ORIGINAL
ARTICLE
The burden of atopic dermatitis in US adults: Health care resource utilization data from the 2013 National Health and Wellness Survey Laurent Eckert, PhD,a Shaloo Gupta, MSc,b Caroline Amand, PhD,a Abhijit Gadkari, PhD,c Puneet Mahajan, PhD,d and Joel M. Gelfand, MDe Chilly-Mazarin, France; Princeton and Bridgewater, New Jersey; Tarrytown, New York; and Philadelphia, Pennsylvania Background: There is a lack of data on the burden of atopic dermatitis (AD) in adults relative to the general population. Objective: To characterize the AD burden in adult patients relative to both matched non-AD controls and matched patients with psoriasis in terms of comorbidities, health care resource utilization (HCRU), and costs. Methods: Adults ($18 years) who self-reported a diagnosis of AD or psoriasis and adult non-AD controls were identified from the 2013 US National Health and Wellness Survey. Patients with AD were propensity scoreematched with non-AD controls and patients with psoriasis on demographic variables. Patientreported outcomes were analyzed between matched cohorts. Results: Patients with AD had a significantly greater risk for atopic comorbidities, as well as significantly greater HCRU and total cost compared with non-AD controls. The burden of AD was generally comparable to that of psoriasis, although patients with AD reported increased use of emergency room visits compared with patients with psoriasis. Limitations: Patient-reported data are susceptible to recall bias and erroneous classification. Conclusions: Adult patients with AD reported a substantial disease burden, suggesting an unmet need for more effective AD treatment options. ( J Am Acad Dermatol http://dx.doi.org/10.1016/j.jaad.2017.08.002.) Key words: Atopic dermatitis; burden of disease; comorbidities; disease severity; health care resource utilization; patient-reported outcomes.
From Sanofi, Chilly-Mazarina; Kantar Health, Princetonb; Regeneron, Tarrytownc; Sanofi, Bridgewaterd; and Department of Dermatology and Department of Biostatistics and Epidemiology, University of Pennsylvania Perelman School of Medicine, Philadelphia.e Supported by Sanofi and Regeneron Pharmaceuticals, Inc. Disclosure: Drs Eckert, Amand, and Mahajan are employees of and stockholders in Sanofi. Dr Gadkari is an employee of and stockholder in Regeneron Pharmaceuticals, Inc. Ms Gupta is an employee of Kantar Health, which has received research funding from Sanofi/Regeneron Pharmaceuticals, Inc. Dr Gelfand is an employee of the University of Pennsylvania Perelman School of Medicine, which has received research funding from Sanofi/ Regeneron Pharmaceuticals, Inc. In the previous 12 months, Dr Gelfand has served as a consultant for Abbvie, AstraZeneca, Celgene Corp, Coherus, Eli Lilly, Janssen Biologics (formerly Centocor), Sanofi, Merck, Novartis Corp, Valeant, and Pfizer Inc,
receiving honoraria; he also receives research grants (to the trustees of the University of Pennsylvania) from Abbvie, Eli Lilly, Janssen, Novartis Corp, Regeneron, Sanofi, and Pfizer Inc, and he has received payment for CME work related to psoriasis. Dr Gelfand is also a coepatent holder of Resiquimod for the treatment of cutaneous T-cell lymphoma. Accepted for publication August 3, 2017. Correspondence to: Laurent Eckert, PhD, Sanofi, 1 Avenue Pierre Brossolette, 91380 Chilly-Mazarin, France. E-mail: Laurent. [email protected]. Published online October 7, 2017. 0190-9622 Ó 2017 by the American Academy of Dermatology, Inc. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). http://dx.doi.org/10.1016/j.jaad.2017.08.002
1
J AM ACAD DERMATOL
2 Eckert et al
n 2017
Atopic dermatitis (AD) is a chronic relapsing, criterion (age $18 years), completed the survey, and inflammatory skin disease characterized by pruritus, gave informed consent for their anonymized data to xerosis, and eczematous lesions and with features be used for research purposes. The NHWS was including erythema, infiltration/papulation, excoriaapproved by the Essex Institutional Review Board tions, and lichenification.1 Although AD most often (Lebanon, NJ).9 2,3 starts in early infancy or childhood, a recent study reported that childhood AD extended into adult life Patient population in approximately half of the The AD cohort comprised cases studied.4 For many adult respondents who indiCAPSULE SUMMARY patients, moderate/severe cated that they had AD is a chronic, lifelong experienced AD in response Limited data exist on the burden of condition that is underreto the question: ‘‘Which of the atopic dermatitis (AD) in adults relative cognized as a major public following conditions have you to the general population and other health concern.2,3,5 experienced in the past chronic skin disorders. Recent publications have 12 months?’’ Possible response Adult patients with AD reported documented the multidimenoptions were to select all significantly increased atopic sional disease burden in paconditions that applied, select comorbidities, health care resource tients with moderate/severe ‘‘I have not experienced any utilization, and costs compared with AD, encompassing increased of the above conditions in the non-AD controls. levels of itch, pain, sleep past 12 months,’’ and select disturbance, anxiety and ‘‘I decline to answer.’’ In The overall burden of AD was similar to depression, and impaired addition, respondents had to that of psoriasis, suggesting an unmet health-related quality of answer yes to the question need for effective treatment options in life.6-9 In addition to the dis‘‘Has your AD been diagnosed adult patients with these common, ease burden, AD is associated by a physician?’’ inflammatory skin disorders. with a substantial economic The non-AD control cohort burden. Costs incurred by comprised adult respondents patients, their families, and payers include physician who did not report experiencing AD, dermatitis, or visits, emergency room (ER) visits, hospitalizations, eczema in the past 12 months. The psoriasis cohort prescriptions, and over-the-counter treatments.10 comprised adult respondents who indicated that they However, comprehensive data on the disease and had experienced psoriasis in response to the question: economic burden of AD relative to the general adult ‘‘Which of the following conditions have you ever population and other chronic skin disorders are experienced?’’ Possible response options were to select limited. To this end, the aim of the present study all conditions that applied, select ‘‘I do not have any of was to evaluate the real-world patient burden, in the above conditions,’’ and select ‘‘I decline to answer.’’ terms of comorbidities, health care resource utilizaIn addition, respondents had to answer yes to the tion (HCRU), and costs, in adults with AD relative to question ‘‘Has your psoriasis been diagnosed by a both adults without AD and adults with psoriasis on physician?’’ Respondents who reported a diagnosis of the basis of data from the 2013 National Health and both AD and psoriasis were excluded from this Wellness Survey (NHWS). An additional aim was to analysis. evaluate the impact of AD severity on these Adults with AD graded their disease severity outcomes. (mild, moderate, or severe) in response to the question ‘‘How severe is your dermatitis/eczema?’’ Adults with psoriasis graded their disease severity in METHODS response to the question ‘‘According to the National Data source Psoriasis Foundation, the palm of the hand equals 1% Patient-level data were obtained from the 2013 US of the skin. Thinking about this, please estimate the NHWS, an Internet-based general population percent of your body surface your psoriasis currently survey.11 The NHWS uses a random stratified affects as follows: mild (\3% body coverage); sampling framework (sex, age, and race/ethnicity) moderate (3%-10% body coverage); or severe to ensure that it is representative of the demographic ([10% body coverage).’’ composition of the US adult population, based on data from the US Bureau of the Census. A total of Propensity score matching 1,183,287 individuals were asked to complete the To correct for potential confounding, the AD 2013 NHWS, 109,592 of whom (9.3%) responded. Of cohort was propensity scoreematched12 to non-AD the responders, 75,000 (68.4%) met the inclusion d
d
d
J AM ACAD DERMATOL
Eckert et al 3
VOLUME jj, NUMBER j
Abbreviations used: AD: CCI: CI: ER: HCRU: NHWS: OR:
atopic dermatitis Charlson comorbidity index confidence interval emergency room health care resource utilization National Health and Wellness Survey odds ratio
controls (1:2) and to patients with psoriasis (1:1 [because of a much smaller AD-to-psoriasis respondent ratio]). The matching variables comprised those variables that were statistically different between the groups at baseline and during iterations of the propensity matching process. Relevant variables were entered into a logistic regression model using the SAS/STATÒ LOGISTIC (SAS Institute, Inc, Cary, NC) procedure to generate the propensity score. Matching was completed by using the ‘‘greedy matching’’ technique. The algorithm proceeds with making the best matches first, and the next-best matches next, and so on in a hierarchical sequence until no more matches can be made. Best matches are those with the highest-digit match on propensity score (5 decimal points). The algorithm proceeds sequentially to the lowest-digit match on propensity score (0 decimal points). Goodness of matched pairs were defined as those with the least absolute difference in matched propensity score. For AD versus non-AD matching, each AD patient was matched with 2 non-AD controls using a 1:2 matching algorithm. For AD versus psoriasis matching, a 1:1 matching algorithm was used. Propensity score matching of patients with AD and non-AD controls was based on the variables age, sex, ethnicity, education, income, insurance status, alcohol consumption, exercise behavior, and Charlson Comorbidity Index (CCI).13 Propensity score matching of patients with AD and patients with psoriasis was based on age, sex, ethnicity, education, employment status, body mass index, insurance status, smoking status, exercise behavior, CCI, and self-rated disease severity (moderate/ severe or mild). After matching, matching variables were re-examined between the matched groups to confirm matching. Comorbidities The CCI was used to evaluate the overall comorbidity burden.13 The self-reported prevalence of arthritis (including osteoarthritis, rheumatoid arthritis, and psoriatic arthritis), atopic-related comorbidities (asthma and hay fever/nasal allergies), hypertension, high cholesterol level, and
osteoporosis/osteopenia was evaluated on the basis of respondents’ answers to the questions ‘‘Which of the following conditions have you experienced in the past 12 months?’’ or ‘‘Which of the following conditions have you ever experienced’’ and ‘‘Has your condition been diagnosed by a physician?’’ HCRU HCRU was defined by the number of visits to a health care provider in the past 6 months (‘‘How many visits did you make to the following traditional health care provider(s) in the past 6 months?’’), ER visits (‘‘How many times have you been to the ER for your own medical condition in the past 6 months?’’), and hospitalizations (‘‘How many times have you been hospitalized for your own medical condition in the past 6 months?’’). Direct costs were estimated using data stratified according to age from the Medical Expenditure Panel Survey 2012.14 For each respondent, the number of each type of visit in the last 6 months was multiplied by 2 to extrapolate to the annual number of visits and then multiplied by its average cost (Supplemental Table I; available at http://www.jaad.org). Analyses The prevalence of comorbidities, HCRU, and costs were analyzed in patients with AD against matched non-AD controls and against matched patients with psoriasis. In addition, these outcomes were examined in the moderate/severe AD cohort versus in the mild AD cohort. A chi-square test or the t test were used to compare differences between groups for categorical or continuous variables, respectively, with the 2-tailed significance level (a) set at .05. A sensitivity analysis of direct cost data was carried out by calculating least squares mean costs after adjusting for the presence of asthma and nasal allergies/hay fever. Analyses were performed with SAS software, version 9.3 (SAS Institute, Inc).
