Research on HIV sexual risk: Social relations-based approach in a cross-cultural perspective

with the collaboration of Philippe Huynen, Michel Hubert and Theo Sandfort
https://doi.org/10.1016/S0277-9536(99)00463-3Get rights and content

Abstract

In opposition to individual-oriented approaches to sexual risk taking, new theoretical approaches have emerged that strive to encompass all of the social relations — that is to say, the connections that structure the components of relations between groups and go beyond simple adaptations to a given interaction — that prevail in a sexual relationship. This paper examines the strengths and weaknesses of such an approach for understanding HIV risk-related behaviour. The analysis is based on data from European surveys of sexual behaviour in the general population, with special attention paid to the data from the Belgian survey, which used a relation-based approach to risk-related behaviour. The analysis shows that sexual behaviour and preventive behaviour patterns of men and women appear to be linked to women’s status in a given society. The data also tend to show that the social and preventive contexts also help structure the relations that develop between partners when it comes to negotiating about using a condom. Finally, the same people behave differently depending on the relationship’s context. In particular, the balance of power within the couple, which is strongly interconnected with gender relations issues, and even characteristics of the former relationship help explain these behaviour differences.

Introduction

The use of individual-oriented approaches and psychologically-based individualistic models of health behaviour have long dominated analyses of behaviour in relation to the risk of the sexual transmission of HIV/AIDS (Aggleton et al., 1994). These approaches are designed to understand how individuals construct their own perceptions of the risk from the epidemiological definition that is used in prevention messages. This domination is partly explained by the fact that social scientists were not greatly involved in developing theories about sexual risk-taking at the start of the AIDS epidemic. Thereafter, other theoretical approaches postulating that risk was not a given that preceded individuals’ social experiences but that social actors constructed their own definitions of risk were developed (Pollak and Schiltz, 1987, Douglas, 1994, Davies and Hickson et al., 1993). Of these new approaches, that proposed by Van Campenhoudt and co-workers (1997) strives to encompass all of the social relations — that is to say, the connections that structure the components of relations between groups or individuals (belonging to a social group: gender, generation, social class, etc.) and go beyond simple adaptations to a given interaction — that prevail in a sexual relationship. This paper examines the merits of such an approach for understanding risk-related behaviour. After a brief review of the strengths and weaknesses of individual-oriented approaches, we shall present the theoretical framework of the relation-based approach. We shall then try to show, through comparative analysis of European data on sexual behaviour and in-depth analysis of data from the Belgian survey, which used a relation-based approach to risk-related behaviour, how social relations shape risk-related sexual behaviour.

In the individual-oriented approach, the decision to protect oneself depends on the individual’s HIV risk perception. The actors’ HIV risk perceptions depend on how they assimilate information about the risk of transmission and behaviour in the face of risk. Protecting oneself is thus seen as an individual, voluntary act, influenced slightly by the social environment. If individuals do expose themselves to the risk and, in this sense, do not act rationally, it is supposedly because they fail to evaluate the risks inherent in a given situation correctly, given that their conceptions act as filters or modifiers of information. Behaviour in the face of risk can therefore be seen as being dependent on the processing of information. Implicit in this type of approach, moreover, is the hypothesis that protecting one’s own health is the overriding priority of any individual. These models actually seek to explain deviations from protective health behaviour. Although other behavioural motivations, such as the fear of being alone or the influence of peer group behaviour, are explored, their roles are interpreted in terms of the health protection rationale.

