Understanding the context of risk practices among injecting drug users: implications for hepatitis C prevention.

Habib, Shah E

Introduction

`Risk' has become a keyword in recent sociological studies of health and related behaviour (Gabe 1995; Lupton 1993; Hayes 1992; Rayner 1992; Douglas 1986), particularly in discourses about HIV/AIDS (Habib et al. 2000-2001; Rhodes 1997; Bloor 1995a, 1995b; Schiller et al. 1994; Connors 1992; Pollak et al. 1992). This keyword is also considered central to the study of the social aspects of hepatitis C (HCV) and to mapping its determinants. The reason for this is that HCV is a behavioural disease which progresses mostly throughout an injecting drug-using population via different interacting patterns and social relationships. As with HIV, the HCV epidemic and associated risk behaviour are of profound sociological significance, not only because of the threat to public health, but also because the onset and course of the disease are interrelated to social behaviour.

The risk of HCV infection among injecting drug users (IDUs) has been the subject of inquiry among behavioural and social scientists, and epidemiologists for many years (MacDonald et al. 2000; Loxley 2000, 1998). Research up until the present has found that high rates of needle sharing have been a common problem for some IDUs (Crofts et al. 1997). It is now well known that sharing injecting equipment puts IDUs at risk of HCV via the transmission of contaminated blood. It is apparent that when the virus is introduced into a community of IDUs, it spreads rapidly and infects a large proportion of individuals who inject drugs. Furthermore, many investigators see IDUs as the epidemiological bridge to other people because of the various mixing patterns within a population.

The context of risk behaviours is important to understanding continued risky injecting practices among drug users (Rhodes 2002, 1995). Previous research shows that sharing injecting equipment occurs most often among people in a relationship or social network; whether family, partner or close friend--wherever members are, at least potentially, mutually oriented to one another and may influence each other's behaviour (Koester 1996; Bloor 1995b; Barnard 1993; Harding and Zinberg 1977). These dynamics of the HCV epidemic among IDUs highlight the importance of a basic knowledge about the context of drug use and risk practices, as well as of the interaction of drug users in their naturally occurring social contexts.

With respect to behavioural change in risk practices, numerous epidemiological studies have been undertaken in the areas of injecting drug use in response to HCV prevention in Australia (MacDonald et al. 2000; Loxley et al. 1997; Crofts et al. 1997). These studies related to the onset of HCV appear too frequently to allow us to ignore the social aetiology of factors that are recognised as more immediately relevant. If changes in risk behaviour influence the course of HCV, it is important to understand the factors that influence behavioural change in general and the onset of risk behaviour in particular. In the absence of efficacious vaccines against this virus, all health promotion strategies to prevent the further spread of HCV infection should focus on changing human behaviour, more specifically injecting behaviour.

No social studies so far have addressed IDUs' risk practices in relation to HCV transmission in Australia, nor the contexts and constraints preventing safer behaviour. This paper investigates the context of HCV-related risk behaviour among IDUs and discusses the behavioural and social determinants of such practices. The aim is to examine how risk is constructed in the context of the everyday lives of IDUs. Although the social `efficiency' of different interacting patterns within the IDU population is not highlighted here, a focus has been placed on regular partners to shed light on the ways in which individuals behave in a risky fashion.

Method

This paper is based on a self-administered survey, conducted by the author between January and June 1998. The sample consisted of 336 people who were IDUs or had a history of injecting drug use. Participants were recruited by a direct approach to persons using needle and syringe programs (NSPs) and by a mail survey sent to members of the New South Wales Users and AIDS Association (NUAA) (1) in the Sydney metropolitan area. The recruitment was conducted through advertising and the distribution of fliers in needle and syringe outlets.

