Differential diagnosis of seizure disorders: A conversation analytic approach
Introduction
In medical and health care settings, the microanalytic approach of conversation analysis (CA) has a strong track record of delivering revealing insights for linguists, sociologists and health workers (e.g. Beach, 2001; Drew, Collins & Chatwin, 2001; Heritage & Maynard (2006a), Heritage & Maynard (2006b); Maynard & Heritage, 2005). While early studies focused on the organisation of doctor–patient interaction (e.g. Heath, 1986; ten Have, 1991), or highlighted the influence of the institutional setting on participation rights (Frankel, 1990; Mishler, 1984), the analytical stance in recent years has increasingly turned the “spotlight on the patient” (Drew, 2001). Coinciding with this has been a trend towards interdisciplinary collaboration, aiming to improve the quality of health care by applying CA in clinical settings (Mangione-Smith, Elliot, Stivers, McDonald, & Heritage, 2006; Schwabe, 2004; Schwabe, 2006; Schwabe, Reuber, Schöndienst, & Gülich, 2007; Stivers, 2002; Stivers, Mangione-Smith, Elliot, McDonald, & Heritage, 2003). This paper is very much in line with these developments. We describe a novel use of CA methodology in the differential diagnosis of seizure disorders.
Neurologists regularly have to differentiate between epileptic and non-epileptic seizures (NES). One recent study found that 57.4% of patients presenting to medical services with a blackout had experienced an epileptic seizure, 18% a NES, and 22.3% had fainted (Kotsopoulos et al., 2003). Whereas there are a number of factors which make the history given by the patient a reliable tool in the diagnosis of faints (Alboni et al., 2001; Colman et al., 2004), the distinction of epilepsy and NES is more difficult: a recent study in over 300 patients with NES showed that it took a mean of seven years to make an accurate diagnosis. Almost all patients had been treated inappropriately with antiepileptic drugs (Reuber, Fernández, Bauer, Helmstaedter, & Elger, 2002).
Whilst epileptic seizures are caused by abnormal electrical activity in the brain, most NES are caused by a dissociative process. A range of predisposing, precipitating and perpetuating aetiological factors have been recognized. A history of significant trauma or an association with anxiety, other dissociative or somatoform symptoms are particularly common (Bowman & Markand, 1996; Reuber, Howlett, Khan & Grünewald, 2007). The accurate distinction between epilepsy and NES is particularly relevant because the treatment is very different. Whereas antiepileptic drugs can control seizures in the majority of patients with epilepsy, the treatment of choice for NES is psychotherapy (Reuber, Howlett, & Kemp, 2005).
The German and English projects were based on the methodology of CA, a well-established qualitative approach for the analysis of social interaction (Hutchby & Wooffitt, 1998; Sacks, 1992; ten Have, 2000). The project in England maintained the general methodological framework of the EpiLing project (Schwabe et al., 2007), a collaboration between the Department of Linguistics and Literature of the University of Bielefeld, Germany, and the Bethel Epilepsy Centre (Gülich, Schöndienst, & Surmann, 2002; Schöndienst & Gülich, 1999; Schwabe et al., 2007; Wolf, Schöndienst, & Gülich, 2000). Drawing on the analytic mentality and methods established in CA, the project focussed specifically on the patients’ descriptions of their subjective seizure experience and treated patients as experts of their own experience and disorder. A number of studies based on the analysis of over 100 recorded consultations in the Bethel Epilepsy Centre in Bielefeld, Germany suggested that patients use a range of different communicative methods when describing their seizures, and that the use of these methods clusters in a non-random way. These communication styles seemed to concord with the medical diagnoses of epileptic or NES (Gülich et al., 2002; Schöndienst, 2002; Schwabe et al., 2007). Although differences were seen most clearly between patients with focal epilepsies and NES, they were also found in the comparison of patients with generalised epilepsy syndromes and NES (Surmann, 2005).
This study was intended to determine whether patients describing their seizures in English and in a somewhat less specialised clinical setting use the same or similar communicative methods. In other words, our initial aim was to establish to what extent the identified communicative methods are specific to one language (in this case German) or a more universal reflection of the way patients feel about and deal with their seizures regardless of their language.
Our first hypothesis was that English-speaking patients would use very similar communicative methods to describe their seizures compared to German-speaking patients, although there are a number of differences between the two languages on lexical and syntactical levels. Whilst English and German are part of one language family (Barber, 2000), English is a verb-orientated language whereas German uses a nominal style, which allows for the insertion of several sub-clauses and extended participial constructions. However, we did not expect such differences to interfere much with the features we were looking at, as the differentiating communicative features described in German patients with seizures essentially represent interactive methods rather than specific lexico-semantic aspects of communication. Our second hypothesis was that patients describing their seizures in a less specialised secondary care setting would use similar communicative methods as the patients previously analysed in a tertiary reference centre (adding further weight to the generalisability of the findings).
Section snippets
Methodology
All patients admitted to a video-electroencephalography (EEG) unit, because a consultant neurologist remained uncertain about the nature of seizure disorder, were considered as potential participants for this study if their previous medical notes did not suggest the occurrence of different seizure types (none of the patients recruited were admitted for video-EEG in the context of epilepsy surgery evaluation or had reduced cognitive ability, for further clinical details see Table 1). Interviews
Formulation effort
One of the most striking observations in the German data was that seizure descriptions by epileptic and non-epileptic patients differ markedly in terms of patients’ formulation effort. The description of seizures seems to pose an ongoing challenge—although most patients interviewed had seizures for several years and must have talked about them many times before. Patients seem to realise that their seizure symptoms are not like any experience they share with their doctor. Although the
Conclusion and future perspectives
Our results suggest that the interactive features previously identified in consultations in which German patients describe their epileptic or NES can also be identified in interviews with English speakers. What is more, these features occur in analogous clusters. Like the German-speaking patients investigated in the EpiLing project (Schwabe et al., 2007), their English counterparts use two distinct communicative styles which are associated with a diagnosis of epilepsy on the one hand or NES on
Acknowledgements
This study was undertaken with financial support from the Frank Moody Bequest administered by the University of Sheffield and by Steven Howell. We are grateful for the advice on earlier version of this manuscript from Elisabeth Gülich, Paul Drew, Leendest Plug and Fabienne Chevalier.
References (41)
- et al.
Diagnostic value of history in patients with syncope with or without heart disease
Journal of the American College of Cardiology
(2001) - et al.
The diagnosis of epileptic and non-epileptic seizures
Epilepsy Research
(2003) Participating in decisions about treatment: Overt parent pressure for antibiotic medications in paediatric encounters
Social Science & Medicine
(2002)The English language. A historical introduction
(2000)Lay diagnosis
Text
(2001)- et al.
The effect of doctor behavior on the collection of data
Annals of Internal Medicine
(1984) - et al.
Psychodynamics and psychiatric diagnoses of pseudoseizure subjects
American Journal of Psychiatry
(1996) - et al.
Diagnostic value of history taking in reflex syncope
Clinical Autonomic Research
(2004) Spotlight on the patient
Text
(2001)- et al.
Conversation analysis: A method for research into interaction between patients and healthcare professionals
Health Expectations
(2001)