The Subject of Functional Foods: Accounts of Using Foods Containing Phytosterols

by Kate Weiner
University of Manchester

Sociological Research Online, 16 (2) 7
<http://www.socresonline.org.uk/16/2/7.html>
10.5153/sro.2343

Received: 18 Jan 2011 Accepted: 12 Mar 2011 Published: 31 May 2011

Abstract

This paper explores the notion of the late modern or reflexive subject, for whom consumption, rationality, autonomy and a reflexive attitude to risk are said to be constitutive. Drawing on an example of 'ordinary' health consumption (Gronow and Warde, 2001), the paper addresses what kinds of consumer identities emerge in people's talk about buying or eating foods containing phytosterols. These are 'functional foods' which are marketed on the basis that they actively lower cholesterol. Based on interviews with people who say that they buy or eat these foods, the analysis focuses on participants' reported trajectories relating to how this came about. Participants' accounts contain a number of explicit and implicit reasons for buying or eating the foods, which I characterise as agential, contextual, or non-agential, depending on the degree to which they draw on the agency of the actual purchaser or eater. These different types of explanations can be ordered in terms of their appeals to rationality, risk consciousness and autonomy. In agential explanations, people talk, for example, of doing something good for themselves, or experimenting with the foods. These explanations explicitly position consumers as health conscious, autonomous and rational to varying degrees. Contextual explanations drew on, for example, the role of doctors or family history in alerting people to a potential problem. These suggest both a different sense of risk consciousness, which may be prompted or contextual, and a less autonomous kind of consumer who is connected to others through a set of family and other relationships. Non-agential explanations, for example, where people attributed their consumption to others or to habit, appeal neither to the rationality, the health consciousness nor the autonomy of the actual consumer. The analysis helps to reinforce the potentially contextual or fluctuating nature of risk consciousness, and the relational and non-instrumental aspects of daily practices.

Keywords: Functional Food; Ordinary Consumption; Phytosterol; Cholesterol; Consumer Subject; Agency; Reflexivity; Food Practices; Non-Instrumental Conduct

Introduction

1.1 This paper explores people's accounts of buying and eating foods containing phytosterols. It is based on 40 semi-structured interviews with self-identified current and former purchasers or eaters of these foods, and other members of their households. Drawing mainly on concepts from the sociologies of health and illness, food, and consumption, the paper aims to shed light on aspects of contemporary consumer identities in an era of health consumption and increasing emphasis on self-care.

1.2 Social theorists such as Giddens (1991) and Beck (1992) have characterised the central organising ideas of the current period as risk, choice and individual agency. As Gronow & Warde (2001:2) summarise, these theories are based on a model of a reflexive actor 'who, having become aware of the plasticity of personal biography, was forced to constantly choose among alternative courses of action through which the self came to be constituted'. Consumption is seen as central to these theories of reflexive or late modernity and integral to identity formation. A reflexive attitude to risk is also suggested, and how to manage risks is seen as central to subjectivity (Lupton, 2000). Indeed, in advanced liberal society, risk management becomes individualised and privatised and citizens are obliged to adopt a calculative and prudent relation to the future (Rose, 1996).

1.3 This model of the reflexive, autonomous and choosing late modern subject has been challenged on a number of levels in discussions of health consumption and consumption more widely. While symbolic aspects to 'lifestyle choices' are recognised, the model has been critiqued for overemphasising rational and reflexive action. For example, in relation to health care, MacDonald et al (2007) have focussed on the symbolic work involved in consumption in this realm. With their research undertaken in the primary care setting, they are interested in the identity work that people undertake when they talk about and access healthcare. They argue that while Giddens emphasises freedom and rationality, consumer identities are constrained by cultural norms and social expectations. Thus, in the context of finite health care resources and an emphasis on self-care, how people approach decisions in relation to healthcare provides messages about their fitness as citizens.

1.4 In a similar vein, sociologists and anthropologists of food have long been cognisant of the symbolic or communicative aspects of eating (Caplan 1997; Murcott 1988, 1998), and critical of understandings of food choices that are premised on a model of an individual and rational eater unconstrained by material or cultural concerns (for a recent example, see Delormier et al, 2008). This argument highlights not only critiques of rationality, but also points to the issue of autonomy. The model of the individual eater or health consumer has been critiqued for neglecting the potentially inter-subjective, collective or gendered aspects of eating habits, food choice and health care (Lupton 1997; Murcott 1983; Petersen and Lupton 1996).

1.5 While so far I have contrasted ideas of rational and calculative action with symbolic communication, both of these may be characterised as instrumental, reflexive or conscious types of conduct (Gronow and Warde, 2001; Warde 2005). In their discussions of 'ordinary consumption' and theories of practice, Gronow and Warde (Gronow and Warde, 2001; Warde 2005) argue for a place for non-instrumental ideas about conduct concerning, for example, the role of routine or affect. This critique is echoed in work on health consumption which recognises a place for emotions (Henwood et al. 2003; Lupton 1997).

