Volume 6 - Number 2 1997
Special Issue: STDs and Sexual/Reproductive Health

Published by SIECCAN
The Sex Information & Education Council of Canada

A Theory-Based Framework for Intervention and Evaluation in STD/HIV Prevention

William A. Fisher
Department of Psychology
Department of Obstetrics & Gynaecology
University of Western Ontario
London, Ontario

Abstract:

This paper examines the fundamental role of social science theory as a basis for the construction of STD/HIV prevention interventions and as a basis for the evaluation of such interventions. Theoretically based approaches to STD/HIV prevention have the potential to specify critical factors to target for change, to suggest methods for effecting change, and to guide us in evaluating STD/HIV prevention intervention process and outcome. Examples of theory-based approaches to the creation and evaluation of STD/HIV prevention interventions involving the Information-Motivation-Behavioural Skills Model, the Theory of Reasoned Action/Theory of Planned Behaviour, and Social Network Theory are presented and discussed.

Keywords:

• HIV prevention, STD prevention , Theory-based interventions, Elicitation research, Evaluation strategies

Acknowledgements:

Work on this manuscript was supported by research grants from Health Canada, the U.S. National Institutes of Health, and Janssen-Ortho Inc.

Correspondence concerning this paper should be directed to William A. Fisher, Ph.D., Department of Psychology and Department of Obstetrics and Gynaecology, Social Science Centre, University of Western Ontario, London, Ontario, Canada, N6A 5C2. email: FISHER@SSCL.UWO.CA.

The Role of Social Science Theory in STD/HIV Prevention

Social science theory is of fundamental importance to STD/HIV prevention because it specifies probable determinants of STD/HIV risk and preventive behaviour. As such, theory can designate factors that should be targeted for change in intervention attempts, and by extension, factors that should be assessed in evaluation research efforts to determine intervention efficacy. In the following sections, theoretical accounts of the determinants of STD/HIV preventive behaviour are outlined and used as a basis for discussing the construction of targeted prevention interventions and the creation of evaluation research strategies appropriate to such interventions.

An Information-Motivation-Behavioral Skills Approach to STD/HIV Prevention

The Information-Motivation-Behavioral Skills model of STD/HIV preventive behaviour (J. Fisher & Fisher, 1992; W. Fisher & Fisher, 1993) proposes that information that is directly relevant to the personal practice of preventive behaviour, motivation to practice prevention, and behavioural skills for practising prevention effectively, are the fundamental determinants of STD/HIV preventive behaviour. The theory suggests that at present, most persons at risk possess inadequate information about the personal practice of preventive behaviour, insufficient personal and social motivation to practice prevention, and inadequate behavioural skills for practising prevention effectively, and that prevailing high levels of STD/HIV risk are the result of this situation. On the basis of this analysis, the theory designates information, motivation, and behavioural skills as critical factors to target for change in intervention efforts to promote preventive behaviour. Evaluation research should therefore monitor these factors, as well as levels of preventive behaviour per se both before and after intervention.

A Theory of Reasoned Action--Theory of Planned Behaviour Approach to STD/HIV Prevention

The Theory of Reasoned Action (Ajzen & Fishbein, 1980; Fishbein & Middlestadt, 1989) and the Theory of Planned Behaviour (Ajzen & Madden, 1986, Basen-Enquist & Parcel, 1992) assert that attitudes towards the personal practice of preventive behaviours, social norms concerning the performance of these behaviours, and perceptions of personal control or ability to successfully enact these behaviours, are critical determinants of STD/HIV prevention. According to these theories, an individual who has positive attitudes about abstaining from vaginal or anal intercourse, or about always using condoms during vaginal or anal intercourse, and who perceives social support for these preventive behaviours from key referent others, and who has the conviction that s/he can carry out these behaviours effectively, will be likely to undertake these STD/HIV peventive actions. By extension, according to these theories, it would be crucial to intervene to try to improve a target populations attitudes towards the personal practice of STD/HIV preventive behaviours, to change perceptions of social support for such preventive behaviours, and to alter individuals perceptions that they can effectively carry out these preventive behaviours, in intervention attempts to promote STD/HIV prevention. This theoretical analysis would also suggest the monitoring of pre- and post-intervention attitudes, social norms, and perceptions of control concerning STD/HIV preventive behaviours, as well as levels of STD/HIV preventive behaviours, as key elements of an evaluation research strategy to determine if and how intervention efforts have an impact on STD/HIV preventive behaviour.