RESULTS AD versus non-AD and moderate/severe versus mild AD Demographics. Demographic characteristics in patients with AD and matched non-AD controls are shown in Table I. The analysis cohort of 1047 adults (mean age, 46.1 years; 68.3% female and 66.8% non-Hispanic white) comprised 349 patients with AD and 698 matched non-AD controls. Self-rated disease severity was available for 306 patients with AD, of whom 52.3% reported moderate (42.8%) or severe (9.5%) disease and 47.7% reported mild disease. Overall, demographic characteristics and CCI score were generally similar between patients
J AM ACAD DERMATOL
4 Eckert et al
n 2017
Table I. Demographic characteristics in patients with AD versus matched non-AD controls and patients with moderate/severe AD versus patients with mild AD Total (N = 1047)
Characteristic
AD (n = 349)
Age, years, mean (SD) 46.1 (15.3) 45.8 Female, n (%) 715 (68.3) 228 Non-Hispanic white, n (%) 699 (66.8) 225 Married/living with partner, n (%) 614 (58.6) 196 College educated, n (%) 571 (54.5) 191 Annual household income $$75000, 334 (31.9) 113 n (%) Employed, n (%) 603 (57.6) 195 Alcohol use, n (%) 759 (72.5) 246 Current smoker, n (%) 201 (19.2) 69 Taking regular exercise, n (%) 727 (69.4) 246 Mean Charlson comorbidity index (SD) 0.81 (1.8) 0.88
Non-AD (n = 698)
P value
Moderate/severe AD (n = 160)
Mild AD (n = 146)
P value
(14.9) 46.3 (15.5) (65.3) 487 (69.8) (64.5) 474 (67.9) (56.2) 418 (59.9) (54.7) 380 (54.4) (32.4) 221 (31.7)
.63 .15 .26 .25 .93 .70
44.7 107 99 84 80 42
(15.3) (66.9) (61.9) (52.5) (50.0) (26.3)
45.4 92 91 87 88 60
(14.0) (63.0) (62.3) (59.6) (60.3) (41.1)
.66 .48 .10 .21 .07 .09
(55.9) 408 (70.5) 513 (19.8) 132 (70.5) 481 (1.8) 0.78
.43 .30 .14 .60 .37
79 106 38 104 1.07
(49.4) (66.3) (23.8) (65.0) (2.2)
95 104 22 118 1.18
(65.1) (71.2) (15.1) (80.8) (3.9)
.006 .35 .15 .002 .76
(58.5) (73.5) (18.9) (68.9) (1.8)
AD, Atopic dermatitis; SD, standard deviation.
Table II. Self-reported comorbid conditions in patients with atopic dermatitis (AD) versus matched non-AD controls and patients with moderate/severe versus mild AD Comorbidity, n (%)
AD (n = 349)
Non-AD (n = 698)
Hypertension 96 (27.5) 178 High cholesterol level* 112 (32.1) 193 Arthritis (osteoarthritis, 103 (29.5) 136 rheumatoid, psoriatic) Osteoporosis/osteopenia 30 (8.60) 51 Asthma 84 (24.1) 62 Nasal allergies/hay fever 169 (48.4) 170
OR (95% CI)
P value
Moderate/severe AD (n = 160)
Mild AD (n = 146)
OR (95% CI)
P value
(25.5) 1.1 (0.8-1.5) .49 (27.7) 1.2 (0.9-1.6) .14 (19.5) 1.7 (1.3-2.3) \.001
49 (30.6) 49 (30.6) 53 (33.1)
36 (24.7) 1.4 (0.8-2.2) 45 (30.8) 1.0 (0.6-1.6) 36 (24.7) 1.5 (0.9-2.5)
.24 .97 .10
(7.3) 1.2 (0.8-1.9) .46 (8.9) 3.3 (2.3-4.7) \.001 (24.4) 2.9 (2.2-3.8) \.001
16 (10.0) 44 (27.5) 82 (51.3)
14 (9.6) 1.1 (0.5-2.2) 36 (24.7) 1.2 (0.7-1.9) 71 (48.6) 1.1 (0.7-1.7)
.90 .57 .65
AD, Atopic dermatitis; CI, confidence interval; OR, odds ratio. *High cholesterol was defined as a total cholesterol serum concentration of $240 mg/dL.
with AD and patients without AD and between patients with moderate/severe AD and patients with mild AD. However, significantly lower proportions of patients with moderate/severe AD than patients with mild AD reported being employed (49.4% vs 65.1% [P = .006]) or exercising (65.0% vs 80.8% [P = .002]). Comorbidities. The self-reported prevalence of comorbid conditions in patients with AD versus matched non-AD controls, and patients with moderate/severe AD versus patients with mild AD is presented in Table II. Compared with non-AD controls, patients with AD had a significantly increased risk for arthritis (odds ratio [OR], 1.7; 95% confidence interval [CI], 1.3-2.3; P \ .001), asthma (OR, 3.3; 95% CI, 2.3-4.7; P \ .001), and nasal allergies/hay fever (OR, 2.9; 95% CI, 2.2-3.8; P \ .001). The risk for comorbidities in patients with moderate/severe AD was numerically higher, although not statistically significant compared with patients with mild AD.
HCRU and costs. Compared with non-AD controls, patients with AD reported using significantly more health care resources (Table III). Mean (SD) health care provider visits were 6.6 (9.8) versus 3.9 (5.4) (P \ .001). The number of ER visits and hospitalizations was more than twice that in non-AD controls. In patients with AD, the mean (SD) number of ER visits was 0.5 (1.2) versus 0.2 (0.8) (P \ .001) and the mean (SD) number of hospitalizations was 0.3 (1.0) versus 0.1 (0.7) (P = .004). Patients with moderate/severe AD used numerically greater resources than patients with mild AD; however, none of the differences were statistically significant. In accordance with the HCRU data, patients with AD incurred significantly higher costs than non-AD controls across all evaluated categories (Table III). Thus, the total mean (SD) annual per patient direct costs in patients with AD were $9782 higher than those of non-AD controls: $24,401 ($37,355) versus $14,619 ($29,799) (P \ .001). Controlling for the
J AM ACAD DERMATOL
Eckert et al 5
.21
presence of asthma and nasal allergies/hay fever reduced the difference between AD and non-AD in total mean direct costs to $7901 (P = .001) (Table III). Patients with moderate/severe AD incurred numerically higher costs than did patients with mild AD across the evaluated categories. Controlling for the presence of asthma and nasal allergies/hay fever increased the difference between moderate/severe versus mild AD in total mean direct costs from $3757 to $5522. None of the differences were statistically significant (Table III).
AD, Atopic dermatitis; CI, confidence interval; ER, emergency room; HCRU, health care resource utilization; LS, least squares; SD, standard deviation. *Direct costs were adjusted for the presence of asthma and nasal allergies/hay fever.
