Study raises questions about conventional theories regarding sexual risk behavior among gay and bisexual menReviewed by Christopher Rosik, Ph.D.
A significant body of research literature related to health behavior theories has assumed that health-related attitudes, beliefs, and behavioral skills precede and causally influence subsequent health behavior. In the area of sexual risk behavior among gay and bisexual men, this has translated into an assumption that beliefs and attitudes toward safe sex practices influence subsequent sexual activity and that interventions need to focus on changing these views (e.g., psychoeducation regarding condom use). However, a recent study has raised serious questions regarding the accuracy of this conventional wisdom (Huebner, Neilands, Rebchook, & Kegeles, 2011).
The authors observe that the vast majority of research in this area has been cross-sectional in nature, meaning that the data are collected at one point in time and thus definitive conclusions about causality between variables cannot be made. Thus, strong associations between sexual risk attitudes toward and concurrent reports of actual risky behaviors may not tell us as much as we thought about the causes of these behaviors. In addition, the limited number of longitudinal studies, in which data is collected from the same sample over two or more time periods (making causal explanations possible), has provided decidedly mixed findings regarding causal pathways. The authors suggest that health behavior theories may have limited value when it comes to complex sexual risk behavior among high risk populations. In spite of all these uncertainties in the literature, the pathway from attitudes and beliefs to sexual risk behavior among gay and bisexual men has frequently been assumed by mental health professionals and public health officials.
Huebner et al. observe that a generally ignored alternative theory may need to be given more serious consideration. From this vantage point, the authors assert, “it is possible that individuals engage in sexual behavior for multiple reasons, some of which have little to do with their health-related attitudes and beliefs, but that they subsequently adjust those attitudes and beliefs accordingly so that they are consistent with their previous behaviors” (p. 112). They note that this situation would explain significant cross-sectional correlations as well as the limited or inconsistent longitudinal effects of thoughts and emotions on actual sexual behavior.
To test this possibility, Huebner and colleagues conducted a sophisticated analysis using structural equation modeling that allows for strong causal inferences to be made with longitudinal data. A sample of 1248 gay and bisexual men were surveyed twice in an 18 month period regarding their frequency of engaging in unprotected insertive and receptive anal intercourse with any nonprimary partner. Peer norms and attitudes for safe sex were also measured. The results were clear: contrary to theories of health behaviors, attitudes and norms did not predict subsequent unprotected anal sex when initial behavior was statistically controlled. Instead, sexual risk behavior predicted subsequent norms and attitudes when initial norms and attitudes were statistically controlled. The authors conclude, “in contrast to the causal predictions made by most theories of health behavior, attitudes and norms did not predict sexual risk behavior over time….These findings are more consistent with a small, but growing body of investigations that suggest instead that engaging in health behaviors can also influence attitudes and beliefs about those behaviors” (p. 114).
Limitations of the study included a convenience sample, significant sample attrition, and an inability to test for other potentially relevant causal pathways. Thus, replication is needed to increase our confidence that causality does indeed flow from sexual risk behavior to changes in attitudes and beliefs about those behaviors. Nonetheless, these findings raise a number of important questions that appear to need much greater consideration.
First, if engaging in sexual risk behavior leads to changes in beliefs and attitudes that legitimize such behavior, is it wise to encourage early self-labeling and sexual activity among male adolescents experiencing same-sex attractions? Could participation in early homosexual risk activity such as unprotected (or even protected) anal intercourse lead some adolescent boys down a path of homosexual activity and identity and away from what might have been an eventual heterosexual adjustment?
Second, how did social scientists and policy makers come to presume that the body of research pertaining to sexual risk behaviors among gay and bisexual men confirm traditional health behavior theories? How did causal assumptions that had no definitive foundation in the methodology of most studies become the basis for public health intervention? It seems plausible that the dynamics of groupthink or the pull of funding pressures have worked against the development of novel and, perhaps, less politically correct theories about sexual risk behavior. If this is even partially correct, it would seem to argue in favor of more sociopolitical diversity in the development of theory and research within the health behavior literature in general and the gay, lesbian, and bisexual literature specifically.
The findings of Huebner and his associates may not provide conclusive information to answer these kinds of questions, but at the very least their research suggests there is an empirical rationale for asking them.
Huebner, D. M., Neilands, T. B., Rebchook, G. M., & Kegeles, S. M. (2011). Sorting through chickens and eggs: A longitudinal examination of the associations between attitudes, norms, and sexual risk behaviors. Health Psychology, 30(1), 110-118.