1.6 In relation to a reflexive attitude towards health risks in everyday practices, empirical studies suggest that health may not be a master discourse, but may only become salient when health problems emerge (Calnan and Williams 1991; Lawton 2002) or may have to be balanced with other day to day priorities (Backett 1992; Keane 1997; Lambert and Rose 1996; Macintyre et al. 1998; Popay and Williams 1996). In the realm of food practices, the balance between 'healthfulness' and 'indulgence' has provided a key organising idea (Niva 2007; Warde 1997). All of this suggests that health risks may only be one of many factors influencing daily practices.

1.7 The foregoing discussion suggests that the model of the rational, calculating and risk-aware late modern subject has been comprehensively critiqued. Yet, many scholars find parallels between this subject and the patients or citizens encoded in contemporary health policy (see for example Henwood et al. 2003; Lupton 1997; McDonald et al. 2007; Nettleton and Burrows 2003). Contemporary health prevention discourses are, for example, premised on the notion that risk factors can be identified and illness avoided, mainly through the actions of individuals (for recent examples, see Cm 6374, 2004; Wanless 2002). Citizens are positioned as autonomous, risk-aware, and rational health consumers who are obliged to make appropriate 'lifestyle' choices (Lupton 1995; Petersen and Lupton 1996). Dietary advice represents a key site of individualised preventative regimes, and assumes a peculiarly rational and calculative approach to eating (Beardsworth and Keil 1997). As McDonald et al. (2007) indicate, these kinds of policy expectations provide the context for, and may be reflected in, people's accounts of their consumption practices.

1.8 Drawing on the foregoing discussion, this paper looks at the kinds of consumer identities that emerge in people's talk about buying and eating foods containing phytosterols. While medical sociology has engaged with similar issues in relation to health consumption, this has largely been in the realm of encounters with health professionals and health service provision (e.g. Lupton 1997; McDonald et al. 2007). Here, I aim to extend such work to the area of ordinary, everyday health consumption, looking at accounts of buying and eating foods containing phytosterols. Such foods are marketed on the basis that they actively lower cholesterol.

1.9 Foods containing phytosterols were first launched in the mid 1990s and a growing range of brands and products is now available worldwide. In the UK, the most prominent foods are margarine-type spreads, yogurt drinks and yogurts. These products are well established and, according to market research, widely used (TNS 2006). They both contribute to an established and long-standing field of commercial cholesterol management (Garrety 1997; Pollan 2007) and represent one of the vanguards of the emerging field of 'functional foods'. These may be thought of as foods marketed on the basis of having health promoting characteristics beyond their nutritional value, and which have been produced through a process of research and development (Heasman and Mellentin 2001; Niva 2008).

1.10 Social scientific research on functional foods has until recently tended to focus on their meaning and acceptability for potential consumers (see Niva 2008). In reaction to the market-orientation of this initial research, Niva (2008) investigates both how these foods (including phytosterols) are understood and consumed, stressing that holding negative views about the foods does not preclude buying or eating them. Recent interest in probiotics has provided analyses of representations in expert and public discourses (Burges et al. 2009; Koteyko and Nerlich 2007; Koteyko 2010) as well as people's accounts of their rationales and practices relating to these products. Drawing on a focus group study, Koteyko (2007) considers the relationship of probiotics to healthy eating in lay accounts. In related work, Crawford et al. (2010) introduce the idea of 'nutritional altruism', whereby people talk of buying and preparing probiotics for the benefit of others. There has been little sociological work of this nature in relation to products containing phytosterols (although see Lehenkari 2003;Niva 2008).

1.11 In an earlier paper, I considered how users of these foods containing phytosterols are imagined and configured within expert discussions and regulatory texts, including the kinds of motivations and practices attributed to them (Weiner 2010). I suggest that, in common with other contemporary health policy, the ideal user is configured as a risk-aware individual consumer who eats the foods in a rational manner. In the current analysis I attend to how consumers, themselves, account for their consumption of these foods.

Methods

2.1 This study is based on interviews with people who identified themselves as currently or formerly buying or eating phytosterol products and other adult members of their households where they were willing and available to participate.

Recruitment and sampling

2.2 Market research suggests that the majority of consumers of foods containing phytosterols are middle-aged and older, with more than half of all consumption accounted for by those aged sixty five and over (TNS 2006). Two main research sites were, therefore, selected on a pragmatic basis in order to access people with a variety of occupational backgrounds of both working and retirement age: a university in the midlands of England; and an older people's network in the north of England. Following institutional research ethics approval, I advertised the study to a variety of academic and non-academic staff groups at the university, and in the quarterly newsletter of the older people's network. Potential interviewees were selected, as far as possible, to provide diversity on age, gender and occupational background.

2.3 Since 'consumers' may be understood both as those who purchase and those who use a product (Warde 2005), recruitment was of those who claimed to buy and/or eat the foods, or had done so in the past. Partners or other members of the household of the initial respondents were invited to participate in interviews on the basis that those who purchase foods do not necessarily eat them and that 'food choices' may be a collective matter.

Interviews

2.4 I undertook a total of 40 semi-structured interviews with 52 participants, including 41 people who currently eat the foods, four who used to, and seven who do not. I devised an interview topic guide, based on my analysis of the biomedical literature concerning phytosterols (Weiner 2010) and on existing literatures on food choice, lay health identities and practices, and functional foods. After establishing which products were purchased currently or in the past, the interviews focused on participants' accounts of: how they came to buy or eat these products, or stop buying/eating the products; their health biographies and interactions with health care professionals; their practices relating to the foods; and more general dietary practices.

