Inter-item correlation matrix.

Objective
To develop and validate a patient/proxy-reported measure of the incidence and severity of pediatric antibiotic-associated diarrhea (PAAD) in inpatient and outpatient settings.
Methods
A patient advisory group, consisting of five parents and two children, was engaged as a research partner. Instrument items were developed from three sources: relevant items from two previously validated instruments; relevant constructs from a newly developed core outcome measurement set; and outcomes identified by parents and clinicians as being the most important. In a prospective observational study, children (birth to 17 years old) newly prescribed antibiotics or on antibiotics for ≤ 7 days, were included and assessed at the time of presentation and daily thereafter until two weeks after antibiotic therapy was completed. Internal consistency and convergent validity of the instrument were examined.
Results
Of 78 patients who agreed to participate and met the eligibility criteria, 30(38%) were lost to follow-up; Data from the remaining 48 were analyzed. By applying four different definitions of diarrhea, we found a broad range of incidence risks of PAAD (27%−83%). PAAD was more likely to develop in younger age groups (≤ 3 years old). Cronbach’s α for the severity scale was less than 0.7. A high correlation was found between the PAAD severity score and numerical rating score of diarrhea severity reported by parents (r > 0.5).
Conclusion
The PAAD instrument is the first designed to measure the incidence and severity of PAAD. The instrument has content and construct validity. For reliability analyses of the severity scale, larger studies are required.