Professional accounts of electroconvulsive therapy: A discourse analysis

Peter Stevens a, , and David J. Harper b,

aOxleas NHS Trust, Dartford, Kent, UK
bUniversity of East London, UK

Available online 8 January 2007.

Electroconvulsive therapy (ECT) is a socially contested psychiatric intervention. However, the accounts of professionals involved in its use have rarely been systematically investigated. This study aimed to examine the accounts of clinicians who have used ECT on a routine basis. Eight health professionals (psychiatrists, anaesthetists and psychiatric nurses from a major city in the United Kingdom) with experience of ECT administration were interviewed about the procedure. Discourse analysis was used to interpret the interview transcripts. Interviewees appeared to draw on a repertoire, which constructed ECT recipients as severely ill. This was used to support claims which had the effect of: defining who should receive ECT; warranting the use of urgent physical psychiatric treatments; reformulating distress in biological terms; and discounting the therapeutic value of alternative, non-physical interventions. The interviewees managed concerns about ECT in a variety of ways, for example by: rendering it as a medical procedure with concomitant risks and benefits; downplaying a lack of clarity over its evidence base; and undermining the legitimacy of criticisms. Implications of these findings are discussed.

Keywords: Discourse analysis; Electroconvulsive therapy; Debates; Persuasive communication; UK

Article Outline
The unconscious motives of professionals: psychoanalytic research
From motives to attitudes, belief and knowledge: cognitive research
From motives, knowledge and attitudes to accounts: the contribution of a discursive approach
Initial reading and coding
The biomedicalˆmedical repertoire
The Œsevere end‚ rhetorical device
Who is appropriate for ECT?
ECT as Œlife-saving‚
Locating the focus of concern at the biological level
Exclusion of non-physical interventions
Managing concern about ECT
Acknowledging grounds for concern: weighing up risks and benefits
Rendering ECT as a medical procedure
Downplaying debate: the evidence is becoming clearer
Managing criticism of ECT: who has rights to speak?
Mental health professionals
Service users
Appendix A. Appendix
A.1. Transcription notation

Electroconvulsive therapy (ECT) remains one of the most widely used but contested and debated interventions in psychiatry. Advocates argue that the procedure is safe, effective and often life-saving (e.g., Abrams, 1997; Fink, 1979) whilst critics argue that it is ineffective and has the potential to cause psychological and neurological harm (e.g., Breggin, 1993; Friedberg, 1977; Johnstone (2000) and Johnstone (2003); Read, 2004).

However, there are two neglected areas in the research literature. The first is the relative lack of interest in the experiences of ECT recipients. Recently, researchers, including those with direct personal experience of receiving ECT have tried to redress this imbalance (Johnstone, 1999; Rose, Wykes, Leese, Bindman, & Fleischmann, 2003; Rose, Fleischmann, & Wykes, 2004), and there is a need for further investigation in this area.

A second neglected area is the empirical investigation of the accounts of those involved in ECT administration. The two most popular investigatory paradigms here have been psychoanalytic and cognitive. We will briefly review studies from these two paradigms, arguing that these studies are both theoretically and methodologically limited, before making the case for the contribution of a discursive approach.

The unconscious motives of professionals: psychoanalytic research
In the two decades following the introduction of ECT in the late 1930s, a variety of theories emerged to explain how Œshock treatments‚ worked and affected patients˜Gordon (1948), for example, reviews 50 of these. Less frequently, theorists examined the way ECT affected the professionals involved in its administration (e.g., Abse & Ewing, 1956; Fenichel, 1945; Wayne, 1955). Both were characterised by a psychoanalytic approach.

Abse and Ewing (1956) analysed Œoff guard‚ statements made by psychiatrists who were experienced in administering Œshock treatments‚. They suggested that such accounts were characterised by themes of hostility and punishment. For example, Œlet’s see if a few shocks will knock him out of it‚; Œwhy don‚t you put him on the assembly line‚; and the description of ECT as Œa mental spanking‚ (all p. 37). The authors suggested that Œthe very nature of the treatment itself can produce the attitudes described‚ (p. 38).

In a more recent example, Levenson and Willett (1982) reported observing Œsplitting‚ (divided clinical opinions), and disruption to the Œtherapeutic alliance‚ (rapport and empathy) in a multidisciplinary team involved in the care of two patients who received ECT. They concluded that ECT was an upsetting procedure because it Œoften produces rapid improvement in patients who had previously not responded to treatment‚ (p. 298) and the staff team experienced the therapeutic success as Œimplying devaluation of their own therapeutic skills‚ (p. 302).

However, this paradigm suffers from a number of limitations. For example, these studies have tended to be based on anecdotal reports rather than a more systematic gathering of material. Moreover, the reported comments are open to a number of competing explanations. Since psychoanalytic interpretations rely for their plausibility on inferred constructs which lie, as it were, within the person’s unconscious and thus cannot be demonstrated in the texts, they remain largely speculative.

From motives to attitudes, belief and knowledge: cognitive research
The second dominant approach in this area is cognitive, with questionnaires as the primary methodology. There have been a number of questionnaire-based studies investigating the attitudes and knowledge of professionals who are involved with ECT (e.g., Finch, Sobin, Carmody, deWitt, & Shiwach, 1999; Janicak, Mask, Trimakas, & Gibbons, 1985; Kalayam & Steinhart, 1981; Lutchman, Stevens, Bashir, & Orrell, 2001). Janicak et al. (1985) compared the knowledge and attitudes of psychiatrists, nurses, psychologists and social workers in relation to ECT. They reported that a positive attitude towards ECT in professional groups was correlated with increased knowledge about the procedure, and they concluded that educating staff about the facts of ECT would increase its utilisation in conditions where they suggested it should be the treatment of choice. Lutchman et al. (2001) used a similar design and drew similar conclusions finding differences in attitudes and knowledge across different disciplines, with psychiatrists being the most positive about ECT and psychologists the least.

