eTable 1. Associations of socio-demographic factors with prevalences of eczema in adults in 2010 and 2012 NHIS
eTable 2. Healthcare costs, lost work days and overall health in eczema and EAH in adults from the 2010 NHIS
eTable 3. Healthcare utilization in US adults with eczema and EAH from the 2010 NHIS
eTable 4. Access to care in US adults with eczema and EAH from the 2010 NHIS
eTable 5. Healthcare costs, lost work days and overall health in eczema and EAH in adults from the 2012 NHIS
eTable 6. Healthcare utilization in US adults with eczema and EAH from the 2012 NHIS
eTable 7. Access to care in US adults with eczema and EAH from the 2012 NHIS
Silverberg JI. Health Care Utilization, Patient Costs, and Access to Care in US Adults With EczemaA Population-Based Study. JAMA Dermatol. 2015;151(7):743-752. doi:10.1001/jamadermatol.2014.5432
Little is known about the health burden of adult eczema in the United States.
To study the out-of-pocket costs, health care access and utilization in adult eczema in the United States.
Design, Setting, and Participants
Two US population-based studies, the 2010 and 2012 National Health Interview Surveys, surveyed 27 157 and 34 613 adults (ages 18-85 years).
History of eczema.
Main Outcomes and Measures
The out-of-pocket costs, lost workdays, days in bed, and access to care.
Adults with eczema had $371 to $489 higher out-of-pocket costs per person-year compared with those without eczema, with higher odds of increased out-of-pocket costs (survey multinomial logistic regression, adjusted odds ratios [ORs] [95% CIs] for NHIS 2012, <$1 to 499: OR, 1.27; 95% CI, 1.05-1.54; $500 to $1999: OR, 1.49; 95% CI, 1.22-1.81; $2000-$2999: OR, 1.74; 95% CI, 1.36-2.21; $3000-$4999: OR, 2.07; 95% CI, 1.56-2.73; ≥$5000: OR, 1.74; 95% CI, 1.34-2.27; P < .001). Adults with eczema were significantly more likely to have at least 6 lost workdays from all causes (OR, 1.53; 95% CI, 1.26-1.84), 1 to 2 half-days (OR, 1.31; 95% CI, 1.14-1.51); 3 to 5 half-days (OR, 1.84; 95% CI, 1.54-2.20), and at least 6 half-days (OR, 2.24; 95% CI, 1.92-2.62) in bed and increased health care utilization with more physician visits (1-3 visits: OR, 1.70; 95% CI, 1.40-2.07; 4-9 visits: OR, 2.45; 95% CI, 2.00-3.00; and ≥10 visits: OR, 3.33; 95% CI, 2.69-4.12), urgent or emergency care visits (1-3 visits: OR, 1.46; 95% CI, 1.29-1.66; 4-9 visits: OR, 1.81; 95% CI, 1.27-2.57; and ≥10 visits: OR, 2.43; 95% CI, 1.19-4.99) and hospitalizations (OR, 1.37; 95% CI, 1.17-1.60). Adults with eczema had significantly limited access to care with inability to afford prescription medications (OR, 2.36; 95% CI, 1.92-2.81), were unable to get an appointment soon enough (OR, 2.04; 95% CI, 1.73-2.41), had to wait too long to see a physician (OR, 1.59; 95% CI, 1.28-1.97), had delayed care (OR, 1.73; 95% CI, 1.49-2.01), and were not able to get care (OR, 1.66; 95% CI, 1.40-1.97) because of worry about the related costs.
Conclusions and Relevance
This study demonstrates a large health burden of eczema in adults and suggests substantial out-of-pocket costs, indirect costs from lost workdays and sick days, and increased health care utilization.
