Wrap up and closing session - The 2nd Global HIV/AIDS ...

skillfulbuyerΠολεοδομικά Έργα

16 Νοε 2013 (πριν από 3 χρόνια και 8 μήνες)

105 εμφανίσεις

Willi McFarland, M.D., Ph.D.

San Francisco Department of Public Health

George W. Rutherford, M.D.

Global Health Sciences, University of California, San Francisco

Report back

Wrap up and closing
session

Addis Ababa Progress Revisited

Area

Recommendations

Theresa’s
杲慤g

䉡湧歯k
杲慤g

Recent
infections

1.
Operations research

2.
Capacity
-
building

1.


2.


Population
-
based
surveys
with HIV testing

1.
Fill

surveillance gaps

2.
Every ~5 years

3.
Only in high prevalence countries

4.
National IRB review



1.


2.


3.
X

4.


Use

of VCT and
PMTCT data

1.
Evaluate utility of VCT data,
standardize analysis

2.
Expand use of PMTCT data to
complement ANC SS

3.
Longitudinal records for linkage to care

1.


2.
X

3.
√ /
-

Addis Ababa Progress Revisited

Area

Recommendations

Theresa’s
杲慤g

䉡湧歯k
杲慤g

Linking biological

and behavioral
data

1.
Ensure

data can contribute to
understanding epidemic dynamics

2.
Encourage linkage for triangulation
and integrated analysis

1.
X

2.


AIDS reporting
and measuring
impact of ART

1.
Develop one AIDS case definition

2.
Improve AIDS case reporting system

3.
Expand TB surveillance to capture
HIV

4.
Capture individual records at entry
into care, determine outcome

5.
Improve mortality data

1.


2.
X

3.
√ /
-

4.
√ /
-

5.
X

Addis Ababa Progress Revisited

Theresa’s grade

11 / 16 = 69%

D+

Bangkok grade

?


More homework (scale
-
up, improve infrastructure, research)


Hire a private tutor (TA, training, capacity
-
building, mentorship)


Punishment (no Facebook, no iPhone, no boys)

Behavioral and

STI


Population
-
based


Surveys*


Most at risk


group surveys


STI surveillance


Prevalence, behaviors,

morbidity


HIV case reporting or


advanced HIV


case reporting


ANC sentinel surveillance*


Population
-
based


Surveys with HIV testing*


Most at risk group


sero
-
surveys


ART outcomes


HIV drug resistance

Mortality


Vital
registration

HIV infection

Advanced HIV

Death

*
Epidemics where infection is driven

by the general population

Incidence


Report early


infant


diagnosis

Key HIV Surveillance Data Points and
the Course of HIV Infection, the Future

Willi

George

Oliver’s BED Conclusions


Viral suppression greatly effects the BED assay


The number of HIV+ virally suppressed individuals
increases


AIDS impacts BED & Avidity estimates


BED and Avidity are resistant to sample handling
problems


Pregnancy does not impact BED or Avidity


A testing algorithm can dramatically impact specificity
for cross sectional incidence testing (and reduce cost)



BED Progress


But, need another assay


New assays under
-
going standard development protocols (
Mireille
)


Field experience and applications


KAIS 2007: Exclusion conditions, confirmation (pending), risk factors for
incident vs. prevalent HIV infection (Tom)


Thailand: Triangulate BED interpretation with other data (
Pasakorn
)


Not perfect… wait or meet certain criteria to go ahead?


Exclusion conditions (e.g., on ART, CD4)


Correction factors (local)


Sample size


What are you using it for?


Still looking for the Holy Grail



Population Size Estimation Methods


Multiplier method


Capture
-
recapture


Census method


Population
-
based
surveys


Components


Compartmental


Truncated Poisson


Scale
-
up


Nomination


Delphi


Consensus


Borrow and adjust
from neighbor


Conventional wisdom
or opinion


Guy was vexed by multiple methods


Tetiana made an impressive case in favor


We will never have a gold standard
-

alas

MSM: RDS Successes
from Johnson et al., 2008

Honduras

China

Zanzibar

Thailand

Nepal

Croatia

Uganda

Indonesia

Bangladesh

Paraguay

Ukraine

Brazil

Vietnam

USA

South
Africa

Kenya

MSM: RDS Failures
from Johnson et al, 2008

Kosovo

Caribbean

Albania

Ninxia

Guangzhou

Anhui

Bangladesh

Beijing

Estonia

Cambodia

Jinan

TLS successes


“Still relevant after all these
years”
-

Frits

San
Francisco

Shenzhen

Salgueiro

Harare

Forteleza

Bangkok

Lao PDR

Russia

Ukraine

MARP Sampling Best Practices:

TLS or RDS?

TLS


“What you see is what
you get”


If not representative, can
make inclusive and
reproducible to what you see

RDS


“Faith
-
based”


If not representative,
reveals truths you
didn’t see


One size does not fit all


All methods need good formative
research (many others not discussed)


Approximations, not Gold Standard


Not Holy Grail

Scary Epi Session!

TLS

(Keith channeling John
Karon
)


Big design effects!


Big adjustments due to
probability of inclusion
based on visits!