However, as Diana Rose and her colleagues (Rose, Fleischmann, & Wykes (2004) and Rose, Wykes, Leese, Bindman, & Fleischmann (2003)) have argued in relation to investigations of ECT recipients‚ experiences, questionnaire-based methods limit and reduce the responses available to participants (see also Rosier, 1974). As a result, the possibility that factors other than knowledge might correspond with attitudes towards ECT has remained unexplored. More importantly, perhaps, this paradigm is methodologically and conceptually limited as a result of its assumptions about what constitutes Œknowledge‚ and Œattitudes‚ which are, again, inferred constructs thought to lie within the individual’s head.

A more fundamental problem with this approach is the way in which knowledge about ECT is presented as uncontested fact when the reality is much more complicated. Both the Janicak et al. (1985) and Lutchman et al. (2001) studies asked participants whether they agreed or disagreed with statements about the procedure. Despite being presented as Œfactually correct‚ or Œfactually incorrect‚ each statement could be interpreted in different ways and their facticity has been disputed in the literature. For example, the first statement from Janicak et al. (1985)˜Œthe therapeutic effect of ECT is related to the induction of a seizure in the brain‚˜is both supported (West, 1981) and contradicted by research demonstrating the effectiveness of placebo ECT (see Medical Research Council, 1965).

Similar assumptions have been made about Œattitudes‚ towards ECT. Janicak et al.’s (1985) participants were asked whether they would have ECT themselves if they were Œsuffering from varying degrees of depression‚ (p. 263). These responses were then regarded as demonstrating positive or negative attitudes toward the procedure. However, as with most attitude research there are fundamental epistemological difficulties with the notion that responses to lists of statements are unproblematically transformed into inferred mental constructs like beliefs or attitudes or unproblematically linked with conduct. Discursive psychologists like Potter and Wetherell (1987) dispute the idea that completing an attitude questionnaire represents a neutral and transparent transfer of an internal cognitive state to a mark on a page. Instead, they argue that people construct what they say in order to serve a range of functions (though not necessarily intentionally) and that this can be seen in intra- and inter-participant response variation. Thus the meaning of a psychiatrist agreeing with the statement that they would undergo the procedure if they were depressed (as 76.6% of Janicak et al.’s participants did) is not straightforward. For example, such agreement might enable participants to counter the potential charge of ethical inconsistency which might accompany disagreement with the statement. Moreover, the statement is abstracted from a context: there might conceivably be situations where these participants would be less likely to agree to have ECT.

From motives, knowledge and attitudes to accounts: the contribution of a discursive approach
Variation, disagreement and contradiction are usually rigorously controlled for in traditional cognitive research but at the cost of focusing on attitudes shorn of their context and this may tell us little about situated Œattitudes‚. In contrast, such variation is a topic of analytic importance for discourse analysts. For example, Harper (1999), in his study of professional and service user accounts of psychiatric medication, reported that a variety of explanations could be used in order to account for events such as the apparent failure of medication. Moreover, these explanations could be seen as having a variety of rhetorical effects˜in other words, they served a persuasive function, though this was not necessarily seen as an intentional action on the part of speakers. For example, when speakers claimed that medication had not worked because of the chronicity of the service user’s Œillness‚, this emphasised the permanence and severity of biological symptoms and shifted responsibility for treatment failure away from the medication and on to the illness.

Discourse analysis (DA) has been used to explore the accounts of health professionals in relation to a number of topics (Griffiths & Hughes, 2000; Parker, Georgaca, Harper, McLaughlin, & Stowell-Smith, 1995; Soyland, 1995). Because of its focus on the inherent variability of accounts, it is particularly useful in understanding how contested issues are constructed by participants. For example, Horton-Salway (2002) has shown how general practitioner’s discussions about ME (myalgic encephalomyelitis or CFS) are often concerned with the management of blame for the condition. Boyle (2002) and Boyle (2004) has suggested that professionals‚ use of rhetorical resources can make some ways of talking about a topic like schizophrenia seem more reasonable than others. In his study of talk about medication, for example, Harper (1999) argued that one of the effects of such talk was to serve certain institutional interests. For example, accounting for medication failure as due to the chronicity of illness could be seen to serve the interests of the Œpsychopharmaceutical complex‚ (Breggin, 1993) because questions about the efficacy of medication were deflected when such responses were deployed.

Johnstone and Frith (2005) have recently used DA to examine Freeman and Kendall’s (1980) study of ECT recipients‚ experiences and attitudes. They describe how patients are rhetorically constructed in this article in a number of ways, for example as passive and compliant or as hostile and unreasonable. They note that one of the effects of such constructions is to Œpreserve the view of ECT as a benign and beneficial procedure, and to validate both psychiatrists and psychiatry in the continued use of this treatment‚ (p. 200).

DA thus provides a novel perspective on the accounts of professionals. However, the kind of accounts presented in published research articles are highly systematised and organised and are likely to differ from those found in more informal settings. The aim of the present study, therefore, was to conduct an empirical qualitative investigation of accounts of the administration of ECT by those who administer it in the more informal context setting of a semi-structured interview.

Eight clinicians involved in administering ECT were interviewed about it by the first author. The interview transcripts were interpreted using DA, which draws on a broadly social constructionist epistemology. Here we will briefly note key discourse analytic concepts on which we drew in the present study. Following Wetherell’s (1998) suggestion, we have drawn on both discursive psychology and Foucauldian discourse analytic traditions (Willig, 2001).

In discursive psychology, the systematic ways of talking about a topic generally found in conversations are termed interpretative repertoires. Edwards and Potter (1992) have described a number of ways in which speakers use rhetorical devices which have the effect of making what they are saying appear factual˜an example is the empiricist forms of accounting found in journal articles, where the agency of the scientist is minimised and agency is implicitly located in the objects of research. Such devices are often deployed when there is some disagreement about the facts and when the speaker has a stake in the outcome.

The analysis also drew on positioning theory (Davies & Harré, 1990), which attempts to re-conceptualise static and essentialist notions like role and identity. It assumes that when people speak they implicitly position themselves and others in relation to the topic under discussion. For example, in a conversation between a doctor and a patient, the doctor is implicitly assumed to have expertise about topics like illness, diagnosis and so on, whilst the patient is implicitly assumed to have expertise about topics like their health concerns. Parker (1997) notes that all positions carry Œcertain rights to speak and specifications for what may be spoken‚ (p. 291). Thus there are certain things which can and cannot be said from particular positions and within particular interpretative repertoires.

