Adherence to Treatment in Children With Epilepsy: Who Follows "Doctor's Orders"?

PURPOSE: The goal of the present study was to examine sociocultural, medical, family environment, and individual cognitive factors that predict adherence to treatment in children with epilepsy. METHODS: The study subjects (4-13 years old) were enrolled in a longitudinal seizure study at the first visit to the seizure clinic, attended at least 6 months, and had at least two appointments. Baseline predictors, which were obtained by interview, chart review, and psychometric testing, included sociocultural and family environment, seizure and previous treatment history, child behavior, cognitive functioning (IQ), and family stress. Four latent factors tapping these indicators of risk (acculturative risk, seizure severity, behavior problems, family environment) and two measured variables (IQ and life events) were hypothesized. Outcomes were visit adherence (proportion of scheduled appointments kept, plus proportion without unscheduled contacts), medication report (proportion of visits at which parent report of medication agreed with records), and medication levels (proportion of serum anticonvulsant levels within expected range for dosage). Two-step analytic procedure included confirmatory factor analysis to validate the hypothetical structure of the baseline risk indicators, followed by structural equation modeling to examine longitudinal relations between baseline risk and subsequent adherence outcomes. RESULTS: Significant prospective relationships included acculturative risk associated positively with visit adherence and medication levels, behavior problems associated negatively with visit adherence and medication levels, family environment associated negatively with medication report, life events associated positively with medication levels and visit adherence, and cognitive functioning (IQ) associated positively with medication levels. Seizure severity was not associated significantly with any adherence outcome. There also were no significant within-time associations between adherence outcomes. CONCLUSIONS: Contrary to clinical expectations, families at higher acculturative risk and with higher life events reported greater adherence. Seizure severity did not influence adherence. The three adherence measures were statistically independent of each other.