Health-related quality of life in adults with epilepsy: the effect of age, age at onset and duration of epilepsy in a multicentre Italian study

Background: The potential effect of age-related factors on health-related quality of life (HRQOL) of patients with epilepsy has rarely been analyzed in the literature. Methods: We examined this association in a selected population of 815 adults with epilepsy recruited in the context of a multicentre study for the evaluation of Epi-QoL, one of the first Italian epilepsy-specific measures of HRQOL for adults with epilepsy. The Epi-QoL is a 46-item self-administered questionnaire focusing on six domains, which was successfully tested for reproducibility and validity. Ordinary least-squares regression models were used to assess the relationships between age-related factors (patient's age, age at seizure onset, and duration of epilepsy) and overall Epi-QoL score, controlling for the effect of potential confounders. We fitted simple regression models including each age-related factor alone to assess the independent role of each factor on the overall Epi-QoL score. We also fitted multiple regression models including pairs of age-related factors solely, as well as one or two age-related factors together with the same set of confounders. Results: Simple regression models showed that age and duration of epilepsy were significant negative predictors of the overall Epi-QoL score: the higher was each age-related factor, the lower was the overall Epi-QoL score; age at onset alone was a nonsignificant predictor of the overall Epi-QoL score. Multiple regression models including two age-related factors solely showed that duration of epilepsy was still a significant negative predictor of the overall Epi-QoL score in both pairwise models, whereas age was a significant negative predictor only in the model including age at onset. Age at onset emerged as a significant positive predictor of the overall Epi-QoL score only in the model including age: the higher was age at onset, the higher was the overall Epi-QoL score. Adjusted regression models including either one or two age-related factors and controlling for the selected confounding variables showed that the age-related factors had no significant effect on the overall Epi-QoL score anymore. Conclusions: If no other known correlates of the overall Epi-QoL score are considered, age and duration of epilepsy can be expected to have a significant negative association with HRQOL in epilepsy (with the effect of duration being stronger and more consistent across models than the one of age), whereas age at onset is a positive predictor of the overall HRQOL of limited significance. However, demographic and clinical factors, such as seizure frequency in the preceding 12 months, may provide a better explanation of HRQOL in epilepsy.