2.5 The topic guide was used in a flexible way, depending on the direction of discussion and on any time constraints indicated by the interviewees. Interviews lasted between approximately 30 and 80 minutes. I recorded all interviews on a digital audio recorder and these were transcribed in full. Interviewees' names referred to in the paper are pseudonyms.

Analysis

2.6 I undertook a thematic analysis of the interview data following the process outlined by Hammersley and Atkinson (1995). It involved an iterative process of identifying data that resonated with existing themes in the literature, and identifying new themes through noticing recurring talk on particular topics or the recurrent use of a particular phrase, and paying attention to surprising aspects. Analysis was an ongoing process initiated at the outset of the fieldwork. The first stages involved compiling summaries and reflections on each completed interview and analytic notes with overarching reflections. On completion of the interviews, I developed a coding schema which I applied to all the data. I used the qualitative data analysis software package NVIVO to facilitate coding, storage and retrieval of the data.

2.7 In analysing these interviews, I have tried not to make assumptions about the veracity or otherwise of people's reported rationales and practices (see Murphy and Dingwall 2003). Rather, in keeping with McDonald et al (2007), the analysis is concerned with what the interview data indicate about the normative frames associated with phytosterol foods. In other words, they reflect what a phytosterol consumer might be expected to say about their rationales and practices in order to be seen as a normal, moral and responsible consumer or satisfactorily account for transgressions. The analysis, therefore, indicates the available narratives about these foods.

Sample characteristics

demographic characteristics: The sample included 21 men and 31 women, of whom 20 and 25 respectively eat/ate the foods. Those who eat/ate the foods were aged from 27-85, but most were aged 40 or over. Interviewees had a variety of occupational backgrounds although, in keeping with available market research (TNS 2006), those in professional and managerial occupations predominated (65%). The majority were white British.
phytosterol products: Households reported consuming or formerly consuming a variety of phytosterol products, sometimes in combinations, including spreads, drinks, yogurts, milk and cheese. The most commonly reported products were spreads and drinks in 34 and 12 households respectively. Interviewees reported eating or having eaten these products for varying durations from 2 months to several years.

Why consume foods containing phytosterols?

3.1 Participants' accounts contain a number of explicit and implicit reasons for buying or eating these foods which I have loosely grouped into three types, depending on the degree to which they appear to draw on the agency of the consumer:

Agential explanations – these centre on the individual agency of the interviewee without reference to other people or influences.
Contextual explanations– where interviewees associated consumption with some kind of external prompt for example interactions with health services or family experiences.
Non-agential explanations – where explanations do not depend on the agency of the actual eater, but where responsibility is firmly placed elsewhere or nowhere at all.

Interviewees might draw on several of these and across the different types during the course of the interview. Each of these types of explanations will be discussed in turn.

Agential explanations:

3.2 Doing something good/within reason: In some instances, people drew on an explicit narrative of health consciousness and of openness to health promoting habits, or of trying to do things that would be good for them. For example, this was the first thing that one retired woman said at the beginning of the interview:

Interviewer: Okay, so could you remind me why you responded to my advert – which of the products you had in mind?
Paula: Well, I thought it was to do with Flora Proactiv. So basically, I started to eat that as soon as it came out because I've always been conscious of health and cholesterol and things like that […] if you're going to put something inside you, you may as well put something inside you that's going to be good for you.

3.3 In this narrative of 'doing something good', the products might be framed as a healthy option or a continuation of healthy eating, in a perhaps more generic way than being linked to specific regimes of cholesterol reduction. Thus Karen, a middle-aged woman, related:

I think it's just in the same vein that maybe we would maybe choose to eat like brown rice instead of white or wholemeal spaghetti or seeded brown bread instead of white. So I guess the Flora Proactiv fits in that category really. It seems like a better food choice somehow.

3.4 Other instances suggested a sense of bounded health consciousness. These may enrol a narrative of doing things within reason, or of fitting around personal limitations. In the following example, for Mike, a man in his fifties, reasonableness is contingent on ease of consumption and taste:

well if I could do anything within reason to do that then that's what I'll do because it's convenient, it's easy […] the actual food content is fine […] it tastes okay.

These narratives of doing something good within reason are in keeping with sociological discussions concerning the balancing of pleasure and health, discussed in the introduction (Niva 2007; Warde 1997). The foregoing narratives work to create an identity as a reasonable actor who shows a commitment to maintaining health.

3.5 Doing something symbolic: The interviews hint at a further, symbolic role of phytosterols, providing a sense of doing something for oneself. Like the narrative of doing something good, these symbolic roles allow people to demonstrate a commitment to health. Here Kate, a middle-aged woman, talks of her husband's relationship to the foods:

that's indicative of all those other values he places on things and that's why we still have it, and having it in the fridge is an important thing because it signifies something. You know I don't think we've got any proof that that particularly has made a difference but it makes him feel like he's actively doing something as opposed to letting things be beyond his control in terms of his health. […] in our house you know 'daddy's spread' is a daily reminder that he's going to make every effort he can.