Atopic dermatitis (AD or eczema) is a chronic inflammatory skin disorder that is a clinically significant cause of morbidity, quality-of-life impairment, and health care costs. A recent population-based study of US adults found the prevalence of adult eczema to be 10.2%.1 This surprisingly high prevalence was remarkably similar to the 10.7% prevalence of eczema found in US children.2 This study suggested that eczema affects many more US adults than previously recognized. Previous epidemiologic studies in children found that the prevalence of eczema has increased from 7.4% in 1997 to 1999 to 12.5% in 2009 to 2011.3 The increasing prevalence of eczema may explain why previous prevalence estimates were much lower. Nevertheless, if the prevalence of childhood eczema and, by extension, adult eczema, is increasing in the United States, it is logical that the cost of eczema care is also increasing. However, little is known about the direct and indirect costs of adult eczema, and recent cost estimates for adult eczema are lacking.
Previous studies estimated the annual costs of eczema care in children to be $364 million in 19904 and $0.9 to $3.8 billion in 1997.5 However, these studies focused on the cost of eczema care to insurers and not on patients’ burden of disease. I hypothesized that eczema is associated with considerable out-of-pocket costs for health care beyond insurer’s cost. Furthermore, I hypothesized that adults with eczema have a considerable burden of disease related to more lost workdays and sick days, as well as the time and effort needed to seek care for their eczema. Previous studies6,7 showed that dermatology patients may experience delays before receiving appropriate care. I hypothesized that adults with eczema, especially those with multiple flares, have delayed or inadequate access to health care. The goal of this study was to determine the patient health burden and patient costs of adult eczema in the United Sates using data obtained from 2 large population-based surveys of adults in the United States.
The 2010 and 2012 National Health Interview Survey (NHIS) questionnaires were collected by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention. The NHIS is the principal source of information on the health of the civilian noninstitutionalized population of the United States. The questionnaires included a separate core module with questions to estimate the prevalence of various adult health issues. The surveys were administered in-person to selected households by the US Census Bureau using approximately 400 trained interviewers with computer-assisted personal interviewing. Subsequently, 1 adult per household was randomly selected for the sample adult questionnaires. Interviews were conducted in English and Spanish. Using data from the US Census Bureau, sample weights were created by the NHIS that factored age, sex, race, ethnicity, household size, and educational attainment of the most educated household member using a multistage area probability sampling design. These sample weights are needed to provide nationally representative frequency and prevalence estimates for each state’s population of noninstitutionalized adults older than 17 years. Raw and weighted frequency and weighted prevalence estimates are presented, which reflect this complex weighting. The study was approved by the institutional review board at Northwestern University.
One-year history of eczema was determined by an affirmative response to the questions, “During the past 12 months, have you had dermatitis, eczema, or any other red, inflamed skin rash?” in the 2010 NHIS and “During the past 12 months, have you been told by a physician or other health professional that you had eczema or any kind of skin allergy?” in the 2012 NHIS. Lifetime prevalence of asthma was determined by affirmative response to the question “Have you ever been told by a physician or other health professional that you had asthma?” History of hay fever was determined by response to the question “During the past 12 months, have you been told by a physician or other health professional that you had hay fever?” Subset analyses were done using a composite variable for no eczema, eczema without asthma or hay fever, and eczema with asthma and/or hay fever (EAH) was created.
A number of associations with eczema were examined, including age, sex, race, Hispanic origin, household income, out-of-pocket cost for health care, number of lost workdays overall and number of lost workdays from eczema (2010 only), number of days in bed for at least half a day, number of physician visits overall, seeing a physician for eczema (2010 only), number of visits to an emergency department and hospitalizations, number of nights in hospital, receiving home health care by nurse or other health care professional, number of home visits, having a usual place for sick visits and the type of place, use of preventive care, inability to afford prescriptions, reasons for delayed care, including being unable to reach the physician’s office by telephone, inability to get an appointment soon enough, having to wait too long to see physician, the office or clinic not being open, not having transportation, and worry about the cost. These outcomes are summarized in the Box.