The individual-oriented approach has undeniably made it possible to identify a number of factors associated with behaviour change (Becker and Joseph, 1988, Coates and Stall et al., 1988, Kelly and Murphy, 1991) or with risk-related behaviour (Moatti et al., 1997). In a review of the international literature Moatti and co-workers (1997) drew up an inventory of the most frequently studied variables associated with increased probability of HIV infection or AIDS-related high-risk behaviours. These were cultural group, socio-economic status, age, alcohol and drug use, knowledge of health guidelines (in the earlier stages of the epidemic), health guidelines’ efficacy, perceived costs of prevention (such as reduced sensation), perceived threat, perceived self-efficacy, and peer support. Whilst the predictive power of some of these variables was not contested, most authors (Ahlemeyer and Ludwig, 1997, Ingham and Van Zessen, 1997, Moatti et al., 1997, Bastard and Cardia-Vonèche, 1997) nevertheless stressed the weak explanatory power of most of the models under consideration. They underscored the limited ability of the analyses to interpret the facts, notably as concerns behaviour changes, and their failure to shed light on the chain of elements leading up to a decision. In particular, these analyses have difficulty accounting for the fact that individuals who change their behaviour due to the risk of AIDS may still occasionally engage in unsafe actions.

Much of the criticism of individual-oriented approaches seemed to address approaches based on economic rationales. True, the individual-oriented approach and economic rationales are associated in some models such as the Health Belief Model. As Van Campenhoudt (1998) has rightly stressed, all the relation-based approaches do not abandon all reference to the economic rationale.

Some authors (Van Campenhoudt, 1998, Ferrand and Snijders, 1997) consider that it is less the postulate of a rational individual assessing the costs and benefits of an action before acting to maximise her/his own interest that is involved, than the idea that the individual can make this assessment alone, in isolation and outside any and all relationships. The criticism levelled at rationality converges with that levelled at individual-oriented approaches.

The collective work by Van Campenhoudt et al. (1997) claims to present ‘new conceptual frameworks that focus on interactions between partners and among social networks’. The interaction- and/or relation-oriented perspectives of sexual behaviour are presented as an emerging scientific field. So, many questions understandably remain unanswered. The basic idea is “that the partners’ interactive behaviours and meanings are the building blocks of the relationship. A relationship-based approach takes into account the partners’ interactive behaviour and meanings rather than the individuals’ own characteristics. By interactive behaviour and meanings we mean what constitutes the relationship and has significance only with respect to the relationship and the other partner. In an interaction-oriented approach these interactive behaviours and meanings are incomprehensible if they are considered separately from each other. They are elements or moments in connected sets of interactions that are called systems or sequences of interactions, depending on the theoretical perspective that is adopted and whether one favours a structural or dynamic explanation. Such a perspective does not necessarily overlook or underestimate contextual elements and individual characteristics; it recomposes them in line with the relationship’s structural or dynamic logic” (Van Campenhoudt, 1997). The authors of Sexual Interactions and HIV Risk. New Conceptual Perspectives in European Research contend that the main dimensions of the relationship between sexual partners that must be taken into account are the situational (or contextual), temporal, emotional, and power dimensions, as well as the meaning of the relationship and the status of risk. Van Campenhoudt (1997) reduced these components of the relationship to three main components, namely, structure, process, and meaning, along classic sociological dividing lines (Berthelot, 1991).

Van Campenhoudt (personal communication) defines the main concepts as follows: ‘By relationship or interaction (in the broad sense of the word) between sexual partners we mean all the communication and action taken in accordance with each other, whether or not the partners are face to face, and that guide, create the framework for and give meaning to their sexual relations with each other. The interaction in the narrow sense, that is, the reciprocal influence that the partners have on each others’ action when they are in each other’s immediate physical presence (Goffman, 1973) is a component of their relationship. Sexual intercourse is part of these interactions’. He suggests talking about a relation-based approach and thus adopting the broader perspective. More specifically, he suggests investigating four levels of the relationship: the institutional and macro-social context, the relationship’s close social context, the interaction between the partners, and the intrapersonal level.

