Integrating benefit and harm in epilepsy treatment: Illustrating the usefulness of the likelihood of being helped versus harmed.

Balancing the potential benefits and harms of antiseizure medications (ASMs) remains a central challenge in epilepsy care. Although randomized trials routinely report efficacy and adverse-event outcomes, these are typically presented separately, limiting their usefulness for shared decision-making. The Likelihood of Being Helped versus Harmed (LHH)-the ratio of the number needed to harm (NNH) to the number needed to treat (NNT)-offers a simple, intuitive way to integrate these dimensions into a single measure. Using data from a previously published comparative analysis, this brief report illustrates how the LHH can summarize the relative probability of achieving a clinical benefit vs experiencing a treatment-related harm. As an example, an ASM with an LHH of 1.9 would be interpreted as being almost twice as likely to produce a ≥50% seizure reduction as to lead to discontinuation due to adverse effects, whereas an LHH below 1 would indicate that harm is more likely than benefit. These examples demonstrate how the LHH can support clearer communication of benefit-harm trade-offs and complement emerging tools for individualized seizure-risk prediction.