Increased out-of-pocket costs
Increased health care utilization
Home health care
Half-days in bed
Delayed access to care
Patient worried about cost
Office or clinic was not open
Patient did not have transportation
Patient had to wait too long to see physician
Patient could not afford prescriptions
Patient was unable to reach physician on telephone
Patient was unable to get an appointment soon
All data processing and statistical analyses were performed with SAS statistical software (version 9.4; SAS Institute Inc). Analyses of survey responses were performed using SURVEY procedures. Weighted frequencies and prevalences were determined (PROC SURVEYFREQ). Total number of sick days and days spent in bed, office visits, and hospitalizations for the US adult population were calculated using the sample weights as noted herein. Out-of-pocket costs were divided by the NCHS into categories (<$500, $500-$1999, $2000-$2999, $3000-$4999, ≥$5000). Therefore, total out-of-pocket costs were calculated by multiplying the weighted frequencies of patients who responded with a particular cost range by the midpoint value of that cost range.
Bivariate and multivariable logistic regression models were constructed with the measures of health care utilization, cost, and access to care as noted, as the binary (binomial logistic regression) and ordinal or categorical (multinomial logistic regression) dependent variables. Multinomial regression was chosen over ordinal regression because none of the outcomes met the proportional odds assumption (score test, P < .01). Eczema was modeled as a binary independent variable. In addition, my a priori hypothesis was that comorbid asthma may modify the cost and burden of disease. Therefore, eczema was also modeled as a categorical independent variable (none, eczema, EAH). Odds ratios (ORs) and 95% CIs were estimated. Multivariable models included significant associations from bivariate analyses as covariates, including age (continuous), sex (male or female), race (African American/black, Asian, Native American/American Indian, white, multiracial/other), Hispanic origin (yes or no), household income (<$35 000, $35 000-$74 999, $75 000-$99 999, ≥$100 000), and current insurance coverage (yes or no). Two-way interaction terms between covariates were tested and included in the final models only if significant (P < .05) and modified the effect size by greater than 20%. Adjusted OR (aOR) and 95% CI were estimated that controlled for these covariates. Complete data analysis was performed; that is, individuals with missing data were excluded. Multicollinearity was tested using variable inflation factors and tolerance.
Data were collected from 27 157 adults for the 2010 NHIS and 34 613 for the 2012 NHIS, representing all age, sex, and racial/ethnic groups. The US prevalence of and associations with adult eczema from the 20101 and 20128 NHIS were previously presented. Briefly, the 2010 US prevalence (95% CI) of eczema in adults was 10.2% (95% CI, 9.7%-10.6%), of whom 7.0% (95% CI, 6.7%-7.4%) had eczema alone and 3.2% (95% CI, 2.9%-3.4%) had EAH, respectively. The 2012 US prevalence (95% CI) of eczema in adults was 7.2% (95% CI, 6.9%-7.6%), of whom 3.7% (95% CI, 3.4%-4.0%) had eczema alone and 3.5% (95% CI, 3.3%-3.8%) had EAH, respectively. Associations between eczema and sociodemographic variables for both studies are presented in eTable 1 in the Supplement.
Adults with eczema paid $37 762 442 054 and $29 341 828 250 in out-of-pocket health care costs in 2010 and 2012, respectively. There were significantly higher out-of-pocket health care costs than for those without eczema overall (an average of $371 and $489 per person-year, respectively) and at all cost strata (Table 1 and Table 2); the associations remained significant in multivariable models. For comparison, I examined the excess out-of-pocket costs in adults with hypertension and diabetes mellitus; these disorders were associated with lower out-of-pocket costs than eczema (hypertension, $206 and $241; diabetes mellitus, $353 and $210, respectively).
Eczema was associated with a total of 68 599 942 and 73 073 034 weighted days of lost work in 2012 and 2010 from any cause, respectively; many of the lost workdays in patients with eczema were due directly to their eczema (5 898 289). In multivariable models, adults with eczema were significantly more likely to have 6 or more lost workdays from any cause than those without eczema.
