occupational, drug : official journal of the American Contact Dermatitis Society, North American Contact Dermatitis Group. 2018;29(1):22–31. [DOI] [PubMed] [Google Scholar]
- 34.Cheng BT, Silverberg JI. Predictors of Hospital Readmission in US Children and Adults with Atopic Dermatitis. Annals of allergy, asthma & immunology : official publication of the American College of Allergy, Asthma, & Immunology. 2019. [DOI] [PubMed] [Google Scholar]
- 35.Silverberg JI, Gelfand JM, Margolis DJ, Boguniewicz M, Fonacier L, Grayson MH, et al. Atopic Dermatitis in US Adults: From Population to Health Care Utilization. The journal of allergy and clinical immunology In practice. 2019;7(5):1524–32 e2. [DOI] [PubMed] [Google Scholar]
- 36.Boguniewicz M, Fonacier L, Guttman-Yassky E, Ong PY, Silverberg J, Farrar JR. Atopic dermatitis yardstick: Practical recommendations for an evolving therapeutic landscape. Annals of allergy, asthma & immunology : official publication of the American College of Allergy, Asthma, & Immunology. 2018;120(1):10–22 e2. [DOI] [PubMed] [Google Scholar]