Injecting drug users

toolkit material

In Europe, behavioural surveillance has mainly been conducted using service-based or community-based sampling.

Table 1. below provides selected features of available methods for collecting behavioural surveillance data among IDU.

The second table presents the best three methods proposed by the experts according to two different contexts.

Table 1. Selected features of available methods for collecting behavioural data among IDU 

Design Advantages Disadvantages Comments
Service & venue based approaches*
  • A pragmatic approach with services & venues providing a sampling frame
  • Well-suited to behavioural surveillance where resources are constrained, or where there is extensive provision of well- used and easy to access services
  • Potentially low cost, as can use existing structures
  • Sample may not be representative of the whole IDU population, and may miss those who are very marginalised or stigmatised
  • Not suitable to situations where most of the population is not in contact with services, and where service provision is poor or limited, or where there are no other venues
Variations on this approach have been adopted for surveillance sys-tems in several countries (inc. France, UK and Australia), relati-vely low cost and sustainable
Community based - outreach sampling 
  • Can access marginalised groups
  • Can sample purposively to target specific sub-groups
  • Marginalised populations difficult to reach by other methods can be targeted 
  • Sample will probably not be representative of the whole IDU population, biases difficult to spot
  • May be difficult and expensive if IDU population not easy to access 

This approach has been used in research studies in many locations worldwide , and used in some surveillance studies

Community based - snowball sampling    
  • Can access marginalised groups
  • Can sample purposively to target specific sub-groups
  • Marginalised populations difficult to reach by other methods can be targeted and may be more easily reached through snowballing than outreach sampling
  • Sample will probably not be representative of the whole IDU population, biases difficult to spot
  • May be difficult and expensive as IDU population can be hard to access
  • Success will be influenced by the extent and strength of IDU networks

This approach has been used in research studies in many locations worldwide , and used in some surveillance studies

Time location sampling 
  • As for Community-based outreach sampling, but with mapping of the population to establish a time- and location- based sampling frame with the aim of improving representativeness
  • If mapping is thorough enough can allocate sampling quotas to sites
  • Sample may not be representative of the whole IDU population, and may miss those marginalised or stigmatised, if these cannot be mapped
  • May be difficult as IDU population can be hard to access
  • Biases difficult to identify
  • Very costly

This approach has mostly been used for research and evaluation studies in some developing and transitional countries

Respondent driven sampling 
  • Aims to recruit a broad-based sample through controlled chain referral, which can then be weighted to reduce recruitment biases
  • Should, in theory, access most of the sub-groups of the population 
  • Sample may not be representative of the whole IDU population, and may still miss those margi¬nalised or stigmatised, if these are not well networked
  • Logistically complex, and only works when IDU are well networked
  • Biases difficult to identify 
  • Costly 
This recent approach has been used in a number of research studies worldwide

Treatment demand indicator 

  • Pragmatic approach using an established system to gather  clinical assess¬ment information from those in contact with addiction treatment services
  • Very low cost and appropriate where treatment services are widely provided and easy to access
  • Can provide comprehensive national data
  • Sample will not be representative of the whole IDU population, and will miss those not in treatment, including those who are very marginalised or stigmatised
  • Not suited to situations where most of the population is not in contact with addiction treatment services, and where provision of such services is poor or limited
  • As relies on clinical reports, data are not directly provided by clients and so may be subject to biases
  • Data collected likely to be limited
Used in a number of European countries, though often accom-panied by other systems (to provide more in depth information and/or corroboration)

*including recruitment through: low-threshold services; treatment services; community venues; and/or social services 

The second table presents the best three methods proposed by the experts according to two different contexts:

  • the population is reachable in known settings and is not severely stigmatised
  • the population is not very well known, not easy to reach, and/or stigmatised

Examples of good practice are provided (references)

Table 2. Best three methods to access IDU according to context

 

   Best methods  Main indication for preference
   A. The population is reachable in known settings and is not severely  stigmatised

 1

Service-based a) b)
  • Where there is extensive provision of well used easy to access services targeted at IDU, such as NEX and OST programmes

 2

Venue based e)
  •  Where there is a range of venues/services through which IDU can be accessed, these could include low and high threshold health services, homeless hostels, drop-in centres and social venues/settings

 3

Those entering addiction treatment g) h) i)
  • Settings where there is extensive provision of addiction tretament services, but more limited provision of other services and few other venues where population can be accessed
   B. the population is not very well known, not easy to reach, and/or stigmatised  

 1

 RDS c) d)
  •  Formative studies, and where population is well networked

 2

 Time location f)
  •  Formative studies, and where population is not well networked

 3

Community outreach sampling j) k)
  • Formative studies, and where population is poorly networked and mapping is difficult

 

 

 

 

 

*including recruitment through: low-threshold services; treatment services; community venues; and/or social services 

The second table presents the best three methods proposed by the experts according to two different contexts:

