Transcript VA Virtual Cohort

Poster # 819
Contact Information:
Dr. Amy C. Justice
Yale University and
VA Healthcare System
Building 35A, Rm 212
950 Campbell Avenue
West Haven, CT 06511
Phone: 203-932-5711
x3541
Email:
[email protected]
Evaluation of Intracranial Hemorrhage in 49,610 HIV-Infected Veterans and
California Medicaid Recipients
Shawn L. Fultz 1, David Zingmond 2, Kirsha S. Gordon 3, Joseph L. Goulet 3, Larry Mole 4, Joseph T King, Jr 3,4, Dawn Bravata 3,4, Hernan Valdez 5,
6
3,4
Michael Kraft , Amy C Justice , and VACS Project Team
Office of Public Health and Environmental Hazards, Veterans Health Administration, Department of Veterans Affairs1~University of California Los Angeles, Los Angeles, CA, USA2~VA Connecticut Healthcare
System, West Haven, CT, USA3~Yale University School of Medicine, New Haven, CT, USA4~VA Palo Alto Health Care System, Palo Alto, CA, USA5~Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, CT,
USA6~Boehringer Ingelheim Pharmaceuticals, Inc., Ingelheim, Germany7
OBJECTIVES
RESULTS
• To examine the rate of intracranial hemorrhage in an HIVinfected population not exposed to TPV during the HAART era.
Table 1. Descriptive Statistics
VA Virtual
Cohort
(n=16,573)
BACKGROUND
• HIV has become a complex chronic disease with extended life
expectancy and concomitant long term exposure to
antiretroviral (ARV) therapy.
Baseline Demographics
• Common long-term drug toxicities with ARV therapy appear
as comorbid conditions.
Male
• In pre-approval trials, 13 of 6840 patients (0.20%) exposed to
tipranavir (TPV) have developed intracranial hemorrhage
(ICH) at an incident rate of 0.26 per 100 person years of
exposure (95% CI 0.09, 0.61). It is unknown whether this rate
exceeds that observed with other, non TPV containing
HAART regimens.
• We identified HIV-infected individuals under care in the U.S.
between October 1997 and December 2003 through the US
Veterans Healthcare System-derived Virtual Cohort 1 and the
California state Medicaid (Medi-Cal) program 2.
• In the Virtual Cohort, patients were identified as being HIV
infected by virtue of having at least two outpatient or one
inpatient codes for HIV. This algorithm was validated against
the VA Immunology Case Registry Data which requires each
VA site to confirm the HIV status of patients included in the
registry.
• HIV infected Medi-Cal enrollees were identified by the Medical
Care Statistical Section of Medi-Cal using case identification
from claims and eligibility data developed for ongoing review of
HIV-related costs and utilization. AIDS was determined by
either two outpatient or one inpatient claims for AIDS or
confirmation in the California AIDS Registry.
• ICH was identified using ICD-9 codes (430.xx, 431.xx and
432.xx, except 432.0) 3. Subjects who had ICD-9 codes
indicating trauma or ischemic stroke complicated with
intracranial hemorrhage were excluded. Other diagnoses were
identified using standard, validated, ICD-9 code groupings.
Age, > 40 years
76%
98%
Race/Ethnicity
Black
White
Hispanic
Other
41%
29%
7%
24%
25%
43%
25%
7%
AIDS Defining Illness
36%
Covariate
Present %
Covariate
Absent %
p
Female Gender
0.00%
0.20%
1
0.50%
0.47%
0 .8
>40 years old
0.00%
0.20%
0.003
0.64%
0.35%
Black Race
0.20%
0.20%
0.9
0.63%
0.34%
0 .09%
0.001
1.80%
Hypertension (HTN)
Diabetes (DM)
0.27%
69%
74%
Hypertension (HTN)
23%
12%
Diabetes (DM)
9%
6%
Coronary Artery Dz (CAD)
6%
4%
Alcohol Abuse /Dependence
20%
Liver Disease
2%
Active Protease Inhibitor
20%
0.14%
0.3
1.30%
Alcohol Abuse or
Dependence
Liver Disease
18%
AIDS Defining
Illness
4%
CD4<200 or
AIDS Illness
29%
Known Dead
20%
20%
Covariate
Present
Covariate
Absent
p
Female Gender
0.09
0.06
0.7
0.18
0.22
.1
<0.001
>40 years old
0.07
0.04
0.2
0.27
0.17
<0.001
0.43%
0.02
Black Race
0.10
0 .03
0.01
0.25
0.20
0.1
0.28%
<0.001
0.16
0.04
0.0001
0.39
0.19
<0.001
0.17
0.05
0.01
0.43
0.20
<0.001
0.09
0.06
0.6
0.35
0.21
0.04
0.15
0.03
0.0001
0.38
0.18
<0.001
0.10
0.05
0.2
0.32
0.19
0.0007
0.13
0.06
0.5
0.56
0.21
<0.001
AIDS Defining Illness
0.10
0.05
0.04
0.26
0.12
,0.0001
0.4
CD4<200 or AIDS
Illness
0.07
0.06
0.6
0.38
0.16
<0.001
0.2
Active Protease
Inhibitor
0.04
0.06
0.5
0.22
0.21
0.8
0.07
0.06
0.9
0.20
0.22
0.5
0.24
0.02
<0.001
0.91
0.12
<0.001
0.43%
<0.001
0.13%
0.01
1.80%
0.42%
<0.001
0.28%
0.10%
0.01
1.50%
0.28%
<0.001
0.24%
0.13%
0.1
1.10%
3.50%
0.00%
0.15%
1
1.90%
4.20%
0.23%
0.08%
0.04%
0.0004
0.01
Active Protease
Inhibitor
0.04%
0.08%
1.0
Hypertension (HTN)
Diabetes (DM)
Coronary Artery
Disease (CAD
<0.001
<0.001
HTN, DM, or CAD
Alcohol Abuse or
Dependence
Liver Disease
0.93%
0.46%
0.27%
0.53%
<0.001
0.40%
0.50%
Active HAART
0.03%
Known Dead
0.54%
0.18%
0.05%
0.04
<0.001
0.38%
1.10%
0.52%
0.33%
0.1
<0.001
Known Dead
*Prevalent intracranial hemorrhage events are identified using ICD-9 codes and were deemed prevalent if they occurred within 12 months prior or 6 months after baseline. Events occurring after this point were considered incident. Active medications
are defined as any drug for which the patient has a prescription that would not have run out on or by the baseline date. Neither database captures nonprescription nonsteroidal anti inflammatory agents such as aspirin. Hypertension, diabetes,
coronary heart disease, alcohol abuse and dependence, and liver disease are determined based upon the presence of ICD-9 diagnostic codes.
