Skip to main content

Archived Comments for: HIV/AIDS in Vancouver, British Columbia: a growing epidemic

Back to article

  1. Unintended results of research

    Michael Goodyear, Dalhousie University

    7 March 2009

    A fundamental principle of ethics in research is that the object of the research, in this case people infected with HIV, should benefit from the research. Those living in Vancouver will already be familiar with the political fallout from sensationalist media coverage of this article, stressing 'more than a quarter of female sex trade workers in city infected'.

    Therefore, although unintended, the research subjects have been harmed, stigmatised and quite probably subjected to increased levels of violence. The health and social problems of Vancouver's Downtown Eastside (DES) are well known in Canada, and it is difficult to see how this research will benefit them.

    This is one of many studies from the B.C. Centre for Excellence in HIV/AIDS, nearly all of which have been very high quality, and ethical and have aided those seeking to improve the lot of the local inhabitants. This long term research has been carried out with the collaboration of sex workers and service agencies in the area. Many of the publications have been models of community research and participatory-action research. Therefore everything should have gone right, but instead went horribly wrong. How did this happen?

    This study took existing data and fitted it to a computer model of how HIV increased in prevalence in Vancouver, but in many ways ignores the geography of the city where HIV is mappable to discrete areas. The investigators used existing data on prevalence rates for what they considered three high risk groups (7% of the population), men who have sex with men, intravenous drug users (IDU) and female sex workers (FSW). They do not provide their raw data, but it can be extracted from their sources.

    For the sex work data they estimated the population (which is fraught with uncertainty) as being 4% of the high risk group (0.2% of the population), however it appears they only counted the highest risk people amongst FSWs, survival workers, who have a high prevalence of iv drug use. They had three data points for prevalence of HIV. One from 1988-1992 was 6.2%, one from 1996-7 was 32% (however we know these were drawn from a sample of IDU, of which they represented 23%), and a more recent sample in 2006 which was 26%. We know the latter group were all survival sex workers from the DES whom we know have a very high prevalence of IDU. It is this last number which has hit the headlines.

    While it is true that of the three groups studied, the estimated prevalence rates for 2006 were 15, 17 and 26%, the FSW represented a tiny proportion, and were almost completely confounded with IDU. Furthermore these were not representative samples but convenience samples from women seeking assistance at outreach centres, and are likely to be overestimates. Of course iv drug using survival sex workers from the DES are highly unrepresentative of Vancouver's sex worker population.

    Therefore the media have focussed on known data from 2006 in which the prevalence rate was 26%, apparently higher than the 1992 data, if obtained from a comparable sample. The issue is really one of IDU not FSW. From other research we know that HIV rates amongst much wider samples of FSWs are actually very low, and almost always associated with iv drug use, and unprotected sex with intimate partners. HIV transmission from clients is uncommon, but as established in another paper from this group, it is the actions of the authorities which make FSWs vulnerable to being coerced into unsafe sex, and this should have been mentioned in this study.

    It would be easy to dismiss this as merely irresponsible reporting, but it is not as simple as this. When women gave permission to be tested for HIV they did not anticipate that it would be used to stigmatise them, and presumably when they gave consent for it to be used for research they were not informed of that. Incidentally there is no mention of ethical oversight on the current paper.

    The other ethical questions raised are the extent to which investigators, ethical committees, editors and reviewers should take responsibility for the dissemination and interpretation of their data, given their duty of care to the women who collaborated with them in this research. Misinterpretation of the paper might have been anticipated with more foresight.

    The outcome unfortunately appears to be moral panic, stigmatisation of a disadvantaged group, likely escalation of violence against a population already subject to extremely high levels, and a destruction of trust. We can do better.

    A broader issue is why there is so much emphasis on HIV rates amongst FSWs and not of their clients. This creates further victimisation of women by depicting them as vectors of disease.

    All of this is in stark contrast with the role of the New Zealand Prostitutes' Collective, co-opted and funded by the Health Deartment as health educators, and widely credited with the control of HIV/AIDs in that country.

