Aboriginal status is a prognostic factor for mortality among antiretroviral naïve HIV-positive individuals first initiating HAART.

Abstract

Although the impact of Aboriginal status on HIV incidence, HIV disease progression, and access to treatment has been investigated previously, little is known about the relationship between Aboriginal ethnicity and outcomes associated with highly active antiretroviral therapy (HAART). We undertook the present analysis to determine if Aboriginal and non-Aboriginal persons respond differently to HAART by measuring HIV plasma viral load response, CD4 cell response and time to all-cause mortality.
A population-based analysis of a cohort of antiretroviral therapy naïve HIV-positive Aboriginal men and women 18 years or older in British Columbia, Canada. Participants were antiretroviral therapy naïve, initiated triple combination therapy between August 1, 1996 and September 30, 1999. Participants had to complete a baseline questionnaire as well as have at least two follow-up CD4 and HIV plasma viral load measures. The primary endpoints were CD4 and HIV plasma viral load response and all cause mortality. Cox proportional hazards models were used to determine the association between Aboriginal status and CD4 cell response, HIV plasma viral load response and all-cause mortality while controlling for several confounder variables.
A total of 622 participants met the study criteria. Aboriginal status was significantly associated with no AIDS diagnosis at baseline (p = 0.0296), having protease inhibitor in the first therapy (p = 0.0209), lower baseline HIV plasma viral load (p < 0.001), less experienced HIV physicians (P = 0.0133), history of IDU (p < 0.001), not completing high school (p = 0.0046), and an income of less than $10,000 per year (p = 0.0115). Cox proportional hazards models controlling for clinical characteristics found that Aboriginal status had an increased hazard of mortality (HR = 3.12, 95% CI: 1.77-5.48) but did not with HIV plasma viral load response (HR = 1.15, 95% CI: 0.89-1.48) or CD4 cell response (HR = 0.95, 95% CI: 0.73-1.23).
Our study demonstrates that HIV-infected Aboriginal persons accessing HAART had similar HIV treatment response as non-Aboriginal persons but have a shorter survival. This study highlights the need for continued research on medical interventions and behavioural changes among HIV-infected Aboriginal and other marginalized populations.

