A Population-Based Survey of Eczema Prevalence in the United States

Jon M. Hanifin; Michael L. Reed; Eczema Prevalence and Impact Working Group


Dermatitis. 2007;18(2):82-91. 

In This Article


This study represents the only epidemiologic survey of eczema and AD for a large and representative sample of the general US population in recent times. The survey covers all age ranges and all regions of the United States and assesses symptoms, self-rated severity, and comorbidities. These data provide strong evidence that eczema and AD are widespread health problems even when recall, misclassification, and reporting biases inherent in such surveys are accounted for.

One study limitation is that this was purely a survey study although a clinical validation substudy used by the author in a previous report[15] helped inform the choice of questions for this survey. A recent study suggested that even within a medically diagnosed population, up to 30% of AD cases in children were not recalled by parents.[22] Our 1-year prevalence rates are 10.7% and 6% for eczema and AD, respectively, based on empirical diagnosis according to self-reported symptoms.

Another limitation is the absence of data regarding potential response bias due to disease or symptom severity. Without these data, there is a risk that prevalence estimates may over- or underestimate true prevalence if those with positive cases are unable or unwilling to respond or (conversely) are more likely to respond owing to an interest in the survey topic. Although nonresponders were not followed up in this study, participants in a recent research study on bipolar disorder prevalence[20] were recontacted by phone. Prevalence rates were slightly higher among nonresponders, which suggests that our reported estimates of eczema prevalence are conservative.

Extrapolation of the prevalence estimates from this study to a total US population of 296 million (according to US census data for July 2005) indicates an affected population of 31.6 million persons (as found with empirical eczema criteria) and 17.8 million persons (found with empirical AD criteria). Taken together, these prevalence estimates are of the same magnitude as those of other chronic conditions such as migraine (which affects approximately 35 million people),[19] hay fever and allergic rhinitis (which affect approximately 36 million people),[23] and arthritis (which affects approximately 40 million people).[24] In roughly two-thirds of individuals with eczema and AD, skin symptoms are moderate to severe; in more than one-third, sleep is sometimes or frequently disturbed; and in more than one-quarter, the skin symptoms are constant and unrelenting. In contrast to what has been found in Singapore for asthma[25] and in Hong Kong for eczema,[26] our study showed prevalence to increase with decreasing income.

Consistent with the greater overall severity of symptoms in the individuals with empirical AD, approximately one-third of these individuals had a self-reported physician diagnosis of asthma or hay fever/allergic rhinitis, as opposed to 20% of those with an empirical diagnosis of eczema. These findings generally support the comorbid nature of asthma/hay fever/allergic rhinitis and atopic dermatitis.[3] Onset of symptoms was reported as being earlier (between birth and 5 years of age) among a larger proportion of individuals with empirical AD than among those with empirical eczema (whose onset occurred between 18 and 29 years of age). Age at AD onset among the patients in our survey is consistently higher than that reported elsewhere[3] and in clinical practice. This may be due to unreliable memory or reporting bias, imprecise diagnosis, or failure of the responder to connect adult and infant forms of the disorder.[22]

Our results indicate that eczema and AD are remarkably underdiagnosed by physicians in the United States. Among individuals reporting skin symptoms during the previous year, only 37.1% reported a physician diagnosis of an eczematous condition. However, the prevalence of symptoms of eczema (itching/scratching, red or inflamed rash, excessive dryness/scaling, and/or thickening of skin) was higher among the undiagnosed individuals. The higher prevalence among low-income individuals contrasts with most studies that indicate higher rates among wealthier classes,[3] but this may reflect selection and response biases in other studies.[27] Also, the higher prevalence in minority populations and low-income individuals may be a factor in the low rate of physician diagnosis. It may also reflect an increasing trend in which individuals with health conditions are seeking advice from nonmedical sources. In a recent survey of a randomly selected population of a rural Australian city, most individuals (71%) with self-reported dermatitis or eczema sought advice from a medical practitioner, but pharmacists (9%), the self-prescribed (16%), and other sources (4%) were also cited.[28]

These results raise important questions. Why is the prevalence of eczema and AD directionally higher among Asians and Native Americans? Why does prevalence decline with increasing income? What factors motivate individuals to seek (or hinder individuals from seeking) a diagnosis (and presumably treatment) from a physician? The low proportion of symptomatic individuals reporting a physician diagnosis and the higher prevalence of eczema symptoms among undiagnosed individuals suggest that symptoms are not the determining factor. Logistic regression analysis of Australian data revealed that an individual with skin symptoms was more likely to seek advice from a medical practitioner than from an alternative source if the skin condition was moderate to severe (eg, dermatitis, psoriasis, or urticaria) but that other variables (sex, age ≥ vs < 60 years, insurance, and income) were not statistically significant.[28] Further research is needed to answer these questions.

Estimates of prevalence provide information regarding the public health toll and societal costs of specific diseases. The results of the current study highlight the important impact of eczema and AD on health and suggest a major economic impact. They also highlight the need for improvement in the identification and treatment of patients with eczematous conditions in order to develop better epidemiologic criteria for diagnosis[22,29] and to achieve increased public awareness of the availability of therapies to control the suffering caused by these ailments.

CLICK HERE for subscription information about this journal.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.