Discourse analysts try to avoid the ascription of intentionality to participants‚ accounts. In other words, it is not assumed that participants strategically design their talk in order to have certain effects. Parker (1992) comments that Œpeople make discourse, but not in discursive conditions of their choosing‚ (p. 32). One significant context-marker for the present study is that the participants were being interviewed by a psychologist who, by virtue of his discipline (see Lutchman et al., 2001), might be thought to have a different position to them on ECT, and this may well have influenced the results. Walkup (1994) has argued, for example, that such situations may well lead to Œquite self-conscious attention to legitimation themes‚ (p. 149). The reader needs to take this into account in judging the quality of the analysis.

All the interviews took place in a major city in the United Kingdom. The interviewees represented the cross-professional involvement in ECT administration, including pre- and post-procedure care (psychiatric nurses), prescription and delivery of the treatment (psychiatrists) and the administration of anaesthetic and muscle relaxant (anaesthetists). The research aim was to identify which positions and repertoires were culturally available to the speakers˜in other words to see what range of things could be said about ECT by professionals. Since discourse analysts are critical of abstract notions like Œattitude‚ and Œview‚, the aim was not to represent participants‚ views nor to compare views within the sample.

We considered that eight participants would produce a reasonable range of possible repertoires and positions, and this is consistent with previous DA research. A purposive sampling strategy was followed in that a range of disciplines were interviewed: four psychiatrists, two psychiatric in-patient nurses, and two anaesthetists. More psychiatrists were interviewed since ECT is primarily a psychiatric intervention.

Potential participants were identified and recruited through liaison with departments of anaesthetics or old age psychiatry˜the latter departments prescribe a substantial portion of the total number of ECT administrations in the UK (Pippard & Ellam, 1981). All were sent a standard invitation letter followed by a telephone call. The first eight professionals invited to take part agreed to do so.

The small number of participants increases the risk that they will be identified and so, in order to protect confidentiality, we will present demographic information for the group as a whole and not for individuals. Four men and four women were interviewed. Three participants were aged in their 30s, three in their 40s and two in their 50s. Five of the six medical professionals were consultants and one was a senior registrar. The two nurse participants had considerable involvement with ECT. All but one participant was involved with ECT administration at the time of the interviews.

Potter and Hepburn (2005) have cautioned against what they see as the inappropriate and over-use of interview methods in DA studies. It is certainly true that interviews create particular kinds of interaction, and they should not be seen as a proxy for more Œnaturally occurring‚ kinds of conversations. However, it seemed that interviews were an appropriate source of data here because the study was interested in how professionals constructed accounts about ECT, for example how they might provide justifications for a contested intervention.

Interviews took place at a time and place of participants‚ choosing, usually at their workplace. The aims of the investigation were explained, anonymity was guaranteed and consent to both audio-taping the interviews and publishing extracts was obtained. A semi-structured interview schedule was followed, covering the participants‚ experiences of ECT and issues that had been debated within the literature. Appropriate ethical permission had previously been granted from a University Ethics Committee and interviews were conducted in 2001.

Initial reading and coding
Following Wood and Kroger’s (2000) guidelines, all the transcripts were read through in their entirety while a separate list of recurring or interesting categories and features were noted and eventually categorised. Extracts seen as related to each category were copied and pasted into correspondingly named computer topic files. Once printed, these became the material for analysis.

Discourse analysts see their interpretation of such texts as only one of a number of possible readings and not the definitive one (Willig, 2001). Our reading of the interview transcripts indicated that different interpretative repertoires and rhetorical devices were drawn on by the interviewees, suggesting that a range of such features were culturally available to them. During the reading, the subject of interest was talk about ECT and, increasingly, as the analysis progressed, about those to whom it was administered. Although a range of features were identified, because of limitations of space, our analysis is structured by a focus on the Œbiomedicalˆmedical‚ interpretative repertoire and the Œsevere end‚ rhetorical device. We will also discuss how interviewees managed concerns about ECT. We will also draw on some of the psychiatric literature in order to place the interview material in context.

Transcription notation is described in the Appendix A. The code underneath each extract refers to the participant (e.g., Dr. A), their discipline (e.g., consultant psychiatrist) and the line numbers of the interview transcript from which the extract derives (e.g., lines 20ˆ22).

The biomedicalˆmedical repertoire
Throughout the interviews, it was common for ECT recipients to be described with diagnostic medical language, which typically referred to their behaviour and the consequences of that behaviour. For example, asked who was most likely to receive ECT, Dr. A responded Œpatients with severe depression who may be in a depressive stupor˜that means that they‚re sort of bed-bound, immobile, not eating not drinking‚ (Dr. A, consultant psychiatrist: lines 54ˆ55). Such lists of behaviour were described elsewhere as Œclear biological features‚ (Dr. E, consultant psychiatrist: lines 753ˆ754) or Œprominent physical symptoms‚ (Dr. B, senior psychiatric registrar: line 117).

A variant of the biomedicalˆmedical conceptualisation of patient behaviour and psychiatric interventions were accounts framed in electro-chemical terms. These are also to be found in some of the professional literature, for example, Œrepeated treatments [of ECT] alter chemical messages in the brain and bring them back to normal‚ (Royal College of Psychiatrists, 1995, p. 104).

The biomedicalˆmedical repertoire was often accompanied by a rhetorical device which focused on the chronicity and severity of problems.

The Œsevere end‚ rhetorical device
In the following extract the characteristics of those who are seen as most likely to be appropriate for treatment with ECT are discussed. Dr. A is discussing the occasions when ECT might be used to save life:

Dr. A: They‚re sort of bed bound, immobile, not eating, not drinking, er at risk of developing um major medical problems such as deep vein thrombosis, pulmonary embolic complications, which will kill them.

Interviewer: Right. Is that as a result of er (.)

Dr. A: Depression. Because they‚re severely depressed, they‚re in a depressive stupor. It’s defined as stuporous, literally.