21,642 (16,768-27,934) 27,164 (21,290-34,659) .0012 14,870 (12,868-17,183) 22,771 (18,509-28,013)
.24 .98 .06 .42 (22,070) (28,837) (3004) (43,252) 13,867 8668 1147 23,682 (20,258) (26,973) (4420) (37,756) 16,735 8741 1963 27,439 \.001 .026 \.001 \.001 (13,882) (21,203) (2339) (29,799) 9252 4759 608 14,619 (19,893) (26,520) (3602) (37,355) 14,797 8145 1459 24,401
.32 .96 .08 6.3 (11.5) 0.3 (1.0) 0.4 (1.0) 7.6 (9.6) 0.3 (1.0) 0.6 (1.4) \.001 .004 \.001 3.9 (5.4) 0.1 (0.7) 0.2 (0.8) 6.6 (9.8) 0.3 (1.0) 0.5 (1.2)
Mean HCRU (SD) Health care provider visits in past 6 mo Hospitalizations in past 6 mo ER visits in past 6 mo Mean annual per patient costs (SD), $, Health care provider visit costs Hospitalization costs ER visit costs Total direct costs LS mean adjusted annual per patient costs, $, (95% CI)* Total direct costs
P value Mild AD (n = 146) Moderate/severe AD (n = 160) P value Non-AD (n = 698) AD (n = 349)
Table III. HCRU and costs in patients with atopic dermatitis (AD) versus matched non-AD controls and patients with moderate/severe versus patients with mild AD
VOLUME jj, NUMBER j
AD versus psoriasis Demographics. The analysis cohort comprised 260 patients with AD and 260 matched patients with psoriasis. Demographic characteristics were similar between the 2 groups (Table IV). Comorbidities. The self-reported prevalence of comorbidities in patients with AD versus matched patients with psoriasis are presented in Table V. Compared with patients with psoriasis, patients with AD had significantly higher odds of reporting asthma (OR, 1.7; 95% CI, 1.1-2.7; P = .01) and nasal allergies/hay fever (OR, 2.3; 95% CI, 1.6-3.2; P \ .001). Conversely, patients with psoriasis had higher odds of developing hypertension (OR, 0.7; 95% CI, 0.5-1.0; P = .03). HCRU and costs. HCRU in patients with AD versus in matched patients with psoriasis is shown in Table VI. Although the number of visits was low, the mean (SD) number of ER visits required by patients with AD in the past 6 months was significantly higher than that of patients with psoriasis: 0.5 (1.1) versus 0.3 (1.0) (P = .02). Accordingly, mean (SD) annual per patient ER visit costs incurred by patients with AD were almost twice those of patients with psoriasis: $1448 ($3421) versus $750 ($3008) (P = .01). The number of hospitalizations and hospitalization costs were also higher in patients with AD, but these differences did not attain statistical significance.
DISCUSSION To improve the morbidity associated with AD in adults, it is necessary to characterize its disease burden in the general population. This comparative analysis of comorbidities, HCRU, and costs indicates that US adult patients with AD have a significantly increased risk for reporting atopic comorbidities such as asthma and nasal allergies/hay fever compared with matched non-AD controls. The association between AD and atopic comorbid conditions has been recognized previously15-17 and is consistent with the prevailing concept of the
J AM ACAD DERMATOL
6 Eckert et al
n 2017
Table IV. Demographic characteristics in patients with AD and matched patients with psoriasis AD (n = 260)
Characteristic
Mean age (SD), y Female, n (%) Non-Hispanic white, n (%) Married/living with partner, n (%) College educated, n (%) Annual household income $$75,000, n (%) Employed, n (%) Alcohol use, n (%) Current smoker, n (%) Regularly exercise, n (%) Mean Charlson comorbidity index (SD)
45.4 169 167 149 148 82 152 181 48 189 0.80
Psoriasis (n = 260)
(14.8) (65.0) (64.2) (57.3) (56.9) (31.5) (58.5) (69.6) (18.5) (72.7) (1.9)
44.7 170 170 156 146 72 159 181 41 188 0.71
P value
(15.2) (65.4) (65.4) (60.0) (56.2) (27.7) (61.2) (69.6) (15.8) (72.3) (1.3)
.56 .93 .72 .53 .86 .73 .53 1.00 .71 .92 .55
AD, Atopic dermatitis; SD, standard deviation.
Table V. Self-reported comorbid conditions in AD and matched patients with psoriasis AD (n = 260)
Comorbidity, n (%)
Hypertension High cholesterol concentration* Arthritis (osteoarthritis, rheumatoid, or psoriatic) Osteoporosis/osteopenia Asthma Nasal allergies/hay fever
70 75 66 22 64 129
(26.9) (28.9) (25.4) (8.5) (24.6) (49.6)
Psoriasis (n = 260)
OR (95% CI)
P value
93 79 73 17 41 79
0.7 0.9 0.9 1.3 1.7 2.3
.03 .70 .49 .41 .01 \.001
(35.8) (30.4) (28.1) (6.5) (15.8) (30.4)
(0.5-1.0) (0.6-1.4) (0.6-1.3) (0.7-2.6) (1.1-2.7) (1.6-3.2)
AD, Atopic dermatitis; CI, confidence interval; OR, odds ratio. *High cholesterol concentration was defined as a total cholesterol serum concentration of $240 mg/dL.
‘‘atopic march’’ from AD to other common allergic conditions.18 In accordance with the high comorbidity burden in patients with AD, HCRU was significantly higher in the AD cohort compared with non-AD controls for all categories evaluated. The difference was largely driven by health care provider visits and associated costs. Although the numbers of reported hospitalizations and ER visits were small, the utilization of these categories by patients with AD was approximately twice that reported by non-AD controls. Overall, mean annual per patient total direct costs in patients with AD were approximately $10,000 higher than those of non-AD controls ($24,401 vs $14,619). Controlling for the presence of asthma and nasal allergies/hay fever reduced the AD versus non-AD difference in total mean direct costs by approximately 20%, which is in accordance with the higher atopic burden in patients with AD.15-17 Patients with more severe AD have been shown to be at a higher risk for atopic comorbidities.19 Also, they are generally associated with significantly higher all-cause HCRU and costs relative to less severe AD.20,21 The nonsignificant differences
between severity levels in this study were likely due to insufficient statistical power to detect small differences in outcomes between the moderate/ severe and mild subgroups. It is important to note that more than 80% of the patients in the moderate/ severe group reported their disease as moderate. Thus, the comparative analysis of burden according to severity was predominantly that between patients with moderate disease and those with mild disease, which partly explains why the differences are not larger. An additional factor that may have had an influence on the severity analyses was that severity was patient rated and the level of concordance between patient- and physician-rated severity was not assessed in the current study. Previous studies have reported significant differences between patient- and physician-rated severity in adult patients.22,23 These results highlight the need for objective assessment of AD severity in patient self-report surveys through validated instruments, such as the Patient-Oriented Severity Scoring of Atopic Dermatitis,24 or the Patient Oriented Eczema Measure.25 The present analysis and the companion article in this issue of the Journal of the American Academy of
J AM ACAD DERMATOL
Eckert et al 7
VOLUME jj, NUMBER j
Table VI. HCRU in patients with AD and matched patients with psoriasis
Mean HCRU (SD) Health care provider visits in past 6 mo Hospitalizations in past 6 mo ER visits in the past 6 mo Mean annual per patient costs (SD), $ Health care provider visit costs Hospitalization costs ER visit costs Total direct costs
AD (n = 260)
Psoriasis (n = 260)
P value
6.2 (8.0) 0.3 (1.0) 0.5 (1.1)
6.3 (9.8) 0.1 (0.8) 0.3 (1.0)
.88 .07 .02
14,058 8177 1448 23,682
(18,252) (27,040) (3421) (37,041)
14,171 4244 750 19,165
(21,495) (21,435) (3008) (34,459)
.95 .07 .01 .15
AD, Atopic dermatitis; ER, emergency room; HCRU, health care resource utilization; SD, standard deviation.
Dermatology (Eckert et al) are, to the best of our knowledge, the first publications to provide comparative data on the disease burden in adults for AD and psoriasis. Like AD, psoriasis is a common inflammatory skin disease, with a prevalence in adults of about 3%.26-28 Patients with chronic plaque psoriasis, the most common form of the disease (about 90% of cases), present with itching and erythematous plaques covered by silver lamellar scales.29 Psoriasis is associated with impairments in health-related quality of life and an excess risk for cardiometabolic disease and mortality, especially in patients with more severe disease.30-35 Our analyses of NHWS real-world data indicate that patients with AD have a significantly higher risk for atopic comorbidities than do patients with psoriasis. With regard to HCRU, the numbers of hospitalizations and ER visits of patients with AD were approximately double those reported by patients with psoriasis. Interestingly, patients with psoriasis reported higher rates of hypertension, confirming the different cardiovascular risk profile compared with AD. A strength of this study is that the NHWS is representative of the demographic composition of the US adult population, and thus, the results presented here are likely to be applicable to the wider population of US adult patients with AD. Propensity score matching of patients with AD and non-AD controls, which was carried out in an effort to minimize confounding due to demographic characteristics, was an additional strength of this study. The inherent limitation of the study is that patient-reported data are susceptible to recall bias36 and erroneous classification, given that such data are not independently verified against clinical records. However, these errors would be expected to be nondifferential and thus unlikely to mask differences in burden between groups. Furthermore, we matched our cohorts on the basis of numerous factors to minimize the potential for confounding to affect the results, with the caveat that
overmatching can result in the masking of true associations.37 The NHWS does not include data on key AD- and psoriasis-specific outcome measures, such as itch and pain; nor does it allow the clinical evaluation of AD severity using SCORAD24 or psoriasis severity using the Simplified Psoriasis Index.38 An additional limitation is that HCRU data could not be definitively attributed to AD or psoriasis because claims specifically linking resource use with these conditions are not captured by the NHWS database.