The majority of data was obtained from clients by staff distributing needles and syringes, via a self-administered questionnaire. The staff at respective outlets asked clients to fill-out the questionnaires while distributing `fits' (injecting equipment). Potential participants were informed that the purpose of the study was to ascertain the context of drug injecting and risk behaviour in relation to the spread of HCV among IDUs. Furthermore, that the study was anonymous and thus names would not be sought, and that participation was voluntary. Most of the survey forms were returned during fit `collection' at the outlets. Of the 336 respondents, 274 persons were recruited at ten NSPs (2) and one private methadone clinic in the Sydney metropolitan area. The mailed survey, sent to members of NUAA with a sealable return-postage prepaid envelope, yielded 62 completed questionnaires. To obtain a geographically representative sample from the Sydney metropolitan area, participants were recruited from the inner, eastern, northern and western suburbs-geographical areas. The subjects were selected on the basis of convenience rather than attempting to obtain a random sample of this ill-defined group.

Both qualitative and quantitative data were collected by means of an anonymous questionnaire, tested in terms of readability and reliability devised for the purposes of this study. The 36-item questionnaire was developed using questions specifically about drug use, HCV knowledge, risk practices, the social context of risk behaviour, needle procurement, disposal of used needles, and harm reduction services provided by the NSPs. The survey focused on producing high-quality, in-depth data on a sample of IDUs. A brief report on the quantitative aspect of the study has recently been published (Habib et al. 2001 a). The analysis reported in this paper is based on the qualitative component of data. In order to retrieve documents and text segments efficiently, data were coded by the computer package NUD.IST. Coding and analysis of qualitative information was undertaken on an ongoing basis throughout the data collection period.

Background characteristics

The sample comprised 336 IDUs, of whom 51% (n = 172) were male, 47% (n = 157) female and 2% (n = 5) classified themselves as transgender. Across the whole sample, 88.4% (n = 297) had injected drugs in the two months prior to participating in the study, while 11.6% (n = 39) had not injected drugs for years. The most popular substance reported by this sample was heroin (72%), however many of them using other drags in conjunction with or in substitution for it.

The age of the respondents ranged from 14 to 64 years, with 17% younger than 25 years of age (mean age 32). The mean age at which respondents had first injected or been injected by someone else was 18.5 years. One-third (33.8%) of the sample had completed 10 years of schooling, 19% had received their higher secondary education (HSC), 15% were at TAFE, and 20.3% had at least some additional tertiary education, while 12% had no high school education. Most subjects (64%) were unemployed or receiving government benefits or pensions, nearly one-third (32%) were in paid employment (full-time, part-time, or casual), 3% were students and 1.2% did not respond to this question. The majority (77%) of the sample were born in Australia, while a considerable number of the rest were born in other western countries, mostly from New Zealand (23%) and England (5%). In addition, most of them had been previously HCV tested (93%), and 66% were HCV positive.

Researching risk: understanding the context of risk practices

The qualitative data presented in Figure 1 focus on a range of risk behaviours and the context within which these behaviours occur. The example of dependency and craving is frequently referred to drug and HIV/HCV literature in understanding risk behaviour. These two concepts often appear to guide the thinking behind contexts of HCV-related risk. This is because dependency or craving is known to be associated with higher HIV/HCV risk-taking behaviour. The effects of drugs can reduce inhibition, cloud judgement, result in memory lapses, and lead to false feelings of safety and less concern about HCV or HIV. It comes as no surprise that IDUs with more dependency on drugs are more likely to share injection equipment. The link between drug use and HCV is that the virus is efficiently transmitted while sharing drug injection paraphernalia, including syringes, needles, cookers or spoons, cotton, and water glasses.

