Mathematical models of patterns of seizures. Their use in the evaluation of drugs.

The analysis of the results of many well-designed, double-blind trials of anticonvulsant drugs has been unsophisticated. We draw attention to the nonrandom occurrence of seizures, which negates the simple comparison of average seizure frequency. We propose a method of taking into account clustering of seizures when deciding on the appropriate length of follow-up after introducing a new treatment. Deterministic and nondeterministic models were used to show why there may be reasons for sometimes using more than one drug in the treatment of epilepsy.