CONCLUSIONS The results from this analysis indicate that adult patients with AD experience a substantial patient burden relative to matched controls without AD, as is evidenced by a significantly increased risk for development of atopic conditions (asthma and hay fever/nasal allergies) and arthritis, as well as increased HCRU and associated costs. Adult patients with AD reported a significantly higher prevalence of atopic conditions and a significantly increased number of ER visits and associated costs relative to matched patients with psoriasis, whereas psoriasis was associated with a greater prevalence of hypertension. Taken together, these data suggest a significant need for enhanced treatment options for patients with AD. The authors would like to thank the NHWS respondents and all investigators involved in this study. Medical writing support, including developing outline and initial drafts and incorporation of author/reviewer comments was provided by Fernando Gibson, PhD (of Prime, London, UK). Editorial support, including formatting and manuscript submission, was provided by Sinead Stewart (of Prime, Knutsford, UK). Medical writing and editorial support was funded by Sanofi/Regeneron Pharmaceuticals, Inc, according to Good Publication Practice guidelines. The authors were responsible for all content and editorial decisions, and they received no honoraria related to the development of this publication.
8 Eckert et al
REFERENCES 1. Weidinger S, Novak N. Atopic dermatitis. Lancet. 2016;387: 1109-1122. 2. Bieber T. Atopic dermatitis. N Engl J Med. 2008;358:1483-1494. 3. Margolis JS, Abuabara K, Bilker W, et al. Persistence of mild to moderate atopic dermatitis. JAMA Dermatol. 2014;150: 593-600. 4. Napolitano M, Megna M, Patruno C, et al. Adult atopic dermatitis: a review. G Ital Dermatol Venereol. 2016;151:403-411. 5. Gittler JK, Shemer A, Suarez-Farinas M, et al. Progressive activation of T(H)2/T(H)22 cytokines and selective epidermal proteins characterizes acute and chronic atopic dermatitis. J Allergy Clin Immunol. 2012;130:1344-1354. 6. Simpson EL, Bieber T, Eckert L, et al. Patient burden of moderate to severe atopic dermatitis (AD): insights from a phase 2b clinical trial of dupilumab in adults. J Am Acad Dermatol. 2016;74:491-498. 7. Simpson EL, Bieber T, Guttman-Yassky E, et al. Two phase 3 trials of dupilumab versus placebo in atopic dermatitis. N Engl J Med. 2016;375:2335-2348. 8. Simpson EL, Gadkari A, Worm M, et al. Dupilumab therapy provides clinically meaningful improvement in patient-reported outcomes (PROs): a phase IIb, randomized, placebo-controlled, clinical trial in adult patients with moderate to severe atopic dermatitis (AD). J Am Acad Dermatol. 2016;75:506-515. 9. Whiteley J, Emir B, Seitzman R, et al. The burden of atopic dermatitis in US adults: results from the 2013 National Health and Wellness Survey. Curr Med Res Opin. 2016:1-7. 10. Drucker AM, Wang AR, Qureshi AA. Research gaps in quality of life and economic burden of atopic dermatitis: the National Eczema Association Burden of Disease Audit. JAMA Dermatol. 2016;152:873-874. 11. Kantar Health. The National Health and Wellness Survey. Available at: http://www.kantarhealth.com/docs/datasheets/ kh-national-health-and-wellness-survey.pdf. Accessed July 5, 2016. 12. Austin PC. A comparison of 12 algorithms for matching on the propensity score. Stat Med. 2014;33:1057-1069. 13. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373-383. 14. US Department of Health and Human Services. Medical Expenditure Panel Survey (MEPS). Available at: https://meps. ahrq.gov/mepsweb/. Accessed August 24, 2017. 15. Boguniewicz M, Leung DY. Atopic dermatitis: a disease of altered skin barrier and immune dysregulation. Immunol Rev. 2011;242:233-246. 16. Illi S, von Mutius E, Lau S, et al. The natural course of atopic dermatitis from birth to age 7 years and the association with asthma. J Allergy Clin Immunol. 2004;113:925-931. 17. Leung DY, Guttman-Yassky E. Deciphering the complexities of atopic dermatitis: shifting paradigms in treatment approaches. J Allergy Clin Immunol. 2014;134:769-779. 18. Shaker M. New insights into the allergic march. Curr Opin Pediatr. 2014;26:516-520. 19. Simpson E, Guttman-Yassky E, Margolis DJ, et al. Chronicity, comorbidity and life course impairment in atopic dermatitis: insights from a cross-sectional study in US adults. Presented at: 25th European Academy of Dermatology and Venereology. September 28-October 2, 2016; Vienna, Austria.
J AM ACAD DERMATOL
n 2017
20. Drucker A, Stemart A, Amand C, et al. Burden of atopic dermatitis: a claims-based analysis. Presented at: American Academy of Dermatology 74th Annual Meeting. March 4-8, 2016; Washington DC. 21. Shrestha S, Miao R, Wang L, et al. Burden of atopic dermatitis: comorbidities, healthcare resource utilization, and costs in US commercial and medicare adult populations. Presented at: Academy of Managed Care Pharmacy (AMCP) Nexus. October 3-6, 2016; National Harbor, WA. 22. Wei W, Anderson P, Gadkari A, et al. Disagreement between physician- and patient-reported disease severity in adults with a history of moderate-to-severe atopic dermatitis. Presented at: 75th Annual Meeting of the Society for Investigative Dermatology. May 11-14, 2016; Scottsdale, AZ. 23. Fivenson D, Arnold RJ, Kaniecki DJ, et al. The effect of atopic dermatitis on total burden of illness and quality of life on adults and children in a large managed care organization. J Manag Care Pharm. 2002;8:333-342. 24. European Task Force on Atopic Dermatitis. Severity scoring of atopic dermatitis: the SCORAD index. Consensus Report of the European Task Force on Atopic Dermatitis. Dermatology. 1993; 186:23-31. 25. Charman CR, Venn AJ, Williams HC. The patient-oriented eczema measure: development and initial validation of a new tool for measuring atopic eczema severity from the patients’ perspective. Arch Dermatol. 2004;140:1513-1519. 26. Helmick CG, Lee-Han H, Hirsch SC, et al. Prevalence of psoriasis among adults in the U.S.: 2003-2006 and 2009-2010 National Health and Nutrition Examination Surveys. Am J Prev Med. 2014;47:37-45. 27. Rachakonda TD, Schupp CW, Armstrong AW. Psoriasis prevalence among adults in the United States. J Am Acad Dermatol. 2014;70:512-516. 28. Kurd SK, Gelfand JM. The prevalence of previously diagnosed and undiagnosed psoriasis in US adults: results from NHANES 2003-2004. J Am Acad Dermatol. 2009;60:218-224. 29. Boehncke WH, Schon MP. Psoriasis. Lancet. 2015;386:983-994. 30. Bhosle MJ, Kulkarni A, Feldman SR, et al. Quality of life in patients with psoriasis. Health Qual Life Outcomes. 2006;4:35. 31. Shutty BG, West C, Huang KE, et al. Sleep disturbances in psoriasis. Dermatol Online J. 2013;19:1. 32. Dalgard FJ, Gieler U, Tomas-Aragones L, et al. The psychological burden of skin diseases: a cross-sectional multicenter study among dermatological out-patients in 13 European countries. J Invest Dermatol. 2015;135:984-991. 33. Gelfand JM, Neimann AL, Shin DB, et al. Risk of myocardial infarction in patients with psoriasis. JAMA. 2006;296:1735-1741. 34. Mehta NN, Azfar RS, Shin DB, et al. Patients with severe psoriasis are at increased risk of cardiovascular mortality: cohort study using the General Practice Research Database. Eur Heart J. 2010;31:1000-1006. 35. Neimann AL, Shin DB, Wang X, et al. Prevalence of cardiovascular risk factors in patients with psoriasis. J Am Acad Dermatol. 2006;55:829-835. 36. Brusco NK, Watts JJ. Empirical evidence of recall bias for primary health care visits. BMC Health Serv Res. 2015;15:381. 37. Bland JM, Altman DG. Statistics notes: matching. BMJ. 1994; 309:1128. 38. Chularojanamontri L, Griffiths CE, Chalmers RJ. The Simplified Psoriasis Index (SPI): a practical tool for assessing psoriasis. J Invest Dermatol. 2013;133:1956-1962.
J AM ACAD DERMATOL
Eckert et al 8.e1
VOLUME jj, NUMBER j
Supplemental Table I. Mean HCRU event costs stratified by patient age (Medical Expenditure Panel Survey 2012) Cost, $ Age, y
ER visit
Hospitalization
Physician visit
18-44 45-64 $65
1470 1851 1285
13,339 22,627 19,604
840 1278 1656
ER, Emergency room; HCRU, health care resource utilization.
