The UK Faculty of Public Health has recently taken ownership of the Health Knowledge resource. This new, advert-free website is still under development and there may be some issues accessing content. Additionally, the content has not been audited or verified by the Faculty of Public Health as part of an ongoing quality assurance process and as such certain material included maybe out of date. If you have any concerns regarding content you should seek to independently verify this.

Section 5. Behaviour in response to illness and treatments


Behavioural risks and responses to illness and treatments

Theorising risk and “risky” health behaviour

The strategy of health promotion represents a central element of government public health interventions. The field of health promotion itself encompasses, along with its more traditional educational campaigns, community development, personal skills development, the control of advertising ‘unhealthy' products’, and the monitoring and periodic screening of sub-populations. These approaches to the promotion of health and prevention of disease utilise a particular discourse of risk and its management which has been termed the ‘at risk self; ‘The encroachment of health promotion into these areas has multiplied the number of sites for preventive action, and given rise to an endless parade of ‘at risk’ populations and ‘risky situations’ (Petersen, 1997:195).

In other words, many of the health promotion and prevention strategies have been concerned with identifying so-called ‘problem’ groups such as adolescent drug-users, smokers, the ‘obese’, etc. Interventions are directed at persuading these ‘at risk’ groups to change or control their behaviour or ‘lifestyle’ so as to reduce the damage they are perceived to be causing to their health through unprotected sex, smoking, unbalanced diet, etc. The assumption underlying many health promotion campaigns is that it is individual volitional behaviour that constitutes the primary ‘risk’ to health.

In contrast, sociologists would generally argue that the notion of a ‘health risk’ cannot be understood outside of the wider context of what has been termed the ‘Risk Society’. That is, the social and environmental context in which the hazards and insecurities of modern industrialised societies occur, and which in large part, are an inadvertent product of scientific and technological development. These developments which include industrial pollution, the safe use of nuclear energy, a transport system centred on the personal use of the car, etc, have brought the concept of risk to the forefront of social and scientific awareness over the past two decades. (Beck, 1992). Beck's notion of the 'risk society' and his argument that ‘late modern’ societies are marked by end of uncritical confidence in scientific knowledge has become hugely influential within the social and political sciences. It points to the status of expert knowledge becoming increasingly contested in such societies.


Risk in the context of health interventions:

What has become known as the ‘risk assessment approach’, emerged largely in response to a heightened public awareness of such social and environmental threats. This approach is about the identification and measurement of risk, and has its origins within the Engineering, Transportation, and Power industries. Within this approach, risk analysis is conceived as a largely technical matter, and therefore potentially controllable.

The ‘risk assessment’ approach now plays an important role in medicine and health care. Skolbekken (1995) pointed out that more than 80,000 articles on risk had been published in medical journals since 1987. He argued that these studies generally lacked any kind of coherence regarding what constitutes a ‘risk’. Petersen & Lupton (1996) have also argued that the notion of risk is essentially constructed out of multi-causal epidemiological models that turn statistical relationships into causal ones through the aggregation of probabilities. This process results in risk factors being seen as realities / causative agents in their own right.

Health interventions logically develop from this essentially epidemiological approach. For example, the targeting of particular ‘at risk’ groups for health promotion ‘behaviour-change’ interventions, or as in the case of the new reproduction technologies, i.e. screening for foetal abnormalities new dilemmas of both a practical and ethical nature are opened-up (Higgs, 1998;183). However, because epidemiological observational studies usually consider several risk exposures, outcomes, as well as subgroups necessitating the use of multiple statistical tests of hypotheses, there is often a high probability of finding associations that are statistically significant but spurious (Pocock et al, 2004).

Hence for example, in the field of AIDS gay men were until relatively recently identified as having a higher risk of getting AIDS compared with heterosexual men because of so-called ‘gay lifestyle’ practices, identified as anal intercourse and multiple sexual partners. Risk has been calculated on the questionable assumption that all gay men ascribed to a single set of cultural practices and values (Glick Schiller et al, 1994  -  cited in Gabe, 1995:3). In the case of the new reproduction technologies, which open-up the potential for screening for foetal abnormalities, new personal dilemmas of both a practical and ethical nature have emerged from the apparent ability to objectively measure the risks of such interventions (Higgs, 1998;183).

