British Medical Bulletin

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (6) Request Permissions Disclaimer Google Scholar Articles by Bonell, C Articles by Imrie, J Search for Related Content PubMed PubMed Citation Articles by Bonell, C Articles by Imrie, J Related Collections HIV/AIDS Social Bookmarking          What's this?

British Medical Bulletin 58:155-170 (2001)
© 2001 The British Council

Behavioural interventions to prevent HIV infection: rapid evolution, increasing rigour, moderate success

C Bonell^* and J Imrie

^* Social Science Research Unit, Institute of Education, University of London, London, UK
Department of Sexually Transmitted Diseases, Royal Free and University College Medical School, University College London, London, UK

	Abstract

Top Abstract Introduction Maximising the likely... Evaluating interventions Implementing effective... Conclusions Acknowledgements References

 
Behavioural interventions aim to alter behaviours that make individuals more vulnerable to becoming infected or infecting others with HIV. Research in this field has developed rapidly in recent years. Increased rigour in the design and conduct of evaluations and moderate successes in bringing about behaviour change in target populations are the key achievements so far. This paper reflects on these developments, addresses recent innovations and highlights likely areas for future work. Discussion focuses on maximising the potential effectiveness of new interventions, methodological issues relating to evaluation and implementation of interventions into practice. The paper concludes there is evidence that interventions deemed effective under evaluation conditions can be implemented in HIV prevention services and that this is the next major challenge. The immediate goal should be consolidation of the learning that has occurred, particularly efforts to maintain theoretical and evaluative rigour whilst encouraging increased collaborative partnerships between researchers, service providers and affected communities.

	Introduction

Top Abstract Introduction Maximising the likely... Evaluating interventions Implementing effective... Conclusions Acknowledgements References

 
The lack of an effective HIV vaccine means prevention through behaviour change is the most important available strategy to reduce new infections. It is well established that behaviour rather than identity determines risk of HIV infection and certain social groups are more vulnerable to infection than others. Effective HIV prevention programmes need to include interventions that target identified groups at increased risk, as well as the general population, with the balance between these being determined by current epidemiology and future projections^1,2.

Behavioural interventions aim to reduce behaviours that make individuals more vulnerable to becoming infected, or infecting others, with HIV. These interventions have generally aimed to increase use of condoms or reduce numbers of partners^3–6. More recently, ‘negotiated safety’ interventions have sought to promote condom protected sex between individuals of different HIV sero-status, acknowledging that partners who know themselves to be of the same status may have unprotected sex, but have suggested this is only an adequate risk reduction strategy if both partners agree to have no unprotected sex with others⁷. Sexual behaviour is not a static phenomenon, but is influenced by many factors, including characteristics of the individual as well as their social and economic context. Thus, while the aim of behavioural interventions is relatively simple, the circumstances in which they operate often necessitates their being complex and multi-dimensional. To date, there has been a huge diversity of interventions, employing methods that variously focus on the individual (e.g. counselling), the community (e.g. community development) and society (e.g. changes in legislation and public policy).

Research into behavioural interventions for HIV prevention has rapidly developed in the course of the last ten or so years. This development is characterized by increasing rigour and moderate success. This paper reflects on developments, addressing recent innovations and highlighting some important areas for future work. The paper is structured around: maximizing the potential effectiveness of new interventions; evaluating interventions; and implementing interventions more widely. The discussion is largely confined to interventions targeting the prevention of HIV, and other poor sexual health outcomes such as sexually transmitted infections (STIs) and unwanted pregnancies in industrialised countries. Although non-industrialised and middle-income countries are hardest hit by the HIV epidemic, the reality of behavioural intervention research so far is that it has been disproportionately situated in the richer countries. This is, however, slowly changing, and there is an emerging literature concerned with behavioural interventions in resource-poor countries.

	Maximising the likely effectiveness of interventions

Top Abstract Introduction Maximising the likely... Evaluating interventions Implementing effective... Conclusions Acknowledgements References

 
Use of theory

There is widespread agreement that the use of theory is central to developing effective behavioural interventions^8–10. Theory in this context refers to either a formal theory of individual or social behaviour (see below), or evidence regarding the social or individual constraints that either promote or inhibit behaviour change¹⁰. Nearly all behavioural interventions for HIV prevention can broadly be classified as complex interventions – that is they consist of separate elements that are combined in order for the intervention to function as intended and to deliver the desired outcome^9,10. People may respond differently to singular components and what constitutes the ‘active ingredient’ may be different for each person. One of the great attractions of theoretical approaches is they can serve as an explicit framework to enable a thorough consideration of what factors must be addressed to bring about the desired behaviour change, and how. At the same time, this framework will guide evaluators' identification of process and outcome measures to be examined.

