Transcript Robin Wood Desmond Tutu HIV Centre University of Cape Town

Cape Town AIDS Cohort-CTAC
Electronic Data Sets
Robin Wood
Desmond Tutu HIV Centre
University of Cape Town
Photograph of Masiphumelele
CTAC Data Sets
• Natural history cohort 1984-2000
• Hospital-based ART trials 1996-2003
• Community-based ART programmes
2002-2003
• Web-based ART Register
Natural History Cohort 1984-2000
• Epi-info based
– Flow charts of 2088 outpatients
– Laboratory, clinical data, OI prophylaxis
• Pros
– Doctors like flow charts of CD4 & WHO stages but
not perceived non-useful data!
• Cons
– Data transfer, LTFU, missed hospitalisations &
deaths required other data sources
• Outputs
– OI frequency, Kaplan-Meier survival
Incidence rate/100 patient yrs
CTAC: Incidence Rates of
Opportunistic Infections
70
60
50
40
30
20
10
0
>350
200-350
<200
S. bact
Toxo
Cryptosp
CMV
MAI
Crypto
Wasting
KS
Enceph
PCP
HSV
OC
TB
CD4+ T-cells/mm3
CTAC 2002
Hospital-based ART Trials 1996-2003
• Excel spread sheet of CRF data
– Lab, clinical events, hospitalisations, deaths
• Pros:
– Trial data good quality, variable CD4 entry criteria,
frequent visit intervals
• Cons:
– Extracting data laborious
• Outputs:
– Changed disease progression
– Resource utilisation
TB incidence rates & cases prevented per
100 pys of HAART
25
Cases /100pys
20
15
HAART
Naïve
10
5
0
CD4 >350
CD4 200-350
CD4<200
WHO3&4
WHO 1&2
Cases saved
95% CI
1.3
9.4
11.3
(0.3-2.9)
(3.8-14.3)
(6.2-19.1)
18.8
(13.2-26.1)
Badri, Wilson, Wood Lancet 2002
Progression-free survival of Triple vs NOART: WHO stage 1&2
Cumulative Proportion Surviving (Kaplan-Meier)
Proportion of patients staying in stage1&2
Complete
Censored
stage1&2, p<0.00001
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0
200
400
600
800
1000
1200
1400
1600
NOART
Triple
Time
Category
N
Patients progressing to
to higher stages
Patients not progressing
To higher stages
P-value
Triple
128
6 (4.7%)
122 (95.3%)
<0.0001
NoART
140
52 (37.1%)
88 (62.9%)
Progression-free survival of Triple vs NOART: WHO stage 3
Cumulative Proportion Surviving (Kaplan-Meier)
Proportion of patients staying in stage3
Complete
Censored Triple vs NOART, stage 3, p<0.001
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
200
400
600
800
1000 1200 1400 1600 1800 2000 2200
Triple
NOART
Time
Category
N
Patients progressing to
higher stages
Patients not progressing to
higher stages
P-value
Triple
85
13 (15.3%)
72 (84.7%)
0.014
No ART
155
46 (29.7%)
109 (70.3%)
Progression-free survival of Triple vs NOART: WHO stage 4
Cumulative Proportion Surviving (Kaplan-Meier)
Proportion of patients staying in stage4
Complete
Censored
Triple vs NOART, p=0.009stage 4,
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0
200
400
600
800
1000
1200
1400
1600
1800
Triple
NOART
Time
Category
N
Patients progressing
to higher stages
Patients not progressing
to higher stages
P-value
Triple
35
4 (11.4%)
31 (88.6%)
0.19
NoART
142
30 (21.1%)
112 (78.9%)
HAART impact on hospitalisation
(1995-2000)
Days/yr (95% CI)
OR (95% CI)
AV cost PY (95% CI)
HAART
0.7 (0.55-0.82)
0.2 (0.14-0.27)
525 (392-694)
No ART
4.4 (3.9-4.9)
1
3,351 (2873-3888)
HAART
1.4 (1.25-1.65)
0.13 (0.11-0.15)
1,113 (980-1293)
No ART
10.9 (10.29-11.46)
1
8,512 (8065-8982)
HAART
6.3 (5.58-6.98)
0.4 (0.38-0.49)
4,899 (4374-5471)
No ART
14.6 (13.81-15.33)
1
11,412 (10824-12016)
WHO Stages 1&2
WHO Stage 3
WHO Stage 4
Usapho Lwethu Community ART
Data sources and follow up
• Patient live in recruitment area & must have
attended local HIV clinic for > 6 months
• Household support by “Sizophila monitor”
• Schedule 1 NNRTI, schedule 2 PI/r
• Single referral hospital
• Post mortem is requested on all deaths
• Access to national mortality data base via “ID
number”
Community ART Programme
2002-2003
• Sequel database, data capture sheets
– Laboratory, clinical, QOL, adherence data
• Pros
– Fixed visit schedule, LTFU rate low but deaths
high, data captured on “lap-top”
• Cons
– Needs data administrator, active search for
missing data
• Outputs
– Programme status reports of CD4, VL, survival etc
Clinical Trials vs Community Clinic
Clinical trials n=299
Community n=103
Median CD4
230 cells/ul
82 cells/ul
WHO stage 3&4
49.8%
92.9%
Treatment
ART Regimen
NNRTI 58.2%
PI 39.1%
Triple NRTI 2.7%
NNRTI 100%
Results at 16 weeks
VL<400 copies/ml
75.5% (CI 71-80)
94.2% (CI 88-100)
LTFU
5
1
Deaths
2 (0.69%)
7 (6.8%)
Monthly drug costs of d4T, 3TC, EFV in
study period from Sept 02-May 03
1400
$159
US$ cost calculated @ R8 per $
1200
Rand
1000
800
$73
600
$41
400
200
0
Initial
4 months
8 months
EFV
3TC
d4T
Proportional Costs for 250 Patients
18%
2%
Drug
Staff
Monitoring
Safety
48%
32%
Cost per patient R8856 per annum
Aerial Photograph of Masiphumele
Pilot National ART Register
• Web-based minimum data from ART
programme linked to drug supply
– ID number
– Clinical stage
– Current drug prescriptions
• Linked to laboratory & pharmacy data
– CD4 & Viral load monitoring
– Drug monitoring
Proposed Register Outputs
• Patient retention, number on Rx, LTFU and
discontinuing
• Drug switching patterns
• Time to first failure (MTCT exposure?)
• Survival (benefit if compare with natural
history model!)
• Drug accountability, dispensing to patients &
“leakage”
• Laboratory outcomes CD4/viral suppression
at time-points, resistance
ACKNOWLEGEMENTS
FUNDING AGENCIES
NIH Cipra-SA
BMS “Secure the Future” Foundation
Hannan Crusaid Trusts
Heiser Foundation
Doris Duke Foundation
UCT AIDS COHORT-CTAC
Motasim Badri, Desmond Tutu HIV Centre, UCT
Eduard Beck, McGill University, Montreal, Canada
Linda-Gail Bekker, Desmond Tutu HIV Centre, UCT
Gary Maartens, Dept of Medicine, Groote Schuur Hospital, UCT
Kwesi Matoti, PAWC Gugulethu
Catherine Orrell, Desmond Tutu HIV Centre, UCT
Larissa Reader, Desmond Tutu HIV Centre, UCT
Eve Sabotski, PAWC South Peninsula
Douglas Wilson, Dept of Medicine, Somerset Hospital, UCT
Photograph of Sizophila Team
Sizophila Team April 2003