A Social Network Theory Approach to STD/HIV Prevention

A social network theory approach to STD/HIV prevention (J. Fisher, 1988; J. Fisher & Misovich, 1990; Kelly, St. Lawrence, Brasfield, Stevenson, & Haugh, 1991; Kelly, St. Lawrence, Stevenson, & Haugh 1992) suggests that individuals function within social networks that establish norms for behaviour, including safer sexual behaviour, and that these social networks enforce adherence to these norms. Social networks may establish and enforce norms that are relevant to safer sexual behaviour by supplying information that is supportive of a particular safer or risky norm, or by directly socially reinforcing or stigmatizing safer or risky norms and behaviours that are related to these norms. According to social network theory, natural opinion leaders -- social network members who are respected widely and who have the power to establish and enforce social norms -- have a critically important role to play in changing STD/HIV preventive behaviours. On the basis of this conceptualization, STD/HIV prevention interventions must address existing social norms regarding safer and risky sexual behaviour in efforts to change such behaviour. If risk is normative and socially rewarded within a social network, safer sexual behaviour interventions will fail unless the norms promulgated by natural opinion leaders and enforced by the social network are modified. If, however, risk is inconsistent with social network norms and is stigmatized by the social network, the social network may be used to gain behaviour change leverage in STD/HIV prevention interventions. A social network analysis would suggest that STD/HIV prevention interventions seek to change social norms that are antagonistic to safer sexual behaviour, or capitalize on existing social network norms that are favourable to preventive behaviour, by enlisting the aid of natural opinion leaders to change or enforce social norms that are relevant to STD/HIV risk and prevention. The social network approach would also suggest that evaluation research associated with such STD/HIV prevention interventions should focus on pre- and post-intervention assessment of natural opinion leaders views and actions with respect to risky and safer sex within their social network, on the perceptions of members of the social network of social norms in this regard within their group, and on levels of STD/HIV preventive behaviour.

Creation of Theory-Based Intervention and Evaluation Strategies

The creation of theory-based intervention and evaluation strategies may be thought of as comprising a multi-level STD/HIV prevention process that is presented in Figure 1.

As can be seen in the figure, elicitation research is conducted with a sample of a target population to assess pre-intervention levels of factors designated by theory as determinants of STD/HIV risk and preventive behaviour. Next, targeted, group-specific interventions are constructed on the basis of elicitation research findings to address deficiencies in the factors that determine STD/HIV preventive behaviour, and to capitalize on assets that may exist within a groupin regard to such factors. Finally, evaluation research is conducted to determine intervention impact on the factors that are presumed to influence STD/HIV preventive behaviour and to assess intervention impact on STD/HIV preventive behaviour, in order to gauge intervention success and to provide an empirical basis for refining and improving future versions of the intervention.

Following the elicitation--intervention--evaluation strategy, if one were applying the Information--Motivation--Behavioural Skills approach to STD/HIV prevention within a particular population, one would conduct elicitation research to assess the level of that populations information about preventive behaviour, their motivation to practice prevention (including attitudes about preventive behaviours, social norms regarding prevention, and perceptions of vulnerability to infection), and their behavioural skills for the practice of key preventive behaviours such as negotiating abstinence from vaginal or anal intercourse, or insisting on condom use, and their most important areas of risk behaviour exposure. Next, on the basis of elicitation research findings, one would create a targeted, group-specific intervention to address information, motivation, and behavioural skills deficits in this population, and to capitalize on existing strengths in these areas, with respect to the populations most frequent or most dangerous risky behaviours. Finally, one would conduct evaluation research that focused on assessment of pre- to post-intervention gains in information, motivation, and behavioural skills, and on corresponding pre- to post-intervention gains in STD/HIV preventive behaviour. (See W. Fisher & Fisher, 1993, and J. Fisher et al., 1996, for detailed discussion of these elicitation, intervention, and evaluation research procedures; see Wasserheit, Aral, Holmes, & Hitchcock, 1991, for a general discussion of research issues in human behaviour and sexually transmitted disease, and Kirby, 1992, for a discussion of evaluation research issues in this area).