3.6 In other instances this symbolic role was framed in psychological terms, even as a 'placebo'. As one young woman, Rachel, explained:

You're doing something to sort of unconsciously help something that could be a problem and it's better than doing nothing and knowing you're doing nothing […] I think it's psychological, but maybe a real benefit of having it'

These narratives are reminiscent of those reported by Koteyko (2009: 598) concerning probiotics, who suggests that their consumption may be 'justified as part of an 'optimizing nutrition' move which may or may not deliver actual health benefits but is a safe option that makes one 'feel better' with respect to one's own health or that of others'. This symbolic role in making oneself feel better or feel like one is doing something partly chimes with recent theories of practice, which have highlighted non-instrumental aspects of conduct (Warde 2005). In other words, the symbolic role described in these interviews may contribute to identity work, demonstrating to self and others a commitment to health, but may also be connected with feelings, emotions and embodiment.

3.7 Experimental consumption: Narratives of 'doing something symbolic', discussed in the previous section, suggest a degree of ambivalence amongst consumers about the efficacy of these foods. In other instances, however, a type of 'experimental' consumption was discernable, which was expressed as a more or less instrumental stance towards the continued consumption or otherwise of the phytosterol products in which proof of efficacy came through personal cholesterol testing. This was exemplified in an interview with Elizabeth, a retired woman in her late seventies:

I saw these things in the shop and I thought well they probably don't do you any good but you never know and might be better than nothing. And so I started taking them and they're quite innocuous […] and so I did take them and as I explained to you the next time I had a blood test the cholesterol had gone down, though only slightly. I mean it's still raised cholesterol but at least there's a little bit of a difference, so I thought well might as well carry on then.

3.8 As a brief methodological aside, Kate, the woman cited in the previous section, who so articulately expressed the symbolic aspect, drew on this experimental narrative elsewhere in the interview:

I remember having taken Flora and possibly changed other things in his lifestyle as well, he noticed a real difference in his cholesterol. So he sort of stopped worrying about that quite so much and felt that he'd just continue sort of eating those sorts of things.

This example underscores the multiplicity of narratives which may be thought of as a stock of ideas and arguments on which interviewees may draw in accounting for the consumption of these foods.

3.9 Consumption as an alternative to medication: Some interviewees explicitly presented the products as an alternative to statin medication, talking of replacing medication, hoping to replace medication, or staving off medication, because of actual experience or fear of 'side effects' or antipathy to taking pharmaceutical medications in general. Like experimental use, this suggests a more instrumental stance to the foods than suggested by the narrative of symbolic consumption.

3.10 Taken together, the different forms of agential explanations (doing something good or symbolic, experimental consumption, or as an alternative to medication) draw on an explicit narrative of health consciousness and/or of rational and calculative action. These largely describe reflexive and conscious types of conduct, although symbolic consumption hints at an additional place for affect. The kinds of subjects that emerge in these narratives roughly align with those described in theories of late modernity and prescribed in health policy.

Contextual Explanations

3.11Medical tests and monitoring: While phytosterol use was associated with a variety of interwoven narratives, it was often directly related to receiving cholesterol testing or ongoing monitoring. Furthermore, cholesterol testing was mostly framed as having been instigated by health care practitioners as part of routine or investigative procedures, rather than at the specific instigation of the interviewee. In the following example, Mary, a woman in her sixties, suggests that it was a series of illnesses, i.e. an embodied consciousness of ill-health, that brought her into contact with medical testing, rather than a more abstract or preventative mode of health consciousness. Cholesterol testing is portrayed as rather incidental to the woman's initial reasons for making contact with health professionals:

Interviewer: How did you come to be buying the Benecol bottles?
Mary: Well in two ways, first of all I decided that I ought to have a health check up because [describes series of respiratory tract infections and generally feeling unwell], so I went to one of these BUPA^[1] things […] then when I went to my doctor over something else she said to me "Oh you're coming up to the age where we do a health check up," and then repeated everything. So, and the BUPA one, one of the things they pointed out that was useful was that my cholesterol was high, because it was six point something and, and then the doctor […] she did a cholesterol test and it was the same and she said I should try to get it down.

3.12 These accounts can be characterised as wholly rational and reflexive, in as much as they relate specifically to knowledge of cholesterol levels. It is noticeable, however, that cholesterol testing emerges as something instigated by others. The interviewees do not frame it as their intention to seek cholesterol testing, but to seek help for other reasons. This suggests a partly reactive model of phytosterol consumption in which interactions with the medical field acts as a prompt. In contrast to some of the agential explanations discussed in the previous section, being a cholesterol – conscious or prevention-orientated type of person is not central to these kinds of accounts.