ARTICLE
The burden of atopic dermatitis in US adults: Health care resource utilization data from the 2013 National Health and Wellness Survey Laurent Eckert, PhD,a Shaloo Gupta, MSc,b Caroline Amand, PhD,a Abhijit Gadkari, PhD,c Puneet Mahajan, PhD,d and Joel M. Gelfand, MDe Chilly-Mazarin, France; Princeton and Bridgewater, New Jersey; Tarrytown, New York; and Philadelphia, Pennsylvania Background: There is a lack of data on the burden of atopic dermatitis (AD) in adults relative to the general population. Objective: To characterize the AD burden in adult patients relative to both matched non-AD controls and matched patients with psoriasis in terms of comorbidities, health care resource utilization (HCRU), and costs. Methods: Adults ($18 years) who self-reported a diagnosis of AD or psoriasis and adult non-AD controls were identified from the 2013 US National Health and Wellness Survey. Patients with AD were propensity scoreematched with non-AD controls and patients with psoriasis on demographic variables. Patientreported outcomes were analyzed between matched cohorts. Results: Patients with AD had a significantly greater risk for atopic comorbidities, as well as significantly greater HCRU and total cost compared with non-AD controls. The burden of AD was generally comparable to that of psoriasis, although patients with AD reported increased use of emergency room visits compared with patients with psoriasis. Limitations: Patient-reported data are susceptible to recall bias and erroneous classification. Conclusions: Adult patients with AD reported a substantial disease burden, suggesting an unmet need for more effective AD treatment options. ( J Am Acad Dermatol http://dx.doi.org/10.1016/j.jaad.2017.08.002.) Key words: Atopic dermatitis; burden of disease; comorbidities; disease severity; health care resource utilization; patient-reported outcomes.
From Sanofi, Chilly-Mazarina; Kantar Health, Princetonb; Regeneron, Tarrytownc; Sanofi, Bridgewaterd; and Department of Dermatology and Department of Biostatistics and Epidemiology, University of Pennsylvania Perelman School of Medicine, Philadelphia.e Supported by Sanofi and Regeneron Pharmaceuticals, Inc. Disclosure: Drs Eckert, Amand, and Mahajan are employees of and stockholders in Sanofi. Dr Gadkari is an employee of and stockholder in Regeneron Pharmaceuticals, Inc. Ms Gupta is an employee of Kantar Health, which has received research funding from Sanofi/Regeneron Pharmaceuticals, Inc. Dr Gelfand is an employee of the University of Pennsylvania Perelman School of Medicine, which has received research funding from Sanofi/ Regeneron Pharmaceuticals, Inc. In the previous 12 months, Dr Gelfand has served as a consultant for Abbvie, AstraZeneca, Celgene Corp, Coherus, Eli Lilly, Janssen Biologics (formerly Centocor), Sanofi, Merck, Novartis Corp, Valeant, and Pfizer Inc,
receiving honoraria; he also receives research grants (to the trustees of the University of Pennsylvania) from Abbvie, Eli Lilly, Janssen, Novartis Corp, Regeneron, Sanofi, and Pfizer Inc, and he has received payment for CME work related to psoriasis. Dr Gelfand is also a coepatent holder of Resiquimod for the treatment of cutaneous T-cell lymphoma. Accepted for publication August 3, 2017. Correspondence to: Laurent Eckert, PhD, Sanofi, 1 Avenue Pierre Brossolette, 91380 Chilly-Mazarin, France. E-mail: Laurent. [email protected]. Published online October 7, 2017. 0190-9622 Ó 2017 by the American Academy of Dermatology, Inc. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). http://dx.doi.org/10.1016/j.jaad.2017.08.002
1
J AM ACAD DERMATOL
2 Eckert et al
n 2017
Atopic dermatitis (AD) is a chronic relapsing, criterion (age $18 years), completed the survey, and inflammatory skin disease characterized by pruritus, gave informed consent for their anonymized data to xerosis, and eczematous lesions and with features be used for research purposes. The NHWS was including erythema, infiltration/papulation, excoriaapproved by the Essex Institutional Review Board tions, and lichenification.1 Although AD most often (Lebanon, NJ).9 2,3 starts in early infancy or childhood, a recent study reported that childhood AD extended into adult life Patient population in approximately half of the The AD cohort comprised cases studied.4 For many adult respondents who indiCAPSULE SUMMARY patients, moderate/severe cated that they had AD is a chronic, lifelong experienced AD in response Limited data exist on the burden of condition that is underreto the question: ‘‘Which of the atopic dermatitis (AD) in adults relative cognized as a major public following conditions have you to the general population and other health concern.2,3,5 experienced in the past chronic skin disorders. Recent publications have 12 months?’’ Possible response Adult patients with AD reported documented the multidimenoptions were to select all significantly increased atopic sional disease burden in paconditions that applied, select comorbidities, health care resource tients with moderate/severe ‘‘I have not experienced any utilization, and costs compared with AD, encompassing increased of the above conditions in the non-AD controls. levels of itch, pain, sleep past 12 months,’’ and select disturbance, anxiety and ‘‘I decline to answer.’’ In The overall burden of AD was similar to depression, and impaired addition, respondents had to that of psoriasis, suggesting an unmet health-related quality of answer yes to the question need for effective treatment options in life.6-9 In addition to the dis‘‘Has your AD been diagnosed adult patients with these common, ease burden, AD is associated by a physician?’’ inflammatory skin disorders. with a substantial economic The non-AD control cohort burden. Costs incurred by comprised adult respondents patients, their families, and payers include physician who did not report experiencing AD, dermatitis, or visits, emergency room (ER) visits, hospitalizations, eczema in the past 12 months. The psoriasis cohort prescriptions, and over-the-counter treatments.10 comprised adult respondents who indicated that they However, comprehensive data on the disease and had experienced psoriasis in response to the question: economic burden of AD relative to the general adult ‘‘Which of the following conditions have you ever population and other chronic skin disorders are experienced?’’ Possible response options were to select limited. To this end, the aim of the present study all conditions that applied, select ‘‘I do not have any of was to evaluate the real-world patient burden, in the above conditions,’’ and select ‘‘I decline to answer.’’ terms of comorbidities, health care resource utilizaIn addition, respondents had to answer yes to the tion (HCRU), and costs, in adults with AD relative to question ‘‘Has your psoriasis been diagnosed by a both adults without AD and adults with psoriasis on physician?’’ Respondents who reported a diagnosis of the basis of data from the 2013 National Health and both AD and psoriasis were excluded from this Wellness Survey (NHWS). An additional aim was to analysis. evaluate the impact of AD severity on these Adults with AD graded their disease severity outcomes. (mild, moderate, or severe) in response to the question ‘‘How severe is your dermatitis/eczema?’’ Adults with psoriasis graded their disease severity in METHODS response to the question ‘‘According to the National Data source Psoriasis Foundation, the palm of the hand equals 1% Patient-level data were obtained from the 2013 US of the skin. Thinking about this, please estimate the NHWS, an Internet-based general population percent of your body surface your psoriasis currently survey.11 The NHWS uses a random stratified affects as follows: mild (\3% body coverage); sampling framework (sex, age, and race/ethnicity) moderate (3%-10% body coverage); or severe to ensure that it is representative of the demographic ([10% body coverage).’’ composition of the US adult population, based on data from the US Bureau of the Census. A total of Propensity score matching 1,183,287 individuals were asked to complete the To correct for potential confounding, the AD 2013 NHWS, 109,592 of whom (9.3%) responded. Of cohort was propensity scoreematched12 to non-AD the responders, 75,000 (68.4%) met the inclusion d
d
d
J AM ACAD DERMATOL
Eckert et al 3
VOLUME jj, NUMBER j
Abbreviations used: AD: CCI: CI: ER: HCRU: NHWS: OR:
atopic dermatitis Charlson comorbidity index confidence interval emergency room health care resource utilization National Health and Wellness Survey odds ratio
controls (1:2) and to patients with psoriasis (1:1 [because of a much smaller AD-to-psoriasis respondent ratio]). The matching variables comprised those variables that were statistically different between the groups at baseline and during iterations of the propensity matching process. Relevant variables were entered into a logistic regression model using the SAS/STATÒ LOGISTIC (SAS Institute, Inc, Cary, NC) procedure to generate the propensity score. Matching was completed by using the ‘‘greedy matching’’ technique. The algorithm proceeds with making the best matches first, and the next-best matches next, and so on in a hierarchical sequence until no more matches can be made. Best matches are those with the highest-digit match on propensity score (5 decimal points). The algorithm proceeds sequentially to the lowest-digit match on propensity score (0 decimal points). Goodness of matched pairs were defined as those with the least absolute difference in matched propensity score. For AD versus non-AD matching, each AD patient was matched with 2 non-AD controls using a 1:2 matching algorithm. For AD versus psoriasis matching, a 1:1 matching algorithm was used. Propensity score matching of patients with AD and non-AD controls was based on the variables age, sex, ethnicity, education, income, insurance status, alcohol consumption, exercise behavior, and Charlson Comorbidity Index (CCI).13 Propensity score matching of patients with AD and patients with psoriasis was based on age, sex, ethnicity, education, employment status, body mass index, insurance status, smoking status, exercise behavior, CCI, and self-rated disease severity (moderate/ severe or mild). After matching, matching variables were re-examined between the matched groups to confirm matching. Comorbidities The CCI was used to evaluate the overall comorbidity burden.13 The self-reported prevalence of arthritis (including osteoarthritis, rheumatoid arthritis, and psoriatic arthritis), atopic-related comorbidities (asthma and hay fever/nasal allergies), hypertension, high cholesterol level, and
osteoporosis/osteopenia was evaluated on the basis of respondents’ answers to the questions ‘‘Which of the following conditions have you experienced in the past 12 months?’’ or ‘‘Which of the following conditions have you ever experienced’’ and ‘‘Has your condition been diagnosed by a physician?’’ HCRU HCRU was defined by the number of visits to a health care provider in the past 6 months (‘‘How many visits did you make to the following traditional health care provider(s) in the past 6 months?’’), ER visits (‘‘How many times have you been to the ER for your own medical condition in the past 6 months?’’), and hospitalizations (‘‘How many times have you been hospitalized for your own medical condition in the past 6 months?’’). Direct costs were estimated using data stratified according to age from the Medical Expenditure Panel Survey 2012.14 For each respondent, the number of each type of visit in the last 6 months was multiplied by 2 to extrapolate to the annual number of visits and then multiplied by its average cost (Supplemental Table I; available at http://www.jaad.org). Analyses The prevalence of comorbidities, HCRU, and costs were analyzed in patients with AD against matched non-AD controls and against matched patients with psoriasis. In addition, these outcomes were examined in the moderate/severe AD cohort versus in the mild AD cohort. A chi-square test or the t test were used to compare differences between groups for categorical or continuous variables, respectively, with the 2-tailed significance level (a) set at .05. A sensitivity analysis of direct cost data was carried out by calculating least squares mean costs after adjusting for the presence of asthma and nasal allergies/hay fever. Analyses were performed with SAS software, version 9.3 (SAS Institute, Inc).