So far, there is little in the way of clear guidance on deciding which theory is most appropriate, or how best to apply it. Examining the results of reported research is rarely helpful. Reported intervention studies seldom provide explanations for the choice of theory. Equally, they rarely describe how a chosen theory guided development of the intervention or choice of outcome measures¹¹. Furthermore, evidence to support the specific application of many theories is limited and of varying quality. Application of the theory of the social construction of sexuality, for example, has little basis in research¹². The application of other theories, such as the theory of planned behaviour, has been extensively investigated^13,14. Most of these studies have used observational designs; there have been remarkably few attempts to test the theories experimentally¹¹.

Health promotion theory
Health promotion theory, unlike the other theories discussed below, does not attempt to determine how various factors affect health outcome, but rather it attempts to categorise and prescribe health promotion approaches¹⁵. Nevertheless, it is often central in determining what factors interventions target and identifying the most appropriate modalities of delivery. Caplan categorises health promotion in terms of two dimensions¹⁶. The first dimension focuses on whether health promotion aims to affect the individual's actions or to modify the environment (material or social) within which they live. For example, HIV prevention programmes might, on the one hand, try to improve gay men's sexual assertiveness, or on the other, try to encourage the development of a less homophobic society. Caplan's¹⁶ second dimension focuses on the extent to which health promotion is driven by the priorities articulated by affected individuals and communities, or by those of ‘experts’.

Health promotion theory is often prescriptive, encouraging interventions that focus on the societal determinants of health, and activities developed and delivered in partnership with affected communities, to redress a perceived existing bias towards individualistic and ‘top-down’ interventions. This prescriptive element is stressed in documents such as the World Health Organization's Global Strategy of Health for All by the Year 2000¹⁷. In practice, however, most health promotion, including that addressing HIV infection, remains individual-focused and expert-driven¹⁸. To date, some interventions have sought the involvement of affected communities¹⁹, and a few have addressed the environmental determinants of risk^3,20. Recognition of the need for this collaborative approach, particularly in dealing with the most vulnerable groups, is slowly being reflected in the literature^3,4.

Social psychological theories
The growing commitment to using theory in intervention development stems partly from the evidence that nearly all sexual health and HIV prevention programmes that have demonstrated some impact on sexual risk behaviour have drawn on social psychological models^6,8,9,11. These theories relate behaviour to individual cognitions (i.e. thoughts, attitudes or beliefs). If these cognitions can be modified, and there is a demonstrable corresponding shift in behaviour, then there is good reason to target these cognitions in behavioural interventions.

There are a number of different models; these have mostly evolved from the Health Belief Model²¹. Only one of these, the AIDS Risk Reduction Model, has been specifically designed to address behavioural change in relation to HIV²². The rest have been borrowed from other areas of behavioural research and adapted to meet the needs of HIV prevention. There is substantial overlap between the various theories with regard to the cognitions on which they focus. The theories do, however, differ from one another in a number of ways, including their range of application, formal structure, and complexity. Table 1 summarises the main cognitions that these models focus on, theories with which they are associated, and examples from the recent literature of interventions employing them.

View this table: [in this window] [in a new window]   Table 1 Overview of the main social psychological theories used in behavioural interventions for HIV prevention with examples of evaluated behavioural interventions

 
A critique that has emerged in the literature is that social psychological models often overlook the situational factors that might shape sexual behaviour, such as the effects of recreational drugs, and negotiation with one's sexual partner²³. Others have argued that the more sophisticated models do take into account how situational factors might impact at the individual level²⁴. Theoretically-based behavioural interventions should be able to overcome any such difficulties by employing modelling and qualitative testing prior to application¹⁰.

Sociological theories
Sociological theories also offer considerable potential for guiding the design of behavioural interventions. These theories relate the actions of individuals to the societies in which they live. Unfortunately, one weakness of some potentially useful sociological theories is that there is as yet little empirical support for their application in interventions. One example is the social construction of sexual identity¹². This theory suggests that sexual identities develop in the course of social interaction with same-sex peers during adolescence, generally being based on prior, non-sexual gender identities. Where conventional gender roles predominate, interactionists suggest that young women view sexual activity as a service to men, while young men see it as a form of personal achievement, and homosexual activity is viewed as antithetical to conventional gender self-identity¹². Behavioural interventions informed by this theory can encourage people to explore how their understanding of sexuality has been formed predominantly through interaction with their peers, influenced by wider society, so that negative views of one's identity, or stereotypical behaviour can be challenged.

