Abstracts

Compliance is known to be a major issue in the management of epilepsy. Thus, the identification of factors associated with poor compliance should lead to improved seizure control. In a French, multi-centric, open, prospective study, neurologists were asked to include newly diagnosed epileptic patients and to ask them to fill out auto-questionnaires assessing: [underline]compliance[/underline] (P Loiseau and C Marchal compliance questionnaire, 1988), [underline]knowledge of epilepsy[/underline] (adapted from Jarvie questionnaire, 1993), [underline]impact of epilepsy[/underline] (Jacoby, 1993), [underline]quality of life[/underline] (QOLIE 31) and [underline]stigmatization[/underline] (from Jacoby stigmatization of epilepsy scale, 1994). Compliance were also assessed using pills count and blood levels of valproic acid. Regression analysis was performed in order to identify determinants associated with compliance. A total of 356 patients (mean age: 38.5 years, sd: 15.78; male: 50%) were included by 124 neurologists and treated with valproate for 26 weeks.
According to the Loiseau questionnaire item 1, at week 26, compliance was globally high. Among patients who have taken the treatment at least once, only 53 patients (21.3%) notified to have forgotten their medication sometimes (n=47) or often (n=6), versus 121 never and 75 exceptionally (missing data: n=81).
Under Loiseau questionnaire item 2, patients who reported having forgotten to take their pills for at least one day, had more seizures than others (11.5% vs 32.0%, Exact Fisher test, p=0,028).
By regression analysis:
- According to Loiseau questionnaire item 1three determinants appeared related to high compliance:
[diams]Being a female (p=0.012, male versus female odds ratio: 0.531, 95%CI: [0.324-0.872])
[diams]Being older (p[lt]0.0001, odds ratio: 1.036, 95%CI: [1.019-1.053]),
- According to pills count, two determinants (only statistical trends) were associated to poor compliance:
[diams] Having a high score at D0 to the QOLIE 31 (p=0.0893).
[diams]Antecedent of anxiety disorder (p=0.0543).
Neither the nature of the epileptic syndrome, nor the impact of epilepsy, nor the stigmatization nor the knowledge about epilepsy influenced compliance. These data emphasize the heavy consequences of poor compliance in epilepsy and plead for developing therapeutic alliance. They also underline that factors related with compliance are highly dependent on the criteria used for assessing compliance. (Supported by sanofi[sim]synthelabo)