Figure 1: Risks associated with injecting drug users' lifestyles

Dependency

`When you need a shot you have to have it now, right now because you
are hanging out' (35 year old female HCV positive IDU)

Illusion/craving

`I thought I was re-using my own, but months later on diagnosis
realised it was my friend's. I've got hep C from one of the very
few times I shared, so am now very careful & don't share anything'
(32 year old female HCV positive IDU)

Sharing needles

`Two years ago caught Hep C immediately after one shared
needle' (38 year old female HCV positive IDU)

Sharing with partner

`One week ago I only shared spoon with a fit with my husband'
(33 year old female HCV positive IDU)

`I only do these things with my lover' (23 year old female HCV
positive IDU)

Injecting in prison

`When I was in jail only a used fit was available' (37
year old female HCV negative IDU)

Misconception about HCV transmission

`I didn't realise I could catch Hep C only from sharing
a spoon, filter and tourniquet'
(23 year old female HCV positive IDU)

Unprotected sex

`A week ago or two at home ... my defacto partner had
unprotected sex--I don't know what she has been doing
lately sad eyes' (35 year old male HCV positive IDU)

Other risks/dangers

`Reused own needles/syringes last week, shared needles
10 months ago, and at the beginning of the year when I
got out of jail in 1994 and stayed at a friend's house
and shared razor and syringe' (33 year old male HCV positive IDU)

`Shared needles 4 years ago, shared spoon yesterday, unsafe sex one
month ago, toothbrush week ago, reused my own equipment usually happen
once or twice a month. When I first started using, I shared with my
partner after only rinsing in H2O. Now I never share fits. This was
easily in my using days--we only had one fit, my partner insisted on
using first and I was stupid--naive so I used it after him' (25 year
old female HCV positive IDU)

There are a number of factors associated with sharing injecting paraphernalia in general that exhibit a greater extrinsic component. Most powerful among these is having a sexual partner who is also an IDU. Having an IDU partner increases risk both for intravenous drug use and sharing of equipment, making risk reduction less likely. In addition to injecting drug use with sexual partners, sharing of injecting equipment inside prisons also represented a case example (Figure 1). Prisoners have a greater chance of getting the HCV virus because of the increased risk of the virus spreading within their confined community, in conjunction with their lack of access to sterile injecting equipment. This is important, given the fact that the proportion of prisoners and sharing among IDUs in the prison environment is much higher in New South Wales than in other States and Territories (Australian Bureau of Statistics 2001).

Figure 1 also shows a case example indicating other intrinsic factors related to risk-taking across the IDU population. These include common misconception among persons who believe the transmission of HCV to be limited in risk while sharing spoons (where blood is not in contact), tooth brushes or razors. Due to the longevity of the HCV virus there is a significant risk of being infected by sharing such domestic items, and the risk increases if there is repeated contact with infected items. This sort of misconception reflects a deficiency of knowledge about HCV transmission that could subject these people to risk. The data presented below on the risks associated with injecting are analysed in detail, in the context of drug users' lifestyles.

Needle sharing with sexual partners

As mentioned earlier, sharing injecting needles or other related equipment occurs most among people involved to some degree in relationships, especially sexual partners. There is evidence in this sample indicating that sharing needles or equipment occurred between lovers in a steady relationship. The following section examines the context of social relationship especially sexual ones, in order to discover the reasoning which underlies unsafe injecting behaviours. Thirty-five respondents (12% of the sharing population)--22 women and 13 men--reported a regular sexual and sharing relationship with their partners where commitment, love and trust were used as the main justification. Of these people, 15 women and 7 men reported that they were HCV positive. For many drug injectors in this study, needle sharing risk behaviour occurred in a steady and monogamous relationship and was associated with drug users' daily lifestyles. The comments made by many sharers emphasised their commitment and trust to their partners, or indicate that they perceive themselves risk free:

 Have shared with partner as we both have been using together half
 and half always. A way of life. (24 year old male HCV positive
 IDU)

 Only shared with partners-never used any one else's. (34 year old
 female HCV negative IDU)

 Shared equipment last week only ever with boyfriend who has
 been tested negative for Hep C and HIV. (25 year old female HCV
 negative IDU)

 Regularly share non-syringe equipment that is fully swabbed and
 new syringes only as we mix up in one batch and divide between
 partner and I. Did so today--used own syringe once only about 2 yrs
 ago. (32 year old male HCV negative IDU)

 Only share when one syringe between my husband & myself--weeks
 ago. (25 year old female HCV negative IDU)