One sociological theory⁴ for which there is some empirical evidence in relation to HIV prevention is the theory of ‘diffusion of innovations’. This theory describes how material or cultural innovations come to be adopted by communities. Through analysis of the take-up of innovations in different fields, Rogers identified factors that influence the rate of adoption²⁵. These include: the characteristics of the social system; the innovation itself; and the characteristics of early and later adopters. According to this theory, early adopters might influence key opinion leaders who in effect become peer-educators within their communities²⁵. One limitation is that the theory assumes innovations remain unchanged as a result of their diffusion²⁵. While this might be true of physical products, it is easy to imagine that cultural products, such as health promotion messages, might become modified and even distorted in the process of diffusion. This model has been used to inform several carefully evaluated peer-education programmes, including at least two targeting gay men^4,26. The contrasting findings of these two studies are discussed below.

Sociological analyses also have much to offer behavioural intervention research in explaining socio-economic and cultural factors that can contribute to a population's vulnerability to infection^27,28. Such theorizing has been an important influence in the targeting of HIV prevention interventions, and in the focus on social and personal empowerment^3,5,27,29.

Use of formative research

Theories can offer general guidance concerning factors addressed by interventions, but there are other aspects they cannot inform. For example, social psychological theory cannot provide guidance on the ideal number and schedule of intervention sessions. Additional research is required to determine which factors need most attention with a specific audience and setting^5,27. ‘Formative research’, often in the form of qualitative inquiry, should occur prior to the development of an intervention. Such research can be analogous to phase I development of interventions in clinical medicine, but also has the added benefit of allowing exploration of acceptability, appropriateness and feasibility of the modes of deployment of an intervention¹⁰.

Despite growing recognition of the important contribution of formative research prior to deployment and evaluation of an intervention, findings from this important research phase are often not widely disseminated in the published literature¹¹. Shain et al report on the formative research they conducted prior to the development and evaluation of an HIV prevention intervention^5,27. Their 18 month study involved 25 focus groups and 102 in-depth interviews with both men and women to collect background information on values and beliefs, behaviours, strategies to motivate behavioural change and barriers to change as well as the logistics of intervention²⁷. They also specify how the background behaviours and beliefs identified, were directly addressed through the intervention²⁷. Subsequently, the researchers demonstrated links between the adoption of risk reduction strategies that targeted these behaviours and a decrease in the acquisition of sexually transmitted infections³⁰. Funders have recognised the value of these research processes^6,10. In future, they will need to demonstrate their commitment to funding formative research processes, which in reality are not optional bolt-ons, but central to the development of successful interventions¹⁰.

	Evaluating interventions

Top Abstract Introduction Maximising the likely... Evaluating interventions Implementing effective... Conclusions Acknowledgements References

 
Rigorous evaluation of effectiveness

While there is little disagreement about the importance of evaluating behavioural interventions, there is still on-going controversy about the most appropriate evaluation methodologies. Evaluation is important to ensure implementation of effective interventions in appropriate settings with appropriate target groups; to stop ineffective or harmful interventions being further deployed^26,31,32; and to determine the cost-effectiveness of interventions³³. In order to conclude that an intervention is effective, we need to establish a significant association between positive outcomes and exposure to the intervention in question. This requires minimising the possibility that such an association reflects the effects of some unacknowledged factor (i.e. confounding), or underlying differences between those who received the intervention and those who do not (i.e. bias).

Many, including ourselves, would argue that experimental evaluations, and randomised controlled trials (RCTs) in particular, provide the best means to achieve these objectives. In an uncontrolled study, if an evaluator detects an improvement in the outcome measures compared with baseline measures this might reflect the effects of the intervention or of some other time-related confounding factor (e.g. secular trends). However, by comparing the outcomes of the intervention group with those of an equivalent control group, the researcher can take account of this confounding and thus make an assessment of the effects of the intervention itself.

Random allocation, and to a lesser extent matching, appear to offer the best means of ensuring that intervention and control groups resemble each other in terms of factors, for example, age and sexual behaviour that, if they varied between the two groups, could affect outcomes and so bias results. Randomisation should distribute both the known and unknown factors that influence outcomes equally between the groups³⁴. It is thus an elegantly simple device for taking account of the multiplicity of social influences on outcomes studied in research³⁵. The problem with matching is that it requires knowing about all the relevant factors that can influence outcome, and this is unlikely to be the case. For this reason, RCTs are generally regarded as the most rigorous evaluation design³⁶.

Increased rigour is a key development in recent behavioural intervention research. The quality of evidence obtained from an evaluation is now likely to be assessed on more than merely random allocation to intervention and control groups. Assessment requires the reporting of baseline characteristics of the intervention and control groups, the mode of randomization, attrition rates in each group at each follow-up, all of the outcomes of both these groups, and sample size calculations in relation to outcomes^11,34.

