Using longitudinal, population-based HIV surveillance to

measure the real-world impacts of ART scale-up in KwaZulu-Natal,

South Africa

Frank Tanser

Presentation at IAS, Melbourne, Australia

22nd July 2014

Outline

• Background• ART coverage and risk of HIV acquisition• Population viral load

Global ART scale-up

UNAIDS Report on the Global AIDS Epidemic 2012; WHO Universal Access Report 2013; Aaron Motsoaledi 2012

South Africa has the worldwide

largest absolute number of

patients on ART, >1.6 million by some estimates

Hlabisa sub-district, with the Africa Centre surveillance area and the TasP trial area

Africa Centre for Health and Population Studies

The Africa Centre

Bor, Herbst, Newell, and Bärnighausen Science 2013

Adult life expectancy over time

13,060 deaths among 101,286individuals aged 15 years and older, contributinga total of 651,350 person-years of follow-up time

HIV incidence by age and sex 2004-2010

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

0 20 40 60 80 100

HIV

Inci

denc

e (p

er Y

ear)

Age (Years)

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

0 20 40 60 80 100HI

V Inc

iden

ce (p

er ye

ar)

Age (years)

Females Males

Tanser, Bärnighausen, Newell CROI 2011

•80% life-time risk of acquiring HIV

Spatial clustering of

new HIV infections

Tanser et al CROI. Boston, MA; 2011.

Distribution of sero-conversions

Tanser et al CROI. Boston, MA; 2011.

Outline

• Background• ART coverage and risk of HIV acquisition• Population viral load

Population-based HIV surveillance

• Since 2003: Population-based HIV cohort– Longitudinal, dynamic cohort

– Entire adult population living in a contiguous geographical area of 438 km2 eligible for testing

– Annual rounds

– 75% of those observed to be HIV-uninfected subsequently retest

– All individuals geo-located

Tanser, Hosegood, Bärnighausen et al. International Journal of Epidemiology 2007

26237

26232

26241

2623826233

26235

26240

26236

2623926234

26231

26243

26242

26244

Adjusted for age, sex, community-level HIV prevalence, urban vs. rural locale, marital status, >1 partner in last 12 months, and household wealth index

<10% 10-20%

20-30%

30-40%

40-50%

>50%0

0.2

0.4

0.6

0.8

1

1.2

Proportion of all HIV positive people receiving ART

Ad

juste

d H

azard

rati

o

p=0.004

P<0.0001

p<0.0001P<0.0001

p=0.171

Survival analysis

> 17,000 HIV-negative individuals followed-up for HIV acquisition over 60,558 person-years

1,573 HIV sero-conversions

Time- (and space-) varying demographic, sexual behavior, economic and geographic controls, including HIV prevalence

Population impact of ART coverage on risk of HIV acquisition (2004-2012)

Conclusion

• We find continued reductions in risk of acquiring HIV infection with increasing ART coverage in this typical rural sub-Saharan African setting

• However, there is suggestion of a “reduction saturation” effect (at coverage of >40%) under treatment guidelines of <350 CD4+ cells/µl

Outline

• Background• ART coverage and risk of HIV acquisition• Population viral load

Population/Community viral load

• Proposed as:–an aggregate measure of the potential for ongoing HIV transmission within a community

–as a surveillance measure for monitoring uptake and effectiveness of antiretroviral therapy.

Figure 1

Miller, Powers, Smith et al Lancet Infectious Diseases 2013

CVL as a measure for assessment of HIV

treatment as prevention

CVL as a measure for assessment of HIV

treatment as prevention“ The issue of selection bias [in community viral load] could be addressed with a population-based survey in a clearly defined target population of all people in a community, including those with and without known HIV infection”

Miller, Powers, Smith et al Lancet Infectious Diseases 2013

Objectives

1. Asses whether viral loads in this population are randomly distributed in space or whether high or low viral loads tend to cluster in certain areas

2. Assess the degree to which different population viral load summary measures highlight known areas of high incidence

3. (Establish the degree to which different population viral load summary measures predict future risk of new HIV infection in uninfected individuals)

Methods

• All 2,456 individuals testing positive in a single round of the population-based HIV testing

• Total number DBS specimens tested was 2,420 (36 specimens excluded due to being insufficient for testing).

• Of these, 30% (726) were below the

detectable limit of 1550 copies/ml (Viljoen et al, JAIDS 2010)

01.

0e-0

62.

0e-0

63.

0e-0

6D

ens

ity

0 2000000 4000000 6000000 8000000 10000000VL Result (copies/ml)

Viral load distribution

Median viral load = 6428 copies per ml

Geometric mean population viral load

Copies /ml

Population prevalence of detectable virus (PPDV)

Prevalence(%)

Conclusion• To measure transmission potential of a

community, any viral load summary index must take into account the size of the uninfected population

• Population prevalence of detectable virus (PPDV) successfully identified known areas of continued high HIV incidence

• We propose the PPDV as a simple community-level index of transmission potential

AcknowledgementsWe thank the field staff at the Africa Centre for Health and Population Studies at the University of KwaZulu-Natal, South Africa, for their work in collecting the data used in this study and the communities in the Africa Centre demographic surveillance area for their support and participation in this study.

Co-authors

Till Bärnighausen, Deenan Pillay

Funding

Frank Tanser & Till Bärnighausen were supported by grant 1R01-HD058482-01 from the National Institute of Child Health and Human Development (NICHD), National Institutes of Health (NIH), USA. Core funding for the Africa Centre's Demographic Surveillance Information System (GR065377/Z/01/H) and Population-based HIV Survey (GR065377/Z/01/B) was received from the Welcome Trust, UK.