RESULTS AD versus non-AD and moderate/severe versus mild AD Demographics. Demographic characteristics in patients with AD and matched non-AD controls are shown in Table I. The analysis cohort of 1047 adults (mean age, 46.1 years; 68.3% female and 66.8% non-Hispanic white) comprised 349 patients with AD and 698 matched non-AD controls. Self-rated disease severity was available for 306 patients with AD, of whom 52.3% reported moderate (42.8%) or severe (9.5%) disease and 47.7% reported mild disease. Overall, demographic characteristics and CCI score were generally similar between patients
J AM ACAD DERMATOL
4 Eckert et al
n 2017
Table I. Demographic characteristics in patients with AD versus matched non-AD controls and patients with moderate/severe AD versus patients with mild AD Total (N = 1047)
Characteristic
AD (n = 349)
Age, years, mean (SD) 46.1 (15.3) 45.8 Female, n (%) 715 (68.3) 228 Non-Hispanic white, n (%) 699 (66.8) 225 Married/living with partner, n (%) 614 (58.6) 196 College educated, n (%) 571 (54.5) 191 Annual household income $$75000, 334 (31.9) 113 n (%) Employed, n (%) 603 (57.6) 195 Alcohol use, n (%) 759 (72.5) 246 Current smoker, n (%) 201 (19.2) 69 Taking regular exercise, n (%) 727 (69.4) 246 Mean Charlson comorbidity index (SD) 0.81 (1.8) 0.88
Non-AD (n = 698)
P value
Moderate/severe AD (n = 160)
Mild AD (n = 146)
P value
(14.9) 46.3 (15.5) (65.3) 487 (69.8) (64.5) 474 (67.9) (56.2) 418 (59.9) (54.7) 380 (54.4) (32.4) 221 (31.7)
.63 .15 .26 .25 .93 .70
44.7 107 99 84 80 42
(15.3) (66.9) (61.9) (52.5) (50.0) (26.3)
45.4 92 91 87 88 60
(14.0) (63.0) (62.3) (59.6) (60.3) (41.1)
.66 .48 .10 .21 .07 .09
(55.9) 408 (70.5) 513 (19.8) 132 (70.5) 481 (1.8) 0.78
.43 .30 .14 .60 .37
79 106 38 104 1.07
(49.4) (66.3) (23.8) (65.0) (2.2)
95 104 22 118 1.18
(65.1) (71.2) (15.1) (80.8) (3.9)
.006 .35 .15 .002 .76
(58.5) (73.5) (18.9) (68.9) (1.8)
AD, Atopic dermatitis; SD, standard deviation.
Table II. Self-reported comorbid conditions in patients with atopic dermatitis (AD) versus matched non-AD controls and patients with moderate/severe versus mild AD Comorbidity, n (%)
AD (n = 349)
Non-AD (n = 698)
Hypertension 96 (27.5) 178 High cholesterol level* 112 (32.1) 193 Arthritis (osteoarthritis, 103 (29.5) 136 rheumatoid, psoriatic) Osteoporosis/osteopenia 30 (8.60) 51 Asthma 84 (24.1) 62 Nasal allergies/hay fever 169 (48.4) 170
OR (95% CI)
P value
Moderate/severe AD (n = 160)
Mild AD (n = 146)
OR (95% CI)
P value
(25.5) 1.1 (0.8-1.5) .49 (27.7) 1.2 (0.9-1.6) .14 (19.5) 1.7 (1.3-2.3) \.001
49 (30.6) 49 (30.6) 53 (33.1)
36 (24.7) 1.4 (0.8-2.2) 45 (30.8) 1.0 (0.6-1.6) 36 (24.7) 1.5 (0.9-2.5)
.24 .97 .10
(7.3) 1.2 (0.8-1.9) .46 (8.9) 3.3 (2.3-4.7) \.001 (24.4) 2.9 (2.2-3.8) \.001
16 (10.0) 44 (27.5) 82 (51.3)
14 (9.6) 1.1 (0.5-2.2) 36 (24.7) 1.2 (0.7-1.9) 71 (48.6) 1.1 (0.7-1.7)
.90 .57 .65
AD, Atopic dermatitis; CI, confidence interval; OR, odds ratio. *High cholesterol was defined as a total cholesterol serum concentration of $240 mg/dL.
with AD and patients without AD and between patients with moderate/severe AD and patients with mild AD. However, significantly lower proportions of patients with moderate/severe AD than patients with mild AD reported being employed (49.4% vs 65.1% [P = .006]) or exercising (65.0% vs 80.8% [P = .002]). Comorbidities. The self-reported prevalence of comorbid conditions in patients with AD versus matched non-AD controls, and patients with moderate/severe AD versus patients with mild AD is presented in Table II. Compared with non-AD controls, patients with AD had a significantly increased risk for arthritis (odds ratio [OR], 1.7; 95% confidence interval [CI], 1.3-2.3; P \ .001), asthma (OR, 3.3; 95% CI, 2.3-4.7; P \ .001), and nasal allergies/hay fever (OR, 2.9; 95% CI, 2.2-3.8; P \ .001). The risk for comorbidities in patients with moderate/severe AD was numerically higher, although not statistically significant compared with patients with mild AD.
HCRU and costs. Compared with non-AD controls, patients with AD reported using significantly more health care resources (Table III). Mean (SD) health care provider visits were 6.6 (9.8) versus 3.9 (5.4) (P \ .001). The number of ER visits and hospitalizations was more than twice that in non-AD controls. In patients with AD, the mean (SD) number of ER visits was 0.5 (1.2) versus 0.2 (0.8) (P \ .001) and the mean (SD) number of hospitalizations was 0.3 (1.0) versus 0.1 (0.7) (P = .004). Patients with moderate/severe AD used numerically greater resources than patients with mild AD; however, none of the differences were statistically significant. In accordance with the HCRU data, patients with AD incurred significantly higher costs than non-AD controls across all evaluated categories (Table III). Thus, the total mean (SD) annual per patient direct costs in patients with AD were $9782 higher than those of non-AD controls: $24,401 ($37,355) versus $14,619 ($29,799) (P \ .001). Controlling for the
J AM ACAD DERMATOL
Eckert et al 5
.21
presence of asthma and nasal allergies/hay fever reduced the difference between AD and non-AD in total mean direct costs to $7901 (P = .001) (Table III). Patients with moderate/severe AD incurred numerically higher costs than did patients with mild AD across the evaluated categories. Controlling for the presence of asthma and nasal allergies/hay fever increased the difference between moderate/severe versus mild AD in total mean direct costs from $3757 to $5522. None of the differences were statistically significant (Table III).
AD, Atopic dermatitis; CI, confidence interval; ER, emergency room; HCRU, health care resource utilization; LS, least squares; SD, standard deviation. *Direct costs were adjusted for the presence of asthma and nasal allergies/hay fever.