Acceptance of RCTs in the field of behavioural interventions for HIV prevention is still by no means unanimous^37,38. Critics have suggested that allocation (random or otherwise) of participants to interventions and control groups is impossible in certain interventions. This, they argue is particularly so in interventions where those from affected communities participate in planning and delivery^37,39; interventions delivered via community networks and structures⁴⁰; and interventions addressing the socio-economic or legislative environment³⁹. There are, however, examples of quasi-experimental evaluations of interventions in all these categories, including a group-work intervention developed collaboratively with local voluntary agencies³; peer-education sexual health promotion interventions employing community networks^4,41; and interventions targeting societal determinants of risk among adolescents^20,42.

Another criticism of RCTs of sexual health promotion is that contamination between intervention and control groups will prevent adequate control of confounding^37,43. In other words, there is more chance that health promotion messages, as opposed to clinical interventions, will come to influence control as well as intervention-group participants. One methodological development that seeks to counter problems of contamination is cluster RCTs. In these, whole sites rather than individuals, are allocated to intervention or control, the belief being that intervention and control site participants are then less likely to interact. Examples of clusters used in experimental evaluations of sexual health promotion include: gyms²⁶ and cities⁴¹. However, the possibility exists that an intervention aimed at an intervention cluster could still affect a control cluster, for example because residents of an ‘intervention’ city travel to a ‘control’ city, and have sex or otherwise interact with its residents. There is, however, evidence that well-designed studies can avoid problems of contamination between intervention and control clusters by employing clusters whose residents do not intermingle⁴¹ or by statistically controlling for the effects of contamination⁴⁴.

These counter-arguments regarding allocation and contamination are not meant to suggest that allocation and contamination difficulties never impede or prevent the use of experimental evaluations in the field of sexual health promotion. Certain interventions, especially those addressing socio-economic and legislative determinants of behaviour, may be difficult or impossible to trial^28,34,42. Allocation to control and intervention groups may be impossible for practical or ethical reasons. It would have been highly problematic, for example, to evaluate the HIV prevention effect of removing discrimination in the age of consent for male homosexual intercourse by doing so in some UK ‘clusters’, but not others. Where experimental evaluations genuinely are impossible, other methods, despite providing less clear evidence on effectiveness, must suffice.

Choice of outcome measures

To demonstrate with confidence the effectiveness of behavioural interventions, evaluators need to use outcome measures that reflect the objectives of the intervention and the aims of key stakeholders, including the affected communities. This makes the choice of outcome measures complex. For example, if the aim of an intervention is to reduce sexual risk taking behaviours among young heterosexual women, the choice of outcome measures is not necessarily clear. Should evaluators focus only on measures of behaviour (either the proportion or the actual number of episodes of sex that are deemed to be less risky)? Do indicators of morbidity and other negative sexual health outcomes (e.g. unplanned pregnancy) offer a better measure? The answer is rarely clear-cut and will depend partly on the behaviour of the target population and partly on the aims of the intervention^11,34. The most appropriate outcomes will therefore be specific to the intervention and the evaluation design, but should always be identified as such at the outset^11,34.

In HIV prevention research, the incidence of new HIV infections in intervention and control groups offers the most valid measure of effect of interventions. However, this has almost never been possible in evaluations of single interventions in industrialised country settings because of low incidence. Nevertheless, other biological outcomes may be suitable proxy indicators of reduced risk¹¹. These might, for example, include incident sexually transmitted infections¹¹. However, the relationship between sexual behaviour and sexually transmitted infections is not direct or simple^11,45,46. Studies have shown that specific sexual behaviours cannot predict reductions in infections^32,45,46.

Biological measures can have their own validity problems, concerning definition and ascertainment. Definitions of sexually transmitted infections that are too broad can include infections acquired through less risky sex (e.g. oral sex). Ascertainment of cumulative diagnosed infections by record review at the end of an evaluation will yield more endpoints, but miss asymptomatic infections^11,32. Ascertainment of infections by screening at a given time post-intervention has the advantage of including asymptomatic infections, but requires large numbers to detect differences between trial groups⁴⁷. Bias in measurement of cumulative diagnosed infections is more likely if the intervention changes the probability that individuals will attend clinics. Another potential bias is that of missing infections diagnosed at other clinics³². Within the context of an RCT, intervention group participants may be more likely than controls to ‘hide’ an infection acquired during the study from researchers by having it treated elsewhere¹¹.

