This chapter will discuss the concept of risk and present data on selected high-risk practices from the European sexual surveys conducted between 1987 and 1993. The prevalence of these practices are analyzed with respect to age group and gender and compared between countries. In addition, one of the practices (having had more than five partners during the past year) is analyzed within countries with respect to cohabitational status, place of residence and gender of the sexual partner; the degree of association with these background variables is then compared between countries. The prevalence of risk-related practices can be seen as estimates of the proportions of the population with a potential risk of becoming infected with HIV. The extent to which such estimates actually correlate with the occurrence of HIV infection is discussed.
In epidemiological terms, the risk of acquiring a disease is the probability that a subject belonging to a particular population may develop a certain disease over a certain time period. In longitudinal studies it is expressed as the number of people who fall ill (or contract an infection) over the total number of subjects under observation. This proportion is called the cumulative incidence. In the case of HIV, the relative risk of infection following a specific type of exposure, for instance a certain behaviour, can be calculated by dividing the cumulative incidence of infection in the exposed group by the cumulative incidence in the unexposed group.
An HIV risk indicator is a variable that has been found to be statistically associated with HIV infection in cohort or case-control studies. The present cross-sectional surveys do not directly relate sexual practices to actual HIV infection. Rather, these look at whether people engage in practices which might put them at risk of contracting HIV. Thus, only potential risk, not actual risk, can be estimated from the surveys. It would have been informative to