If one adopted the Theory of Reasoned Action--Theory of Planned Behaviour approach to STD/HIV prevention, elicitation research would focus on the assessment of a target populations existing attitudes towards the personal practice of preventive behaviours, on their perceptions of social norms or social support for the practice of these preventive behaviours, and on their perception of personal control or ability to carry out these preventive behaviours successfully. Targeted, group-specific intervention activities would then be constructed to focus on the modification of attitudes, social norms, and perceptions of control of preventive behaviour that appeared to be deficient. Taking its cue from theory, evaluation research would focus on the assessment of pre- to post-intervention changes in attitudes, norms, and perceptions of control regarding STD/HIV preventive behaviour, and on corresponding changes in STD/HIV preventive behaviour. (For a detailed discussion of such theory-based intervention activities, see Ajzen & Fishbein, 1980, and Fishbein and Middlestadt, 1989).

Finally, if one adopted the social network approach to promoting STD/HIV preventive behaviour, elicitation research would focus on a target populations existing social norms concerning STD/HIV risk and preventive behaviour, and on the identification of natural opinion leaders who would be in a position to change or to enforce social network norms regarding STD/HIV risk and prevention. A targeted, group-specific intervention, based on the theory and guided by elicitation findings, would seek to recruit natural opinion leaders, to train them in STD/HIV prevention information and influence techniques, and to contract with them to engage in STD/HIV prevention contacts, involving change or activation of social norms that are relevant to risky and safer sex, with members of their social networks. Evaluation research yoked to these activities would focus on verifying that natural opinion leaders have actually had prevention contacts with social network members, on assessing pre- to post-intervention changes in social network members perceptions of the acceptability and desirability of STD/HIV preventive behaviour, and on corresponing changes in STD/HIV preventive behaviour. (For a detailed discussion of these intervention and evaluation procedures, see Kelly et al., 1991, 1992).

Evaluation Research Issues

Although theory specifies factors that should be assessed in intervention and evaluation research, there are a number of specific issues in the conduct of evaluation research that cut across theoretical approaches and that must be carefully addressed in evaluating intervention-induced changes in STD/HIV preventive behaviour.

Design of Evaluation Research

At its most basic level, evaluation research seeks to assess whether an intervention has caused change in pre- to post-intervention levels of STD/HIV preventive behaviour. While simply conducting pre- and post-intervention assessments would seem to be a straightforward means for assessing intervention-induced changes in STD/HIV preventive behaviour, it is critical to note that pre- and post-intervention assessments alone cannot document that an intervention has caused any particular change. In addition to the fact that an STD/HIV prevention intervention has occurred between pre- and post-testing, it must be recognized that other events in the surrounding environment are occurring as well, and that some of them, rather than the intervention itself, may actually be responsible for pre- to post-intervention shifts in STD/HIV preventive behaviour. Although we may have conducted a pre-intervention assessment, delivered an intervention, and conducted a post-intervention assessment, we would have no assurance that other events in the environment (for example, Magic Johnson's announcement that he is HIV+, or Newsweeks announcement of "The end of AIDS") were not actually responsible for improvement or deterioration in STD/HIV preventive behaviour that we may have observed. To control for such possibilities, pre- and post-intervention assessments must be conducted with individuals who have received the prevention intervention, as well as with a control group of individuals who have not. If pre- to post-intervention change occurs uniquely among those who have received the intervention and no change occurs among those who have not received the intervention, we can have confidence that it is our intervention which is responsible for observed improvements in STD/HIV preventive behaviour, and not extraneous events in the surrounding environment.