 Shared last week--I only share [with] my sexual partners, equipment
 etc. (32 year old female HCV positive IDU)

 Recently shared, but usually only with my partner of 14 years, ours
 is a strictly monogamous relationship. (27 year old female HCV
 positive IDU)

There were indications from some individuals in this study that a positive HCV test result increased the likelihood of needle sharing, and that few individuals had knowingly shared with someone whom they knew was HCV positive. Thus, having a positive HCV test result increases the likelihood of needle sharing between partners. As indicated by one injector: `I only share equipment with my close friend and long-term partner. We are both Hep C positive.' This was substantiated by another user who was HCV positive: `I do share needles and other equipment with my friend and girlfriend. We are in a monogamous relationship. Neither of us have AIDS, but we do have Hep C.' This may suggest that fear of further infection (in terms of genetic variation of HCV) was not perceived as a risk factor by users who had already been infected with the virus. (3) The genetic variations of HCV are important as they may have an effect on response to treatment (Crofts, Thompson and Kaldor 1999). It is also important given the fact that 15 to 22 percent of people infected with HCV can get rid of the virus within two to six months (Wightman and Toomey 1998).

These findings support the perception that if users are infected with HCV, they can share injecting equipment with another person who is also infected with HCV and it will not make any difference. In such situations, an extreme commitment to share common feelings with steady partners can be seen to be a function of `accepted risk'. It is also possible that these users were heavily dependent on drugs and were not aware of the implications of viral load, or the types of HCV, or were in the acute (short-term) or chronic (long-term) state of infection.

Risk in prison settings

Risk behaviours, such as sharing needles and syringes, are often much more prevalent in prisons. There is a strong connection between incarceration, or incarceration history, and the risk of getting blood-borne or sexually transmitted diseases. Since prisoners lack sterile equipment and the means to disinfect injecting paraphernalia inside correctional facilities, they cannot take adequate precautions to act on their risk awareness. Therefore, their behaviour often results in unsafe practices in terms of the transmission of blood-borne viruses. Many of the cases in this study (48%) relating to HCV risk behaviour by IDUs occurred in prison settings. Their general situation was summed up by one respondent in answer to the question: `Did you ever share needles/syringes while you were in gaol?': `Two years ago in gaol and again in gaol late '96. Very hard to get a new fit in gaol although it was possible for some.'

Situational effects and constraints on safer behaviour

Much of this study's emphasis has been given to exploring the barriers to safe injecting and changes toward safer behaviour to reduce HCV transmission. The following examples highlight situations in which respondents thought unsafe injecting was most likely to occur. According to the respondents' description at least one or two of the following conditions accounted for needle sharing.

1. The user was `hanging out' (desire a drug strongly) and was reluctant to wait to use drugs.

2. The user found it difficult to get a sterile needle at the time of injecting.

3. There was no needle and syringe program during late 70s, and the user was unaware of any communicable diseases.

Drug dependency or the state of hanging out

In response to the question `why did you share injecting equipment', the first condition was typified by the statement of a woman who was HCV positive: `Because when you are sick you need it now--right now--straight away'. The next comment is from a young man who expressed the same urgency to use drugs when asked why he shared injecting equipment: `Used half a friend's shot of heroin after him while waiting to score as I was hanging out'.

In some situations the illusion of control was feigned, especially when the users were impatiently hanging out. Situational effect and perceived risk might be the major factors in predicting these behaviours. The following comment by a woman who was HCV positive suggests that sharing might occur if craving is out of control:

 I didn't realise I could catch Hep C from sharing a spoon, filter &
 tourniquet. One person was mixing up for everyone. And I wasn't
 thinking straight (this person has Hep C). Please find a cure! I
 feel that I may have a positive result next test. (23 year old
 woman)

Another man underestimated the probability of getting infected by sharing injecting equipment, although he understood that sharing needles with an infected person was extremely dangerous: `There is no risk [of] sharing if blood not in contact--no fits shared, just [a] spoon'. In addition to the state of hanging out, concern about police harassment was also reported by a few respondents. In one man's words: `Impatiently hanging out. Also [at] other times worried about police harassment. What's this where's the dope'?