21,642 (16,768-27,934) 27,164 (21,290-34,659) .0012 14,870 (12,868-17,183) 22,771 (18,509-28,013)
.24 .98 .06 .42 (22,070) (28,837) (3004) (43,252) 13,867 8668 1147 23,682 (20,258) (26,973) (4420) (37,756) 16,735 8741 1963 27,439 \.001 .026 \.001 \.001 (13,882) (21,203) (2339) (29,799) 9252 4759 608 14,619 (19,893) (26,520) (3602) (37,355) 14,797 8145 1459 24,401
.32 .96 .08 6.3 (11.5) 0.3 (1.0) 0.4 (1.0) 7.6 (9.6) 0.3 (1.0) 0.6 (1.4) \.001 .004 \.001 3.9 (5.4) 0.1 (0.7) 0.2 (0.8) 6.6 (9.8) 0.3 (1.0) 0.5 (1.2)
Mean HCRU (SD) Health care provider visits in past 6 mo Hospitalizations in past 6 mo ER visits in past 6 mo Mean annual per patient costs (SD), $, Health care provider visit costs Hospitalization costs ER visit costs Total direct costs LS mean adjusted annual per patient costs, $, (95% CI)* Total direct costs
P value Mild AD (n = 146) Moderate/severe AD (n = 160) P value Non-AD (n = 698) AD (n = 349)
Table III. HCRU and costs in patients with atopic dermatitis (AD) versus matched non-AD controls and patients with moderate/severe versus patients with mild AD
VOLUME jj, NUMBER j
AD versus psoriasis Demographics. The analysis cohort comprised 260 patients with AD and 260 matched patients with psoriasis. Demographic characteristics were similar between the 2 groups (Table IV). Comorbidities. The self-reported prevalence of comorbidities in patients with AD versus matched patients with psoriasis are presented in Table V. Compared with patients with psoriasis, patients with AD had significantly higher odds of reporting asthma (OR, 1.7; 95% CI, 1.1-2.7; P = .01) and nasal allergies/hay fever (OR, 2.3; 95% CI, 1.6-3.2; P \ .001). Conversely, patients with psoriasis had higher odds of developing hypertension (OR, 0.7; 95% CI, 0.5-1.0; P = .03). HCRU and costs. HCRU in patients with AD versus in matched patients with psoriasis is shown in Table VI. Although the number of visits was low, the mean (SD) number of ER visits required by patients with AD in the past 6 months was significantly higher than that of patients with psoriasis: 0.5 (1.1) versus 0.3 (1.0) (P = .02). Accordingly, mean (SD) annual per patient ER visit costs incurred by patients with AD were almost twice those of patients with psoriasis: $1448 ($3421) versus $750 ($3008) (P = .01). The number of hospitalizations and hospitalization costs were also higher in patients with AD, but these differences did not attain statistical significance.
DISCUSSION To improve the morbidity associated with AD in adults, it is necessary to characterize its disease burden in the general population. This comparative analysis of comorbidities, HCRU, and costs indicates that US adult patients with AD have a significantly increased risk for reporting atopic comorbidities such as asthma and nasal allergies/hay fever compared with matched non-AD controls. The association between AD and atopic comorbid conditions has been recognized previously15-17 and is consistent with the prevailing concept of the
J AM ACAD DERMATOL
6 Eckert et al
n 2017
Table IV. Demographic characteristics in patients with AD and matched patients with psoriasis AD (n = 260)
Characteristic
Mean age (SD), y Female, n (%) Non-Hispanic white, n (%) Married/living with partner, n (%) College educated, n (%) Annual household income $$75,000, n (%) Employed, n (%) Alcohol use, n (%) Current smoker, n (%) Regularly exercise, n (%) Mean Charlson comorbidity index (SD)
45.4 169 167 149 148 82 152 181 48 189 0.80
Psoriasis (n = 260)
(14.8) (65.0) (64.2) (57.3) (56.9) (31.5) (58.5) (69.6) (18.5) (72.7) (1.9)
44.7 170 170 156 146 72 159 181 41 188 0.71
P value
(15.2) (65.4) (65.4) (60.0) (56.2) (27.7) (61.2) (69.6) (15.8) (72.3) (1.3)
.56 .93 .72 .53 .86 .73 .53 1.00 .71 .92 .55
AD, Atopic dermatitis; SD, standard deviation.
Table V. Self-reported comorbid conditions in AD and matched patients with psoriasis AD (n = 260)
Comorbidity, n (%)
Hypertension High cholesterol concentration* Arthritis (osteoarthritis, rheumatoid, or psoriatic) Osteoporosis/osteopenia Asthma Nasal allergies/hay fever
70 75 66 22 64 129
(26.9) (28.9) (25.4) (8.5) (24.6) (49.6)
Psoriasis (n = 260)
OR (95% CI)
P value
93 79 73 17 41 79
0.7 0.9 0.9 1.3 1.7 2.3
.03 .70 .49 .41 .01 \.001
(35.8) (30.4) (28.1) (6.5) (15.8) (30.4)
(0.5-1.0) (0.6-1.4) (0.6-1.3) (0.7-2.6) (1.1-2.7) (1.6-3.2)
AD, Atopic dermatitis; CI, confidence interval; OR, odds ratio. *High cholesterol concentration was defined as a total cholesterol serum concentration of $240 mg/dL.
‘‘atopic march’’ from AD to other common allergic conditions.18 In accordance with the high comorbidity burden in patients with AD, HCRU was significantly higher in the AD cohort compared with non-AD controls for all categories evaluated. The difference was largely driven by health care provider visits and associated costs. Although the numbers of reported hospitalizations and ER visits were small, the utilization of these categories by patients with AD was approximately twice that reported by non-AD controls. Overall, mean annual per patient total direct costs in patients with AD were approximately $10,000 higher than those of non-AD controls ($24,401 vs $14,619). Controlling for the presence of asthma and nasal allergies/hay fever reduced the AD versus non-AD difference in total mean direct costs by approximately 20%, which is in accordance with the higher atopic burden in patients with AD.15-17 Patients with more severe AD have been shown to be at a higher risk for atopic comorbidities.19 Also, they are generally associated with significantly higher all-cause HCRU and costs relative to less severe AD.20,21 The nonsignificant differences
between severity levels in this study were likely due to insufficient statistical power to detect small differences in outcomes between the moderate/ severe and mild subgroups. It is important to note that more than 80% of the patients in the moderate/ severe group reported their disease as moderate. Thus, the comparative analysis of burden according to severity was predominantly that between patients with moderate disease and those with mild disease, which partly explains why the differences are not larger. An additional factor that may have had an influence on the severity analyses was that severity was patient rated and the level of concordance between patient- and physician-rated severity was not assessed in the current study. Previous studies have reported significant differences between patient- and physician-rated severity in adult patients.22,23 These results highlight the need for objective assessment of AD severity in patient self-report surveys through validated instruments, such as the Patient-Oriented Severity Scoring of Atopic Dermatitis,24 or the Patient Oriented Eczema Measure.25 The present analysis and the companion article in this issue of the Journal of the American Academy of
J AM ACAD DERMATOL
Eckert et al 7
VOLUME jj, NUMBER j
Table VI. HCRU in patients with AD and matched patients with psoriasis
Mean HCRU (SD) Health care provider visits in past 6 mo Hospitalizations in past 6 mo ER visits in the past 6 mo Mean annual per patient costs (SD), $ Health care provider visit costs Hospitalization costs ER visit costs Total direct costs
AD (n = 260)
Psoriasis (n = 260)
P value
6.2 (8.0) 0.3 (1.0) 0.5 (1.1)
6.3 (9.8) 0.1 (0.8) 0.3 (1.0)
.88 .07 .02
14,058 8177 1448 23,682
(18,252) (27,040) (3421) (37,041)
14,171 4244 750 19,165
(21,495) (21,435) (3008) (34,459)
.95 .07 .01 .15
AD, Atopic dermatitis; ER, emergency room; HCRU, health care resource utilization; SD, standard deviation.
Dermatology (Eckert et al) are, to the best of our knowledge, the first publications to provide comparative data on the disease burden in adults for AD and psoriasis. Like AD, psoriasis is a common inflammatory skin disease, with a prevalence in adults of about 3%.26-28 Patients with chronic plaque psoriasis, the most common form of the disease (about 90% of cases), present with itching and erythematous plaques covered by silver lamellar scales.29 Psoriasis is associated with impairments in health-related quality of life and an excess risk for cardiometabolic disease and mortality, especially in patients with more severe disease.30-35 Our analyses of NHWS real-world data indicate that patients with AD have a significantly higher risk for atopic comorbidities than do patients with psoriasis. With regard to HCRU, the numbers of hospitalizations and ER visits of patients with AD were approximately double those reported by patients with psoriasis. Interestingly, patients with psoriasis reported higher rates of hypertension, confirming the different cardiovascular risk profile compared with AD. A strength of this study is that the NHWS is representative of the demographic composition of the US adult population, and thus, the results presented here are likely to be applicable to the wider population of US adult patients with AD. Propensity score matching of patients with AD and non-AD controls, which was carried out in an effort to minimize confounding due to demographic characteristics, was an additional strength of this study. The inherent limitation of the study is that patient-reported data are susceptible to recall bias36 and erroneous classification, given that such data are not independently verified against clinical records. However, these errors would be expected to be nondifferential and thus unlikely to mask differences in burden between groups. Furthermore, we matched our cohorts on the basis of numerous factors to minimize the potential for confounding to affect the results, with the caveat that
overmatching can result in the masking of true associations.37 The NHWS does not include data on key AD- and psoriasis-specific outcome measures, such as itch and pain; nor does it allow the clinical evaluation of AD severity using SCORAD24 or psoriasis severity using the Simplified Psoriasis Index.38 An additional limitation is that HCRU data could not be definitively attributed to AD or psoriasis because claims specifically linking resource use with these conditions are not captured by the NHWS database.