Full text

Page 1

BioMed Central
Page 1 of 9
(page number not for citation purposes)
AIDS Research and Therapy
Open Access
Research
Aboriginal status is a prognostic factor for mortality among
antiretroviral naïve HIV-positive individuals first initiating HAART
Viviane D Lima1, Patricia Kretz1, Anita Palepu2, Simon Bonner1,
Thomas Kerr1, David Moore1, Mark Daniel3, Julio SG Montaner1,2 and
Robert S Hogg*1,2
Address: 1British Columbia Centre for Excellence in HIV/AIDS, St. Paul's Hospital; Vancouver, Canada, 2Departments of Medicine and/or
Healthcare and Epidemiology, Faculty of Medicine, University of British Columbia, Vancouver, Canada and 3Département de médecine sociale et
préventive, Université de Montréal et Centre de recherche du Centre hospitalier de l'Université de Montréal, Québec, Canada
Email: Viviane D Lima - vivianed@interchange.ubc.ca; Patricia Kretz - pkretz@cfenet.ubc.ca; Anita Palepu - anita@sm.hivnet.ubc.ca;
Simon Bonner - sbonner@stat.sfu.ca; Thomas Kerr - tkerr@cfenet.ubc.ca; David Moore - dmoore@cfenet.ubc.ca;
Mark Daniel - mark.daniel@umontreal.ca; Julio SG Montaner - jmontaner@cfenet.ubc.ca; Robert S Hogg* - bobhogg@cfenet.ubc.ca
* Corresponding author
Abstract
Background: Although the impact of Aboriginal status on HIV incidence, HIV disease progression, and access
to treatment has been investigated previously, little is known about the relationship between Aboriginal ethnicity
and outcomes associated with highly active antiretroviral therapy (HAART). We undertook the present analysis
to determine if Aboriginal and non-Aboriginal persons respond differently to HAART by measuring HIV plasma
viral load response, CD4 cell response and time to all-cause mortality.
Methods: A population-based analysis of a cohort of antiretroviral therapy naïve HIV-positive Aboriginal men
and women 18 years or older in British Columbia, Canada. Participants were antiretroviral therapy naïve, initiated
triple combination therapy between August 1, 1996 and September 30, 1999. Participants had to complete a
baseline questionnaire as well as have at least two follow-up CD4 and HIV plasma viral load measures. The
primary endpoints were CD4 and HIV plasma viral load response and all cause mortality. Cox proportional
hazards models were used to determine the association between Aboriginal status and CD4 cell response, HIV
plasma viral load response and all-cause mortality while controlling for several confounder variables.
Results: A total of 622 participants met the study criteria. Aboriginal status was significantly associated with no
AIDS diagnosis at baseline (p = 0.0296), having protease inhibitor in the first therapy (p = 0.0209), lower baseline
HIV plasma viral load (p < 0.001), less experienced HIV physicians (P = 0.0133), history of IDU (p < 0.001), not
completing high school (p = 0.0046), and an income of less than $10,000 per year (p = 0.0115). Cox proportional
hazards models controlling for clinical characteristics found that Aboriginal status had an increased hazard of
mortality (HR = 3.12, 95% CI: 1.77–5.48) but did not with HIV plasma viral load response (HR = 1.15, 95% CI:
0.89–1.48) or CD4 cell response (HR = 0.95, 95% CI: 0.73–1.23).
Conclusion: Our study demonstrates that HIV-infected Aboriginal persons accessing HAART had similar HIV
treatment response as non-Aboriginal persons but have a shorter survival. This study highlights the need for
continued research on medical interventions and behavioural changes among HIV-infected Aboriginal and other
marginalized populations.
Published: 24 May 2006
AIDS Research and Therapy 2006, 3:14doi:10.1186/1742-6405-3-14
Received: 29 September 2005
Accepted: 24 May 2006
This article is available from: http://www.aidsrestherapy.com/content/3/1/14
© 2006 Lima et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Page 2