    Competing interests

    Reviewer for BMC

  2. RE: Unintended results of research

    Eric Druyts, British Columbia Centre for Excellence in HIV/AIDS

    14 March 2009

    We thank Dr. Goodyear for his response to our article. We fully agree with his concerns surrounding the recent coverage of our work on HIV prevalence in British Columbia, Canada. Dr. Goodyear has expressed difficulty in seeing how this study will benefit the individuals who participated in the research. Of note, estimates of HIV prevalence among at-risk groups are vital in planning for the development and provision of appropriate policy and programmatic responses. We wish to affirm that it is our overarching goal to ensure that there are adequate services for all individuals living with HIV infection in Vancouver. The WHO has consistently shown that less than 10% of sex workers have adequate access to HIV prevention and care resources.

    Our paper did not aim to highlight HIV infection among sex workers in particular. Instead, we sought to model the estimate of HIV prevalence at the city level and related gaps in services in Vancouver. Also of note, all the studies considered in our paper received institutional ethical approval.

    We acknowledge that prevalence estimates are rarely perfect and are limited by uncertainty surrounding population size and potential biases inherent in source data. We would like to clarify that the estimate of HIV prevalence among female sex workers in 2006 is based on data collected among survival sex workers predominantly located in Vancouver’s Downtown Eastside, who live in poverty and all who inject and/or smoke illicit drugs. This estimate therefore does not reflect indoor sex workers, such as sex workers in establishment-based venues, bars, or escort services. We are fully aware that female sex workers in Vancouver do not constitute a homogeneous group. This could have been further stressed in the published paper.

    Perhaps most importantly, we recognize that sex workers have been unfairly stigmatized in the past by medical research as vectors of disease, and it was not our intention to perpetuate this in any way. We have acknowledged in our article that detailed data on sex work clients were not available. As a global assessment of HIV prevention needs, our article did not seek to review the factors that enhance vulnerability to HIV infection among marginalized populations, such as survival sex workers. However, as mentioned by Dr. Goodyear, we feel it is important to acknowledge that many pivotal studies both in Canada, including some of our own, and globally have demonstrated that criminalized sex work legislation, enforcement-based strategies and violence greatly reduces sex workers’ ability to safely negotiate condom use with clients as well as other HIV risk reduction strategies.

    Finally, we concur with UNAIDS and WHO that structural approaches to HIV prevention are crucial both for the health of sex workers and clients. This includes policy changes such as the removal of criminal sanctions targeting sex workers.

    Eric Druyts, Robert Hogg and Julio Montaner

    Competing interests

    None declared

  3. Engendering Knowledge about Sex Workers, Drug Use and HIV/AIDS

    Cecilia Benoit, University of Victoria

    27 March 2009

    Cecilia Benoit, Centre for Addictions Research of BC & Dept of Sociology, University of Victoria

    Jody Paterson, freelance writer, Victoria, BC.

    Mikael Jansson, Centre for Addictions Research of BC, University of Victoria

    We appreciate the comment of Dr. Goodyear, “Unintended results of research,” and the follow up response from Drs. Druyts, Hogg and Montaner. We are impressed that the authors acknowledged that their original article should have stated their data source for estimates of HIV/AIDS prevalence rates for sex workers was confined to female survival sex workers from Vancouver’s Downtown Eastside, who use addictive substances and live in improvised conditions (i.e., the authors note their “estimate therefore does not reflect indoor sex workers, such as sex workers in establishment-based venues, bars, or escort services. We are fully aware that female sex workers in Vancouver do not constitute a homogeneous group”).

    We would like to add that work venue (on or off street) is not the only characteristic the authors and Dr. Goodyear should recognize when considering the heterogeneity of the sex work population in British Columbia and elsewhere, their use of addictive substances and their prevalence rates of HIV/AIDS. There is no dispute in the academic literature that the sex trade is highly gendered: estimates range from 75-80 percent identify as female, depending on time and place. On the other hand, we must be careful not to assume that all sex workers are female, just as we must not assume the sub-group of sex workers with substance use problems or carriers of HIV/AIDS are female. There is in fact a substantial minority of sex workers who identity as male and a smaller portion as transgendered. This includes male sex workers who have sex with other men for pay, a sub-category of one of the other groups of focus in the original article, men who have sex with men (MSM). It is estimated about 20% of street sex workers in Canada are males (Shaver, 1993).