3.13 Family experiences: References to family history or family experiences were also a relatively frequent part of these accounts. Interviewees talked either explicitly of an heredity aspect to heart troubles or high cholesterol, or indicated that family events, affinities or conversations had prompted an awareness of the possibility of their own fallibility. These types of narratives were sometimes drawn on immediately to provide a direct explanation for buying phytosterol foods, or were interwoven with other explanations at different points in the course of the interview. The clearest example was provided by a middle age man, who reported his own cholesterol levels to be normal:

Interviewee: Can you just tell me how the cholesterol lowering things fetched up coming into your household?
David: Right, I think in terms of, if you look at my family history, three of my four grandparents died of a heart attack, the fourth died of pneumonia having had a heart attack […] my dad has had open heart surgery, triple bypass when he was fifty five, and has had a number of heart problems, so I think it would be fair to say I'm in a high risk group in terms of heart disease. I mean there may well be a number of differences in terms of lifestyle and things […], but I still think hereditary-wise, I've got to be in a fairly high risk group.

3.14 The fairly prominent place of family history and family experiences in participants' narratives is congruent with previous studies on lay understandings of heart disease. These have shown that heredity is an important, but not decisive element of these understandings and argued that family experiences are paramount in the development of lay theories of health and illness (Davison et al. 1989; Hunt et al. 2000; Hunt and Emslie 2001). The data support the idea that risk consciousness is situated or contextual; it is not necessarily a universal feature of the late modern citizen. In other words, health risks might not be a master discourse shaping people's everyday practices, but might only become salient under specific circumstances.

3.15 Life course and life events: Changes in life circumstances were also sometimes linked with the decision to purchase the products. These might be characterised as points at which one might take stock or reappraise one's habits. Changed circumstances including starting or ending a relationship, 'trying for a baby', starting a family, starting the menopause or retiring were all linked to changed eating and lifestyle habits, and periods of renewed health consciousness. These narratives again suggest a view of risk consciousness as fluctuating and contextual rather than constant and universal.

Non-agential explanations

3.16 Delegated responsibilities:in more than one quarter of the interviews there was reference to responsibility for consuming phytosterol foods being attributed, at least in part, to people other than those who actually eat them. There were both cases of interviewees attributing their own consumption to other people or of claiming responsibility for other people's consumption. So, for example, one woman in her eighties attributed her own consumption to her daughter:

Moira: Well just my daughter told me about it, she said you need to get these sort of things instead of whatever I was you know? […]
Interviewer: So I'll come back to the Flora Proactiv spread, so do you remember, did you buy it yourself the first time?
Moira: No my daughter bought it for me. The first time, she said this is what you need to use now.

3.17 These narratives were in some cases a continuation of more general gendered narratives of household eating patterns which charged women with looking out for the health of their male partners or their children, as Barry, a middle aged man, explained:

My wife does most of the shopping and she bought, if I remember it's because I'd had a recent health check which had shown my cholesterol was not high, but it was on the top of where they like it to be […]. So she worries more than I do about these things, thought we ought to try this, if it says it lowers your cholesterol.

This idea of delegated responsibility places into question the notion of the individual or autonomous rational consumer. This has been repeatedly challenged by work on food practices (e.g. Beagan et al. 2008; Crawford et al. 2010; Henson et al. 1998; Murcott 1983).

3.18 Habit: The idea of habit or being used to certain products seems to imply a dissipation or absence of responsibility for selecting these foods. In keeping with the idea of delegated responsibility, interviewees might attribute the original agency for purchasing the foods with other people or say that they cannot recall the origin. For example, Robert, a man in his late seventies, attributed the decision to buy phytosterols spreads to his late wife. On checking with him that he had decided to carry on buying it he responded, nodding to the mantelpiece on which there were a number of china ornaments:

Oh yeah, I carry on with everything. Some of the ornaments I probably wouldn't like too much, but I still carry on with those. No, I think it works you see yeah.

3.19 This implicit reference to habit became explicit later in the interview, when I asked Robert why he buys phytosterol products, despite receiving free prescriptions for cholesterol-lowering medicines (called statins):

well I have to say it's simply habit because we've always done it and probably it would be before I was on the statins.

These narratives of habit again challenge the notion of the rational and individual phytosterol consumer. As discussed in the introduction, theorists of modernity have been critiqued for overemphasising the role of reflexive, instrumental action, at the expense of routine non-instrumental conduct (Gronow and Warde 2001; Warde 2005). The data suggest that users themselves can be reflexive about the non-reflexive and routinised aspects of their daily habits.

Discussion

4.1 Participants in this study express a variety of narratives about foods containing phytosterols. I have characterised their accounts of why they buy or eat these foods as agential, contextual or non-agential. Agential explanations explicitly position consumers as health conscious, autonomous and doing something good for themselves. In this way, these explanations suggest a highly instrumental and conscious type of conduct. They vary in the degree to which they may be characterised as rational, with experimental use and use as an alternative to medication providing the most explicitly rational stances and symbolic consumption the most explicitly communicative stance. However, symbolic consumption was also partly concerned with affect (i.e. how it makes one feel about oneself). Contextual explanations do not have health consciousness as the starting position, but are concerned with how a person becomes health conscious through context i.e. through interactions with the health care system, family, friends or through life changes. These explanations provide both a different sense of risk consciousness, which may be prompted or situated, and also suggest a less autonomous kind of consumer who is connected to others through a set of family and other relationships. Such explanations may be characterised as highly rational where they are connected with medical test results that indicate raised cholesterol, or perhaps appeals to family history, although the latter may follow a lay calculus of risk that does not necessarily match formal expert assessments. Non-agential explanations appeal neither to the health consciousness nor the autonomy of the actual consumer. Furthermore, in instances of delegated responsibility, consumption could not be characterised as a rational project of the person who eats the foods, although perhaps the rational project of someone else, and habit represents a form of routine, non-instrumental conduct. In sum, these different types of narratives about why people buy and eat these foods can be clearly ordered, from agential to non-agential, in terms of their appeals to autonomy, reflexivity about risk, and (although not quite so neatly) rational or instrumental actions.