AIDS Research and Therapy 2006, 3:14http://www.aidsrestherapy.com/content/3/1/14
Page 2 of 9
(page number not for citation purposes)
Introduction
There are trends for global health concerns to have a
greater impact on ethnic minorities [1-5]. For example,
the human immunodefeciency virus (HIV) epidemic has
disproportionately affected Aboriginal persons through-
out North America. American Indians and Alaska Natives
make up 6% of all new HIV infections in the USA, yet they
represent less than 1 percent of the population [6]. Studies
attempting to explain this disparity have shown that both
Aboriginal men and women are at an increased risk of
antecedent risk factors for HIV infection including sexual
abuse, a history of IDU, poverty, poor mental health and
involvement in the sex trade [7-9] Furthermore, within
risk groups, for example amongst individuals with a his-
tory of injection drug use (IDU), Aboriginal men [10,11]
and women [11,12] are at a higher risk of being infected
with HIV.
In addition to a higher risk of HIV infection, it has been
shown that Aboriginal persons were more likely to suffer
from other morbidities that can complicate HIV disease
progression. For example, HIV-infected Aboriginal
youths, were more likely to be co-infected with hepatitis C
virus [13-15], being depressed [16,17], and having ane-
mia [18], all of which have been independently associated
with morbidity or mortality in HIV-infected individuals
[19-21].
Further disparity for Aboriginal persons and other mar-
ginalized groups is found when considering access to
treatment for HIV. In the USA, assessment of state surveil-
lance and claims data revealed that access to HIV therapy
was influenced by state policies which showed racial ine-
quality in pharmaceutical access [22]. In British Colum-
bia, Canada, where antiretrovirals are distributed at no
cost, Aboriginal ethnicity was negatively associated with
receiving HIV treatment before death [23] and positively
associated with leaving the hospital against medical
advice [24].
Although the impact of Aboriginal status on HIV inci-
dence, HIV disease progression, and access to treatment
has been investigated previously, little is known about the
relationship between Aboriginal ethnicity and outcomes
associated with highly active antiretroviral therapy
(HAART). There have been studies on racial/ethnic (black,
white and Hispanic) differences in response to HAART
[25-28], yet, to our knowledge, there has been only one
small study examining this issue with Aboriginal persons
in Greenland [29], but none including Aboriginal persons
in urban areas. Therefore, the present study examined
whether Aboriginal and non-Aboriginal persons respond
differently to HAART by measuring HIV plasma viral load
response, CD4 cell response and time to all-cause mortal-
ity. We also examined potential confounder effects of var-
ious demographic and clinical characteristics on the
association between Aboriginal ethnicity and HIV disease
progression.
Methods
HIV/AIDS drug treatment program
The distribution and the population-based monitoring of
antiretroviral therapy in British Columbia have been
extensively described in the literature [30-32]. The Centre
distributes antiretroviral drugs based on guidelines gener-
ated by the Therapeutic Guidelines Committee, which is
made-up of physicians, pharmacists, virologists, health
service researchers and economists. The Providence
Health Care Ethics Committee for Human Experimenta-
tion has approved use of the data generated from the pro-
gram for research purposes.
Data collection
All antiretroviral treatment recipients in the province are
entered into an Oracle-based monitoring and evaluation
reporting system that uses standardized indicators to pro-
spectively tract the antiretroviral use and clinical and
health status of HIV-1 positive individuals. Physicians
enrolling an HIV-1 infected individual into the system
must complete a drug request enrolment prescription
form, which compiles information on the applicant's
address, past HIV-specific drug history, CD4 cell counts,
plasma HIV-1 RNA, current drug requests, and enrolling
physician data. Typically, persons receiving antiretroviral
therapy are monitored by physicians at intervals no longer
than three months at which time prescriptions are
renewed or modified. At the time of the initial dispensa-
tion, participants are asked to provide informed consent
for accessing medical electronic records (which may be
used for health utilization studies, but is not relevant to
the analyses in this study), and complete a participant sur-
vey, which elicits information on socio-demographic
characteristics, clinical and health status, and alternative
therapy use. Both the consent form and the participant
survey are optional and participant's refusal to do either
will not limit his or her access to free antiretroviral ther-
apy. At the same time, the treating physicians are asked to
complete a clinical staging form using the World Health
Organization (WHO) clinical staging system.
The Centre recommends that plasma HIV plasma viral
load levels and CD4 cell counts be monitored at baseline
(time of enrolment), at four weeks after starting antiretro-
viral therapy and every three months thereafter. HIV
plasma viral load were determined using the Roche
Amplicor Monitor assay (Roche Diagnostics, Laval, Que-
bec, Canada) using either the standard method or the
ultrasensitive adaptation. CD4 cell counts were measured
by flow cytometry, followed by fluorescent monoclonal