Adults with eczema were significantly more likely to have 1 to 2, 3 to 5, and 6 or more days where they were in bed for half a day from any cause. The total weighted number of days in bed was 221 607 545 in 2010 and 808 857 142 in 2012 with an average of 6.3 and 5.0 additional days per person-year, respectively.
In multivariate models, eczema was associated with increased odds of 1 to 3, 4 to 9, and 10 or more physician visits per year (Table 3 and Table 4). In the 2010 NHIS, approximately 3 of 4 adults with eczema (6 686 047) had seen a physician for their eczema in the previous 12 months.
Adults with eczema had a weighted total of 9 702 968 (for 2010) and 7 790 657 (for 2012) visits to an urgent care center or emergency department in the previous 12 months, respectively. In multivariable models, eczema was associated with higher odds of 1 to 3, 4 to 9, and/or 10 or more visits (Table 3 and Table 4). Eczema was also associated with higher odds of hospitalization in both studies and at least 4 nights in the hospital in the 2010 NHIS. In the 2010 NHIS, adults with eczema had a higher odds of receiving home health care by a nurse or other health care professional; however, this was not significant in the 2012 NHIS.
Adults with eczema had higher odds of having 1 or more usual place for sick visits (Table 3 and Table 4). There were no differences in the types of locations used for sick visits by adults with eczema compared with those without eczema. There were increased odds of preventive care usage in patients with eczema compared without those with eczema in the 2012 NHIS but not the 2010 NHIS.
There were multiple differences of access to care in adults with eczema compared with those without eczema, including not being able to afford prescription medications, higher odds of delayed care from being unable to reach the physician’s office by telephone, inability to get an appointment soon enough, lack of transportation, having to wait too long to see the physician, and the office not being open when they could get there (Table 5 and Table 6). Finally, adults with eczema were more likely to have delayed care and not get care because of worry about the related costs.
There were significant statistical interactions, such that adults with EAH had even higher odds of out-of-pocket costs and increased utilization (eTables 2-7 in the Supplement).
Using a population-based approach, the present study found that eczema was associated with (1) higher out-of-pocket health care costs, (2) more lost workdays owing to eczema and other causes, (3) poorer overall health, (4) more physician visits due to eczema and other causes, (5) more urgent care and emergency department visits, (6) higher odds of prolonged hospitalizations, and (7) higher odds of home health care visits. Finally, adults with eczema reported considerably impaired access to care with (8) inability to afford prescription medications, (9) delayed care, and (10) not receiving care when needed. This study demonstrates that adults with eczema have a major health burden with significantly increased health care utilization and costs (eg, out-of-pocket health care expenses and more physician visits owing to eczema), as well as indirect costs (eg, lost workdays owing to eczema). Furthermore, eczema is associated with an increased public health burden given its high prevalence in the United States and the substantial number of US adults with increased costs and utilization, despite having very low mortality. Similarly, the Global Burden of Disease Study9 performed in 2010 found that skin conditions were the fourth leading cause of nonfatal burden, of which eczema was the leading cause of disability-adjusted life-years.
A previous study4 of emergency department visits in a children’s hospital as well as national data sets conservatively estimated the total cost for the treatment of childhood AD in the United States in 1990 to be $364 million. A retrospective claims-based study from a private insurer in the central United States and state Medicaid program in the eastern United States found the cost of AD in 1997 to 1998 to be $0.9 to $3.8 billion in direct payer costs.5 The advantage of that claims-based study was the ability to get detailed information about provider (eg, physicians, nurses, patient care assistants, nurse practitioners, therapists) costs and inclusion of costs related to comorbid problems. However, those estimates are more than 10 year out of date, out-of-pocket costs were not assessed, and no information was available for accessibility and affordability of care. Because the prevalence of eczema has steadily increased over the past few decades, it is likely that health care costs in AD have also increased. The present study found an average of $371 to $489 higher out-of-pocket health care costs and increased utilization in adults with eczema compared with those without eczema, including more physician, urgent, and home care visits, and hospitalizations. Future studies are needed to determine the cost of eczema care per se and how it has changed over the years and identify approaches to improve cost-efficiency of eczema care.