These recent advances have contributed to a change in the basic attitudes where matters of life and death are concerned. The risk acceptance that is internalised in a fatalistic attitude to these matters is being replaced by an ideology whose primary goal is to gain control over life and death, where identification of and the struggle to reduce/eliminate risk factors have become activities of considerable importance and prestige within the health professions’ (Skollbekken, 1995;297).

Typically, the concept of risk behaviour in relation to health outcomes is generally based upon an epidemiological understanding of disease incidence and frequency, and often neglects the less easily measurable social contexts within which these ‘risky behaviours’ occur (Williams et al, 1995:120). This epidemiological use of the concept is present in the bulk of the health inequalities literature.


Public perceptions of risky health behaviour

Public perceptions of risk often bear little relationship to what professional experts say about probability of certain events taking place. In other words, there is often a mismatch between the notions of risk underlying state health interventions and the health concerns and self-defined needs of the lay public. With such dissimilar frames of reference, risk information however detailed, may do little to narrow such differences. Whilst the social distribution of risk does differ from the social distribution of wealth, risks are often distributed in a class specific way and as such they reinforce existing social inequalities.


Sociological & social psychological approaches to theorising risk behaviour

Bloor (1995) recognises four distinct approaches to the analysis of individual/group risk behaviour within the social sciences:

The first approach is represented by social psychological models utilising the principles of social cognition to emphasise the volitional character of risk behaviour. This is an approach which privileges an individual’s perceptions or beliefs over group influence or the actual social situation in which the behaviour occurs. Examples of this approach include the ‘Health beliefs model’ and the ‘Expectancy-value models’ of behaviour, one example being the theory of reasoned action (TRA). The former analyses the perception of vulnerability or susceptibility of individuals to particular health threats, i.e. smoking and lung cancer, but in practice has found to be a poor predictor of health behaviour. The TRA was revised by its original author Ajzen (1988) to accommodate the criticism that behaviours are often not under volitional control, by adding an additional component ‘perceived behavioural control’ (essentially the concept of ‘self-efficacy’). The major criticisms of these cognitive models (i.e. the use of social schemas) of how people process information is that they too often ignore the affective components of behavior.  N.B. until relatively recently, affect and cognition were seen as separate information processing systems. Secondly, that they assume that risks have an independent objective existence separate from the more complex social contexts in which people experience them.

A second approach utilises a social psychological cost/benefit analysis. It draws upon social learning theory in order to explain the benefits to individuals of engaging in risky behavior, i.e. risking pregnancy or HIV infection to gain intimacy with a sexual partner. This approach has been termed situated rationality. It assumes that such health behaviour is rational to an individual given their immediate social situation, however risky it might be to their long-term physical health. The problem with this approach is that it assumes a type of economic rationality in decision-making by individuals, a weighing-up of all the potential costs to see if the benefits of a particular behaviour are worth the risks. In practice all behaviour is social and involves others, and risks may therefore be unconsidered, a taken-for-granted aspect of a person`s daily life.

A third approach is to see risk behaviour as a culturally variable product. Here, variations in risk recognition, assessment and response within and between social groups are seen to reflect their differential socialisation in sub-cultures and within complex social institutions. This approach emphasises the point that what society calls ‘risky’ is essentially determined by social and cultural factors. In support of this position we know from qualitative sociological research on the impact of lay health beliefs that the extent to which people feel they have control over their daily lives, including their health, is shaped by position within the social structure. In other words, individual cognitions such as expectations and behaviour have a basis in differential cultural experiences of the social structure and social processes.

