Serial monogamy and the spread of HIV
15
Serial monogamy and the spread of HIV How explosive can it get? Wim Delva MD, PhD Carel Pretorius MSc, PhD Stijn Vansteelandt MSc, PhD Marleen Temmerman MD, PhD Brian Williams MSc, PhD
description
Serial monogamy and the spread of HIV. How explosive can it get? Wim Delva MD, PhD Carel Pretorius MSc, PhD Stijn Vansteelandt MSc, PhD Marleen Temmerman MD, PhD Brian Williams MSc, PhD. 3. Serial monogamy and the spread of HIV. Department of Health , South Africa , 2009. - PowerPoint PPT Presentation
Transcript of Serial monogamy and the spread of HIV
Serial monogamy and the spread of HIV
How explosive can it get?
Wim Delva MD, PhDCarel Pretorius MSc, PhDStijn Vansteelandt MSc, PhDMarleen Temmerman MD, PhDBrian Williams MSc, PhD
3. Serial monogamy and the spread of HIV
Department of Health, South Africa, 200926
2
The compartmental model
R10C
R10L
Sero-discordant casual relationship
Sero-discordant long-term relationship
27
3
R10C
R10L R11
L
R11Cν(τ)
ν(τ)Sero-concordant casual relationship
Sero-concordant long-term relationship
The compartmental model
27
ν = HIV transmission rateτ = times since infection 4
R10C
R10L R11
L
R11Cν(τ)
ν(τ)
f CσC
f LσL
σ = partner turnover ratef = assortativeness parameter
The compartmental model
27
5
R10C
R10L R11
L
R11Cν(τ)
ν(τ)
f CσC
f LσL
(1-f C)σC
(1-f
L)σL
σ = partner turnover ratef = assortativeness parameter
The compartmental model
27
6
R10C
R10L R11
L
R11Cν(τ)
ν(τ)
f CσC
f LσL
(1-f C)σC
(1-f
L)σL
(1-f C)σC (1-f
L)σL
The compartmental model
27
σ = partner turnover ratef = assortativeness parameter 7
μ(τ) μ(τ)
μ(τ)μ(τ)R10
C
R10L R11
L
R11Cν(τ)
ν(τ)
f CσC
f LσL
(1-f C)σC
(1-f
L)σL
(1-f C)σC (1-f
L)σL
μ = HIV-related mortality rate
The compartmental model
27
8
Parameter Base-case value (Peak)
Range
Transmission probability per sex act
Acute 0.92% 0.28%-3.60%
Chronic 0.10% 0.05%-0.20%
Late 0.73% 0.20%-1.53%
Co-factor effect of other STIs
Relative risk of HIV transmission 4 2-9
STI prevalence in casual relationships 50% 20%-80%
STI prevalence in long-term relationships 20% 10%-50%
Frequency of sex acts 2 / week 1-3
Average duration of relationships
Casual relationships 3 months 26 days-6 months
Long-term relationships 1 year 6 months-2 years
Fraction of relationships that are casual 30% 10%-50%
Uncertainty analysis
28
9
Results
Doubling time (years)
Frac
tion
of s
imul
atio
ns
29
10
R0
Doubling time (years)
Frac
tion
of R
0 in
1st y
ear
Base case
Results
30
11
Sensitivity analysis
31
12
How to get a DT of 1.2 years?
31
13
Variable ValueSTI cofactor 20.2Frequency of sex 7.6 sex acts / weekPSTP Acute / (DT > 1.9 years) Chronic 0.55% Late 40%Fraction casual / (DT > 2.0 years)Partner turnover rate 2 / weekSTI prevalence In casual relations / (DT > 2.4 years) In long-term relations / (DT > 2.1 years)
Conclusions Serial monogamy unlikely to result in doubling
time of 1.2 years or less
The South African HIV epidemic is unlikely to have emerged from a serially monogamous sexual network
Serial monogamy reduces effect of high acute infectiousness
New models needed to capture the effect of concurrency realistically
Other potential (partial) explanations: Anal sex, MSMW, unsafe medical injections
32
14
50
Thanks
15
IWT (Flemish Institute for Innovation through Science and Technology)
FWO (Flemish Research Fund)