4.2 Gronow and Warde (2001) suggest that studies of consumption have tended to focus on instrumental forms of conduct, particularly the communicative or symbolic aspects, linking this to the types of examples studied. They appeal for scholars to pay greater attention to 'ordinary' as opposed to conspicuous or spectacular forms of consumption. Yet as McDonald et al (2007) point out, even less conspicuous forms of consumption, such as making 'choices' about health care within primary practice, may be associated with identity work. In the current study, I see the research interview as a place where identity work may occur. Here, agential accounts largely present people as autonomous, reflexive, rational and calculative actors. This type of actor most closely conforms to sociological descriptions of the late modern subject and may also be thought of as the most preferred or sanctioned kind of actor within health policy. Yet, several other kinds of actors and types of conduct can be seen in these accounts. It is clear that these participants did not necessarily feel obliged to present themselves foremost as rational, reflexive or autonomous. Indeed, resonating with previous studies, the analysis finds space for non-rational and non-instrumental elements of phytosterol consumption.

4.3 The current analysis finds parallels with recent work on probiotics. For example, the idea of nutritional altruism (Crawford et al, 2010) also highlights the interconnected or relational aspects of food practices. There are also parallels with the interpretive repertoires identified by Koteyko (2010), in their differential appeals to rationality and affect. There, the balanced diet repertoire, which sees probiotics as equivalent to supplements and generally unnecessary except on specific occasions, provides the most rational approach. Like the current study, Koteyko suggests that eating for extra health is partly concerned with the symbolic benefits of feeling like one is doing something for one's own health, while eating for pleasure is entirely to do with affect, providing the least instrumental position. The symbolic aspects identified by Koteyko equate, perhaps, with the 'regimes of hope' relating to probiotics discussed by Burges et al. (2009), which they contrast with a scientific focus on rationalist 'regimes of truth'.

4.4 It is notable that the current interviews have circumscribed a number of types of rationales for consuming foods containing phytosterols that were either proscribed or unanticipated in expert discussions (see Weiner, 2010), such as seeing them as a generic healthy food, symbolic consumption, eating as an alternative to medication, and habitual consumption. Like work on prescription and lifestyle medicines (Fox and Ward 2006; Pound et al. 2005), the analysis brings to light potential uses of these foods that conform to lay rather than expert rationales. The current work, however, contrasts with that of medical sociologists interested in the use of prescription medicines, who have been at pains to demonstrate the rationality of lay practices, in the face of expert discourses about 'compliance'. For example, Pound et al. (2005) have characterised these practices as a largely rational form of cost-benefit analysis, describing an array of ways that people modify drug regimens to mitigate putative harms. Studies of health-related beliefs and practices, more widely, have also pointed to the rationality of these in their own terms, for similar reasons (e.g. Backett et al. 1994). Nevertheless, these studies suggest that lay people weigh-up and balance a range of imperatives and considerations, including rationales such as social acceptability (e.g. neither too much nor too little focus on health) and pleasure. These, in Ward's (2005) terms, might be considered, respectively, as instrumental, but non-rational aspects of practice which are concerned with communicating values, and non-instrumental aspects concerned with emotion and embodiment. It is notable that in lay narratives, pleasure may be framed as health promoting (Backett 1992; Niva 2008), thus, what Ward considers to be non-instrumental may be reconfigured as instrumental behaviour.

4.5 This paper represents an attempt to bring medical sociology into conversation with sociology of consumption in the area of mundane health consumption. Overall, the analysis reinforces critiques of sociological theories of late modernity, which are said to privilege rational and autonomous action at both the expense of other forms of instrumental action, and also non-instrumental conduct. The idea of contextual and non-agential narratives helps bring to light the potential granularity of risk consciousness, and the relational/inter-dependent and non-instrumental aspects of daily practices.

Acknowledgements

This work was supported by the Leverhulme Trust Early Career Fellowship Award No 40344. I am particularly indebted to Catherine Will for her incisive comments and suggestions during the development of the paper. I would also like to thank Joseph Burridge, Anne Murcott, Wendy Wills and the two anonymous reviewers for their helpful comments on earlier drafts of the paper, and colleagues at the Institute for Science and Society and at the BSA Food Study Group Conference 2010 for their comments and discussion at presentations of this work.

Notes

¹ BUPA is a private health insurer and provider in the UK which provides health assessments.

References

BACKETT, K. (1992), 'Taboos and excesses: lay health moralities in middle class families', Sociology of Health & Illness, Vol. 14, No. 2, pp. 255-73. [doi:10.1111/1467-9566.ep11343709]

BACKETT, K., DAVISON, C., and MULLEN, K. (1994), 'Lay evaluation of health and healthy lifestyles: evidence from three studies', British Journal of General Practice, Vol. 44, pp. 277-80.