The emphasis on biological endpoints here is not intended to undermine the importance of behavioural measures. However, empirical research has questioned the validity and the accuracy of self-reported sexual behaviour^11,32,48,49. Use of new technologies, such as computer-assisted self-interview (CASI), instead of traditional face-to-face interviews, are an important methodological advance that may go some way toward addressing the underlining problem⁵⁰. However, unblinded randomised evaluations of behavioural interventions will continue to be vulnerable to biased self-reporting. Even with careful explanation of equipoise, it is difficult to exclude the possibility that people who know they are receiving a certain intervention will feel obliged to under-report risk behaviours more than people who know they are in a control group. Stephenson et al's review of rigorous behavioural intervention trials found that all reported some improvement in one or more self-reported behavioural outcomes, but only two interventions reported declines in sexually transmitted infection rates¹¹. It cannot, therefore, be assumed that interventions having a beneficial effect on reported sexual behaviour will have a similar effect on sexually acquired infections^11,32.

The question of which measures constitute the most appropriate outcomes in a given evaluation is unlikely to be resolved quickly, and continued research into the complex epidemiological relationship between behaviours and sexually transmitted infections is needed^45,46,48. However, as suggested at the outset, the key is that outcomes should be realistic and reflect the aims of the intervention. Where a complex intervention aims to address various determinants of risk in order to reduce HIV transmission, it is perfectly reasonable to measure biological outcomes. Failure to measure biological measures means that, as evaluators, we will have no means to know whether ultimately we have affected them¹¹. However, more modest interventions, such as educational projects, may not in isolation hope to influence biological end-points and so including biological outcome measures might be unrealistic.

A number of behavioural interventions have been reported as effective in increasing rates of condom use, reducing numbers of partners and reducing incidence of STIs within a population^3–5,51. However, the outcomes of ‘negotiated safety’ interventions remain unevaluated. While the effects of these interventions have not been as dramatic as the reported effects of STI treatment programmes in reducing HIV infection⁵², behavioural and STI treatment approaches to HIV prevention are complementary.

Process evaluation

All outcome evaluations, whether they employ control groups or not, should be conducted alongside process evaluations so that effective interventions can be carefully described, the processes by which an intervention exerts effects (or fails to do so) can be explored, and the context within which an intervention will and will not be effective can be examined⁵³. A recent example of such a study is a controlled trial of peer education for gay men in London, which had no apparent impact on HIV risk behaviours²⁶. Because the study had an integral process evaluation, the researchers were able to identify possible explanations for intervention ‘failure’, such as the unacceptability of the intervention to the peer educators who were asked to provide the it²⁶.

	Implementing effective behavioural interventions in HIV prevention practice

Top Abstract Introduction Maximising the likely... Evaluating interventions Implementing effective... Conclusions Acknowledgements References

 
The overarching goal of behavioural intervention research is to ensure interventions, particularly those for which there is good evidence of effectiveness, are implemented widely in HIV prevention services. A number of interventions have been identified as ‘effective’ and ready for implementation⁶, yet service providers have been slow to take up these interventions and incorporate them into their services⁵⁴. This is an area where we feel there is likely to be intensified interest in the future⁵⁵. This section examines some of the obstacles to widespread implementation of evidence-based interventions and some of the potential responses.

One of the principal obstacles to wider implementation of interventions may be the degree to which they are targeted and developed for a specific audience⁵. As discussed above, theories used to develop interventions may not be directly transferred from one cultural setting to another. Kelly et al's application of the theory of diffusion of innovations with gay men in the US was demonstrably effective in reducing risk-taking behaviour, while Elford et al's application in the UK was not^4,26. A recent review of the most rigorously evaluated interventions concluded that an explicit understanding of behaviours, beliefs and risk perceptions seems to be key to developing successful interventions and that the generalisability of interventions may be determined by contextual specificities indicated in the formative research stage¹¹. If true, this suggests that an intervention shown to be effective with one target group, cannot be applied with another, or in different setting, with the same anticipated benefit without some new formative research to assess its appropriateness and viability. One strategy that avoids this is to develop interventions, informed by theory and formative research, and to conduct their subsequent evaluation, across diverse populations and in different geographical and possibly, cultural settings^47,51,56. However, evaluations on this scale are costly, require major personnel resources and are rarely financially or practically feasible for a single intervention.

A second obstacle, critics of randomised trials argue, is that the generalisability of interventions evaluated using trials is especially questionable because of the unusual level of control and care exercised by those conducting the evaluation³⁷. Rather than supporting arguments against trials, this suggests that, in order to maximise the likely generalisability of an intervention, similar development processes should be used for interventions that will undergo a trial evaluation as for those that would not. In other words, we should involve affected communities as well as the intended providers of behavioural interventions in the planning and undertaking of intervention studies from the outset⁵⁵. The involvement of affected communities and providers in the development and evaluation of interventions is not inimical either to the use of theory in developing interventions or the employment of rigorous methods of evaluation⁵³.

