Assessing daily management of childhood diabetes using 24-hour recall interviews: reliability and stability.

Conducted 24-hr recall interviews concerning daily diabetes management with seventy-eight 6- to 19-year-old patients and their parents. Patients and parents were interviewed independently nine times over 3 months. Data obtained were used to construct 13 adherence measures. All measures yielded statistically significant estimates of parent-child concordance. Parent-child agreement was higher for weekday versus weekend behaviors and when based on nine versus three interviews. For the sample as a whole, parent-child concordance remained stable over the course of the study. Compared to the older patients, the 6- to 9-year-olds exhibited poorer parent-child agreement on measures involving time (e.g., injection and exercise-duration measures). This deficit disappeared, however, as the children became more practiced with the interview procedure. The dietary and glucose-testing measures exhibited moderate stability over the 3-month study. Lower stability estimates were obtained for the exercise and injection measures.