BEAGAN, B., CHAPMAN, G., D'SYLVA, A., and BASSET, R. (2008), ''It's just easier for me to do it': rationalizing the family division of foodwork', Sociology, Vol. 42, No. 4, pp. 653-71. [doi:10.1177/0038038508091621]

BEARDSWORTH, A. and KEIL, T. (1997), 'Sociology on the Menu: Invitation to the Study of Food and Society' London: Routledge.

BECK, U. (1992), 'Risk Society: Towards a New Modernity'; London: Sage.

BURGES, D., MOREIRA, T., and MURTAGH, M. (2009), 'Little bottles and the promise of probiotics', Health, Vol. 13, No. 2, pp. 219-34.

CALNAN, M. and WILLIAMS, S. (1991), 'Style of life and the salience of health: an exploratory study of health related practices in households from differing socio-economic circumstances', Sociology of Health & Illness, Vol. 13, No. 4, pp. 506-29. [doi:10.1111/j.1467-9566.1991.tb00525.x]

CAPLAN, P. (1997), 'Approaches to the study of food, health and identity', in P. CAPLAN (ed.), Food, health and identity London: Routledge, pp. 1-31.

CM 6374 (2004), 'Choosing Health: Making Healthy Choices Easier', London: TSO.

CRAWFORD, P., BROWN, B., NERLICH, B., and KOTEYKO, N. (2010), 'Nutritional altruism and functional food: lay discourses on probiotics', Sociology of Health & Illness, Vol. 32, No. 5, pp. 745-60. [doi:10.1111/j.1467-9566.2010.01238.x]

DAVISON, C., FRANKEL, S., and DAVEY SMITH, G. (1989), 'Inheriting heart trouble: the relevance of common-sense ideas to preventative measures', Health Education Research, Vol. 4, No. 3, pp. 329-40. [doi:10.1093/her/4.3.329]

DELORMIER, T., FROHLICH, K. L., and POTVIN, L. (2009), 'Food and eating as social practice – understanding eating patterns as social phenomena and implications for public health', Sociology of Health & Illness, Vol. 31, No. 2, pp. 215-28. [doi:10.1111/j.1467-9566.2008.01128.x]

FOX, N. and WARD, K. (2006), 'Health identities: from expert patient to resisting consumer', health, Vol. 10, No. 4, pp. 461-79.

GARRETY, K. (1997), 'Social worlds, actor-networks and controversy: the case of cholesterol, dietary fat and heart disease.' Social Studies of Science, Vol. 27, No. 5, pp. 727-73. [doi:10.1177/030631297027005002]

GIDDENS, A. (1991), Modernity and Self-Identity: Self and Society in the Late Modern Age; Cambridge: Polity Press.

GRONOW, J. and WARDE, A. (eds.) (2001), Ordinary Consumption London: Routledge.

HAMMERSLEY, M. and ATKINSON, P. (1995), Ethnography: Principles in Practice; London: Routledge.

HEASMAN, M. and MELLENTIN, J. (2001), The functional foods revolution. Healthy people, healthy profits? ; London: Earthscan Publications.

HENSON, S., GREGORY, S., HAMILTON, M., and WALKER, A. (1998), 'Food choice and diet change within the family setting', in A. MURCOTT (ed.), The Nation's Diet: the Social Science of Food Choice London: Longman, pp. 183-96.

HENWOOD, F., WYATTS, S., HART, A., and SMITH, J. (2003), ''Ignorance is bliss sometimes': constraints on the emergence of the 'informed patient' in the changing landscapes of health information', Sociology of Health and Illness, Vol. 25, No. 6, pp. 589-607. [doi:10.1111/1467-9566.00360]

HUNT, K. and EMSLIE, C. (2001), 'Commentary: The prevention paradox in lay epidemiology-Rose revisited', International Journal of Epidemiology, Vol. 30, pp. 442-46. [doi:10.1093/ije/30.3.442]

HUNT, K., EMSLIE, C., and WATT, G. (2000), 'Barriers rooted in biography: how interpretations of family patterns of heart disease and early life experiences may undermine behavioural change in mid-life', in H. GRAHAM (ed.), Understanding Health Inequalities Buckingham: Open University Press, pp. 113-26.

KEANE, A. (1997), 'Too hard to swallow? The palatability of health eating advice', in P. CAPLAN (ed.), Food, health and identity London: Routledge, pp. 172-92.

KOTEYKO, N. (2009), ''I am a very happy, lucky lady, and I am full of Vitality!' Analysis of promotional strategies on the websites of probiotic yoghurt producers', Critical Discourse Studies, Vol. 6, No. 2, pp. 111-25. [doi:10.1080/17405900902749973]

KOTEYKO, N. (2010), 'Balancing the good, the bad and the better: a discursive perspective on probiotics and healthy eating', health, Vol. 14, No. 6, pp. 585-602.

KOTEYKO, N. and NERLICH, B. (2007), 'Multimodal Discourse Analysis of Probiotic Web Advertising', The International Journal of Language, Society and Culture, Vol. 23, pp. 20-31.

