The purpose of attributing cause: beliefs about the causes of myocardial infarction

Abstract

Patients’ beliefs concerning the causes of their illnesses are core to a number of theoretical models of illness behaviour. The current study explored the themes that emerged from patients’ accounts of the causes of their first myocardial infarctions (MIs). Semi-structured interviews were conducted with 12 patients within a week of their MI, at a UK district hospital. Transcripts were analysed using interpretative phenomenological analysis. Three researchers noted themes that emerged from each interview, then common themes were selected and refined on the basis of an aggregation of interviews. Three dominant themes emerged: (1) single versus multiple causation, (2) causes as triggers versus underlying dispositions, and (3) the potentially conflicting motives of avoiding blame whilst at the same time, seeking control. Whilst many participants had complex ideas concerning what caused their MI, and could often name several causes, they tended to emphasise the importance of a single cause, which often related to their symptoms. Further, several participants interpreted “cause” in terms of an acute trigger of MI, rather than as a chronic causal factor. Participants were apparently attempting to answer the question about why they had an MI now, leading to talk about single causes that trigger MI. By contrast, much previous research has been concerned with patients’ knowledge of “risk factors”, considering the production of only a few causes to reflect ignorance. A key process in participants attributing cause appeared to be attempting to avoid blaming themselves or others for their MI, whilst simultaneously seeking to assert control over future recurrence. Analysis of the functions and purposes of causal attribution suggests that patients’ focus on blame and control may be both emotionally and behaviourally adaptive, if not necessarily epidemiologically precise. These findings suggest that interventions to change causal attributions may be misguided, and may even be harmful.

Introduction

Causal attributions are the common-sense causal explanations people give for events, and have been extensively studied by psychologists for decades (see Hewstone, 1989). There is a large research literature concerned with the causal attributions people make for illness or other unexpected negative events, and how these relate to a variety of outcomes (reviewed by Turnquist, Harvey, & Andersen, 1988; Benyamini, Leventhal, & Leventhal, 1997; Roesch & Weiner, 2001; Hall, French, & Marteau, 2003). These reviews show that there is good evidence for a weak to moderate relationship between causal attributions made and the emotional impact of the illness or event, as well as the coping strategies employed.

There is also a large literature on the nature of the beliefs people hold about the causes of coronary heart disease (CHD), and CHD events, such as myocardial infarction (MI), more commonly known as heart attacks (reviewed by French, Senior, Weinman, & Marteau, 2001). A recent review identified 47 papers and reports, containing 54 discrete data sets, concerned with causal beliefs for CHD (French et al., 2001), and further reports continue to be published (e.g. Arefjord, Hallaraker, Havik, & Maeland, 2002; French, Marteau, Senior, & Weinman, 2002; Furze, Lewin, Roebuck, Thompson, & Bull, 2001). A number of clear findings consistently emerge from this literature. First, the most popular attributions in these studies as a whole are to chronic stress (most prevalent attribution in 22 of the 54 data sets included in the French et al., 2001 review) and to lifestyle factors such as smoking and diet (most prevalent in 16 of the 54 data sets). Second, in most studies, several such attributions are reported by most participants. For instance, 143 first-time MI patients in Auckland, New Zealand were presented with a list of 24 putative causes, and asked which were involved in the causation of their own MI (Weinman, Petrie, Sharpe, & Walker, 2000). These MI patients either “agreed” or “strongly agreed” with a mean of 7.3 causal factors, with “stress or worry”, “high levels of cholesterol”, and “eating fatty foods” the most strongly endorsed factors.

A criticism that has been made of research on casual attributions in general is that it has been overly reductionist: it is not interested in the whole of the causal explanations people give, but prefers to reduce these explanations to simple categories (Antaki (1988), Antaki (1994); Hewstone, 1989). Antaki (1988) argues that two potentially important sources of information are lost by this reductionism. First, the reasoning behind an attribution is lost: the participants’ evidence, backing and justification of their causal beliefs are ignored. Second, by ignoring the purposes for which attributions are made, and the context in which these statements are made, inferences about what is occurring may be misleading (Harré, 1993).

An understanding of the purposes and reasoning involved in making causal attributions, rather than just the outcomes, may in itself lead to a richer understanding of the attribution process, particularly why people make the attributions they do. These phenomena have been relatively neglected due to the focus on the outcome of the process by which people form attributions about illness events. That is, although it is clear that people tend to attribute the causes of CHD to chronic stress and lifestyle factors, there is little information on why they make these attributions, and the purposes these attributions serve for the individuals who make them.

The aim of the present study, therefore, was to investigate the beliefs of people who had experienced an MI in the past week, about the causes of their MI. Specifically, given the wealth of research that has identified the most prevalent causal attributions for CHD, the particular aims of the current study were to explore the reasoning involved in causal attributions, and the possible purposes served by such causal attributions. In contrast to the majority of research in this area, this was achieved using semi-structured interviews with people recovering from their first MI, rather than the more usual quantitative approaches employed. Transcripts of these interviews were analysed using interpretative phenomenological analysis (IPA, Smith & Osborn, 2003). IPA was employed because it allows the researcher to investigate the process of explanation, including the reasoning and justification that participants make for particular attributions. The other feature of IPA that was helpful in this study was its critical realist approach to how participants view their illnesses, with the aim of saying something about how participants think about the world (Smith, 1996). Consequently, IPA has the dual virtues of allowing an examination of the processes involved in attributing cause, whilst retaining a philosophical outlook that is congruent with the large existing body of quantitative work in this area.