Why is HIV Prevalence SoSevere in Southern Africa?, and
“What Works” (and Doesn’t)for AIDS Prevention?
Daniel Halperin, PhD, MS Senior Advisor for Behavior
Change/Primary Prevention, USAID
With assistance from colleagues at USAID, World Bank, Global Fund, UCSF, etc.
Key Points* Back to the Basics: using a “basic public health”
approach to prevention
• Using evidence (vs. politics, ideology, etc.) to set priorities
• *Prevention works (Uganda today compared to 10-15 yrs. ago)
[Please see the “notes” under most of the slides!]
HIV PREVALENCE BY REGION
Source: Adapted from WHO AFRO 2003 Report
0
5
10
15
20
25
30
35
1990199119921993199419951996199719981999200020012002
Southern Africa
Eastern Africa
Central AfricaWest Africa
% H
IV p
revale
nce
2004 Report on the Global AIDS Epidemic (Fig 8)
TRANSMISSION DYNAMICS
Epidemic concentrated - if transmission occurs largely among vulnerable groups and vulnerable group interventions would reduce overall infection
Epidemic generalized - if transmission occurs primarily outside vulnerable groups and would continue despite effective vulnerable group interventions
Epidemics DON’T inevitably keep escalating
HIV TRANSMISSION PATTERNSIN ACCRA, GHANA, 2003
0%
10%
20%
30%
40%
50%
60%
70%
80%
ANC prevalence Sex worker
prevalence
Adult male
infections from
sex workSources: GAC/NAP and Cote et al data, 2004
HIV TRANSMISSION PATTERNS IN MASHONALAND CENTRAL, ZIMBABWE
0%
10%
20%
30%
40%
50%
60%
Pregnant women
HIV prevalence
Sex worker
prevalence
Male infections
from sex workSources: Wilson and Cowan et al data, 2003
Economic Status and HIV prevalence (Tanzania)
0
2
4
6
8
10
12
lowest second middle fourth highest
womenmen
Source: 2003-2004 AIS
Education Status and HIV prevalence in Tanzania
012345678910
noeducation
primaryincomplete
primarycomplete
secondary+
womenmen
Source: 2003-2004 AISS
Why is HIV so much Higher in Southern Africa??
• Multiple concurrent partnerships (“nyatsi,” “lishende,” “small house,” “second office”...)
• Lack of male circumcision
• And various other factors, such as relatively developed/highly mobile societies, income inequality, gender dynamics, "dry sex,” etc.
Source: 2002 Lesotho BSS
Source: 2002 Lesotho BSS
“Concurrent” Partnerships
*Source M. Carael, 1995; Halperin and Epstein, 2004
“Concurrent” Partnerships
*Source M. Carael, 1995; Halperin and Epstein, 2004
Proportion of 15-24 year-olds reporting more than one current sexual partner, South Africa
2005
05
101520253035404550
Males Females Total
15-19
20-24
Source: South African National HIV Prevalence, HIV Incidence, Behavior and Communication Survey, 2005
“Map” of the largest component of a sexual network in Likoma, Malawi
Sour
ce: K
ohle
r H a
nd H
elle
ringe
r S. T
he S
truc
ture
of S
exua
l Net
wor
ks a
nd th
e Sp
read
of
HIV
in S
ub-S
ahar
an A
fric
a: E
vide
nce
from
Lik
oma
Isla
nd (M
alaw
i). P
ARC
Wor
king
Pap
er
Serie
s: W
PS 0
6-02
A NATION AT WAR WITH HIV&AIDS
Low degree networks create a transmission core
In largest component: 2% 41%
64%10%
Mean:
1.74Mean:
1.80Mean:
1.86
Largestcomponents
Mean:
1.68Number ofPartners
Bicomponentsin red
Source: Martina Morris, Univ. of Washington, used with permission from a presentation given at a meeting on concurrent sexual partnerships and sexually transmitted infections at Princeton University, 6 May 2006.
Transmission efficiency“Mathematical models estimate the average probability of male–female transmission of HIV-1 per unprotected coital act to be between 0.0005 and 0.003% during chronic HIV infection, which in itself would not sustain an epidemic.”
-Pao et al, AIDS (2005)
“Acute Infection” and Concurrence
Source: WHO/GPA surveys
BEHAVIOURAL AND HIV TRENDS IN UGANDA
0
10
20
30
40
1989
1995
1989 35 16 15
1995 15 6 3
Men with one or more "casual" partners in
past year
Women with one of more "casual" partners
in past year
Men with three or more "non-regular" partners
in the past year0
5
10
15
20
1991 15
2003 4.1
Adult HIV prevalence
CONCLUSION
• MCP is a driver• Behavior can be changed• Challenge – HOW• Social Movement