LAMBERT, H. and ROSE, H. (1996), 'Disembodied knowledge? Making sense of medical science', in A. IRWIN and B. WYNNE (eds.), Misunderstanding science? The public reconstruction of science and technology Cambridge: Cambridge University Press, pp. 65-83. [doi:10.1017/CBO9780511563737.004]

LAWTON, J. (2002), 'Colonising the future: temporal perceptions and health-relevant behaviours across the adult lifecourse', Sociology of Health & Illness, Vol. 24, No. 6, pp. 714-33. [doi:10.1111/1467-9566.00315]

LEHENKARI, J. (2003), 'On the bordeline of food and drug: constructing credibility and markets for functional food product', Science as Culture, Vol. 12, No. 4, pp. 499-525. [doi:10.1080/0950543032000150337]

LUPTON, D. (1995), The Imperative of Health: Public Health and the Regulated Body; London: Sage Publications.

LUPTON, D. (1997), 'Consumerism, reflexivity and the medical encounter', Social Science & Medicine, Vol. 43, No. 3, pp. 373-81. [doi:10.1016/S0277-9536(96)00353-X]

LUPTON, D. (2000), 'Food, risk and subjectivity', in M. CALNAN, J. GABE, and S. WILLIAMS (eds.), Health, Medicine, and Society: Key Theories, Future Agendas London: Routledge, pp. 205-18.

MACINTYRE, S., REILLY, J., MILLER, D., and ELDRIDGE, J. (1998), 'Food choice, food scares, and health: the role of the media', in A. MURCOTT (ed.), The nation's diet: the social science of food choice Harlow: Addison Wesley Longman, pp. 228-49.

MCDONALD, R., MEAD, N., CHERAGHI-SOHI, S., BOWER, P., WHALLEY, D., and ROLAND, M. (2007), 'Governing the ethical consumer: identity, choice and the primary care medical encounter', Sociology of Health & Illness, Vol. 29, No. 3, pp. 430-56. [doi:10.1111/j.1467-9566.2007.00493.x]

MURCOTT, A. (1983), '"It's a pleasure to cook for him". Food, mealtimes and gender in some South Wales households', in I. E. GAMARNIKOV, et al. (eds.), The public and the private: social patterns of gender relations London: Heinemann, pp.78-90.

MURCOTT, A. (1998) 'Sociological and social anthropological approaches to food and eating', World Review of Nutrition and Dietetics, Vol. 55, pp. 1-40.

MURCOTT, A. (1998), 'Food choice, the social sciences and 'The Nation's Diet' research programme', in A. MURCOTT (ed.), The Nation's Diet: The Social Science of Food Choice London: Longman, pp. 1-22.

MURPHY, E. and DINGWALL, R. (2003), Qualitative Methods and Health Policy Research; New York: Aldine De Gruyter.

NETTLETON, S. and BURROWS, R. (2003), 'E-scaped medicine? Information, reflexivity and health', Critical Social Policy, Vol. 23, No. 2, pp. 165-85. [doi:10.1177/0261018303023002003]

NIVA, M. (2007), ''All foods affect health': understandings of functional foods and healthy eating among health-oriented Finns', Appetite, Vol. 48, pp. 384-93. [doi:10.1016/j.appet.2006.10.006]

NIVA, M. (2008), Consumers and the conceptual and practical appropriation of functional foods; Helsinki: National Consumer Research Centre.

PETERSEN, A. and LUPTON, D. (1996), The New Public Health: Health and Self in the Age of Risk; London: Sage.

POLLAN, M. (2007), 'Unhappy meals', The New York Times, 28 January.

POPAY, J. and WILLIAMS, G. (1996), 'Public health research and lay knowledge', Social Science and Medicine, Vol. 42, No. 5, pp. 759-68. [doi:10.1016/0277-9536(95)00341-X]

POUND, P., BRITTEN, N., MORGAN, M., YARDLEY, L., POPE, C., DAKER-WHITE, G., and CAMPBELL, R. (2005), 'Resisting medicines: a synthesis of qualitative studies of medicine taking', Social Science & Medicine, Vol. 61, pp. 133-55. [doi:10.1016/j.socscimed.2004.11.063]

ROSE, N. (1996), 'Governing "Advanced" Liberal Democracies', in A. BARRY, et al. (eds.), Foucault and Political Reason: Liberalism, Neo-Liberalism and Rationalities of Government, London: UCL Press, pp. 37-64.

TNS (2006), 'Consumer research on the consumption of phytosterols: Prepared for the COI and the Food Standards Agency', London: TNS.

WANLESS, D. (2002), 'Securing our Future Health: Taking a Long-Term View', London: HM Treasury.

WARDE, A. (1997), Consumption, Food and Taste: Culinary Antinomies and Commodity Culture; London: Sage.

WARDE, A. (2005), 'Consumption and theories of practice', Journal of Consumer Culture, Vol. 5, No. 2, pp. 131-53. [doi:10.1177/1469540505053090]

WEINER, K. (2010), 'Configuring users of cholesterol lowering foods: A review of biomedical discourse', Social Science & Medicine, Vol. 71, No. 9, pp. 1541-47. [doi:10.1016/j.socscimed.2010.06.048]