Page 3

AIDS Research and Therapy 2006, 3:14http://www.aidsrestherapy.com/content/3/1/14
Page 3 of 9
(page number not for citation purposes)
antibody analysis (Beckman Coulter, Inc., Mississauga,
Ontario, Canada).
Study participants
All HIV-infected men and women in the current study
were entered into the Centre's monitoring and evaluation
system when they were first prescribed antiretroviral
agents. Eligible study participants are persons who were
≥18 years old, naïve to antiretroviral therapy when they
started HAART, and first dispensed triple combination
therapy consisting of two nucleoside reverse transcriptase
inhibitors and either a protease inhibitor (unboosted) or
a non-nucleoside reverse transcriptase inhibitor, between
August 1, 1996 and September 30, 1999. Eligible partici-
pants must have completed at least one questionnaire and
undertaken at least two follow-up CD4 cell count and HIV
plasma viral load tests.
Outcome measures and explanatory variables
There were three primary endpoints of interest in this
study: (1) time to HIV plasma viral load suppression as
measured from the time of starting HAART to the first of
two consecutive plasma HIV viral loads of < 500 copies/
ml; (2) CD4 cell response as measured by the time of start-
ing HAART to the first CD4 cell test with an increase of
100 cells over baseline; and (3) all cause mortality.
Event-free subjects were right censored as of June 30,
2003. Participants included in this analysis were not fol-
lowed after this date and those lost to follow-up were cen-
sored at the date of last known contact with the HIV/AIDS
Drug Treatment Program. For the first outcome, individu-
als we also censored if the last available HIV plasma viral
load test result was the first to be < 500 copies/ml, and the
censoring was applied at the time of the prior test. Deaths
occurring during the follow-up period were identified on
a continuous basis from physician reports and through
annual record linkages carried out with the British Colum-
bia Division of Vital Statistics.
The predictor variable of interest was Aboriginal status. A
number of potential confounders have been considered:
baseline CD4 cell count, baseline plasma HIV viral load,
physician's experience with HIV-infected patients, gender,
age, income, education, AIDS diagnosis at baseline, pro-
tease inhibitor use, year of initiation of therapy, history of
IDU and adherence. Physician experience was estimated
for the first follow-up physician of each patient. It was
defined as the cumulative number of HIV-infected
patients receiving antiretroviral therapy within their prac-
tice, by the date of subject's first known eligibility [33]. In
all analysis, in order to get a more parsimonious fit we
transformed this variable by dividing the original value by
10. Income was defined as an average yearly income of
≥$10,000 or <$10,000 (in Canadian dollars). Education
was defined as having completed high school. History of
IDU was defined as "ever-injected drugs" (yes or no),
which was physician or self-reported. Adherence was esti-
mated using pharmacy refill compliance. In brief, we
divided the number of months of HAART dispensed by
the number of months of follow-up in the first year of
therapy. This measure of adherence has been found to be
independently associated with HIV viral suppression and
survival among HIV-infected persons enrolled in the HIV/
AIDS Drug Treatment Program [34,35]. Patients were
defined a priori as non-adherent if they received antiretro-
viral medications for less than 95% of the follow-up
period during the first year of therapy, as in previously
published work [36,37].
Statistical analyses
Cox-proportional hazard regression was used to model
the effect of Aboriginal status and other potential con-
founders on the time to virologic suppression, time to
CD4 response and survival time [38]. Categorical varia-
bles were analyzed using Pearson chi squared statistics,
normally distributed continuous variables were analyzed
using t-tests for independent samples, and skewed contin-
uous variables were analyzed using Wilcoxon rank sum
tests. The assumptions of proportional hazards were
examined graphically.
A number of potential confounders have been included in
our analyses. In the selection of important confounders,
we applied a method based on the magnitude of change
in the coefficient of Aboriginal status [39,40]. Age, sex and
adherence were forced into all models. A stepwise
approach was employed to select additional confounders
[41]. Starting with the full model, variables were dropped
one at a time, using the relative change in the coefficient
for Aboriginal status as a criterion, until the maximum
change from the full model exceeded 5%, which is a more
conservative approach than the one suggested by Maldo-
nado and Greenland [39].
Analyses were performed using SAS software version 8.02
(SAS, Cary, NC). All tests of significance were 2-sided,
with a P value of less than 0.05 indicating that an associa-
tion was statistically significant.
Results
Between August 1st, 1996 and September 30, 1999, a total
of 1,312 antiretroviral naïve participants aged 18 years
and over initiated triple combination therapy consisting
of two nucleosides plus a protease-inhibitor or a non-
nucleoside reverse transcriptase inhibitor. Of these, 121
(9.2%) were excluded in this analysis for not having both
baseline CD4 and plasma HIV-1 RNA levels measures
available within six months prior to the start of antiretro-
viral therapy or having initiated therapy as part of a clini-