Multiple Convergent Measures

Evaluation research estimates of intervention impact will be improved to the extent that there are multiple convergent measures of change in the theorized determinants of STD/HIV preventive behaviour and multiple convergent measures of change in STD/HIV preventive behaviour. Such multiple convergent assessments of change should involve both self-reports of change and objective observations of change. Further, assessment of intervention-induced change should be conducted, at least in part, in a fashion that appears to be independent of the intervention itself, so that intervention participants, seeing the link between the well-intentioned intervenors and their evaluation research, do not simply attempt to be kind to the intervenors and to provide evidence, however spurious, that the intervention has worked. An example of a self-report assessment of intervention impact that appears to be unconnected with an STD/HIV prevention intervention might involve a "wellness survey" assessing a variety of wellness behaviours (e.g., seatbelt use, dietary intake, and condom use), administered in a mall or in a high school or in a teen clinic by personnel not identified with the intervention. An example of objective observations of intervention impact, also apparently unconnected with an STD/HIV prevention intervention, might involve the monitoring of "sentinel" condom machines (how many condoms were purchased pre- and post-intervention in high school condom machines in intervention versus non-intervention high schools?), "sentinel" drugstores (how many condoms were purchased during sample after school hour periods before and after an intervention at intervention and non-intervention sites?), "sentinel" measures of persons presenting for STD treatment at a local clinic, assessments of STD/HIV prevention questions raised by teens in visits to "sentinel" family doctors in intervention and nonintervention sites, etc. Note that while STD/HIV prevention interventions may sometimes appear to initially inflate measures of STD/HIV risk (e.g., an intervention may paradoxically produce an immediate increase in STD diagnoses among teens at a local clinic because of their increased knowledge about signs and symptoms of STD and because of the increased social acceptability of seeking STD- related care that the intervention may have induced), such assessments should show more positive long-term effects. Finally, with a view toward the conservation of financial and personnel resources, it is not always necessary to assess each and every participant in an intervention or each and every person at a non-intervention comparison site. Provided that the number of persons assessed is sufficient to make valid inferences about the sample as a whole, a randomly selected subsample can be used for evaluation purposes. For example, it may only be necessary for every second or fourth study participant to complete a pre- or post-intervention wellness questionnaire.

Process and Outcome Evaluation

It is critical to assess the process by which an intervention may work, as well as the outcome of the intervention. Within the current theory-based approach, this implies the necessity of evaluating intervention-induced changes in factors that theory designates as determinants of STD/HIV preventive behaviour (e.g., information, motivation, behavioural skills; attitudes and social norms; social network prescriptions and proscriptions), as well as the outcome of such changes (e.g., increases in STD/HIV preventive behaviour, decreases in incidence of STD/HIV infection). In this fashion, we can clarify whether our intervention is having a significant impact on each theoretically designated determinant of STD/HIV preventive behaviour, and we can strengthen our interventions if necessary to bolster effects on factors that have not been sufficiently affected. Moreover, we can begin to examine whether change in one particular "process" factor (e.g., information or motivation or behavioural skills; attitudes or social norms) may have the most impact in creating changes in important STD/HIV prevention outcomes. Such information can be particularly helpful if we choose to "downsize" our intervention so as to focus on its most influential active ingredient, expending only those resources necessary to produce desired improvement in STD/HIV preventive behaviour.

Assessing the Export Potential of Interventions

It is critical to assess the degree to which interventions to change STD/HIV preventive behaviour may be easily exported and deployed in the diverse settings in which such interventions are needed. An example of a highly exportable intervention would be one that is easy to deploy, is cost effective, is accompanied by a manual, readily adopted by variably motivated personnel already in the field, and possible to implement within existing time, financial, and institutional constraints. An example of a difficult-to-export intervention would involve one which is expensive, is not accompanied by a manual, is difficult to persuade personnel in the field to adopt, and cannot be deployed without changes in time, financial, and institutional resources. When we consider an intervention that has been well developed, that has an intervention manual, videos, worksheets, and can be implemented within four classroom periods by a health education teacher, we have a potentially generalizable intervention which would be well worth evaluating. When considering an intervention that requires 16 classroom sessions, content that teachers resist, and two teacher training weekends, we have an intervention that may be very difficult to export to the diverse settings with which we are concerned and may not be worth evaluating or implementing. Similarly, when we are considering an intervention that has been offered by the best, most motivated, volunteer intervenors, to the best, most motivated, volunteer participants, we may conjecture that this intervention may be far less effective than one which has been conducted and evaluated with the entire span of available intervenors and with non-selected participants.