(Dr. A, consultant psychiatrist: lines 54ˆ60)

Who is appropriate for ECT?
One feature of the extract is the way in which Dr. A constructs those patients who are seen as appropriate for ECT. They are Œbed bound, immobile, not eating, not drinking‚˜and located at what another interviewee described as Œthe severe end‚ of the clinical spectrum (Dr. E, consultant psychiatrist: line 56). Examples of the use of the Œsevere end‚ device can also be found in the professional literature:

It is recommended that electroconvulsive therapy (ECT) is used only to achieve rapid and short-term improvement of severe symptoms after an adequate trial of other treatment options has proven ineffective and/or when the condition is considered to be potentially life-threatening, in individuals with: severe depressive illness; catatonia; a prolonged or severe manic episode.

(National Institute for Clinical Excellence, 2003, p. 1)

ECT as Œlife-saving‚
A formulation of ECT recipients as Œsevere‚ appeared to warrant the use of ECT on occasions when the person was viewed as at risk of dying either because of their attempts to actively kill themselves or because they were not eating or drinking. Both these kinds of behaviour were seen as symptoms of depression with failure to eat or drink viewed more seriously if the person was older, thereby potentially increasing the risk of physical health complications. The claim that ECT was life-saving was common across the interviews, regardless of participants‚ professional orientation. One effect of the Œlife-saving‚ repertoire was to sanction the use of ECT under conditions of compulsory psychiatric treatment (i.e., without requiring recipients‚ consent). This was termed Œemergency ECT‚ by participants. The grounds for such treatment are highly contested: both Johnstone (2003) and Read (2004) argue that there are no studies to support the claims of advocates for Œemergency ECT‚. Read comments that ECT Œdoes not prevent suicide and for a small number may precipitate it‚ (2004, p. 95).

Locating the focus of concern at the biological level
Another feature of interest in Dr. A’s extract is that there appears to be a causal chain in the location of the cause of problems. Causal agency initially appears to be located in physical health problems but these are then framed as caused by the patients‚ self-destructive behaviour, which in turn is seen as caused by depression. Ultimately, these biological signs and symptoms are presented as the threat to life. One effect of using technical descriptions (Œpulmonary embolic complications‚) is to construct the patient as both the cause of, and passive victim of, biological processes.

A second effect of constructing the problem at the biomedicalˆmedical level, rather than, say the psychological, or societal, is that it then becomes difficult to argue for alternative, especially non-biomedical, explanations and interventions. ECT was thus constructed not only as a reasonable intervention, but also an obvious one because it was viewed as fast-working and effective. One effect of this was that non-physical interventions were often discounted.

Exclusion of non-physical interventions
In the following extract, a participant describes the importance of early treatment when service users are extremely Œill‚:

Dr. E: ∑ if you delay effective treatment, you often have a tougher fight to fight.

(Dr. E, consultant psychiatrist: lines 789ˆ709)

In this extract Œtreatment‚ is implicitly seen as synonymous with ECT and the notion of time is foregrounded. It is implied that the use of other interventions would mean delaying the use of ECT, which is constructed as an Œeffective treatment‚. Elsewhere in the interviews the term Œtreatment‚ was seen as synonymous with pharmaceutical interventions. The ECT fact sheet for patients published by the UK’s Royal College of Psychiatrists (1995) states, Œthe alternative [to ECT] is drug therapy which also has risks and complications‚ (p. 104). Non-physical interventions such as psychotherapy or increased social support are absenced in these accounts, and one effect of this is to preclude them from being offered as alternative interventions. Arscott (1999) points out, Œit is likely to be difficult for a patient to refuse treatment with ECT if they believe that this is their only chance of cure‚ (p. 106).

Managing concern about ECT
All of the interviewees recognised that there was public concern about ECT and that it was seen as controversial, and they managed this in a number of ways. They acknowledged grounds for concern but framed ECT within a risks and benefits calculus similar to other medical procedures, sometimes accompanying this with claims that the evidential basis was becoming clearer. The interviewees acknowledged criticisms of ECT but in some of their accounts, more rights to speak were accorded to service user critics who had undergone it.

Dr. E: And I‚m actually quite scrupulous about g-, giving the pros and cons I, I, I very often clear with people, I really think this is the treatment you should be having, but I‚m also very clear ab-, that many people don‚t agree with this treatment, it’s still regarded as controversial, there are some things you probably will experience which you will not like. You probably will find your memory’s adversely affected. Er there are, you know, having an anaesthetic is at, at some level a risk, oh, the figure that’s quoted in the College of Psychiatrists Handbook is three deaths per one hundred thousand anaesthetics.

(Dr. E, consultant psychiatrist: lines 193ˆ202)

Acknowledging grounds for concern: weighing up risks and benefits
One way in which concern about ECT was managed was by drawing on a notion of informed consent. In this extract Dr. E notes how ECT recipients are presented with Œthe pros and cons‚. He includes some of the key concerns that critics of ECT have raised: that many disagree with it; that memory problems may occur; and that the administration of the anaesthetic itself is a risk. However, at the same time Dr. E says that he states ŒI really think this is the treatment you should be having‚. Also, the risk from the anaesthetic is slightly downplayed by noting that it is a risk Œat some level‚, although it was the risk seen as raising the most concern. This informed consent account is commonly encountered in discussions about medical procedures like major surgery, and so one effect of employing such a risk and benefits calculus is that ECT was seen as similar to other medical procedures.