Finally, at the level of technology transfer, there are structural obstacles to the implementation of evidence-based interventions. Somlai et al's findings from a survey of community-based prevention organisations in the US showed that few offered workshops on risk self-appraisal and risk-reduction skills even though there was evidence to suggest these approaches were effective with some groups⁵⁷. In the US, a number of programmes are now in place to support and enhance technology transfer between researchers and prevention providers^6,55. There is also a very small but growing literature that provides the descriptions and evidence for the effectiveness of such work at the level of the HIV prevention providers^55,58.

	Conclusions

Top Abstract Introduction Maximising the likely... Evaluating interventions Implementing effective... Conclusions Acknowledgements References

 
Development, evaluation and implementation of behavioural interventions has progressed rapidly in the last decade or so. Yet most critical reviews conclude that such interventions have only been modestly successful in reducing HIV transmission risk¹¹. However, this should not be a cause for despondency. The necessary rapidity of such developments, generally with limited funding, means that evidence of success should be celebrated. Moreover, the evidence that with appropriate support interventions demonstrated effective under evaluation conditions can be implemented into HIV prevention services suggests that there is real potential for success. What is now required is a consolidation of all the learning that has occurred, and in particular an effort to maintain and further develop theoretical and evaluative rigour, whilst encouraging a participative ethos and multidimensional focus within HIV prevention behavioural interventions.

	Acknowledgements

Top Abstract Introduction Maximising the likely... Evaluating interventions Implementing effective... Conclusions Acknowledgements References

 
We would like to thank colleagues in the Department of Sexually Transmitted Diseases and the Social Science Research Unit for helpful comments on preliminary drafts of this paper. Our thoughts on psychological and sociological theory have been informed and stimulated by those of Dr Daniel Wight of the MRC Social and Public Health Sciences Unit, Glasgow University, Scotland.

	Footnotes

 
Correspondence to: John Imrie, Department of Sexually Transmitted Diseases, Mortimer Market Centre, Mortimer Market, off Capper Street, London WC1E 6AU, UK

	References

Top Abstract Introduction Maximising the likely... Evaluating interventions Implementing effective... Conclusions Acknowledgements References

 