Page 4

AIDS Research and Therapy 2006, 3:14http://www.aidsrestherapy.com/content/3/1/14
Page 4 of 9
(page number not for citation purposes)
cal trial. Among the remaining 1,191 participants a total
of 622 individuals (560 men and 62 women) had
reported their aboriginal status.
Study participants were very heterogeneous when com-
pared to those excluded from the analysis. Participants in
the study: (1) enrolled mainly in 1996–1997 whilst non-
participants enrolled mainly in 1998–1999 (57% versus
63%; p-value: < 0.001); (2) were older (median age 36.3
versus 37.9 years; p-value 0.0003); (3) had lower baseline
CD4 cell count (median CD4 cell count 260 versus 290
cells/mm3; p-value: 0.0091); (4) had higher adherence
than non-participants (71% versus 42%; p-value: <
0.0001); (5) were treated by more experienced physicians
(median experience 55 versus 37 patients; p-value:
0.0003). No other differences were noted between the two
groups.
At baseline, for all participants, the median age was 38
years (interquartile range (IQR): 33–44 years), CD4 cell
count was 260 cells/mm3 (IQR: 110–410 cells/mm3),
and HIV plasma viral load was 130,000 copies/ml (IQR:
42,000–320,000). A total of 50.0% of participants had an
average annual income of less than $10,000 dollars,
38.9% did not complete high school, 34.2% had a history
of IDU, 17.2% had an AIDS diagnosis, and 14.0%, 41.3%,
24.1% and 20.6%, respectively, started therapy in 1996,
1997, 1998 and 1999. At baseline, 77% initiated a pro-
tease inhibitor based regime as compared to a non-nucle-
oside reverse transcriptase regime. During the first year of
follow-up 29.4% of participants had adherence level less
than 95%.
Most (N = 568; 91.3%) participants achieved 2 consecu-
tive plasma HIV viral loads of < 500 copies/ml before the
end of study. Median time to suppression among the 568
participants was 3.4 months (IQR: 1.8–7.3 months), and
3.6 months (IQR: 1.8–9.0 months) among all patients.
Similarly, 560 (90.0%) of participants achieved a CD4 cell
recovery of 100 cells over baseline during follow-up
period. Median time to recovery among the 560 partici-
pants was 5.0 months (IQR: 1.9–12.3 months), and was
6.0 months (IQR: 2.0–16.1 months) among all patients.
As of June 30, 2003, a total of 67 deaths were identified in
the study population over the follow-up period with an
overall crude mortality rate of 10.8%. Among all causes of
death, HIV disease resulting in infectious and parasitic
diseases (excluding acute HIV infection syndrome) (N =
32, 47.8%) and injuries due to intentional self-harm (N =
13, 19.4%) were responsible for over 67% of all deaths.
The remaining deaths (N = 22, 33%) were attributable to
a total of twelve causes of deaths. The overall median time
of follow-up was 62.5 months (inter quartile range: 50.3,
71.8 months).
Among the 622 individuals, 88 (14.1%) described them-
selves as being Aboriginal. As noted in Table 1, 8 (9.1%)
commenced therapy in 1996, 32 (36.4%) in 1997, 21
(23.9%) in 1998, and 27 (30.7%) in 1999. Over half of
these participants (59; 67.0%) initiated therapy with a
protease-inhibitor, while the rest of the study participants
(29; 33.0%) had a regimen that included a non-nucleo-
side reverse transcriptase inhibitor. Aboriginal status was
associated with a history of injection drug use, not com-
pleting high school, an annual income < $10,000, not
having an AIDS diagnosis at baseline, protease inhibitor
use, having lower baseline plasma HIV viral load, and less
experienced HIV physicians (p < 0.05). Age, gender, year
of initiation of therapy, baseline CD4 cell count, and
adherence were not significantly associated with being
Aboriginal.
Table 2 shows the univariate and multivariate associations
of aboriginal status, clinical and socio-demographic con-
founders with time to the first CD4 cell test with an
increase of 100 cells over baseline. Aboriginal status was
not associated with CD4 cell count response, but only
baseline HIV plasma viral load, age, physician experience
and adherence.
Table 3 shows the results for the univariate and multivar-
iate associations of Aboriginal status, clinical and socio-
demographic confounders with time to the first of two
consecutive plasma HIV viral loads of < 500 copies/ml. In
these analyses Aboriginal status was not associated with
HIV plasma viral load response. The multivariate analysis
shows that only baseline HIV plasma viral load, age, edu-
cation, adherence, and history of IDU were associated
with HIV plasma viral load response.
The univariate and multivariate associations of aboriginal
status, clinical and socio-demographic confounders with
time to death are displayed in Table 4. Aboriginal status
was strongly associated with mortality in both univariate
(hazard ratio (HR) = 2.87, 95% confidence interval (CI):
1.70–4.84) and multivariate analyses (HR = 3.12, 95%CI:
1.77–5.48). In addition to Aboriginal status, the multivar-
iate analysis shows that age, income, and adherence were
also associated with mortality.
Discussion
We found that there is no significant difference between
HIV-infected Aboriginal persons and non-Aboriginal per-
sons regarding the time to HIV plasma viral load suppres-
sion of < 500 copies/ml and time to CD4 cell response of
100 cells over baseline after initiation of HAART. There
was however, a significant higher mortality risk for Abo-
riginal persons after the initiation of HAART. After adjust-
ment for confounder variables, Aboriginal persons had
mortality rates 3.12 times higher than non-Aboriginal per-