Rendering ECT as a medical procedure
Within the interviews, ECT was often discussed through analogies with effects or processes associated with general medical procedures, and this also had an effect of managing concern about ECT. If ECT can be seen as a medical procedure with concomitant risks and benefits, then its particularly controversial status within the popular imagination (e.g., from films like One Flew over the Cuckoo’s Nest) can be minimised. Differences between the work of psychiatrists administering ECT and that of their medical colleagues carrying out surgery are de-emphasised and the procedure is presented as reasonable˜as Johnstone and Frith (2005) put it, ECT is thus presented as a Œbenign and beneficial procedure‚ (p. 200). Pilgrim and Rogers (1993) have argued that linking the content of psychiatric procedures with those of other medical procedures is one way of increasing the institutional alignment between psychiatry and general medicine˜an enterprise which Baruch and Treacher (1978) have suggested began in the 1960s when psychiatrists shifted their site of operation from separate institutions to district general hospitals. Equating the physical processes involved in psychiatric treatment with those involved in general medical treatment allows the status of the problem to be more easily conceptualised as biomedicalˆmedical. Moreover, Bracken and Thomas (2001) note that attempts to assert the equivalence of psychiatric and medical illness ignore the power of psychiatrists to use psychiatric treatments coercively. They remark, Œpatients and the public know that a diagnosis of diabetes, unlike one of schizophrenia cannot result in their being forcibly detained in hospital‚ (p. 725) and note, Œit is hard to imagine the emergence of Œanti-paediatrics‚ or Œcritical anaesthetics‚ movements‚ (Bracken & Thomas, 2001, p. 724).

Downplaying debate: the evidence is becoming clearer
A common theme in many psychiatric accounts of contested treatments is that the grounds for debate are lessening with increased research (see, for example Clare, 1976). This was evidenced in the interview material. In the transcript just before the next extract, Dr. B had noted that patient concerns about the lack of knowledge about how ECT works was understandable.

Dr. B: If the case for ECT was (.) crystal clear, that it does definitely work, then that wouldn‚t happen. (.) But as it happens, the case is not crystal clear, but I think it’s clearing up.

(Dr. B, senior psychiatric registrar: lines 447ˆ449)

Here Dr. B acknowledges that the case for ECT is Œnot crystal clear‚. Left at this point, Dr. B could potentially be challenged on why he continues to prescribe ECT. However, he then goes on to claim that it is Œclearing up‚. This could be seen as a form of Œrhetorical inoculation‚ (Sorenson, 1991) where a speaker deflects anticipated criticism.

In addition to these ways in which ECT recipients‚ concerns were managed within the interviews, the interviewees also drew on a range of other responses to direct challenges from ECT critics including groups of patients who have received ECT.

Managing criticism of ECT: who has rights to speak?
In the following extract, a participant discusses the confusion which they suggest can be experienced by patients and relatives regarding the behaviour of service users after they have received ECT. Dr. B is discussing the case of a patient who had Œlost her memory of her holidays‚ (line 89):

Dr. B: ECT is given to people who are very ill. And it’s very easy for relatives and patients (.) to (.) mix the two, sort of to (.) blame (.) ECT, the ECT for things that are to do with the illness.

(Dr. B, senior psychiatric registrar: lines 93ˆ95)

Here Dr. B suggests that behaviour could be viewed both as the direct result of ECT or, instead, Œto do with the illness‚. In claiming that it is due to the latter, Dr. B draws on the Œsevere end‚ device by describing those receiving ECT as Œvery ill‚. If this statement was made by a non-professional, it might be seen simply as an assertion. However, discourse analysts note that the force of a statement comes not only from its content but who makes it. Edwards and Potter (1992) describe the Œcategory entitlement‚ device which refers to the way in which the veracity of a report can be warranted by the entitlement of a speaker in a particular category to know certain things or have certain skills. Dr. B’s professional status confers rights to speak about what might constitute both unwanted effects of ECT and symptoms of illness. Others, like the relatives and patients mentioned here are positioned as less informed about the medical understanding of emotional distress and the effects of treatment. Thus it becomes possible to construct them as less authorised to make such judgements. One effect of this is to undermine the legitimacy of criticisms made by non-experts and especially recipients of ECT and their relatives. Johnstone and Frith (2005) have noted similar strategies at work in their DA of Freeman and Kendall’s (1980) article.

Of course, service users can attempt to claim certain kinds of category entitlements. As recipients of ECT they might understandably argue that they had expertise in understanding the effects of ECT. Indeed, such arguments have enabled users‚ experiences of ECT to gain a foothold in research conducted by ECT recipients themselves (Rose, Fleischmann, & Wykes (2004) and Rose, Wykes, Leese, Bindman, & Fleischmann (2003)). However, in the next extract, we see how attempts to gain such rights to speak can be challenged:

Interviewer: What, and when you talk about er (.) lay views of depression, many people underestimate how profoundly depressed people can be, what, what is the lay view of depression do you think? What is the standard view?

Dr. F: Well, I, it could be people who don‚t feel like going to work, who don‚t feel like going out, who don‚t feel like getting out of bed in the morning (.) er and they‚re a bit low (.) er it’s really, they stop socialising. I don‚t think people realise that depression is also, you stop eating, (.) that you‚ve stopped having baths, that you don‚t have a bath for three months and they find you at home. You know, that’s what they don‚t realise. You know, because as I said, most people have felt a bit low, but they don‚t really realise that depression is a serious illness. I don‚t know if you‚re a patient. Did, did you talk to the patient groups?

Interviewer: Well I‚ve read some of their literature.

Dr. F: I don‚t know, do they really know how bad these people are (.) those, those who are against ECT? (.) Do they ever come across, them, do they ever see how bad they are, do they/

Interviewer: /Do you mean people in the anti-ECT groups?

Dr. F: That’s right, do they really see them? You see it’s so easy for them to say when they get better Œoh ECT is a cruel treatment‚, but do they remember how bad they were to start with? Do they really know how severely ill?

(Dr. F, consultant anaesthetist: lines 342ˆ364).

Dr. F’s account here follows a similar trajectory to that of the previous extract from Dr. B in that lay views of depression are challenged: Œthey don‚t really realise that depression is a serious illness‚. In this extract, the Œsevere end‚ device is drawn on by describing Œhow bad‚ and how Œseverely ill‚ patients can be before they have ECT. However, Dr. F goes on to use a similar formulation to challenge the legitimacy of accounts of Œpatient groups‚. These critical accounts are implicitly challenged for being selective or lacking in objectivity in some way: Œdo they really see them?‚

There is an interesting shift in this extract, from describing individuals who are against ECT, but who have not Œcome across‚ those who receive it, to describing those who have received it (Œpatient groups‚), but who are Œbetter‚ and might have forgotten Œhow bad they were to start with‚. This account appears to challenge the legitimacy of the criticisms of both groups by drawing on the implied category entitlement of Œdoctor‚, which is set against the category entitlement of Œpatient‚. Thus, the knowledge and claims of ECT recipients who have experienced adverse effects are positioned as secondary to the knowledge and claims of the professionals who administer the procedure. These service users are positioned as poorly informed and forgetful. However, those who are critical of the procedure, but who have not been recipients, are implicitly positioned as even less authorised to criticise ECT. Elsewhere this is put more explicitly:

Nurse A: And especially if the person is really anti-ECT and had ECT then I haven‚t got too much of a problem. (.) It’s the ones that haven‚t had ECT that are against it that I‚ve got a bit of a problem with.