    The underlying methodological problem is that most sex work research is carried out with sub-groups and typically involves small samples – often involving fewer than 50 respondents – recruited from a particular area of a city or metropolitan area (Benoit and Shaver, 2006). To date, we know little about how sex workers’ health status, including drug use and prevalence rates of HIV/AIDS, varies by gender, work venue and location within and as well as across geographical settings. Without such knowledge it is virtually impossible to develop evidence-based interventions to reduce risk and promote the health, safety and well-being of male, transgendered, as well as female sex workers in the long term. Without reliable generalizable data, moreover, researchers are in danger of perpetuating myths about the typical sex worker -- almost always seen as a female prostitute – who tends to be variously and sometimes simultaneously cast as victim of abusive circumstances, illicit drug addict, wanton entrepreneur, and vector of disease. Indeed, it has been said that female sex workers epitomize the image of the "fallen", "criminal" or “diseased” woman and provide the negative standard against which feminine virtue and characteristics of the good and healthy woman are defined (Hallgrimsdottir et al., 2008). Such stigmas of the female sex worker also become the motifs the dominant media use to report on the sex industry and help determine what the public comes to believe as “truth” about those who work within it.

    Works cited:
    Benoit, C., Shaver, F. M. Critical Issues and New Directions in Sex Work Research. Canadian Review of Sociology and Anthropology Special Issue 2006, 43.3: 243-252.

    Hallgrimsdottir, H., Phillips, R., Benoit, C., Walby, K. Sporting Girls, Streetwalkers, and Inmates of Houses of Ill-Repute: Media Narratives and the Historical Mutability of Prostitution Stigmas. Sociological Perspectives. 2008, 51 (1): 119-138.

    Shaver, F. M. Prostitution: A Female Crime? In Ellen Adelberg & Claudia Currie (eds.), In Conflict with the Law. 1993, 153-173. Vancouver: Press Gang Publishers.

    Competing interests

    No competing interests to declare.

  4. Limitations and fallacies in sex work research

    Michael Goodyear, Dalhousie University

    20 April 2009

    Benoit and Paterson provide a welcome opportunity to discuss the many issues to be considered when designing and interpreting research on sex work. Even the definition of sex worker requires careful consideration, since an act does not define a person. Much of sex work is private, invisible, and a part-time occupation across a very diverse background of education, and professional lives. By default then researchers have focussed on the more visible parts of the spectrum of sex work.

    Researchers are influenced by many factors including their own personal and moral values, and privilege, which may also influence the answers obtained from participants. Available funding and sponsor philosophy may be heavily weighted to dominant discourses, and access is likely to be very skewed. Access is often very dependant on gate keepers, and representative sampling virtually impossible. Therefore very close attention to sampling, context, and the framework in which responses are elicited is required, and any inferences require strict limitation to the sampling frame used. Demographic characteristics can, for instance, vary widely from neighbourhood to neighbourhood within the same community. (Bernstein 1999, Porter and Bonilla 2000)

    Much of sex work research is problematic from these perspectives, and its interpretation even more prone to misapprehension. Thus the data from which public policy is informed is frequently faulty, and often itself informed by the dominant discourse. This helps to explain some of the widely disparate literature on the subject.

    Poorly designed and inappropriately designed research in the social sciences can cause a great deal of harm, further entrench prejudices and stereotypes, and add to the burden of stigmatisation. As Benoit and Paterson correctly point out, this in turn creates a much wider spectrum of harm by feeding moral dualism.


    O'Connell Davidson J., Layder D. Methods, Sex and Madness. Routledge, London 1994
    Vanwesenbeeck I. Another decade of social scientific work on sex work 1990-2000. Ann Rev Sex Res 12: 242 (2001)
    Benoit C., Shaver F. Critical issues and new directions in sex work research. Can Rev Soc Anthr 43: 243 (2006)
    Sanders T. Methodological nuances in researching the female sex industry. Sexualities 9: 249 (2006)
    Weitzer R. Flawed theory and method in studies of prostitution. Violence Against Women 11: 934 (2005)
    Bernstein E. What’s Wrong with Prostitution? What’s Right with Sex-Work? Comparing Markets in Female Sexual Labor. Hastings Women’s Law Journal.10: 91 (1999)
    Porter J, Bonilla L. Drug use, HIV, and the ecology of street prostitution, in Weitzer R (ed.) Sex for Sale, Routledge, NY, 2000. 103.

    Competing interests