For example, the work of Mary Douglas (1992) draws attention to the symbolic nature of boundaries across cultures. The human body is seen as a bounded system in which we regulate what goes ‘in’ and ‘out’ based on cultural rules about what is considered to be ‘food’ and ‘non-food’. Central to these definitions is the binary opposition between ‘purity’ (or ‘clean’) and ‘contamination’ (or ‘dirty’). ‘Contamination’ occurs when boundaries are transgressed; when a substance that is understood to be properly ‘outside’ the body goes inside. Douglas argues these definitions are socially constructed and differ between societies and historical periods. If an individual cannot find a cultural place for a substance as an edible food, then regardless of its nutritional value, it cannot be eaten because it is viewed as dangerous, and takes on the status of ‘risky’.

Douglas argues for a ‘cultural theory of risk’ in order to explain why blame appeared to figure so highly when humans, confronted by some risk, sought to make sense of it. She saw this as reflecting and reinforcing power relations between different social groups within a society. In this sense, risk perception is essentially political, and that modern public debates about risk are about politics (1992;79). 

A fourth possible sociological explanation for risk behaviour focuses on its frequently unconsidered nature, on the understandings that are taken for granted, i.e. unprotected sex cannot kill you, only the elderly get cancer, being involved in construction work inevitably involves some physical danger. This explanation is rooted within phenomenological philosophy and social theory. The work of Schutz (1972) for example, draws a distinction between the social world of routine activities, sharing a set of meanings about everyday events on the one hand, and a world of considered alternatives and calculated action when faced by novel events - two distinct modes of cognition and social action.


Individual and collective responses to symptoms and illness

Responses to symptoms and illness are hugely varied. Edward Suchman (1965) outlined a helpful framework of five stages of experience to explain, in broad terms, why people behave in the way that they do in response to perceived illness. The first phase includes the experience of the symptoms themselves, and an assessment of severity and a judgement regarding care seeking. Thereafter, the individual may assume a “sick role”, may make contact with health services in search for a solution, adopt a dependent “patient” role in the course of treatment, and ultimately experience recovery and rehabilitation.

In sociological research, by some way the greatest attention has been devoted to the first of Suchman’s stages – and in particular motivations for care-seeking behaviour under different conditions (Young, 2004). These can broadly be grouped into the following categories:

  • Structural or functional explanations derive primarily from the work of Talcott Parsons – who viewed illness as behaviourally “deviant” (rather than being related to a physical or psychological condition) in the sense that it disrupts normal behaviour and social relations. Parsons’ theory has been very influential, although it has limitations – including a dominant focus on acute illness.
  • Social psychological perspectives are provided principally by Suchman’s work, which attempts to explain why an individual’s behaviour might vary at different points in their illness journey, and account for the influencing role of the doctor-patient relationship.
  • Economic theory: this approach, after a study by Avedis Donabedian and Leonard Rosenfeld of care-seeking behaviour for pre-natal care in the United States, emphasises the role of the individual’s income and economic agency in their decision on whether or not to seek help from health services.
  • Socio-cultural theory: the two best known works on socio-cultural dimensions of illness behaviour come from Zola (1973) and Mechanic (1985). Zola who argued that people’s approach to illness was governed by a combination of cultural values and health beliefs in response to symptoms. Under most circumstances, people were able to reach a (perhaps temporary) accommodation with their symptoms, but that this accommodation might break down as a result of one or more “triggers” to health seeking behaviour, namely: (1) an interpersonal crisis; (2) perceived interference of symptoms with workplace activity; (3) perceived interference of symptoms with social/leisure activities; (4) sanctioning by others (who may insist that help be sought); or (5) symptoms persisting beyond some arbitrary time deadline set by the individual. Mechanic developed a fuller list.
  • Social network theory: recent work by Pescosolido (1992) shifts focus away from individual choice, and instead emphasises the extent to which decisions to seek care are made in consultation with others within close social networks (of family, friends and broader connections including work colleagues).

Some authors have highlighted the importance of cultural differences in explaining approaches to symptoms. This is a particular feature of the work of Cecil Helman (an anthropologist), but discrete studies elsewhere have examined differing cultural attitudes towards pain and pain management among other symptoms (e.g. Zborowski, 1969).    


                                                             © I Crinson 2007, S Ismail 2017