(Nurse A, Psychiatric Nurse: lines 364ˆ367)

By singling out those who are critical but who Œhaven‚t had ECT‚, Nurse A’s account appears to imply a hierarchy of authority to speak about ECT with those who have received ECT seen as having more rights to speak than those who have not. However, as we have seen, even those service users who have received ECT can still have the validity of their views challenged.

We have presented a reading of professionals‚ accounts of ECT in which interviewees appeared to draw on a repertoire which constructed ECT recipients as severely ill. This was used to support claims which: circumscribed who should receive ECT; warranted the use of urgent physical psychiatric treatments; reformulated distress in biological terms; and discounted the therapeutic value of alternative, non-physical treatments. The interviewees managed concerns about ECT by: rendering it as a medical procedure with concomitant risks and benefits; downplaying a lack of clarity over its evidence base; and undermining the legitimacy of criticisms.

The use of the Œsevere end‚ rhetorical device is interesting in the light of evidence that the strongest predictor of ECT prescription is gender and age, rather than severity of illness: women and those aged over 65 (again, predominantly women for demographic reasons) are most likely to receive ECT (Read, 2004; Salford Community Health Council, 1998). Moreover, the viewing of memory problems as resulting from depression rather than ECT seems a little dated˜the consensus view from a range of studies does appear to demonstrate a range of cognitive problems following ECT administration (McElhiney et al., 1995; Neylan et al., 2001; Robertson & Pryor, 2006).

In interviewees‚ accounts, the agency of psychiatrists in prescribing ECT was downplayed with the decision portrayed as flowing from factors relating to the severity of illness. However, this does not account for the variability of ECT prescription over time, in different regions and also between psychiatrists. For example, the study of ECT prescription in Salford and Greater Manchester (Salford Community Health Council, 1998) detailed wide variations both between individual psychiatrists and over time. In his analysis of discourse concerning paranoid delusions, Harper (1994) reported that professionals used empiricist forms of accounting to explain how they reached diagnoses but he noted that they drew on a more contingent and subjective form of accounting to explain variations in diagnoses between professionals. This could be explored in further research focused on professionals‚ accounts.

There are epistemological challenges associated with applying findings from discourse analytic research (Willig, 1999). However, some discourse analysts explicitly seek to promote Œsubversive discursive practices and spaces of resistance‚ (Willig, 1999, p. 12). Here we will suggest some implications consistent with our analysis for different interest groups in mental health: researchers, professionals and service users.

DA can provide a useful way both of examining some of the assumptions implicit in professional accounts and of detailing their effects. This can be most useful in relation to topics like mental health where conceptualisations are contested. Future research could explore whether the strategies employed here can be seen in other accounts both of ECT and of other psychiatric interventions. However, there is also a need for further research into the experiences of the recipients of psychiatric interventions, including studies conducted by service user researchers.

Mental health professionals
It may be possible to draw on some of the insights of studies like this to develop training packages to help mental health professionals become aware of the ways in which particular rhetorical resources may foreclose the offering of choices about interventions, like ECT, to service users. Such training could be aimed not only at psychiatrists, but also other health professionals (e.g., psychiatric nurses and social workers, psychologists and anaesthetists, etc.).

It is important to note that ECT recipients, like other users of mental health services, are not a homogeneous group of individuals (see Campbell, 1999). For example, some (e.g., ECT Anonymous, 1999a) have called for ECT to be banned, whilst others (e.g., Perkins, 1994) have made a case for its benefits. Service users could be provided with specific information about the type of accounts which they might expect prescribing clinicians to draw on during discussions about ECT. Robertson and Pryor (2006) have provided a useful format for discussing the benefits and risks associated with ECT with mental health service users. Informing service users about the ways in which health professionals can use language to construct ECT as Œordinary‚ or Œurgent‚, and patients as Œseverely ill‚ for example, could allow them to recognise that this is only one possible way of understanding their situation. Simply knowing that other accounts are available could empower service users to ask more questions about ECT and alternative treatments.

Service users
Service users could also develop training packages˜similar to assertiveness training˜in managing psychiatric interviews by role-playing possible responses. Such training would need to be accessible and to acknowledge both the emotional distress experienced by potential ECT recipients and the difficulties associated with challenging professional power. An example of what such training might look like can be seen in the ECT Anonymous (1999b) factsheet which pre-empts a list of questions and suggests possible responses with the suggestion that Œthere are simple answers to all these queries, even ones with technical terms. If there is a deep reluctance to answer these questions you should suspect the likely quality of your treatment‚ (p. 1).

Abrams (1997) R. Abrams, Electroconvulsive therapy (3rd ed), Oxford University Press, Oxford (1997).

Abse & Ewing (1956) D.W. Abse and J.A. Ewing, Transference and countertransference in somatic therapies, Journal of Nervous and Mental Diseases 123 (1956), pp. 32ˆ40. Abstract-MEDLINE

Arscott (1999) Arscott, ECT: The facts psychiatry declines to mention. In: C. Newnes, G. Holmes and C. Dunn, Editors, This is madness: A critical look at psychiatry and the future of mental health services, PCCS Books, Ross on Wye (1999), pp. 97ˆ118.

Baruch & Treacher (1978) G. Baruch and A. Treacher, Psychiatry observed, Routledge and Kegan Paul, London (1978).

Boyle (2002) M. Boyle, Schizophrenia: A scientific delusion? (2nd ed), Routledge, London (2002).