CONCLUSIONS The results from this analysis indicate that adult patients with AD experience a substantial patient burden relative to matched controls without AD, as is evidenced by a significantly increased risk for development of atopic conditions (asthma and hay fever/nasal allergies) and arthritis, as well as increased HCRU and associated costs. Adult patients with AD reported a significantly higher prevalence of atopic conditions and a significantly increased number of ER visits and associated costs relative to matched patients with psoriasis, whereas psoriasis was associated with a greater prevalence of hypertension. Taken together, these data suggest a significant need for enhanced treatment options for patients with AD. The authors would like to thank the NHWS respondents and all investigators involved in this study. Medical writing support, including developing outline and initial drafts and incorporation of author/reviewer comments was provided by Fernando Gibson, PhD (of Prime, London, UK). Editorial support, including formatting and manuscript submission, was provided by Sinead Stewart (of Prime, Knutsford, UK). Medical writing and editorial support was funded by Sanofi/Regeneron Pharmaceuticals, Inc, according to Good Publication Practice guidelines. The authors were responsible for all content and editorial decisions, and they received no honoraria related to the development of this publication.
8 Eckert et al
REFERENCES 1. Weidinger S, Novak N. Atopic dermatitis. Lancet. 2016;387: 1109-1122. 2. Bieber T. Atopic dermatitis. N Engl J Med. 2008;358:1483-1494. 3. Margolis JS, Abuabara K, Bilker W, et al. Persistence of mild to moderate atopic dermatitis. JAMA Dermatol. 2014;150: 593-600. 4. Napolitano M, Megna M, Patruno C, et al. Adult atopic dermatitis: a review. G Ital Dermatol Venereol. 2016;151:403-411. 5. Gittler JK, Shemer A, Suarez-Farinas M, et al. Progressive activation of T(H)2/T(H)22 cytokines and selective epidermal proteins characterizes acute and chronic atopic dermatitis. J Allergy Clin Immunol. 2012;130:1344-1354. 6. Simpson EL, Bieber T, Eckert L, et al. Patient burden of moderate to severe atopic dermatitis (AD): insights from a phase 2b clinical trial of dupilumab in adults. J Am Acad Dermatol. 2016;74:491-498. 7. Simpson EL, Bieber T, Guttman-Yassky E, et al. Two phase 3 trials of dupilumab versus placebo in atopic dermatitis. N Engl J Med. 2016;375:2335-2348. 8. Simpson EL, Gadkari A, Worm M, et al. Dupilumab therapy provides clinically meaningful improvement in patient-reported outcomes (PROs): a phase IIb, randomized, placebo-controlled, clinical trial in adult patients with moderate to severe atopic dermatitis (AD). J Am Acad Dermatol. 2016;75:506-515. 9. Whiteley J, Emir B, Seitzman R, et al. The burden of atopic dermatitis in US adults: results from the 2013 National Health and Wellness Survey. Curr Med Res Opin. 2016:1-7. 10. Drucker AM, Wang AR, Qureshi AA. Research gaps in quality of life and economic burden of atopic dermatitis: the National Eczema Association Burden of Disease Audit. JAMA Dermatol. 2016;152:873-874. 11. Kantar Health. The National Health and Wellness Survey. Available at: http://www.kantarhealth.com/docs/datasheets/ kh-national-health-and-wellness-survey.pdf. Accessed July 5, 2016. 12. Austin PC. A comparison of 12 algorithms for matching on the propensity score. Stat Med. 2014;33:1057-1069. 13. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373-383. 14. US Department of Health and Human Services. Medical Expenditure Panel Survey (MEPS). Available at: https://meps. ahrq.gov/mepsweb/. Accessed August 24, 2017. 15. Boguniewicz M, Leung DY. Atopic dermatitis: a disease of altered skin barrier and immune dysregulation. Immunol Rev. 2011;242:233-246. 16. Illi S, von Mutius E, Lau S, et al. The natural course of atopic dermatitis from birth to age 7 years and the association with asthma. J Allergy Clin Immunol. 2004;113:925-931. 17. Leung DY, Guttman-Yassky E. Deciphering the complexities of atopic dermatitis: shifting paradigms in treatment approaches. J Allergy Clin Immunol. 2014;134:769-779. 18. Shaker M. New insights into the allergic march. Curr Opin Pediatr. 2014;26:516-520. 19. Simpson E, Guttman-Yassky E, Margolis DJ, et al. Chronicity, comorbidity and life course impairment in atopic dermatitis: insights from a cross-sectional study in US adults. Presented at: 25th European Academy of Dermatology and Venereology. September 28-October 2, 2016; Vienna, Austria.
J AM ACAD DERMATOL
n 2017
20. Drucker A, Stemart A, Amand C, et al. Burden of atopic dermatitis: a claims-based analysis. Presented at: American Academy of Dermatology 74th Annual Meeting. March 4-8, 2016; Washington DC. 21. Shrestha S, Miao R, Wang L, et al. Burden of atopic dermatitis: comorbidities, healthcare resource utilization, and costs in US commercial and medicare adult populations. Presented at: Academy of Managed Care Pharmacy (AMCP) Nexus. October 3-6, 2016; National Harbor, WA. 22. Wei W, Anderson P, Gadkari A, et al. Disagreement between physician- and patient-reported disease severity in adults with a history of moderate-to-severe atopic dermatitis. Presented at: 75th Annual Meeting of the Society for Investigative Dermatology. May 11-14, 2016; Scottsdale, AZ. 23. Fivenson D, Arnold RJ, Kaniecki DJ, et al. The effect of atopic dermatitis on total burden of illness and quality of life on adults and children in a large managed care organization. J Manag Care Pharm. 2002;8:333-342. 24. European Task Force on Atopic Dermatitis. Severity scoring of atopic dermatitis: the SCORAD index. Consensus Report of the European Task Force on Atopic Dermatitis. Dermatology. 1993; 186:23-31. 25. Charman CR, Venn AJ, Williams HC. The patient-oriented eczema measure: development and initial validation of a new tool for measuring atopic eczema severity from the patients’ perspective. Arch Dermatol. 2004;140:1513-1519. 26. Helmick CG, Lee-Han H, Hirsch SC, et al. Prevalence of psoriasis among adults in the U.S.: 2003-2006 and 2009-2010 National Health and Nutrition Examination Surveys. Am J Prev Med. 2014;47:37-45. 27. Rachakonda TD, Schupp CW, Armstrong AW. Psoriasis prevalence among adults in the United States. J Am Acad Dermatol. 2014;70:512-516. 28. Kurd SK, Gelfand JM. The prevalence of previously diagnosed and undiagnosed psoriasis in US adults: results from NHANES 2003-2004. J Am Acad Dermatol. 2009;60:218-224. 29. Boehncke WH, Schon MP. Psoriasis. Lancet. 2015;386:983-994. 30. Bhosle MJ, Kulkarni A, Feldman SR, et al. Quality of life in patients with psoriasis. Health Qual Life Outcomes. 2006;4:35. 31. Shutty BG, West C, Huang KE, et al. Sleep disturbances in psoriasis. Dermatol Online J. 2013;19:1. 32. Dalgard FJ, Gieler U, Tomas-Aragones L, et al. The psychological burden of skin diseases: a cross-sectional multicenter study among dermatological out-patients in 13 European countries. J Invest Dermatol. 2015;135:984-991. 33. Gelfand JM, Neimann AL, Shin DB, et al. Risk of myocardial infarction in patients with psoriasis. JAMA. 2006;296:1735-1741. 34. Mehta NN, Azfar RS, Shin DB, et al. Patients with severe psoriasis are at increased risk of cardiovascular mortality: cohort study using the General Practice Research Database. Eur Heart J. 2010;31:1000-1006. 35. Neimann AL, Shin DB, Wang X, et al. Prevalence of cardiovascular risk factors in patients with psoriasis. J Am Acad Dermatol. 2006;55:829-835. 36. Brusco NK, Watts JJ. Empirical evidence of recall bias for primary health care visits. BMC Health Serv Res. 2015;15:381. 37. Bland JM, Altman DG. Statistics notes: matching. BMJ. 1994; 309:1128. 38. Chularojanamontri L, Griffiths CE, Chalmers RJ. The Simplified Psoriasis Index (SPI): a practical tool for assessing psoriasis. J Invest Dermatol. 2013;133:1956-1962.
J AM ACAD DERMATOL
Eckert et al 8.e1
VOLUME jj, NUMBER j
Supplemental Table I. Mean HCRU event costs stratified by patient age (Medical Expenditure Panel Survey 2012) Cost, $ Age, y
ER visit
Hospitalization
Physician visit
18-44 45-64 $65
1470 1851 1285
13,339 22,627 19,604
840 1278 1656
ER, Emergency room; HCRU, health care resource utilization.
