Conversation analysis can help to distinguish between epilepsy and non-epileptic seizure disorders: a case comparison

Factual items in patients' histories are of limited discriminating value in the differential diagnosis of epilepsy and non-epileptic seizures (NES). A number of studies using a transcript-based sociolinguistic research method inspired by Conversation Analysis (CA) suggest that it is helpful to focus on how patients talk. Previous reports communicated these findings by using particularly clear examples of diagnostically relevant interactional, linguistic and topical features from different patients. They did not discuss the sequential display of different features although this is crucially important from a conversation analytic point of view. This case comparison aims to show clinicians how the discriminating features are displayed by individual patients over the course of a clinical encounter.

CA-inspired brief sequential analysis of two first 30-min doctor-patient encounters by a linguist blinded to all medical information. A gold standard diagnosis was made by the recording of a typical seizure with video-EEG.

The patient with epilepsy volunteered detailed first person accounts of seizures. The NES patient exhibited resistance to focusing on individual seizure episodes and only provided a detailed seizure description after repeated prompting towards the end of the interview. Although both patients also displayed some linguistic features favouring the alternative diagnosis, the linguist's final diagnostic hypothesis matched the diagnosis made by video-EEG in both cases.

This study illustrates the importance of the time point at which patients share information with the doctor. It supports the notion that close attention to how patients communicate can help in the differential diagnosis of seizures.