In Reply We thank Van der Linden and colleagues for their interest in our article,1 which describes the long-term cardiovascular risk associated with continued use of enzyme-inducing antiseizure medications (eiASMs). With respect to our mediation analysis, we considered incident dyslipidemia as a binary mediator variable, as opposed to a continuous variable of low-density lipoprotein cholesterol or total cholesterol, given its direct clinical relevance as the threshold over which treatment is initiated. However, we agree that future efforts at exploring a dose-dependent mediation between absolute and relative increases in low-density lipoprotein cholesterol and total cholesterol compared with baseline (pre-eiASM) treatment levels is warranted to further elucidate this potential mechanism. To address confounding, the mediation analysis indeed controlled for baseline hypertension, along with age at epilepsy diagnosis, sex, type 2 diabetes, atrial fibrillation, social deprivation (Index of Multiple Deprivation), former or current smoker, and duration of epilepsy. We opted not to adjust for baseline use of lipid-lowering agents in this analysis because the mediating variable was an incident diagnosis of dyslipidemia. Also, by definition, no person had a myocardial infarction or a stroke at the point of exposure to dyslipidemia, given the outcome (which occurs temporally after the mediator) was incident cardiovascular disease.