Boyle (2004) M. Boyle, Preventing a non-existent illness? Some issues in the prevention of Œschizophrenia‚, Journal of Primary Prevention 24 (2004), pp. 445ˆ469. Abstract-EMBASE

Bracken & Thomas (2001) P. Bracken and P. Thomas, Postpsychiatry: A new direction for mental health, British Medical Journal 322 (2001), pp. 724ˆ727. Abstract-Elsevier BIOBASE | Abstract-EMBASE | Abstract-MEDLINE | Full Text via CrossRef

Breggin (1993) P.R. Breggin, Toxic psychiatry. Drugs and electroconvulsive therapy: The truth and the better alternatives, HarperCollins, London (1993).

Campbell (1999) P. Campbell, The service user/survivor movement. In: C. Newnes, G. Holmes and C. Dunn, Editors, This is madness: A critical look at psychiatry and the future of mental health services, PCCS Books, Ross on Wye (1999), pp. 195ˆ209.

Clare (1976) A. Clare, Psychiatry in dissent: Controversial issues in thought and practice, Tavistock/Routledge, London (1976).

Davies & Harré (1990) B. Davies and R. Harré, Positioning: The discursive production of selves, Journal for the Theory of Social Behaviour 20 (1990), pp. 43ˆ63.

ECT Anonymous (1999a) ECT Anonymous. (1999a). Newsletter. Summer 1999. Riddlesden: Author.

ECT Anonymous (1999b) ECT Anonymous. (1999b). Research information: Don‚t believe a word we say about shock treatment∑. Riddlesden: Author.

Edwards & Potter (1992) D. Edwards and J. Potter, Discursive psychology, Sage, London (1992).

Fenichel (1945) O. Fenichel, The psychoanalytic theory of neurosis, Norton, New York (1945).

Finch, Sobin, Carmody, deWitt, & Shiwach (1999) J.M. Finch, P.B. Sobin, T.J. Carmody, A.P. deWitt and R.S. Shiwach, A survey of psychiatrists‚ attitudes toward electroconvulsive therapy, Psychiatric Services 50 (1999), pp. 264ˆ265. Abstract-MEDLINE | Abstract-EMBASE

Fink (1979) M. Fink, Convulsive therapy: Theory and practice (2nd ed), Raven Press, New York (1979).

Freeman & Kendall (1980) C.P.L. Freeman and R.E. Kendall, ECT 1: Patients‚ attitudes and experiences, British Journal of Psychiatry 137 (1980), pp. 8ˆ16. Abstract-EMBASE | Abstract-MEDLINE

Friedberg (1977) J. Friedberg, Shock treatment, brain damage and memory loss: A neurological perspective, American Journal of Psychiatry 134 (1977), pp. 1010ˆ1014. Abstract-EMBASE

Gordon (1948) H.L. Gordon, Fifty shock therapy theories, Military Surgeon 103 (1948), pp. 397ˆ401.

Griffiths & Hughes (2000) L. Griffiths and D. Hughes, Talking contracts and talking care: Managers and professionals in the British National Health Service internal market, Social Science & Medicine 51 (2000), pp. 209ˆ222. SummaryPlus | Full Text + Links | PDF (771 K)

Harper (1999) D. Harper, Tablet talk and depot discourse: Discourse analysis and psychiatric medication. In: C. Willig, Editor, Applied discourse analysis: Social and psychological interventions, Open University Press, Buckingham (1999), pp. 125ˆ144.

Harper (1994) D.J. Harper, The professional construction of Œparanoia‚ and the discursive use of diagnostic criteria, British Journal of Medical Psychology 67 (1994), pp. 131ˆ143. Abstract-MEDLINE

Horton-Salway (2002) M. Horton-Salway, Bio-psycho-social reasoning in GPs‚ case narratives: The discursive construction of ME patients‚ identities, Health: An Interdisciplinary Journal for the Social Study of Health, Illness and Medicine 6 (2002), pp. 401ˆ421.

Janicak, Mask, Trimakas, & Gibbons (1985) P.G. Janicak, J. Mask, K.A. Trimakas and R. Gibbons, ECT: An assessment of mental health professionals‚ knowledge and attitudes, Journal of Clinical Psychiatry 46 (1985), pp. 262ˆ266. Abstract-MEDLINE | Abstract-EMBASE

Johnstone (1999) L. Johnstone, Adverse psychological effects of ECT, Journal of Mental Health 8 (1999), pp. 69ˆ85. Abstract-EMBASE | Full Text via CrossRef

Johnstone (2000) L. Johnstone, Users and abusers of psychiatry (2nd ed), Routledge, London (2000).

Johnstone (2003) L. Johnstone, A shocking treatment?, The Psychologist: Bulletin of the British Psychological Society 16 (2003), pp. 236ˆ239.

Johnstone & Frith (2005) L. Johnstone and H. Frith, Discourse analysis and the experience of ECT, Psychology and Psychotherapy: Theory, Research and Practice 78 (2005), pp. 189ˆ203. Abstract-MEDLINE | Abstract-EMBASE | Full Text via CrossRef

Kalayam & Steinhart (1981) B. Kalayam and M.J. Steinhart, A survey of attitudes on the use of electroconvulsive therapy, Hospital and Community Psychiatry 45 (1981), pp. 298ˆ306.

Levenson & Willett (1982) J. Levenson and A. Willett, Milieu reactions to ECT, Psychiatry 45 (1982), pp. 298ˆ306. Abstract-MEDLINE | Abstract-EMBASE

Lutchman, Stevens, Bashir, & Orrell (2001) R.D. Lutchman, T. Stevens, A. Bashir and M. Orrell, Mental health professionals‚ attitudes towards and knowledge of electroconvulsive therapy, Journal of Mental Health 10 (2001), pp. 141ˆ150. Abstract-EMBASE

McElhiney et al. (1995) M.C. McElhiney, B.J. Moody, B.L. Steif, J. Prudic, D.P. Devanand and M.S. Nobler et al., Autobiographical memory and mood: Effects of electroconvulsive therapy, Neuropsychology 9 (1995), pp. 501ˆ517. Abstract | Abstract + References | PDF (1604 K) | Full Text via CrossRef

Medical Research Council & Psychiatric Committee (1965) Medical Research Council, Psychiatric Committee, Clinical trial of the treatment of depressive illness, British Medical Journal 178 (1965), pp. 881ˆ886.