In the present study, the increased utilization observed in adults with eczema was only partially due to their eczema per se. There is likely a multitude of comorbid disorders associated with eczema that contributes toward the increased utilization, aside from EAH, or food allergy. Indeed, there were significant statistical interactions, such that adults with EAH had even greater burden of disease, out-of-pocket costs, and health care utilization. In addition, previous studies found associations between childhood eczema and increased extracutaneous infections,10 dental health problems,11 and mental health comorbidity.2 Future research is needed to identify other comorbidities in eczema and determine their impact on the cost and burden of disease.
The impaired access to care reported by adults with eczema may have clinical ramifications. Clinicians routinely treat flares as they appear with use of topical corticosteroids and calcineurin inhibitors, sedating antihistamines, and a variety of systemic agents aimed at rapid resolution of flares. However, the results of the present study suggest that this paradigm is not adequate. Rather, long-term strategies aimed at preventing flares should be used, such as twice-weekly application of topical steroids12 and diligence with gentle skin care even during quiescent periods. Furthermore, new therapies that are effective and sage for long-term use are needed.
The strengths of this study include using 2 large-scale, US population–based surveys of adults with minimal selection bias and population-based weighted frequencies and prevalences and controlling for confounding demographic variables in multivariable models. However, the study also has some limitations. Two different years of NHIS were assessed with consistent results, which shows reliability. However, similar population-based sampling and household survey methods were used in both studies. Thus, the validity of the results is still dependent on the methods used by the NHIS. I do not believe this to be a major problem given the rigorous sampling methods and quality controls used by the NCHS and the complex sample weights generated by NCHS to allow generalizability of results to the general population. Self-reported utilization may not be as accurate as claims-based assessments at estimating clinician and insurer costs. This limitation is offset by the ability to assess out-of-pocket costs and the individual burden of disease. The etiologies responsible for lost workdays, half-days in bed, and other health care utilization outcomes were not fully assessed in NHIS. However, the 2010 NHIS did assess for lost workdays and seeing a physician specifically related to eczema. Together, the results demonstrate that adults with eczema have a high burden of disease from their eczema per se, as well as other comorbidities. However, the cross-sectional nature of the study precludes assessing the direction of association between eczema and increased utilization. It may be that adults seeking care for other chronic diseases are more likely to be diagnosed as having eczema. Studies with longitudinal assessment of adults with eczema would be needed to exclude this possibility. Eczema was assessed by self-report on questionnaire. Self-report and parental report of AD and other types of dermatitis have previously been validated for a wide array of study designs.13- 15 Moreover, the 10.2% prevalence of eczema observed in adults was virtually identical to the 10.7% prevalence observed in US children.2 Finally, self-reported out-of-pocket health care costs in the past year have not been previously validated. Future studies, including up-to-date claims-based approaches and more detailed cost-analyses, are warranted to confirm the findings of this study. In addition, more studies of patient burden of disease and other patient-centered outcomes are needed.
The present study found that eczema was associated with higher out-of-pocket health care costs, more lost workdays, poorer overall health, increased health care utilization, and impaired access to care. This study demonstrates that adults with eczema have a major health burden with significantly increased health care utilization and costs. Future studies are needed to identify the determinants of health care utilization and access in adults with eczema.
Corresponding Author: Jonathan I. Silverberg, MD, PhD, MPH, Northwestern University Feinberg School of Medicine, 676 N St Clair St, Ste 1600, Chicago, IL 60611 (JonathanISilverberg@gmail.com).
Accepted for Publication: December 10, 2014.
Published Online: March 4, 2015. doi:10.1001/jamadermatol.2014.5432.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was made possible with support from the Agency for Healthcare Research and Quality (AHRQ), grant No. K12HS023011.
Role of the Funder/Sponsor: The AHRQ had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.