1. Anderson R. Prevention works. Plenary paper presented at the XIIIth International AIDS Conference, Durban, South Africa. 9–14 July 2000
2. Wasserheit JN, Aral SO. Dynamic topology of sexually transmitted disease epidemics: implications for prevention strategies. J Infect Dis 1996; 174 (Suppl 2): S201–13
3. Kegeles SM, Hays RB, Coates TJ. The Mpowerment project: a community-level HIV prevention intervention for young gay men. Am J Public Health 1996; 86: 1129–36[ISI][Medline]
4. Kelly JA, Murphy D, Sikkema K et al. Outcomes of a randomized controlled community-level HIV prevention intervention: effects on behaviour amongst at-risk gay men in small US cities. Lancet 1997; 350: 1500–5[ISI][Medline]
5. Shain RN, Piper JM, Newton ER et al. A randomised, controlled trial of a behavioral intervention to prevent sexually transmitted disease among minority women. N Engl J Med 1999; 340: 93–100[Abstract/Free Full Text]
6. CDC. HIV/AIDS Prevention Research Synthesis Project, Compendium of HIV Prevention Interventions with Evidence of Effectiveness. Atlanta, GA: Centers for Disease Control and Prevention, 1999
7. Dowsett G. Practising Desire: Homosexual Sex in the Era of AIDS. Stanford, CA: Stanford University, 1996
8. NIH. Interventions to Prevent HIV Risk Behaviors. NIH Consensus Statement Online 1997 Feb 11–13 [cited 2001, June, 5]; 15(2): 1-41. http://odp.od.nih.gov/consensus/cons/104/104_statement.htm
9. King R. Sexual Behavioural Change for HIV: Where Have Theories Taken Us? Geneva: UNAIDS, 1999
10. Campbell M, Fitzpatrick R, Haines A et al. Framework for design and evaluation of complex interventions to improve health. BMJ 2000; 321: 694–6[Free Full Text]
11. Stephenson J, Imrie J, Sutton S. Rigourous trials of sexual behaviour interventions in STI/HIV prevention: what can we learn from them? AIDS 2000; 14 (Suppl 3): S115–24
12. Gagnon JH, Simon W. Sexual Conduct: The Social Sources of Human Sexuality. London: Hutchinson, 1974
13. Ajzen I. The theory of planned behavior. Organizational Behavior Hum Decision Processes 1991; 50: 179–211
14. Sutton S. Using the theories of reasoned action and planned behaviour to develop health behaviour interventions: Problems and assumption. In: Rutter DR, Quine L. (eds) Changing Health Behaviour: Research and Practice with Social Cognition Models. Buckingham: Open University Press, 2001 (In press)
15. Rawson D. The growth of health promotion theory and its rational reconstruction: lessons from the philosophy of science. In: Bunton R, MacDonald G. (eds) Health Promotion: Disciplines and Diversity. London: Routledge, 1992; 202–24
16. Caplan R. The importance of social theory for health promotion: from discussion to reflexivity. Health Promotion Int 1993; 8: 147–57[Abstract/Free Full Text]
17. World Health Organization. Global Strategy of Health for All by the Year 2000. Copenhagen: WHO Regional Office for Europe, 1981
18. Parish R. Health promotion: rhetoric and reality. In: Bunton R, Nettleton S, Burrows R. (eds) The Sociology of Health Promotion. London: Routledge, 1995; 13–23
19. Altman D. Expertise, legitimacy and the centrality of community. In: Aggleton P, Davies P, Hart G. (eds) AIDS: Facing the Second Decade. London: Falmer, 1993; 1–12
20. Olsen RJ, Farkas G. Employment opportunity can decrease adolescent childbearing within the underclass. Eval Program Planning 1991; 14: 27–34
21. Becker MH. The health belief model and personal health behaviour. Health Educ Monographs 1974; 2: 324–508
22. Catania JA, Kegeles SM, Coates TJ. Towards an understanding of risk behavior: an AIDS risk reduction model (ARRM). Health Educ Q 1991; 17: 53–72
23. Davies P. Acts, sessions and individuals: a model for analyzing sexual behaviour. In: Boulton M. (ed) Challenge and Innovation. Methodological Advances in Social Research on HIV/AIDS. London: Taylor & Francis, 1994; 57–68
24. Abraham C, Sherran P. Modelling and modifying young heterosexuals' HIV-preventive behaviour; a review of theories, findings and educational implications. Patient Educ Counsel 1994; 23: 173–86[ISI][Medline]
25. Rogers EM. The Diffusion of Innovations. New York, NY: Free Press, 1983
26. Elford J, Bolding G, Sherr L. Peer education has no significant impact on HIV risk behaviours among gay men in London. AIDS 2000; 15: 535–8
27. Ramos R, Shain R, Johnson L. ‘Men I mess with don't have anything to do with AIDS’: using ethno-theory to understand sexual risk perception. Sociol Q 1995; 36: 483–504[ISI]
28. Parker RG, Easton D, Klein CH. Structural barriers and facilitators in HIV prevention: a review of international literature. AIDS 2000; 14 (Suppl 1): S22–32
29. Homans H, Aggleton P. Health education, HIV infection and AIDS. In: Aggleton P, Homans H. (eds) Social Aspects of AIDS. Lewes: Falmer, 1988; 154–76
30. Shain RN, Perdue S, Piper JM et al. Sexual risk reduction behaviors amenable to intervention: The importance of context. Conference Proceedings, 13th Meeting of the International Society for Sexually Transmitted Disease Research, 11–14 July 1999. Denver CO, USA. Abstract 001
31. Christopher FS, Roosa MW. An evaluation of adolescent pregnancy prevention program: is ‘just say no’ enough? Fam Relat 1990; 39: 68–72
32. Imrie J, Stephenson JM, Cowan FM et al. A cognitive behavioural intervention to reduce sexually transmitted infections among gay men: randomised trial. BMJ 2001; 322: 1451–6[Abstract/Free Full Text]
33. Holtgrave DR, Kelly JA. Preventing HIV/AIDS among high risk urban women: the cost effectiveness of behavioral group intervention. Am J Public Health 1996; 86: 1442–5[Abstract/Free Full Text]
