Decreasing initiation of HAART and treatment delay among injection drug users, 19962006, Baltimore,

1
Decreasing initiation of HAART and treatment
delay among injection drug users, 1996-2006,
Baltimore, MD USA

• DD Celentano1, J Astemborski1, SH Mehta1, GD
  Kirk1,2
• 1 Johns Hopkins Bloomberg School of Public Health
• 2 Johns Hopkins School of Medicine

2
Background

• Injection drug users (IDUs) have lagged behind
  other groups with respect to initiating highly
  active antiretroviral therapy (HAART)
• We previously demonstrated that HAART initiation
  was slowly increasing among a cohort of IDUs in
  Baltimore, MD, USA between 1996 and 1999
  (Celentano et al, JAMA 2000)
• Inconsistent health care access utilization,
  use of suboptimal (non-HAART) regimens and
  continuing drug use were identified as barriers
  to HAART initiation

3
Objective

• As treatments have been simplified and concerns
  related to adherence diminished, IDUs would be
  expected to more readily initiate HAART
• We characterized temporal trends and predictors
  of HAART initiation among treatment eligible IDUs
  from January 1996 through December 2006

4
Methods

• Study population
• AIDS Linked to the IntraVenous Experience (ALIVE)
  
• 2946 IDUs enrolled in 1988 (24 HIV positive)
• Additional recruitments in 1994-95, 1998
• A total of 1206 HIV positive recruited
• Study sample
• 737 HIV positive in follow-up in January 1996
• 629 HAART eligible (CD4lt350 cells/?l) IDUs with gt
  2 follow-up visits

5
Statistical Analysis

• Date of HAART initiation defined as midpoint of
  last visit where HAART was not reported and first
  visit where HAART was reported
• HAART was considered to be a regimen of 3 drugs
  consisting of at least a PI, NNRTI or Abacavir
• Annual rates of HAART initiation per 100
  person-years of observation were calculated to
  characterize temporal trends
• Poisson regression was used to identify
  predictors of HAART initiation

6
Results Characteristics at HAART eligibility

• Median age, 41 years (IQR 36-45)
• 97 African American
• 25 Female
• 71 prior AIDS diagnosis
• 87 were ART experienced
• Median CD4 at eligibility, 230 (IQR 134, 294)
• Median follow-up 4.3 years (IQR 1.6,7.8)

7
Results Outcomes
8
Temporal trends in HAART initiation, 1996-2006
9
Temporal trends in initial HAART regimen,
1996-2006
10
Temporal trends in treatment delay, 1996-2006
Plt0.001 for both (test for trend)
11
Adjusted predictors of HAART initiation, 1996-2006
12
Conclusions

• In Baltimore IDUs, overall rates of HAART
  initiation have remained relatively stable but
  are not improving.
• Delays between eligibility and HAART initiation
  are increasing
• This is consistent with data from the Johns
  Hopkins HIV clinic (Baltimore, USA) which
  suggests that patients are presenting for care at
  later stages and lower CD4 cell counts
  (Keruly J et al, 2008)

13
Conclusions

• Identified barriers to HAART included daily
  injection and use of other drugs
• Facilitators of HAART initiation were medical
  care access and continuity
• A significant number of eligible persons appear
  intransigent to initiation of therapy despite
  counseling, examination, routine testing and
  active referral

14
Acknowledgements

• ALIVE co-investigators
• ALIVE study staff participants
• National Institute on Drug Abuse
• (DA04334, DA12568)