Disease Trajectories in Childhood Atopic Dermatitis

An Update and Practitioner's Guide

A.D. Irvine; P. Mina-Osorio


The British Journal of Dermatology. 2019;181(5):895-906. 

In This Article

Abstract and Introduction


Background: Atopic dermatitis (AD) is a heterogeneous disease with a multifactorial aetiology and complex pathophysiology. This heterogeneity translates into different trajectories of disease progression with respect to severity, persistence and risk of development of atopic comorbidities. Determining which possible disease trajectories or comorbidities any individual child might develop is challenging in clinical practice. Tools that help identify paediatric patients at higher risk of disease progression would greatly aid clinicians.

Methods: We reviewed recent cohort studies to synthesize and simplify the epidemiological data to try to identify shared clinically relevant characteristics that may help physicians estimate the risk of disease progression in paediatric patients with AD.

Results: Despite the variability in data collection and methods of analysis and their limitations, there are common patterns of early-childhood AD that may aid in the estimation of risk for disease progression. Factors associated with risk of AD progression include younger age of onset, family history of atopy, greater AD severity, filaggrin mutations, urban environment and polysensitization and/or allergic multimorbidity. Based on these factors, we provide a practitioner's guide for identifying, counselling and/or referring infants and children with AD at potentially higher risk of developing persistent AD and atopic comorbidities. We also present clinical scenarios to illustrate how these data relate to real-life situations.

Conclusions: Useful insights are provided for physicians and patients to inform them better about the risk of AD progression and to help guide care pathways for the paediatric population with AD.


Atopic dermatitis (AD), the most common skin disorder in infants and children,[1,2] is a complex disease with multifactorial aetiology, involving immune system dysregulation and epidermal barrier dysfunction, which are both influenced by genetic and environmental factors.[3–6] This complex pathophysiology translates into a heterogeneous clinical presentation (phenotype) with differences in age of onset, lesion localization, severity, sensitization profiles, disease persistence, presence of comorbid atopic and nonatopic conditions and longitudinal trajectories of disease progression (Figure 1).[7,8]

Figure 1.

Atopic dermatitis (AD) has a multifactorial pathophysiology involving genetic and environmental factors leading to immune dysregulation and skin barrier dysfunction. This complexity translates into heterogeneous clinical presentations (phenotypes). The term 'endotype' encompasses all these clinical, pathological and aetiological aspects of a disease and has been applied to asthma and other complex diseases to refer to the molecular mechanisms underlying observable disease characteristics (phenotypes).

While the consensus in the population at large and among many practitioners is that most paediatric patients with AD will eventually 'outgrow' the disease, a significant proportion of paediatric cases of AD will develop persistent AD and/or other atopic conditions such as asthma or allergic rhinitis. In an effort to shed some light on the trajectories of disease progression of allergic conditions in the paediatric population, numerous cohort studies have been conducted in the last two decades (Figure S1; see Supporting Information).[9–43] More recently, data-driven approaches to analyse complex epidemiological data have led to the identification of potentially discrete clusters of patients who present with relatively common phenotypes.[19,44,45]

Figure S1.

Multiple birth cohort studies on asthma and allergic diseases have been conducted over the last few decades. This figure lists the key characteristics of some of the studies included in this review.

Although there are no known factors that conclusively predict the risk of disease progression for each of the known AD phenotypes, understanding the characteristics of these patient clusters and their distinct disease trajectories could help clinicians make an early determination of whether a paediatric patient with AD should be closely monitored for their potential to develop comorbidities and/or be referred to a specialist for evaluation and follow-up.