Article Text


Perceived transmissibility of STIs: lack of differentiation between HIV and chlamydia
  1. B Knäuper,
  2. R Kornik
  1. McGill University, Montreal, Canada
  1. Correspondence to:
 Bärbel Knäuper
 Department of Psychology, McGill University, 1205 Dr Penfield Avenue, Montreal, QC, Canada, H3A 1B1;

Statistics from

Sexually transmitted infections (STIs), such as HIV and chlamydia, differ widely in their transmissibility. The estimated probability of HIV transmission from an infected heterosexual man to a woman in one act of unprotected vaginal intercourse is 0.1%,1 whereas the same probability for chlamydia is 35%.2 This research examines college students’ knowledge about the per act transmission probabilities for HIV and chlamydia.

Previous studies reported median perceived transmission probabilities of 50%3 and 33.4%4 for HIV for one act of unprotected receptive vaginal intercourse with an infected man. These findings were interpreted as demonstrating “badly overestimated per act transmission probabilities” (Pinkerton et al4 p 19). However, the distributions of the estimates were not provided. If estimates are widely dispersed across the entire probability range from 0% to 100%, interpretations of averages are meaningless and interpreting the data as indicating a systematic overestimation of the transmission probability would be unfounded. We studied this possibility in a sample of college students.

In all, 234 undergraduate university students (145 women, 85 men, mean age 21.14 years, SD 2.82, four did not report their age and sex) enrolled in a variety of academic programmes were randomly selected and individually approached after classes. Aside from their age and sex, participants were asked in two separate questions: “What do you think is the probability, in percentages, of a woman becoming infected with HIV (chlamydia) from one unprotected act of vaginal intercourse with an infected man?” The order in which people were asked the two questions was counterbalanced.

Figure 1 presents the distribution of the estimates, showing that they are widely and quite equally dispersed across the entire range from 0% to 100%, and that distributions do not differ between the two infections (Kolmogorov-Smirnov Z = −0.73, p = 0.46). No age or sex differences were found. Only 3.9% and 5.6% of the estimates for HIV and chlamydia, respectively, come close to the correct probabilities if “correct” is defined as smaller than 0.5% for HIV and between 30% and 40% for chlamydia. In all, 34.8% of the participants falsely estimate that chlamydia has a lower transmission probability than HIV, 39.5% correctly estimate that chlamydia has a higher transmission probability than HIV, and 25.8% provide exactly the same percentage estimate for both STIs.

The data show that a large majority of college students clearly lacks knowledge of the transmission probabilities of HIV and chlamydia and does not know that chlamydia is more infectious than HIV. Previous reports of statistical averages of the perceived transmissibility3,4 and their interpretation as indicating a systematic overestimation bias may be unfounded. The results highlight the importance of inspecting response distributions and restraining from reporting statistical averages when distributions are widely dispersed. Furthermore, they highlight that information about transmission probabilities should be incorporated into sexual health programmes in order to make people more aware of STIs that are considerably easier to contract than HIV.

Figure 1

Distribution of estimates for HIV and chlamydia for one act of unprotected vaginal intercourse with an infected man.


The reported research was funded by a grant from the Social Science and Humanities Research Council of Canada (SSHRC, 410-2002-09) and a New Opportunities Fund from the Canadian Foundation for Innovation (CFI, 4015) to Bärbel Knäuper. We thank Surkhraj Cheema for her help with the data collection, as well as Irv Binik and Sandi Byers for helpful comments on an earlier version of the manuscript.

 The study was jointly conceptualised and designed by BK and RK; data were collected by RK, with the assistance of Surkhraj Cheema; BK analysed the data and led the writing; both authors jointly interpreted the findings, reviewed drafts of the manuscript, and approved the final version.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.