Temporal trends in highly active antiretroviral therapy initiation among injection drug users in Baltimore, Maryland, 1996-2008

Abstract

Background: We characterized temporal trends in highly active antiretroviral therapy (HAART) initiation (1996-2008) among treatment-eligible persons in a community-based cohort of current and former injection drug users (IDUs) in Baltimore, Maryland.

Methods: The AIDS Linked to the IntraVenous Experience (ALIVE) cohort has been observing human immunodeficiency virus (HIV)-positive IDUs since 1988. HAART eligibility was defined as the first visit after 1 January 1996 at which the patient's CD4(+) cell count was <350 cells/microL. Temporal trends and predictors of HAART initiation were examined using chi(2) tests for trend and lognormal survival models.

Results: The median age of 582 HAART-eligible IDUs was 41 years; 75% of the subjects were male, 97% were African American, and 60% were active IDUs. Of these 582 individuals, 345 initiated HAART over 1803 person-years (19.2 subjects per 100 person-years; 95% confidence interval, 17.2-21.3 subjects per 100 person-years); there was no statistically significant temporal trend in HAART initiation. Independent predictors of delayed initiation included heavy injection drug use; having a prior AIDS diagnosis, having a lower CD4(+) cell count, having a usual source of care, and having health insurance were predictors of more-rapid initiation. The delay between eligibility and initiation decreased among those who became eligible most recently (2003-2007), compared with those in earlier periods (1996-2002); however, a substantial number of patients who became eligible in recent years either initiated HAART after a substantial delay or did not initiate HAART at all.

Conclusions: We failed to observe substantial improvement in HAART initiation among current and former IDUs over a 12-year period; heavy use of injection drugs remains the major barrier to HAART initiation and to consistent HIV care. The fact that many IDUs initiate HAART after a significant delay or do not initiate it at all raises concern that disparities in HIV care for IDUs remain at a time of simplified antiretroviral regimens and increasing adoption of earlier treatment.

Figure 1. Temporal trends in HAART initiation in the ALIVE cohort, 1996–2008

Incidence rates per 100 person years (PY) and 95% confidence intervals (CI) among treatment eligible IDUs (n=582) by calendar year are shown. There was no statistically significant linear trend (solid line). Dashed lines in 1998 and 2002 reflect changes in guidelines for antiretroviral therapy [–22].

Figure 2. Temporal trends in initial HAART regimen among 345 HAART initiators in the ALIVE cohort, 1996–2008

Percents are calculated as the number of persons on a particular type of regimen in a calendar year divided by the number of HAART initiators in that calendar year.

Figure 3. Temporal trends in treatment initiation in the ALIVE cohort, 1996–2008

This figure presents data on trends by calendar year of HAART initiation (x axis) and calendar period of eligibility. Calendar period of eligibility is defined as 1996–98, 1999– 02, 2003–07. Lines represent the median months between eligibility for treatment and initiation for each of the three periods of eligibility. Bars represent the median CD4 cell count at the time of initiation for each of the three periods of eligibility. Listed at the bottom of the figure are (1) the numbers of individuals who initiated HAART in each year according to the calendar period of when they became eligible; and (2) the numbers of individuals who were still under study follow-up but had not yet initiated HAART by that calendar year according to the calendar period of when they became eligible.