Background: Longitudinal modelling of the presence/absence of current eczema through childhood has identified similar phenotypes, but their characteristics often differ between studies.
Objectives: We propose that more comprehensive description of longitudinal pattern of symptoms may better describe trajectories than binary information on eczema presence.
Methods: We derived 6 multi-dimensional variables of eczema spells from birth to age 18 years (including duration, temporal sequencing, and the extent of persistence/recurrence). Spells were defined as consecutive observations eczema separated by no-eczema across 5 epochs in five birth cohorts: infancy (first year); early childhood (age 2-3 years); pre-school/early school age (4-5 years); middle childhood (8-10 years); adolescence (14-18 years). We applied Partition-Around-Medoids clustering on these variables to derive clusters of the temporal patterns of eczema. We then investigated the stability of the clusters, within-cluster homogeneity, and associated risk factors, including filaggrin mutations.
Results: Analysis among 7,464 participants with complete data identified five clusters: 1) No eczema (NOE) (51%); 2) Early transient (ETE) (21.6%); (3) Late-onset (LOE) (8.1%); 4) Intermittent (INT) (7.5%); and 5) Persistent (PE) (11.8%) eczema. There was a very high agreement between assignment of individual children into clusters when using complete or imputed (n=15,848) data (adjusted Rand index=0.99), i.e., the clusters were very stable. Within-individual symptom patterns across clusters confirmed within−cluster homogeneity, with consistent patterns of symptoms among participants within each cluster, and no overlap between the clusters. Clusters were characterised by differences in associations with risk factors (e.g., parental eczema was associated with all clusters apart from LOE; sensitisation to inhalant allergens was associated with all clusters, with highest risk of PE). All clusters apart from LOE were associated with filaggrin mutations. Of note, the strongest association was for PE (relative risk ratio (RRR), 2.70; 95% CI, 2.24-3.26; P<0.0001) followed by INT (RRR, 2.29; 95% CI, 1.82-2.88; P<0.001).
Conclusions: Clustering of multi-dimensional variables identified stable clusters with different genetic architecture. Using multidimensional variables may capture eczema development and derive stable and internally homogenous clusters. However, deriving homogenous symptom clusters does not necessarily mean that these are underpinned by completely unique mechanisms.