Frequency of visiting not
usually measured

RDS

(Lisa)


Big dependence on
fragile network size
question!

Multiple concurrent partnerships

W

W

W

W

W

W

W

W





W

Acute high
infectivity,
and/or long
-
term inter
-
connections

Vinod: Seldom measured, no consensus on how to measure, complex analysis
and interpretation


missing major transmission dynamic in surveillance

More things not measured: STIs


The “fiction” of STIs in Second
Generation HIV Surveillance?


No progress since 1999


Data needs in all areas


Guidance failure (George)


Some country examples on applications
and uses (Thailand, Kenya, San Francisco)

New questions, new themes, new
technologies for data collection


Words of caution: Do participants actually
answer the question being asked? (Kristin)


Way forward and lifeline to help


Include younger ages (Olive, Rachel,
others)


Include older ages (Olive, Rachel, Tom,
others)


Partner by partner data (
Vinod
, Tom,
others)


Violence, alcohol use, perception of
government response, perception of risk
(Olive)

Need to ask:

HIV status, known duration, in care, on
ART, disclosure to and by partners!


Miss unmet need for care (Tom)


Uninterpretable

HIV prevalence trend:
survival vs. incidence (many)


Difficult interpreting case reporting:
treatment failures vs. incidence (
Txema
,
Irum
, others)


Uninterpretable

HIV incidence measures:
BED, other assays (Oliver)

•Help interpret HIV DR (Diane)

Miss major transmission dynamic
(ART suppression of VL)


Miss major source of new infections
(from unknown HIV+)


Miss the next wave of the epidemic


Uninterpretable

risk behavioral data

Need to ask:

HIV status, known duration, in care, on
ART, disclosure to and by partners!

14
14
16
14
3
7
14
19
0
2
4
6
8
10
12
14
16
18
20
No sex
Oral sex only
100% condom use
Only HIV+ partners
Only oral sex with HIV-
Condoms when discordant
Receptive when discordant
No Strategy
%
Behavior of known HIV
-
positive MSM, San
Francisco, 2008

Any unprotected sex: 70%

High HIV transmission risk: 19%

85% HIV+ MSM know status

Scale up of ART coverage in generalized and concentrated
epidemic countries, 2004 to 2007

100

0

20

40

60

80

Percent Coverage (%)

2004

2005

2006

2007

Malawi now >50%

Persons living with
HIV/AIDS, San
Francisco, 1980
-
2007

0
2000
4000
6000
8000
10000
12000
14000
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
2000
2001
2002
2003
2004
2005
2006
2007
Year of Diagnosis/Death
No. of Cases/Deaths
AIDS deaths
Living with AIDS
On ART
Living with HIV
ART scale
-
up
1995
-
2001

Death drops

Living with AIDS rises

Living w
ith

HIV/AIDS

Estimated number of people newly infected with HIV globally

0

1

2

3

4

5

6

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

Millions

Year

Number
of people

newly
infected
with HIV

Upper range

Estimate

Lower range

“HIV incidence is stabilizing”

-

Peter, Anton, Mike, Nguyen, and many country
presentations

0
500
1000
1500
2000
2500
3000
3500
4000
4500
67
72
78
83
88
93
98
2003
2008
2013
2018
2023
2028
2032
2037
2042
2047
The Coming Endemic…

HIV incidence in San Francisco in
Willi’s

lifetime

The cohort of 20 year
-
old MSM with HIV incidence at
1.9% will reach 60% prevalence by 60 years old

Age-Specific HIV Prevalence
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
20
25
30
35
40
45
50
55
60
Age
HIV Prevalence
Homosexual

Intravenous

Drug Users

Commercial

Sex Workers

Children

Pregnant Women

Male with

Multiple Partners


1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
-
2004

2005
-

Children

The Next Big

Things ?

(
Triangulation

of) Warning Signs of
Rising HIV Prevalence in Thailand


Non
-
access,
non
-
venue based (e.g., street, call girls)
sex workers


HIV
prevalence
as high as
18
-
30% were reported


Un
-
safe
sex behaviors, especially among youth


About 20
-
50% using condom with regular and casual
partners


Multiple sexual partners


Exchange sex for money or gifts


Persistently high
HIV prevalence
among IDU and MSM


Indirect evidence on rising trend of HIV incidence by
BED
-
CEIA HIV
seroincidence

surveillance



Thailand gets and
A+!

In hospitality and
surveillance


c

Conclusions


Evidences indicated that the HIV/AIDS problems remained huge
problem in Thailand, and there are the new overlapping concentrated
epidemics in certain group of population.


Effective surveillance is critical to determine HIV epidemic status, as well
as monitoring of prevention efforts. New approach and technology
must be workout to fit the new context of HIV/AIDS dynamism.


Risk of new wave of epidemic is high


Thailand (GOVT, civil society) needs to increase prevention efforts,
especially among high
-
risk groups such as men who have sex with men,
injecting drug users and PLHA who have HIV discordant couples etc.

HIV surveillance needs a steady course ...


Philippe Petit, funambulist in the WTC 1974…

Stay ahead of the wave!