Difficulty in getting needles or syringes

A wide variety of problems were mentioned by users (40% of the total sample) concerning access to clean injecting equipment. Transport was a problem for some respondents who lived in remote rural areas. The closest source of equipment was liable to be in the nearest town. In describing these experiences, one man who was HCV positive said: `I would have had to go for about 20 minutes to 1 hour, the dope may have been used by then'.

The other major reasons given relating to the difficulty of obtaining a new fit was the opening hours of exchanges or pharmacies, and the unavailability of equipment at the time of drug use. Several respondents made comments on these issues:

 Only 1 chemist in my area still sells fits, vending machine only
 recently installed, needle exchange at hospital usually out of fits
 or says they are ... (42 year female HCV positive IDU)

 Have only shared needles with regular partners (not in last year).
 Have not been tattooed in last two years or pierced. No clean fits:
 no exchange available at time of night. Shared immediately after
 regular partner of 8 years. (26 year old male HCV negative IDU)

A man who was HCV positive said he would not share a needle if he could get new fits from either a vending machine or from a chemist. In his words: `Only share when there is no way to get to vending machine or chemist that sells these'. In some cases the unavailability of equipment during weekends was noted as a difficulty. This was more common, however, for respondents who lived in country areas. A typical comment was: `Not enough equipment handed out for the weekend to last all through'. Despite the unavailability of equipment some respondents claimed that they had taken some measures for safe injecting. As one man remarked:

 Being fully aware at the risk of BBV's I was not hasty in sharing, I
 spent over 1 hour driving around the Gold Coast on Sunday morning
 before out of feeling so ill, I attempted to clean someone else's
 fit. (30 year old male HCV negative IDU)

Needle sharing prior to the needle & syringe program

Needle sharing experiences prior to the establishment of syringe distribution programs in New South Wales and other Australian states was common, and problems in country areas were frequently cited by IDUs as particular experiences. In describing their experiences in relation to the questions of past risk practices, one user who was HCV positive said: `All "yes" answers happened in the late 1970s through to the middle eighties except for unprotected sex with unknown partner'. Some were unaware of the consequences of needle sharing in their early injecting career. As noted by a 41 year old female injector who was HCV positive: `7 years ago didn't realise the potential problems. So when one was unavailable thought if cleaned well it was OK. Also had problems (twice) when walking out of chemist was stopped by police'. Indeed, several respondents experienced a difficulty in getting injecting fits early in their injecting careers. The following are the extracts from some users who had such difficulties:

 Most of my sharing was prior to needle exchanges and the easier
 availability of syringes. (40 year old male HCV positive IDU)

 When I first started using not many chemists on the central coast
 would sell fits 15 years ago. (34 year old male HCV positive IDU)

 Chemists in Cairns wouldn't sell fits back in the 80's. I don't know
 what happens in Cairns now a day. (42 year old female HCV positive
 IDU)

These examples illustrate how HCV-related risk behaviours are interdependent on the availability of harm reduction services, particularly with the NSP scheme. For many users, it was difficult to get hold of new needles and syringes during late 70s and early 80s. Thus risk behaviour was shaped by the lack of general availability of sterile injecting equipment.

Discussion

Risk practices are socially patterned and associated for the most part by close relationships (Rhodes and Cusick 2000; Neaigus et al. 1994). Those respondents who shared needles and syringes were much more likely to have shared within their immediate social circle (regular sexual partner, close friend or family members) than with a comparative stranger (Habib et al. 2001b). A proportion of those shared injecting equipment within a monogamous social relationship and knew their partners' HCV status--which documents a version of negotiated safety. This indicates that a larger proportion are simply accepting the risk of transmitting blood-borne viruses, and the continuing incidence of HCV among IDUs is evidence that these risks are real. If the recent prevalence in self-reported sharing is real and is related to HCV, there is a very real need to support and enhance further such behaviour changes through educational efforts targeting all blood-borne viruses.

Some case examples illustrate that IDUs had shared due to the urgency of injecting, particularly when they were hanging out. This suggests that risks appear less significant to IDUs at such times. Experience has taught us that hanging out has a powerful effect on users to share needles or other injecting equipment in order to alleviate their craving. Some respondents claimed they could control their chances of becoming infected, through better judgement, but eventually could not even remember whether they had used someone's fits or not. Simultaneous drug use, often characterised by situational factors, might heighten the risk of rendering conscious control more fragile. Clearly, sharing used equipment happens and its likelihood is dependent upon perceived risks and not wishing to disrupt immediate pleasure.

The harm reduction model, retrospectively understood in terms of a health belief model, can be applied to understand HCV risk behaviour (Becker 1984). It maintains that when individuals are threatened by logistic and legal barriers, they resist behavioural changes. Previous experience shows that IDUs with limited access to sterile needles and syringes or the methadone maintenance program are at risk of contracting the HCV virus (Taylor et al. 2000). However, the provision of these means does not alone necessarily change risky behaviour, as many of the IDUs in this study, who were regular clients of the NSP or methadone program, were found to be engaged in risk practices. Harm reduction strategies such as the needle and syringe programs and methadone maintenance therapy are not, by themselves, enough, even where they achieve complete coverage. A possible factor in determining health behaviour outcomes is that young IDUs repeatedly perceive others, not themselves, to be at risk. This is an area of consideration for policy makers and service providers of any drug related health intervention.

Some misconceptions were identified in relation to sharing injecting equipment (e.g. spoons, filters), and sharing razors and toothbrushes. These sorts of misconceptions imply a deficiency in knowledge about HCV transmission that could subject these people to risk. Interventions for these IDUs should include educational programs to correct their misconceptions about HCV infection and to promote safer drug use practices. Moreover, the development of skills to negotiate safe drug use should be an important part of these interventions.

The findings of this study are derived from data based upon self-reported behaviour. Although the questions asked often required subjects to write their HCV status and other socially stigmatised activities, efforts were made to ensure that valid data were obtained. Subjects were given strong assurances that any information they divulged would be treated as strictly confidential and anonymous. Other research (based on a telephone survey) on injecting drug use and HCV has shown that when subjects are given such guarantees, the data obtained is reasonably valid and reliable (Van De Ven et al. 1999; Watson 1999).

Conclusion

The word risk remains central to health research, particularly on the social aspects of HCV. Although it is a key concept in sociology of health and illness, there is a serious lack of sociological and anthropological research into the IDU population in relation to HCV transmission. The social construction of risk behaviour needs further and more extensive work, in particular social and organisational contexts (such as the physical environments where injecting occurs) to discover the meanings and contexts of various behaviours.

This paper has explored the context of risk behaviour related to HCV transmission among IDUs in Sydney. Findings from these data imply that social factors are embedded in trusting relationships that may play an important role in minimising HCV-related risk behaviour. Risk behaviour, especially sharing needles and syringes with a HCV positive user, requires closer investigation. Apart from needle sharing, there are a host of other risks associated with HCV transmission that need to be explored. Unless these challenges and risks are taken up successfully, HCV has the potential to have a serious impact on many IDUs' lives.

Acknowledgments

This study was carried out under the auspices of NUAA, Sydney. I am grateful to NUAA's former coordinator Annie Madden for her support and assistance with data collection. Discussions with Susan McGuckin were especially helpful in clarifying the issues involved. Thanks are also due to all the respondents for their participation in the study. Finally, I wish to thank Lester Adorjany for his helpful comments and editorial assistance on this article.

Endnotes

(1) As the peak drug users' body in NSW, NUAA presents a state-wide membership of drug users providing the health and social needs for users, and the majority of its members are either current drag users or had a history of injecting drug use.

(2) NUAA NSP, Newtown NSP, KRC, Ryde NSP, Langton Centre, St. George NSP, St. Marys CHC, Foley House, Liverpool Mobile Exchange, Canterbury HIV Prevention Unit, Kobi Methadone Clinic.

(3) Although the influence of HCV genotype on the severity of subsequent infection is very difficult to establish, risk of infection is thought to be influenced by a person's genotype.

References

Australian Bureau of Statistics (2001) Prisoners in Australia (Cat. No. 4517.0). Canberra: Australian Bureau of Statistics.

Barnard, M. A. (1993) `Needle sharing in context: patterns of sharing among men and women injectors and HIV risk', Addiction, 88 (6), pp. 805-812.

Becker, H. S. (Ed.). (1984) The Health Belief Model and Personal Health Behaviour. Thorofare, NJ: Charles B. Slack.

Bloor, M. (1995a) `A user's guide to contrasting theories of HIV-related risk behaviour'. In J. Gabe (Ed.), Medicine, Health and Risk: Sociological Approaches (pp. 19-30). Oxford: Blackwell.

Bloor, M. (1995b) The Sociology of HIV Transmission. London: Sage.

Connors, M. (1992) `Risk perception, risk taking and risk management among intravenous drug users: implications for AIDS prevention', Social Science & Medicine, 34 (6), pp. 591-601.

Crofts, N., Thompson, S. and Kaldor, J. (1999) Epidemiology of the hepatitis C virus (Technical Report Series No. 3). Canberra: Commonwealth Department of Health and Aged Care, Publications Production Unit.

Crofts, N., Jolley, D., Kaldor, J., Van Beck, I. and Wodak, A. (1997) `Epidemiology of hepatitis C virus infection among injecting drug users in Australia', Journal of Epidemiology and Community Health, 51 (6), pp. 692-697.

Douglas, M. (1986) Risk Acceptability According to the Social Sciences. London: Routledge and Kegan Paul.

Gabe, J. (1995) `Health, medicine and risk: the need for a sociological approach'. In J. Gabe (Ed.), Medicine, Health and Risk: Sociological Approaches (pp. 1-17). Oxford: Blackwell.

Habib, S. E., Amanullah, A. S. M., Daniel, A. and Lovejoy, F. (2000-2001) `Risking the Future: Unprotected Intercourse and AIDS Knowledge Among Female Commercial Sex Workers in Central Bangladesh', International Quarterly of Community Health Education, 20 (3), pp. 343-358.

Habib, S. E., Lovejoy, F. H. and Aspin, C. (2001a) `Hepatitis C Prevalence and Risk Behaviour of Injecting Drug Users in Sydney: A Continuing Concern', Southeast Asian Journal of Tropical Medicine and Public Health, 32 (4), pp. 1-12.

Habib, S. E., Bhuiyan, A. Q., Hossain, K. T. and Lovejoy, F. (2001b) `Drug use and Social Circumstances: A Study Among Different Target Groups in Northwest Bangladesh', Japanese Journal of Alcohol Studies & Drug Dependence, 36 (2), pp. 107-123.

Harding, W. M. and Zinberg, N. E. (1977) `The effectiveness of the subculture in developing rituals and social sanctions for controlled drug use'. In Du Toit BM (Ed.), Drugs, rituals and altered states of consciousness (pp. 111-134). Rotterdam: Balkema.

Hayes, M. V. (1992) `On the epistemology of risk: language, logic and social science', Social Science & Medicine, 35 (4), pp. 401-407.

Koester, S. (1996) `The process of drug injection: applying ethnography to the study of HIV risk among IDUs'. In T. Rhodes, R. Hartnoll (Eds.), AIDS, Drugs and Prevention: Perspectives on Individual and Community Action (pp. 133-148). London: Routledge.

Loxley, M., Philips, M., Carruthers, S. J. and Bevan, J. S. (1997) `The Australian Study of HIV and Injecting Drug Use. Part I: Prevalence for HIV, hepatitis B and hepatitis C among injecting drug users in four Australian cities', Drug and Alcohol Review, 16 (3), pp. 207-214.

Loxley, W. (1998) Young Injecting Drug Users and the Risk of HIV/AIDS: Psychosocial and Environmental Constraints on Safer Drug Using and Sexual Behaviour. Perth: Curtin University of Technology, National Centre for Research into the Prevention of Drag Abuse.

Loxley, W. (2000) `Doing the possible: harm reduction, injecting drug use and blood borne viral infections in Australia', International Journal of Drug Policy, 11 (6), pp. 407-416.

Lupton, D. (1993) `Risk as moral danger: The social and political functions of risk discourse in public health', International Journal of Health Services, 23 (3), pp. 425-435.

MacDonald, M. A., Wodak, A. D., Dolan, K. A., Van Beek, I., Ali, R., Cunningham, P. H. and Kaldor, J. M. (2000) `Hepatitis C virus antibody prevalence among injecting drug users at selected needle and syringe programs in Australia, 1995-1997. Collaboration of Australian NSPs', Medical Journal of Australia, 172 (2), pp. 57-61.

Neaigus, A., Friedman, S. R., Curtis, R., Des Jarlais, D. C., Furst, R. T., Jose, B., Mota, P., Stepherson, B., Sufian, M., Ward, T. and Wright, J. W. (1994) `The relevance of drug injectors' social and risk networks for understanding and preventing HIV infection', Social Science & Medicine, 38 (1), pp. 67-78.

Pollak, M., Paicheler, G. and Pierret, J. (1992)AIDS: A Problem for Sociological Research. London: Sage.

Rayner, S. (1992) `Cultural theory and risk analysis'. In S. Krimsky & D. Golding (Eds.), Social Theories of Risk (pp. 83-115). Westport, CT: Praeger.

Rhodes, T. (1995) `Theorizing and Researching `Risk': Notes on the Social Relations of Risk in Heroin Users' Lifestyles'. In P. Aggleton, P. Davies and G. Hart (Eds.), AIDS: Safety, Sexuality and Risk (pp. 125-143). London: Taylor & Francis.

Rhodes, T. (1997) `Risk theory in epidemic times: sex, drags and the social organisation of `risks behaviour', Sociology of Health & Illness, 19 (2), pp. 208-227.

Rhodes, T and Cusick, L. (2000) `Love and intimacy in relationship risk management: HIV positive people and their sexual partners', Sociology of Health & Illness, 22 (1), pp. 1-26.

Rhodes, T. (2002) `The `risk environment': a framework for understanding and reducing drag-related harm', International Journal of Drug Policy, 13 (2), pp. 85-94.

Schiller N. G., Crystal, S. and Lewellen, D. (1994) `Risky Business: The Cultural Construction of AIDS Risk Groups', Social Science & Medicine, 38 (10), pp. 1337-1346.

Taylor, A., Goldberg, D., Hutchinson, S., Cameron, S., Gore, S. M., McMenamin, J., Green, S., Pithie, A. and Fox, R. (2000) `Prevalence of hepatitis C virus infection among injecting drug users in Glasgow 1990-1996: are current harm reduction strategies working?', Journal of Infection, 40 (2), pp. 176-183.

Van De Ven, P., Kippax, S., Crawford, J. and Rodden, P. (1999) `Injecting drug use and knowledge and self-report of hepatitis C among Australian gay and homosexually active men', Drug and Alcohol Review, 18 (3), pp. 271-277.

Watson, R., Crofts, N., Mitchell, C., Aitken, C., Hocking, J. and Thompson, S. (1999) `Risk factors for hepatitis C transmission in the Victorian population: a telephone survey', Australian and New Zealand Journal of Public Health, 23 (6), pp. 622-626.

Wightman, A. and Toomey, M. (1998) Contact: hepatitis C diagnosis booklet. Queensland: Hepatitis C Council of Queensland.

Shah E Habib *

* Dr Shah E Habib, School of Sociology, The University of New South Wales, Sydney 2052, Australia