Planning Evaluation Research

Given the significance of evaluation research, it would seem useful to plan and to budget for theory-based evaluation research, and evaluation research consultation, at the outset of any STD/HIV prevention intervention.

Conclusion

Social science theory provides a systematic framework for the construction of STD/HIV prevention interventions and for the evaluation of such interventions. Simply stated, sound social science theory tells us what to target for change, provides useful clues concerning how to effect change, and guides us in evaluating change. Other advantages of social science theory that differentiate it from the "best guesses" of well-intentioned practitioners seeking to reduce STD/HIV risk include the fact that theory often offers a systematic and comprehensive approach to STD/HIV prevention that has the benefit of sustained conceptual development and empirical verification, while the "best guesses" that so often guide our interventions are often fragmentary and have limited conceptual foundation or empirical support. Arguably, the first decade and a half of AIDS prevention research was dominated by such "best guesses" and by the apparently compelling intervention activities that have, by and large, failed to produce sustained change in STD/HIV risk behaviour (J. Fisher & W. Fisher, 1992), and it is time to take more seriously the role of theory in providing a sound conceptual basis for STD/HIV prevention intervention and evaluation activities.

Figure 1

The Elicitation - Intervention - Evaluation process

References

• Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall.

• Ajzen, I., & Madden, T.J. (1986). Prediction of goal- directed behavior: Attitudes, intentions, and perceived behavioral control. Journal of Experimental Social Psychology, 22, 453-474.

• Basen-Enquist, K., & Parcel, G.S. (1992). Attitudes, norms, and self-efficacy: A model of adolescents HIV-related sexual risk behavior. Health Education Quarterly, 1, 263-277.

• Fishbein, M., & Middlestadt, S.E. (1989). Using the theory of reasoned action as a framework for understanding and changing AIDS-related behaviours. In V. Mays, G.W. Albee, & S.F. Schneider (Eds.), Primary prevention of AIDS: Psychological approaches (pp. 93-110). Newbury Park, CA: Sage.

• Fisher, J.D. (1988). Possible effects of reference group- based social influence on AIDS-risk behavior and AIDS prevention. American Psychologist, 43, 914-920.

• Fisher, J.D., & Fisher, W.A. (1992). Changing AIDS risk behavior. Psychological Bulletin, 111, 455-474.

• Fisher, J.D., Fisher, W.A., Misovich, S.J., & Kimble, D.L. (1996). Changing AIDS risk behavior: Effects of an intervention emphasizing AIDS risk reduction information, motivation, and behavioral skills in a college student population. Health Psychology, 15, 114-123.

• Fisher, J.D., & Misovich, S.J. (1990). Social influence and AIDS-preventive behavior. In J. Edwards, R.S. Tindale, L. Heath, & E.J. Posavac (Eds.), Social influence processes and prevention (pp. 39-70). New York: Plenum.

• Fisher, W.A., & Fisher, J.D. (1993). A general social psychological model for changing AIDS risk behavior. In J. Pryor & G. Reeder (Eds.), The social psychology of HIV infection (pp. 127-153). Hillsdale, NJ: Erlbaum.

• Kelly, J.A., St. Lawrence, J.S., Brasfield, T.L., Stevenson, L.Y., & Haugh, A.C. (1991). HIV risk reduction following intervention with key opinion leaders of population: An experimental analysis. American Journal of Public Health, 81, 168-171.

• Kelly, J.A., St. Lawrence, J.S., Stevenson, Y.L., & Hauth, A.C. (1992). Community HIV/AIDS risk reduction: The effects of endorsements by popular people in three cities. American Journal of Public Health, 82, 1483-1489.

• Kirby, D. (1992). School-based prevention programs: Design, evaluation, and effectiveness. In R.J. DiClemente (Ed.), Adolescents and AIDS: A generation in jeopardy (pp. 159-180). Newbury Park, CA: Sage.

• Wasserheit, J.N., Aral, S.O., Holmes, K.K., & Hitchcock, P.J. (1991). Research issues in human behavior and sexually transmitted diseases in the AIDS era. Washington, DC: American Society for Microbiology.

[Previous] [Table of Contents] [Next]

Last Updated: November 26, 1997