Page 5

AIDS Research and Therapy 2006, 3:14http://www.aidsrestherapy.com/content/3/1/14
Page 5 of 9
(page number not for citation purposes)
sons. It is interesting to note that the only clinical charac-
teristic associated with mortality risk in this population
was adherence during first year of follow-up. The other
variables associated with mortality were socio-demo-
graphic characteristics of the participants (age and
income). We observed that clinical factors were only a sig-
nificant predictor when we looked at virologic and immu-
nologic responses. Note that among the clinical factors,
poor adherence was the strongest predictor of adverse out-
comes in all analyses. To explain this paradox, we consid-
ered potential confounding effects, given that race/
ethnicity are closely intertwined with socioeconomic sta-
tus and behavioural factors [1,4]. Our method of selecting
confounding factors ensured that the final models were
parsimonious while at the same time the estimates were
not substantially affected by any potential confounders
available in the data.
Other studies have examined the effect of ethnicity on
response to HAART but to our knowledge, this was the
first study to specifically examine this issue for Aboriginal
persons living in large urban areas. Prior studies looked at
Table 1: A comparison of baseline socioeconomic and clinical characteristics of Aboriginal and non-Aboriginal participants
Variable Aboriginal Not Aboriginal P-value
(n = 88) (n = 534)
Year of initiation of therapy,
no. (%)
1996
1997
1998
1999
Gender, no. (%)
Female
Male
Age
Median
Interquartile range
AIDS diagnosis, no. (%)
Yes
No
Protease inhibitor use
Yes
No
Baseline CD4 cell count (cells/
mm3)
Median
Interquartile range
Baseline plasma HIV viral load
(copies/ml)
Median
Interquartile range
Adherence (<95%), no. (%)
Yes
No
HIV physician experience
Median
Interquartile range
History of IDU, no. (%)
Yes
No
Completed high-school, no.
(%)
Yes
No
Missing
Income, no. (%)
< $10,000
>= $10,000
Missing
8 (9)
32 (36)
21 (24)
27 (31)
79 (15)
225 (42)
129 (24)
101 (19)
0.0588
13 (15)
75 (85)
49 (9)
485 (91)
0.1044
37 380.9669
33 – 4533 – 44
8 (9)
80 (91)
99 (19)
435 (81)
0.0296
59 (67)
29 (33)
418 (78)
116 (22)
0.0209
275260 0.6551
150 – 425 110 – 410
89,350140,000< 0.001
21, 750 – 185,000 45,000 – 340, 000
58 (66)
30 (34)
381 (71)
153 (29)
0.2995
29600.0133
3 – 1166 – 166
52 (59)
36 (41)
161 (30)
373 (70)
< 0.001
38 (43)
43 (49)
7 (8)
298 (56)
171 (32)
65 (12)
0.0046
49 (58)
36 (42)
3 (3)
211 (43)
281 (57)
42 (8)
0.0115