  • the population is reachable in known settings and is not severely stigmatised
  • the population is not very well known, not easy to reach, and/or stigmatised

Examples of good practice are provided (references)

Table 2. Best three methods to access IDU according to context

   Best methods  Main indication for preference
   A. The population is reachable in known settings and is not severely  stigmatised

 1

Service-based a) b)
  • Where there is extensive provision of well used easy to access services targeted at IDU, such as NEX and OST programmes

 2

Venue based e)
  •  Where there is a range of venues/services through which IDU can be accessed, these could include low and high threshold health services, homeless hostels, drop-in centres and social venues/settings

 3

Those entering addiction treatment g) h) i)
  • Settings where there is extensive provision of addiction tretament services, but more limited provision of other services and few other venues where population can be accessed
   B. the population is not very well known, not easy to reach, and/or stigmatised  

 1

 RDS c) d)
  •  Formative studies, and where population is well networked

 2

 Time location f)
  •  Formative studies, and where population is not well networked

 3

Community outreach sampling j) k)
  • Formative studies, and where population is poorly networked and mapping is difficult

 

 

Examples
(a) Unlinked Anonymous Programme’s Survey of IDU in the UK.
http://www.hpa.org.uk/web/HPAweb&HPAwebStandard/HPAweb_C/1202115519183. See also: 
Sustained increase in the sharing of needles and syringes among drug users in England and Wales.

VD Hope, PA Rogers, L Jordan, T Paine, S Barnett, JV Parry, ON Gill. AIDS 2002; 16:2494-2496. 

Failure to vaccinate injecting drug users against hepatitis B in England and Wales. T Lamagni, VD Hope, K Davison, JV Parry, ON Gill. Communicable Disease and Public Health, 2001; 4:71-2.

(b) The Australian Needle and Syringe Programme Survey
http://www.med.unsw.edu.au/NCHECRweb.nsf/page/Australian+NSP+Survey See also:

Representativeness of injecting drug users who participate in HIV surveillance: results from Australia's Needle and Syringe Program Survey. Topp L, Iversen J, Wand H, Day C, Kaldor J, Maher L; Collaboration of Australian Needle Syringe Programs. J Acquir Immune Defic Syndr. 2008;47(5):632-8

(c)  National HIV Behavioral Surveillance System (NHBS) IDU survey in the USA. Lansky A, Abdul-Quader AS, Cribbin M, et al. Development of an HIV behavioral surveillance system for injecting drug users: the National HIV Behavioral Surveillance System. Public Health Rep 2007;122(Suppl 1):48--55. See also: HIV-Associated Behaviors Among Injecting-Drug Users, 23 Cities, United States, May 2005-February 2006. MMWR April 10, 2009 / Vol. 58 / No. 13 / Pg. 329 - 356

(d)  RDS surveys in Estonia
http://www.tai.ee/failid/IDU_risk_behaviour_and_HIV_prevalence_study_2005.pdf See also:

High-prevalence and high-estimated incidence of HIV infection among new injecting drug users in Estonia: need for large scale prevention programs. Anneli Uusküla, Mart Kals, Kristiina Rajaleid, Katri Abel, Ave Talu, Kristi Rüütel, Lucy Platt, Tim Rhodes, Jack DeHovitz, and Don Des Jarlais. J Public Health 2008 30: 119-125.

(e) Coquelicot Study in France
http://www.invs.sante.fr/beh/2006/33/index.htm See also:

Impact of a harm-reduction policy on HIV and hepatitis C virus transmission among drug users: recent French data--the ANRS-Coquelicot Study. M Jauffret-Roustide, J Emmanuelli, M Quaglia, F Barin, P Arduin, A Laporte, J-C Desenclos. Subst Use Misuse. 2006 ;41 (10-12):1603-21 17002994.

(f) Assessment of HIV Testing of Urban Injection Drug Users: Implications for Expansion of HIV Testing and Prevention Efforts. Robert Heimer, PhD, Lauretta E. Grau, PhD, Erin Curtin, MPH, Kaveh Khoshnood, PhD and Merrill Singer, PhD. American Journal of Public Health 2007; 97:110-116.

(g) Luxemburg National Drug Monitoring System: RELIS. http://www.relis.lu/

(h) French TDI system: RECAP
http://www.ofdt.fr/ofdtdev/live/english-tab/engpubli/tends/tend54eng_fr.html

(i) EMCDDA Treatment Demand Indicator.
http://www.emcdda.europa.eu/themes/key-indicators/tdi

(j) Behavioural monitoring of intravenous drug users (IDU) in Catalonia
http://www.ceescat.org/Index_Ing.htm

(k)  Repeated Community Surveys in London. HIV prevalence and risk behaviour among female injecting drug users in London, 1990 to 1996. Judd, Ali; Hunter, Gillian M.; Maconochie, Noreen; Hickman, Matthew; Parry, John V.; Renton, Adrian M.; Stimson, Gerry V. 1999 AIDS; 13:833-837.