Decline in HIV incidence and injecting, but not in sexual risk behaviour, seen in drug users in Amst

1
Decline in HIV incidence and injecting, but not
in sexual risk behaviour, seen in drug users in
Amsterdam a 19-year prospective cohort study

• C Lindenburg1, A Krol1, C Smit1, M Buster1, R
  Coutinho2,3, M Prins1,2

Publicated in AIDS. 2006 Aug 2220(13)1771-5.
1 Health Service of Amsterdam, Cluster
Infectious Diseases, Department of Research,
Amsterdam 2 Center for Infection and Immunity,
Amsterdam 3 National Institute for Public Health
and the Environment, Center for Infectious
Disease Control, Bilthoven, The Netherlands

2
Introduction

• Injecting Drug Users (IDU) at high risk for human
  immunodeficiency virus (HIV) infection
• Data on HIV incidence and risk behaviour among DU
  in the highly active antiretroviral therapy
  (HAART) era are scarce

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
3
Study Objective

• Trends in HIV incidence and injecting and sexual
  risk behaviours among HIV-negative DU of the
• Amsterdam Cohort Study (ACS)

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
4
ACS

• Initiated in 1985 (n1650)
• Follow-up visits every 4-6 months
• Standardised questionnaires (clinical,
  epidemiological, sexual, drug use, putative route
  of HIV transmission) and blood sample collection

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
5
Outcome variables

• During HIV-negative follow-up time
• Injecting (yes/no)
• Borrowing needles (yes/no)
• Use of needle exchange programs (yes/no)
• Any unprotected sex (yes/no)
• Having a sexual transmitted infection (yes/no)

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
6
Outcome variables

• Putative route of HIV transmission
• Injecting drug use
• Homosexual
• Heterosexual

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
7
Methods

• HIV incidence per 100 person-years using
  person-time techniques
• Trends in risk behaviour OR per calendar year
  increase before and after 1996 using logistic
  regression and GEE
• OR for sexual versus injecting HIV transmission
  route before and after 1996

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
8
Results

• 1315 HIV-negative DU
• 63 male
• 77 European
• Median age at cohort entry 30
• Median year of cohort entry 1991
• 6970 HIV-negative person years of follow-up
• 93 HIV seroconversions

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
9
HIV incidence and proportion of visits per
calendar year at which injecting and sexual risk
behaviour was reported among 1315 DU who were
HIV-negative on ACS entry, 1986-2004.
Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
10
OR per calendar year increase for injecting and
sexual risk behaviour among 1315 HIV-negative DU
in the ACS, 1986-2004.
OR odds ratio, adjusted for age, gender and
calendar year of ACS cohort entry p-value of
the difference between the presented ORs before
and after 1996 STI Sexually transmitted
infection (syphilis, gonorrhoe, herpes simplex or
other STI
Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
11
Results Non-injecting hard drug use

• Reported use of any non-injecting hard drugs in
  the ACS
• 1986 54
• 2004 83

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
12
Exchanged needles in Amsterdam 1983-2003
Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
13
HIV transmission route

• 65 HIV seroconverters included in analysis
• Before 1996
• 50 HIV seroconversions through IDU 8
  HIV seroconversions through heterosexual contact
• After 1996
• 2 HIV seroconversions through IDU 5
  HIV seroconversions through heterosexual contact

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
14
HIV transmission route

• Unadjusted OR of HIV transmission through
  unprotected heterosexual contact after 1996
  compared with period before
• 15.6 (95 CI 2.6-94.6)

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
15
Proportion of HAART users and non-HAART using DU
who reported injecting risk behaviour. C. Smit,
2006.
16
Proportion of HAART users and non-HAART using DU
who reported sexual risk behaviour. C. Smit,
2006.
17
Conclusions

• Steep decline in HIV incidence among DU in ACS
• Strong decline in injecting drug use and
  borrowing needles
• Decline in use of needle exchange units and
  reduction of absolute number of exchanged needles
  per calendar year in Amsterdam
• Shift of injecting to non-injecting hard drugs
  use

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5
18
Conclusions

• DU on HAART do not increase their risk behaviour
• Sexual risk behaviour plays most important role
  in recent HIV transmissions in ACS
• Stable trend in sexual risk behaviour and
  possible increase in STI after 1996
• HIV prevention programmes for DU should pay
  specific attention to the importance of safe sex
  practices

Lindenburg et al. AIDS 2006 Aug 2220(13)1771-5