Ap. Orange et al., SCORING OF ATOPIC-DERMATITIS BY SCORAD USING A TRAINING ATLAS BY INVESTIGATORS FROM DIFFERENT DISCIPLINES, Pediatric allergy and immunology, 8(1), 1997, pp. 28-34
The ETAC (Early Treatment of the Atopic Child) study, a multi-national
double-blind placebo-controlled randomized trial, has been in progres
s since 1994. Fifty-six centers in Europe and Canada participate in th
is study, A total of 817 children with atopic dermatitis [AD] were rec
ruited. The severity of AD was scored using the SCORAD (objective crit
eria). Ninety-eight investigators (mostly pediatricians) were trained
by three members of the European Task Force on Atopic Dermatitis [ETFA
D] to standardize their objective SCORAD scoring (system developed by
the ETFAD). The experts selected photographs and prepared a training a
tlas. The percentages of photographs assessed by the 98 non-expert inv
estigators below, within and above the range of evaluations by the thr
ee experts were calculated. Taking over- and underscoring together, ed
ema/papulation was the easiest intensity item to score (82% within the
range by the experts). The global symptom score, as well as lichenifi
cation, edema/papulation, oozing and excoriation registered by physici
ans with dermatological experience were not statistically significantl
y different from those by others. Erythema was statistically significa
ntly better scored by those with dermatological experience. The result
s of the Euclidean Distance method showed that the item excoriations g
ave the largest distance, Erythema and excoriations were scored better
by dermatologically experienced physicians (t-test, p=0.042 and p=0.0
63 respectively), but lichenification was better scored by non-dermato
logically experienced physicians (p=0.013). The extent of surface area
involved in the disease was calculated on 3 sets of photographs. Most
evaluations by the 98 nonexpert investigators were within the range o
f the experts. Dermatologically experienced physicians scored signific
antly better-than the others (t-test, p=0.006). This training program
is useful for standardizing the scoring in AD and indicates that SCORA
D can be used by investigators from different disciplines.