FINDINGS
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
• Sample included 16,573 veterans and 33,037 California Medicaid
recipients for a total of 49,610 individuals with HIV infection
• Veterans were older (>40 years of age); more likely to be male
(98% vs. 74%), and less likely to have developed AIDS (36% vs.
74%) compared to Medi-Cal recipients (Table 1, p<0.001).
• Within 12 months before baseline and 6 months after baseline,
there were 25 ICH prevalent cases 0.2% (95% CI 0.1-0.2%) within
the VA and 173 ICH prevalent cases 0.5% (95% CI 0.4-0.6%)
within Medi-Cal.
• The VA Virtual Cohort observed 27 incident cases 0.07 (95% CI:
0.05-0.10) per 100 PY and Medi-Cal observed 274 incident cases
0.23 (0.5 CI: 0.20-0.25) per 100 PY.
VA AIDS
• Overall and in every stratified analyses, the prevalence and
incidence of ICH was lower among veterans than among those
receiving Medi-Cal coverage (Table 2 and 3, p<0.001).
VA >40 yrs
• Stratified analyses showed similar associations in both cohorts.
• Prevalence and incidence of ICH were higher among those
Medi-Cal >40
yrs
p
0.54%
0.32%
Active HAART
ICH Events/100 PY
Medi-Cal AIDS
Covariate
Absent
26%
21%
Figure 1. Boehringer Ingelheim (BI), VA, and Medi-Cal ICH Incidence Rates
Medi-Cal Men
Only
Covariate
Present
Covariates
16%
Active HAART
• Poisson regression was used to model incidence rates after
adjusting for covariates.
Medi -Cal
Overall
Medi-Cal Cohort
29%
CD4<200 or AIDS Illness
28%
VA Virtual Cohort
p
HTN, DM, or CAD
HTN, DM, or CAD
Medi-Cal Cohort
Covariate
Absent %
45%
74%
Table 3. Incidence (events/100 PY) of ICH by Important Covariates*
Covariate
Present %
Coronary Artery
Disease (CAD)
• Incidence (per 100 person-years) was defined as events
occurring after this time frame.
VA Men Only
VA Virtual Cohort
Covariates
• Prevalence was defined as ICH events occurring from 12
months before to 6 months after date of initial presentation for
HIV care.
VA Overall
Medi-Cal
(n=33,037)
Clinical Characteristics
METHODS
BI Data
Table 2. Prevalence (%) of ICH by Important Covariates*
•Over 40 years of age
•With hypertension, diabetes, or coronary artery disease
•With alcohol abuse
•With an AIDS defining illness
• Prevalence and incidence of ICH did not differ by active receipt of
protease inhibitors or HAART.
CONCLUSIONS
REFERENCES
• VA incidence rates of ICH were lower than those observed
in pre marketing trials of Tipranavir, but Medi-Cal rates
were similar (Figure 1).
1. Fultz SL, Skanderson M, Mole LA et al. Development and
verification of a ‘virtual’ cohort using the National VA
Health Information System. Medical Care 2006; 44(8
Suppl 2): S25-S30.
• ICH events may be less completely captured:
•In observational databases using ICD-9 codes than in
prospective clinical trials
•in the VA because patients with acute life threatening
conditions may go to the nearest hospital rather than a
VA facility for emergency care.
• While many risk factors for ICH established outside HIV
infection appear to apply, we do not see increased rates of
ICH among women compared to men.
2. Zingmond, DS,Ettner SL, Cunningham WE. The impact
of managed care on access to highly active antiretroviral
therapy and on outcomes among HIV-infected Medicaid
beneficiaries. Medical Care Research and Review. In
press.
3. Rosamond WD, Folsom AR, Chambless LE et al. Stroke
incidence and survival among middle-aged adults: 9-year
follow-up of the Atherosclerosis Risk in Communities
(ARIC) cohort. Stroke 1999; 30(4): 736-743.
• We also see no evidence of increased rates among those
on protease inhibitors or on HAART compared to those
not on these treatments.
Funding: National Institute on Alcohol and Alcohol Abuse (2U10
AA 13566); and the Veterans Health Administration; the VHA
Office of Research and Development; and, VHA Public
Health Strategic Health Care Group; and the University AIDS
Research Program (ID-05-LA-034). Dr. Zingmond is
supported by a NIA Mentored Clinical Scientist Aware
(AG023024-01A1).
Conflict of Interest: No financial or in kind support was
provided by any pharmaceutical company for these
analyses.