DISCUSSION
Our study is the first to validate the use of the UCT in the evaluation
of pediatric CU. We demonstrated that the UCT has adequate internal
consistency, convergent validity, known-groups validity, and screening
accuracy when compared to the CDLQI at study entry and at follow-up.
Similar results were observed in subgroup analyses of CSU and CIndU
patients, although we were unable to assess the convergent validity of
the UCT in CIndU cases because of inadequate sample size. Furthermore, a
greater increase in UCT scores was observed in patients who started or
increased their dose of sgAHs, which exemplifies the clinical relevance
of the UCT.
Results of the validation of the UCT in adult CU were largely consistent
with our study. Internal consistency of the UCT in adult CU was
excellent (Cronbach’s α=0.84)14, whereas our results
revealed respectable internal consistency in pediatric CU (Cronbach’s
α=0.73). Validation of the Turkish UCT revealed lower internal
consistency of the UCT in CIndU compared to CSU (0.89 versus 0.68,
respectively), whereas values were similar in our
results17. Although DLQI and CDLQI scores are not
directly comparable24, in adult CU, UCT scores
strongly negatively correlate with DLQI scores (Spearman’s
rho=-0.72)25. Consistent with our results in children,
the UCT can distinguish between known-groups of disease severity in
adults17,26. Screening accuracy (i.e., AUC) of
the UCT in adult CU was also excellent, although slightly higher than in
our findings (0.89 versus 0.82, respectively)14. We
found a slightly lower optimal cut-off for poorly controlled CU in
children (UCT≤10) than was determined for adults
(UCT≤11)14. A cut-off of 10 in children had a poorer
specificity (63.3% versus 70.8%) but greater sensitivity (95.5%
versus 89.9%) than a cut-off of 11 in adults14. The
discrepancy between the cut-off of poorly controlled CU in adults and
children may be because no studies exist formally validating a CDLQI
cut-off for poorly controlled CU. Despite using a previously published
CDLQI cut-off21, misclassification bias may persist.
No tools exist to evaluate CU symptom control and severity in children,
thereby limiting the evaluation of outcomes, progression, and management
in pediatric CU trials5,13. Patient reported outcome
tools, like the UCT, have been recommended as endpoints in clinical
trials27 and in the assessment of adult
CU15. Although the UAS7 has not been specifically
validated in pediatric CSU, current pediatric guidelines suggest that it
may be used in the evaluation of pediatric CSU3,5.
However, the UAS7 has several limitations compared to the UCT. Firstly,
the UAS7 is only designed to assess CSU and not
CIndU28. Secondly, the UAS7 must be completed
prospectively, which poses logistical and compliance challenges.
Thirdly, the UAS7 does not evaluate angioedema or disease control.
Therefore, the retrospective nature of the UCT, the ability to assess
CIndU, and the evaluation of angioedema and disease control strongly
favours the use of the UCT in the evaluation of pediatric CU and in
future clinical trials14.
Our study is subject to some limitations. The low sample size precluded
the evaluation of the test-retest probability of the UCT. Therefore, we
recommend future multi-center studies validating the use of the UCT in
children. The recall periods of the UCT and the CDLQI are not the same
(28 versus 7 days, respectively). Therefore, it is possible that acute
exacerbations of CU were captured by the UCT and not the CDLQI. This
discrepancy could reduce the convergent validity and the known-groups
validity of the UCT and may lead to the misclassification of patients in
the assessment of screening accuracy. Although the CDLQI is only
validated in children aged 4 to 16 years8, our study
included children of any age, which may bias the results. However,
subgroup analysis of only patients aged 4 to 16 years yielded largely
consistent results with the entire sample. Furthermore, the CLDQI is not
a CU-specific tool; therefore, concomitant dermatologic diseases
(i.e., atopic dermatitis) may have elevated the CDLQI scores,
which would not have been detected by the UCT. Despite the flaws of the
CLDQI, it is the only validated questionnaire that can be used in
pediatric CU.
In conclusion, our findings validate the use of the UCT as a patient
reported outcome in the assessment of pediatric CU and CSU. Larger
studies are necessary to validate the convergent validity of the UCT in
pediatric CIndU.