National Institute for Clinical Excellence (2003) National Institute for Clinical Excellence. (2003). Guidance on the use of electroconvulsive therapy: Summary information. Technology appraisal no. 59. London: NICE. (, accessed 14 September 2004).

Neylan et al. (2001) T.C. Neylan, J.D. Canick, S.E. Hall, V.I. Reus, R.M. Sapolsky and O.M. Wolkowitz, Cortisol levels predict cognitive impairment induced by electroconvulsive therapy, Biological Psychiatry 50 (2001), pp. 331ˆ336. SummaryPlus | Full Text + Links | PDF (55 K)

Parker (1992) I. Parker, Discourse dynamics: Critical analysis for social and individual psychology, Routledge, London (1992).

Parker (1997) I. Parker, Discursive psychology. In: D.R. Fox and I. Prilletensky, Editors, Critical psychology: An introduction, Sage, London (1997), pp. 284ˆ298.

Parker, Georgaca, Harper, McLaughlin, & Stowell-Smith (1995) I. Parker, E. Georgaca, D. Harper, T. McLaughlin and M. Stowell-Smith, Deconstructing psychopathology, Sage, London (1995).

Perkins (1994) R. Perkins, Choosing ECT, Feminism and Psychology 4 (1994), pp. 621ˆ627.

Pilgrim & Rogers (1993) D. Pilgrim and A. Rogers, A sociology of mental health and illness, Open University Press, Buckingham (1993).

Pippard & Ellam (1981) J. Pippard and L. Ellam, ECT in Great Britain, Gaskell, London (1981).

Potter & Hepburn (2005) J. Potter and A. Hepburn, Qualitative interviews in psychology: Problems and possibilities, Qualitative Research in Psychology 2 (2005), pp. 281ˆ307. Full Text via CrossRef

Potter & Wetherell (1987) J. Potter and M. Wetherell, Discourse and social psychology: Beyond attitudes and behaviour, Sage, London (1987).

Read (2004) J. Read, Electroconvulsive therapy. In: J. Read, L.R. Mosher and R.P. Bentall, Editors, Models of madness: Psychological, social and biological approaches to schizophrenia, Brunner-Routledge/ISPS, London (2004), pp. 85ˆ99.

Robertson & Pryor (2006) H. Robertson and R. Pryor, Memory and cognitive effects of ECT: Informing and assessing patients, Advances in Psychiatric Treatment 12 (2006), pp. 228ˆ238.

Rose, Fleischmann, & Wykes (2004) D. Rose, P. Fleischmann and T. Wykes, Consumers‚ views of electroconvulsive therapy: A qualitative analysis, Journal of Mental Health 13 (2004), pp. 285ˆ293. Abstract-EMBASE | Full Text via CrossRef

Rose, Wykes, Leese, Bindman, & Fleischmann (2003) D. Rose, T. Wykes, M. Leese, J. Bindman and P. Fleischmann, Patients‚ perspectives on electroconvulsive therapy: Systematic review, British Medical Journal 326 (2003), pp. 1363ˆ1367.

Rosier (1974) M. Rosier, Asking silly questions. In: N. Armistead, Editor, Reconstructing social psychology, Penguin, Harmondsworth (1974), pp. 101ˆ114.

Royal College of Psychiatrists (1995) Royal College of Psychiatrists, The ECT handbook, Royal College of Psychiatrists, London (1995).

Salford Community Health Council (1998) Salford Community Health Council. (1998). Electro-convulsive therapy, its use and effects. Salford: author. (Available at:, accessed 9 August 2006).

Sorenson (1991) J. Sorenson, Mass media and discourse on famine in the Horn of Africa, Discourse and Society 2 (1991), pp. 223ˆ242.

Soyland (1995) A.J. Soyland, Analyzing therapeutic and professional discourse. In: J. Siegfried, Editor, Therapeutic and everyday discourse as behavior change: Towards a micro-analysis in psychotherapy process research, Ablex, Norwood, NJ (1995), pp. 277ˆ300.

Walkup (1994) J. Walkup, Commentary on Harper, Œthe professional construction of paranoia and the discursive use of diagnostic criteria‚, British Journal of Medical Psychology 67 (1994), pp. 147ˆ151.

Wayne (1955) G.J. Wayne, Some unconscious determinants in physicians motivating the use of particular treatment methods˜with special reference to electroconvulsive treatment, Psychoanalytic Review 42 (1955), p. 83. Abstract-MEDLINE

West (1981) E.D. West, Electrical stimulation therapy in depression: A double blind controlled trial, British Medical Journal 282 (1981), pp. 355ˆ357. Abstract-MEDLINE | Abstract-EMBASE

Wetherell (1998) M. Wetherell, Positioning and interpretative repertoires: Conversation analysis and post-structuralism in dialogue, Discourse and Society 9 (1998), pp. 387ˆ413.

Willig (1999) C. Willig, Introduction. In: C. Willig, Editor, Applied discourse analysis: Social and psychological interventions, Open University Press, Buckingham (1999), pp. 1ˆ21.

Willig (2001) C. Willig, Introducing qualitative research in psychology, Open University Press, Buckingham (2001).

Wood & Kroger (2000) L.A. Wood and R.O. Kroger, Doing discourse analysis: Methods for studying action in talk and text, Sage, London (2000).

Appendix A.
A.1. Transcription notation
Noticeable pauses of less than one second were indicated by a full stop in brackets (.) while pauses of between 1 and 3 s are indicated by the word Œpause‚. A slash (/) indicates interruption and triangular brackets (< and >) indicate incidental contributions that were made by speakers during the other speaker’s turn. Transcripts were punctuated to facilitate reading. Identifying details were changed to preserve the anonymity of participants.

Corresponding author. Tel.: +44 1322 297151; fax

Leave a comment

Your comment:

Subscribe without commenting