Impact of seizure detectors on treatment and costs of epilepsy

Epilepsy management relies heavily on the accurate detection of seizures. Unreliable seizure readouts can lead to missed seizures and erroneous physician visits due to false positives. These inaccuracies can also result in delayed treatment or inappropriate medication adjustments, ultimately increasing healthcare costs. Here, we introduce a model to simulate the effects of an imperfect seizure readout on the (perceived) efficacy of treatment, the number of physician visits, and associated health-care costs.

To assess the effects of the reliability of seizure detection systems on treatment, we introduce a simple model that generates seizures with a probability P, which can be altered by medication. In addition, we simulate different degrees of reliability of seizure detectors, where changes in medication are based on the perceived seizure frequency. We quantify the resulting costs in terms of recognizing true seizures, physician visits, and the number of medication changes as a function of detector reliability. We also differentiate between the impacts of low sensitivity (missed seizures) and low specificity (false positives) in the detection systems.

We show that the degree of unreliability in seizure readouts significantly affects overall healthcare costs. In scenarios where sensitivity and specificity are both between 70% and 85% and thus accuracy is similarly low costs can increase by a factor of three to five compared to systems with higher diagnostic accuracy. For detectors with sensitivity and specificity ranging between 70% and 80%, the additional costs are primarily driven by the combined impact of missed seizures (low sensitivity) and unnecessary physician visits or medication adjustments (low specificity). Furthermore, unreliable systems may cause delayed or inappropriate changes in medication, complicating effective treatment.

We demonstrate that reliable seizure detection systems can reduce healthcare costs and improve epilepsy treatment.