In this paper, we shall use a few examples to discuss the relevance of distinguishing between levels whilst showing how social relations participate in the construction of the social rationales that underpin risk-related behaviour. We shall tackle the discussion of the first level, that is, the institutional and macro-social context in which the relation take place, through an analysis of preventive behaviour according to the gender-specific sexual socialisation process. It will be achieved by analysing the European survey data on sexual behaviour in the general population. In this first analysis, elements of the social context — women’s social status in society — are used as an interpretative framework in order to help us understand risk-related behaviour. The discussion with regard to the second level — the relationship’s close social context — will be tackled through an analysis of preventive behaviour at the start of the relationship according to the circumstances of the encounter. We shall explore the third level — the interaction between the partners — through an analysis of risk management strategies as a function of the ‘balance of power’ within the couple. The fourth level — the intrapersonal level — will be illustrated through an analysis of preventive behaviour after the breaking up of an important relationship. The discussions of the second, third and fourth levels will be based on analyses of the data collected by the Belgian national survey on sexual behaviour and attitudes towards HIV risk.

Section snippets

European surveys on sexual behaviour and HIV risk

In surveys on sexuality as on all other subjects, theoretical and methodological approaches have a considerable impact on the answers obtained (Catania and Gibson et al., 1990, Boulton, 1994). A recent review of national surveys made by Catania et al. (1996) shows clearly that survey comparisons are all the more limited when indicators have not been defined before data collection or at least at the time of their analysis. This paper deals with indicators built a posteriori (representative

Regarding the institutional and macro-social level

The first level is tackled by an analysis of preventive behaviour as regard to sexual socialisation process.

Human behaviour is the result of socialisation processes and differing reports from men and women express the socially determined gender roles that prevail in a given society. To study sexual activity in a population, we shall study — for the sake of an example — multipartnership over the 12 months. Such a choice is due to the fact that it involves a sexual activity variable that is

Individual-oriented approach versus relations-based approach

We began with a brief critical presentation of the individual-oriented approach and an introduction to the relations-based approach, before going on to discuss the four levels of analysing relationships proposed by Van Campenhoudt (personal communication): institutional and macro-social relations, the relationship’s close social context, the interaction between the partners, and the intrapersonal level. One might believe from this that the individual-oriented and relations-based approaches are

Acknowledgements

This article is based primarily on two papers that were presented at the ‘Deuxième Conférence Européenne sur les Méthodes et les Résultats des Recherches en Sciences Sociales sur le Sida: Nouveaux Enjeux pour les Sciences Sociales’ (Paris, 12–15 January 1998): Bajos, N., Hubert, M., Sandfort T., Ways of coping with HIV sexual transmission risks in Europe: the value of a comparison analysis; Marquet, J., Huynen, P., Relation-based approach in research into sexual interactions and AIDS/HIV risk:

References (38)

  • N Bajos et al.

    Sexual risk taking, socio-sexual biographies and sexual interaction: elements of the French national survey on sexual behaviour

    Social Science and Medicine

    (1997)
  • P Aggleton et al.

    Risking everything, risk behavior, behavior change, and AIDS

    Science

    (1994)
  • H.W Ahlemeyer et al.

    Norms of communication and communication as a norm in the intimate social system

  • B Bastard et al.

    From rational individual to actor ensnared in a web of affective and sexual relationships

  • M.H Becker et al.

    AIDS and behavioral change to reduce risk: a review

    Am. J. Public Health

    (1988)
  • J.M Berthelot

    La construction de la sociologie

    (1991)
  • M.F Bochow et al.

    Sexual behaviour of gay and bisexual men in eight European countries

    AIDS Care

    (1994)
  • M Bozon et al.

    Sexual initiation and gender in Europe

  • J.A Catania et al.

    Methodological problems in AIDS behavioral research: influence on measurement error and participation bias in studies of sexual behavior

    Psychol. Bull.

    (1990)
  • J.A Catania et al.

    A review of national AIDS-related behavioural surveys

    AIDS

    (1996)
  • T.J Coates et al.

    Behavioral factors in the spread of HIV infection

    AIDS

    (1988)
  • M Cohen et al.

    The place of time in understanding sexual behaviour and designing HIV/AIDS prevention programs

  • P Davies et al.

    Theorising sex

  • M Douglas

    Risk and Blame Essays in Cultural Theory

    (1994)
  • F Dubois-Arber et al.

    Condom use in Europe

  • A Ferrand et al.

    Social networks and normative tensions

  • Cited by (0)

    View full text