34. Stephenson J, Imrie J. Why do we need randomised controlled trials to assess behavioural interventions? BMJ 1998; 316: 611–3[Free Full Text]
35. Oakley A. Who's afraid of the randomised controlled trial? In: Roberts H. (ed) Women's Health Counts. London: Routledge, 1990; 167–94
36. Chalmers I, Enkin M, Keirse MJNC. Effective Care in Pregnancy and Childbirth. Oxford: Oxford University, 1990
37. Kippax S, van de Ven P. An epidemic of orthodoxy? Design and methodology in the evaluation of the effectiveness of HIV health promotion. Crit Public Health 1998; 8: 371–86
38. van de Ven P, Aggleton P. What constitutes evidence in HIV/AIDS education? Health Educ Res 1999; 14: 461–71[Abstract/Free Full Text]
39. Nutbeam D. Oakley's case for using randomised controlled trials is misleading. BMJ 1999; 318: 944–5[Free Full Text]
40. Downie RS, Tannahill C, Tannahill A. Health Promotion: Models and Values. Oxford: Oxford University, 1996
41. Williamson LM, Hart GJ, Flowers P, Frankis JS, Der GJ. The gay men's task force: the impact of peer-education on the sexual health behaviour of gay men in Glasgow. Sex Transm Infect 2001; In press
42. Rotheram-Borus MJ. Expanding the range of interventions to reduce HIV among adolescents. AIDS 2000; 14 (Suppl 1): S33–40
43. Webb D. Measuring Effectiveness in Health Promotion. Southampton: University of Southampton, 1997
44. Torgerson DJ. Contamination in trials: is cluster randomisation the answer? BMJ 2001; 322: 355–7[Free Full Text]
45. Peterman TA, Lin LS, Newman DR et al. Does measured behavior reflect STD risk? An analysis of data from a randomized controlled behavioral intervention study. Sex Transm Dis 2000; 27: 446–51[ISI][Medline]
46. Fishbein M, Jarvis B. Failure to find a behavioral surrogate for STD incidence – what does it really mean? Sex Transm Dis 2000; 27: 452–5[ISI][Medline]
47. National Institute of Mental Health Multisite HIV Prevention Trial Group. The NIMH multisite HIV prevention trial: reducing HIV sexual risk behaviour. Science 1998; 280: 1889–94[Abstract/Free Full Text]
48. Aral SO, Peterman TA. Do we know the effectiveness of behavioural interventions? Lancet 1998; 351 (Suppl 3): 33–6[ISI][Medline]
49. Zenilman JM, Weisman CS, Rompalo AM et al. Condom use to prevent incident STDs: the validity of self-reported condom use. Sex Transm Dis 1995; 22: 15–21[ISI][Medline]
50. McQueen D, Campostrini S. Monitoring behavioural change in the population: a continuous data collection approach. In: Boulton M. (ed) Challenge & Innovation. Methodological Advances in Social Research on HIV/AIDS. London, Taylor & Francis, 1994; 39–56
51. Kamb ML, Fishbein M, Douglas JM et al. Efficacy of risk-reduction counselling to prevent human immunodeficiency virus and sexually transmitted diseases: a randomized control trial. JAMA 1998; 280: 1161–7[Abstract/Free Full Text]
52. Hayes R, Nicholl A, Grosskurth H et al. A community trial of the impact of improved sexually transmitted disease treatment on the HIV epidemic in rural Tanzania. AIDS 1995; 9: 1–26[ISI][Medline]
53. Peersman G, Harden A, Oliver S. Effectiveness of Health Promotion Interventions in the Workplace: A Review. London: Health Education Authority, 1998
54. DiFranceisco W, Kelly JA, Otto-Salaj L et al. Factors influencing attitudes with AIDS service organizations toward the use of research-based HIV prevention interventions. AIDS Educ Prev 1999; 11: 72–86[ISI][Medline]
55. Anon. Turning HIV prevention research into practice (special supplement). AIDS Educ Prev 2000; 12 (Suppl A): 1–145
56. CDC AIDS Community Demonstration Projects Research Group. Community-level HIV intervention in five cities: final outcome data from the CDC AIDS Community Demonstration Projects. Am J Public Health 1999; 89: 336–45[Abstract/Free Full Text]
57. Somlai AM, Kelly JA, Otto-Salaj L et al. Current HIV prevention activities for women and gay men among 77 ASOs. J Public Health Manage Pract 1999; 5: 23–33
58. Kelly JA, Somlai AM, DiFranceisco W et al. Bridging the gap between the science and service of HIV prevention: transferring effective research-based HIV prevention interventions to community AIDS service providers. Am J Public Health 2000; 90: 1082–8[Abstract/Free Full Text]
59. Ford K, Wirawan D, Fajans P et al. Behavioral interventions for reduction of sexually transmitted disease/HIV transmission among female commercial sex workers and clients in Bali, Indonesia. AIDS 1996; 10: 213–22[ISI][Medline]
60. Jemmott JB, Jemmott LS, Fong GT. Reductions in HIV risk-associated sexual behaviors among black male adolescents: effects of an AIDS prevention intervention. Am J Public Health 1992; 82: 372–7[Abstract/Free Full Text]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				D J Ward, B Rowe, H Pattison, R S Taylor, and K W Radcliffe Reducing the risk of sexually transmitted infections in genitourinary medicine clinic patients: a systematic review and meta-analysis of behavioural interventions Sex. Transm. Inf., October 1, 2005; 81(5): 386 - 393. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (6) Request Permissions Disclaimer Google Scholar Articles by Bonell, C Articles by Imrie, J Search for Related Content PubMed PubMed Citation Articles by Bonell, C Articles by Imrie, J Related Collections HIV/AIDS Social Bookmarking          What's this?

Online ISSN 1471-8391 - Print ISSN 0007-1420

Copyright © 2008 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics