THE 'BOULE DE NEIGE' PROJECT

LOWERING THE THRESHOLD FOR AIDS PREVENTION AMONG INJECTING DRUG USERS

Charles Kaplan and colleagues examine projects in four cities in Europe which use the 'snowball' approach to AIDS prevention

INTRODUCTION

The spread of HIV and related diseases, e.g. AIDS, is the most critical public health problem of this century. Globally, the numbers of infected people are rising exponentially. In the West, the largest numbers of infected had been in the gay community, but this is rapidly being surpassed by injecting drug users (IDU). By the time you read this paper the predicted total accumulated cases of IDUs will have surpassed the gay population. Thus, in a surveillance overview of AIDS in the European Community (EC), Downs et al. (1990) have concluded that although the incidence of AIDS in the gay male population is a function of linear growth, even higher growth rates are reported for IDUs and those persons infected by heterosexual contact. However, on the optimistic side, the EC projections show interesting Member State differences that can be accounted for, in part, by specific policies and programmes which imbue optimism in the IDU community producing positive changes in behaviour (Hart, 1990).

There are necessary preconditions to signal this optimism. One, of course, is the provision of non moralistic treatment services that fit with the articulated needs of the IDU community. Creative methadone programmes are a good example. Public health and social services such as needle exchange which increase the availability of clean syringes are also essential tools which spread optimism. However, experience has shown that these first-line services are necessary, but not at all sufficient. They must be 'bundled' with both treatment and services that lower the threshold towards the 'zero-line', i.e. require more involvement and responsibility from the IDU community itself in AIDS prevention. Thus, the evaluation of needle exchange in Amsterdam showed limited effect and recommended intensive counselling (Hoek et al., 1989). In Rotterdam, experiments as wide-ranging as needle dispensing machines and 'collective' needle distribution, with the active participation of IDUs and house dealers, have been assessed (Kaplan et al.,1989; Grund etal.,1992).

Going beyond first-line service conceptions requires a systematic effort in 'lowering the threshold' between the resources needed to effect change and the IDU community itself. Outreach and empowerment are now common strategies for accomplishing these goals. Thus, the principal investigator of the US National Institute on Drug Abuse-supported AIDS outreach programme in Miami, Clyde McCoy, answered the question, 'What was the most significant finding of your programme ?' in one word, 'People' . Not only is there an extensive population out there in need, but 'there is so much more that can be done' (Joseph, 1991-1992,8-9). In this regard experiments involving person-to-person approaches represent the frontiers of AIDS prevention for drug users.

THE SNOWBALL: A METAPHOR FOR AIDS PREVENTION

One metaphor for imagining such a person-to-person approach is a snowball. A snowball grows by dynamically linking individual units through tracing their relationships. The snowball sampling design has been given increasing attention in the research of hidden populations of drug users (see Bieleman et al., 1993 ). What has been given less attention is the snowball as a form of 'self-organisation' of drug users to 'animate' the community for AIDS prevention. As a scientific concept, self-organisation has a recent origin in the study of self-regulating ecosystems. Recently, this concept has been applied to the evaluation of human service systems (Odum, 1988). Self-organisations can be viewed as basic ecological units that produce, store and copy essential information. A self-organisation of drug users may be conceived of as a special case of these processes, with reference to information about drugs.

In this paper, a project applying the snowball metaphor for AIDS prevention in the European region or Euro-region of Belgium, Germany and the Netherlands will be described. The snowball metaphor is useful in highlighting the structure and function of drug user self-organisations as information transformers. IDUs may be assumed to have the most information about the processes that are related to the spread of HIV in their group. What is needed is a means of animating and transforming this information into realistic beliefs, attitudes and cognitions, reducing the risks-f HIV infection. The snowball project described in this paper responds to this need.

The theoretical background of this approach rests on the great potential of paraprofessionals in health promotion programmes. Research in health education supports the commonly held belief that the effective ness of the health promotion message increases with similarity and trust between educator and target (Bracht and Gleason,1990). In AIDS prevention among homosexual men this approach has been successfully applied ( Bye, 1990) . Those who are culturally sensitive, and therefore able to frame messages in meaningful language in a realistic social context, are most likely to make an impression on people. Thus, a current or former IDU who tells other users that it is both desirable and possible to stop sharing unclean needles will be vastly more convincing than the local health department professional educator. Therefore, the rationale of the 'Boule de Niege' outreach programme is to employ IDUs or other 'street-wise' people

to provide face-to-face information about AIDS to other IDUs with the goal to animate and sensitise them for safe injection and safe sex. This prevention strategy also uses elements of the sociological conceptualisation of drug abuse in that the ex-addicts may not only be better at communication with current addicts, but also serve as 'role-models' for successful behaviour (Friedman et al., 1986).

THE SELF-ORGANISATION OF THE 'BOULE DE NEIGE'

'Boule de Neige' (BN ) in the French language literally translates into 'snowball'. The model was first developed in 1989 in the Belgium cities of Brussels and Charlois. The experiences of the Belgium programme were positive in involving ex-addicts or street-wise people. Furthermore, animating and transforming information by using IDUs was not politically threatening. A first idea was to implement a Euro-region needle-exchange programme analogous to the Netherlands South Limburg installations. However, because of national differences in this politically sensitive area, a common programme politically acceptable throughout the Euro-region was necessary. An intensive counselling alternative was adopted as a feasible, relevant and timely alternative.

The strength of the BN is the partnership between professionals from mental health agencies and drug users, the so-called 'jobists'. The potential value of the jobist in AIDS prevention efforts for IDUs is based upon the assumption that people drawn from a targeted organisation (or cultural subgroup) will be familiar with the cultural environment in which health behaviour change is to take place. Such paraprofessionals should help assure compatibility between the health promotion programme and organisational elements. They should be able to translate concepts in terms that can be understood by peers and avoid cultural pitfalls that professional staff may not be aware of. Paraprofessionals may also have credibility and immediate access to their peers as members of the target social networks, and thus provide immediate entry and points of change. Lastly, the paraprofessional may foster feelings of local ownership and commitment to both the programme and its goals.

Not only does the project bridge the gap between professionals and target client populations, it is the first collaborative project among drug health services in the Euro-region. Thus, an important by product of this collaboration is the increased knowledge among the respective European Community Member States of each other's problems, target groups and development.

The face-to-face setting of the BN provides for extensive information transfer by clarification through questions. This is a basic property of effective self organisation in any ecosystem. Essential to this mode of information transfer is the provision of modulations of the emotional tone of the message in order to provoke sufficient discomfort about sharing injection equipment to motivate behaviour change, but not so much that anxiety prevents behaviour change. The paraprofessional provides reliable information about how used drug injection equipment can be decontaminated and sex can be safer.

DESIGN AND IMPLEMENTATION OF THE 'BOULE DE NEIGE'

Six cities in the Euro-region of Belgium, Germany and the Netherlands, consisting of four distinct cultural areas, were selected for this project: Aachen (Germany); Hasselt (Flemish Belgium); Maastricht, Heerlen, Sittard (the Netherlands); Liege (French Belgium). Key people with knowledge of the drug scenes in the Euro-region were contacted and asked to help in identifying the members of the target population who comprised the pool of paraprofessionals. These key people were designated 'animators'. Potential paraprofessional 'jobists' were recruited by the animators. The selection criteria for the jobists were experience-based knowledge of the drug scene and the desire to be a positive role-model to drug users. The personal motivation of the paraprofessional is crucial for the success of the programme. The reasons for participating in the project varied, the two predominant ones being commitment to the target group and financial rewards. A four-session training course was given to the paraprofessionals. The learning topics included: objectives of the BN programme; target groups; review of prevention material; safe drug use; AIDS and addiction; sexuality (attitude, how to question it, risk behaviour ); questionnaire (content, formulation); and payment.

Professional health service workers were engaged with the organisation, administration, research tasks, development of the teaching outlines and training of the jobists. The role of the jobist was specifically defined and included the following functions: promote safe sex among drug addicts; administer the questionnaires collecting information on knowledge, attitude and behaviour concerning AIDS among the target group; provision of a prevention 'kit' containing written material, condoms and clean needles; identify any hidden subgroups in the local drug scene; supply reports to the programme co-ordinator regarding process and problems; play an active part in the development of the questionnaire and prevention materials.

METHOD

As a result of intensive field work, hundreds of drug users in the Euro-region were contacted, questioned and otherwise involved. A database has been constructed quantifying the output of the field work. The l database consists of the records of a total of 318 inter views of drug users contacted by the prevention team in the four cities. The interview instrument was primarily quantitative and registered responses to items about attitudes, risk behaviours, background demographic characteristics, drug and sexual history and relationship to the prevention programme. A total of 60 variables made up the database. The data are stored in an SPSS-X system file. Data processing required the cleaning of 32 variables before analysis could be undertaken.

The instrument was generally uniform in all the cities. However, 'site-specific' variation on several items existed. This required small amounts of recoding to improve the consistency of the database. For example, a new variable had to be constructed because the numbering of the interviews was sometimes the same or each jobist had his or her own sequence. In Liege, the variable concerning 'frontloading' had a trichotomous response (sometimes, often, never) whereas the rest of the sites had a dichotomous response structure (yes/no). The Flemish instrument included three more subquestions than existed in the standard instrument. It should be noted that 'perfect' standardisation of instruments in multisite studies is neither practical nor desirable. Some local variation is necessary in order to improve the general validity of the instrument (see Bielemanetal., 1993).

Applications for support of a full process and effect evaluation of the BN programme are pending. In this paper, the results are presented of a comparison of each city regarding the distribution of specific classes of variables indicating background, risk behaviour and services relevant to AIDS prevention. The three Dutch cities are taken together and described as one region: South Limburg. The results of this descriptive exercise are discussed - more from the point of view of suggesting interesting themes for further evaluation than for the arrival at any definitive conclusions.

RESULTS

Table 1 presents the frequency distributions of specific key background, risk behaviour and services variables across the Euro-region cities.TABLE 1: BACKGROUND OF DRUG USERS, RISK BEHAVIOUR OF DRUG USERS AND DRUG RELATED SERVICES

DISCUSSION

These data suggest some interesting themes for future evaluation. Although the six cities are loosely integrated in a single European region, there are certain key variables where impressive variation can be seen, most notable being the contact with the mental health services. South Limburg has over two-thirds of the target population in contact. This is more than twice the percentage in the next highest city, Liege (30 per cent). Aachen and Hasselt fall to below the 20 per cent level. This difference is undoubtedly to do with south Limburg being integrated into the general Dutch low threshold public mental health system. The Consultation Bureau for Alcohol and Drugs (CAD) in South Limburg is an integral part of a national system of CADs. Through this system, innovations such as needle exchange can be rapidly implemented throughout the country once a critical threshold of acceptance is passed. No such systematic links at the practice level exist in the other two countries.

Furthermore, the South Limburg scene has been the field of a significant network study on-going since 1982. This study has provided important practical feedback to the CAD which, in turn, improved their outreach capabilities. For example, the study showed that the CAD had been under-reporting its contacts with the South Limburg drug scene. Whilst the official figures registered a 28 per cent level of contact based upon accepted prevalence estimates, the fieldwork showed an actual 45 per cent level (Baars et al., 1989). The researchers argue critically that this under-reporting is the result of a systematic preference in the CAD reporting system for psychotherapeutically served clients. This psychological 'a priori' obscures the registration of the other kinds of contacts the CAD has in the heroin scene through the social networks of their clients. Since the publication of the research article in 1989, adjustments have been implemented to make the South Limburg drug services more sensitive to social network factors and strategies. It can be hypothesised that a similar psychological 'a priori' may be limiting the contacts of the public mental health services in the German and Belgium cities.

High contact with the public mental health services co-varies with relatively high positive outcomes in key AIDS prevention variables. Thus, in South Limburg, on three of the four key risk behaviour variables, drug users have the most positive levels. Support for this hypothesis comes from the low contact case of Hasselt which has, correspondingly, the most negative levels of risk behaviour. These associations provide firm support of Hart's ( 1989) conclusions that differences in European policy and programme can have significant influence on differences in AIDS prevention outcomes in Europe.

An interesting and contrasting case is presented by Aachen. Here, contact with services seems, from the table, to be relatively low, but AIDS risk-reduction behaviour is relatively high. However, this interpretation is misleading as many Aaachen IDUs go across the border to Heerlen to exchange their syringes. Therefore, syringes are relatively easily available even though contact with German public health services is relatively low. This case provides an interesting clue to the interaction of self-organisation and public mental health services. In the Aachen case there are the highest positive percentages in two of three key background variables related to AIDS prevention, i.e. the lowest percentage still using drugs and the highest awareness of having no HIV infection. On the key risk behaviour variable of changed behaviour because of AIDS, the percentage indicating a positive outcome is virtually identical to the South Limburg case where contact is the highest. The Aachen data provide support for the view that, under conditions where both syringes (even if the IDUs have to cross a open border to get them) and condoms are readily available, the self-awareness and self control of AIDS seropositivity and drug use facilitated by self-organisation leads to positive prevention behaviour complementing the contacts to the public mental health services.

This line of reasoning finds support in considering that both Aachen and South Limburg have the highest negative outcomes on the variable of an increase in shooting style in the last 4 weeks. Both cities seem to have drug-using populations which are relatively heterogeneous in terms of their drug-using lifestyles. Extremely problematic drug-using behaviour, in relation to AIDS, seems constantly present in both scenes. The limitations of self-organisation may therefore lie in their relatively low capacity to suppress negative behaviour of extremely problematic target group members. This may be the appropriate role for professional services - helping to lower the threshold on negative behaviour which cannot be effectively controlled by existing self organisations.

CONCLUSION

The research and development situation posed by the Euro-region provides a compact social laboratory where the 'ecosystem' formed by the state of drug abuse services and the state of self-organisations can develop a new 'synergy'. Cities such as Liege which represent moderate levels of risk behaviour with moderate levels of service are especially interesting for in-depth description and specification of latent factors lying outside the ecosystem of services and self-organisation in this paper. Clearly, the 'Boule de Neige' project was a concerted attempt at a level beyond national borders to animate the entire ecosystem, in all its cultural variation, to respond to an epidemic that is just starting (and therefore preventable). More significantly, the project represents a clear European case of the spirit of optimism and creativity, in trying to meet the AIDS problem head-on at an early stage. Problem denial has been the norm in similar situations. In the case of the Euro-region, however, the 'Boule de Neige' project has come to symbolise a true community approach to drug abuse and AIDS control. Projects based upon the community approach underline the fundamental importance of a multistage strategy of social rehabilitation, characterised by the outreach work that precedes the later options of drug treatment and recovery (Sell, 1990). Projects such as these concretely point to horizons beyond the threshold-lowering acts of condom distribution and needle exchange in AIDS prevention.

REFERENCES

Baars, H.M.J., Schaeken, M.G., Mevius, F.J., Uffing, J.l.F. and Driessen, G.A.M. ( 1989) Het bereik van de ambulante verslavingszorg in een heroine scene. Tijdschrift voor alcohol, drugs en andere psychotrope stoffen, 15 (3), 91-105.

Bieleman B.. Diaz, A., Merlo, G. and, Kaplan, C.D. (1993) Lines: over Europe: A Comparative Study of the Nature and Extent of Cocaine Use in Barcelona, Rotterdam and Turin. Amsterdam Lisse: Swets and Zeitlinger.

Bracht, N. and Gleason, J. (1990) Strategies and structures for citizen partnerships. In: Bracht, N. (ed.) Health Promotion at Community Level, pp. 109-139. London: Sage.

Bye, L.L. (1990) Moving beyond counseling and knowledge enhancing interventions. In: Ostrow, D.G. (ed.) Behavioural Aspects of AIDS, pp. 157-161. New York and London: Plenum Medical.

Des Jarlais, D.C., Friedman, S.R. and Wood, J.S. (1990) Intravenous drug use and AIDS. In: Ostrow, D.G. (ed.) Behavioural Aspects of AIDS. New York and London: Plenum Medical.

Downs, A.M., Ancelle-Park, R.A., and Brunet, J.B. (1990) Surveillance of AIDS in the European Community: Recent trends and predictions to l991.AIDS, 4, 1117-1124.

Friedman, S.R., Des Jarlais, D.C. and Sortheran, J.L. (1986) AIDS health education for intravenous drug users. Journal of Health Education Quarterly, 13,383-393.

Grund, J.P., Blanken, P., Adriaans, N.F.P., Kaplan, C.D., Barendregt, C. and Meeuwsen, M. (1992,> Reaching the unreached: An outreach model for' on the spot' AIDS prevention among active, out of treatment drug addicts. In: O'Hare, P.A., Newcombe, R., Mathews, A., Buning, E.C. and Drucker, E. (eds) Reduction of Drug-related Harm. London: Routledge.

Hart, G. (1990). Drug users - epidemiology. AIDS Care, 2(4),395-397.

Hoek, J.A.R., van den, Haastrecht, H.J.A. and van Coutinho, R.A. ( 1989) Risk reduction among intravenous drug users in Amsterdam under the influence of AIDS. American Journal of Public Health, 79, 1355-1357.

Joseph, L. (1991-1992) Reaching a hidden population, Winter, 8-9.

Kaplan, C.D., Grund, J.P. and Adriaans, N.F.P. (1989) Beyond needle exchange: Hidden risks in behavioural patterns of IV drug users for HIVinfection and the prospects for prevention. In: van Steijn, P. (ed.) AIDS: A Combined Environrnental and Systems Approach, pp. 167-176 Amsterdam/Lisse: Swets and Zeitlinger

Odum, H.T. (1988) Self-organization, transformity, and information. Science, 242, 1132-1139.

Sell, H. (1990) The community approach to drug abuse control. Paper presented at the ESCAP/ SEARO/ WPRO meeting on the rehabilitation of drug .dependent persons. Bangkok, January 25-26.