Reuben Granich HIV/AIDS Department World Health Organization

Early access to voluntary HIV counseling and testing, bed nets and water filters through an

integrated campaign in Kenya: potential impact on HIV and TB transmission

Reuben GranichHIV/AIDS Department

World Health Organization

International AIDS Conference, Vienna

July 18-23, 2010

Outline:

• Background

• Methods

• Results

• Conclusions

Smallpox eradication 1796 to 1977:Edward Jenner to Merca Town, Somalia

Campaigns against Guinea Worm

Guinea worm eradication

Campaigns against preventable blindnessPreventable blindness

Community counseling and testing is feasible and works in a wide variety of settings

Photos courtesy of Bunnell R, Marum E, and Vestergaard Frandsen

Late initiation of ART and mortality

Source: Egger M, CROI 2007

Havlir, Getahun et al. 2008 JAMA 300(4):423-430

CD4 level is associated with TB incidence:earlier start may decrease TB risk

"TB death zone"

Slide adapted by Dr. Abhishek Sharma

Coverage of ART among eligible people living with HIV

Kenya (2007 KAIS)

HIV test

57% Unaware of status, not

on ART

4% know status, not on ART

39% know status,

on ART

Among those who knew status and were eligible 92% were on ARTMohammed, CROI 2009

57%

39%

• ~1,300,000 persons living with HIV (15-64yrs)

• Prevalence 7.1% • range (1% to 15%)

• Nyanza and Rift Valley home to >50% of HIV-infected adults (15-64 yrs)

Kenya Aids Indicator survey (KAIS) 2007

Population – 37,538,000

Kisii

Kisii (Nyanza) campaign sites

Kisii Level 5 Hospital

Mosocho

Keumbu Kiogoro

Methods• Kisii District, Nyanza Province (population 4.7

million; target population 5000)• September 2009 three days day, three site

campaign• Campaign objective:

– Lessons from the 2008 Kakamega campaign– Feasibility of scaling up onsite CD4 cell counts – Testing the campaign in a peri-urban setting – Determine potential benefits of early HIV

identification• Interventions:

– LLIN, water filters, 60 condoms– Health education information

Methods: logistics and laboratory• Ministry of Health and Vestergaard Frandsen collaboration• Participants with HIV received a 3-month supply of

cotrimoxazole and referral for further care and treatment• Campaign cohort used Guava AUTOCD4 flow cytometers

for CD4 counts • Hospital reference cohort used Becton Dickinson FACS

Calibur• External quality control of CD4 counts

– 5% campaign samples sent to Kisii Hospital for quality control

– Kisii hospital works in partnership with CDC Kisumu for quality control of CD4 counts

Methods: Hospital cohort

• Kisii Hospital 6 month historical reference cohort abstracted from medical records (March to August 2009)

• First CD4 counts of all patients aged 15 and above diagnosed with HIV

• Historical cohort of 1284 patients• Modeling of potential benefits was done using

similar parameters and model – 2007 KAIS CD4 data

Results– Campaign reached 5203 individuals

• Package given to 5203 (100%) of people whether or not they decided to have HIV test

• 100% tested for HIV• 329 (6.3%) tested HIV positive• 255 (78%) had CD4 count determination • median of 536 cells/ µL (IQR 350;759)

– Kisii Hospital reference cohort• 1284 first CD4 counts • Median 348 cells/ µL (IQR 185;551)

CD4+ cell count distribution in Nyanza Province (KAIS survey) and the Kisii campaign

0.0

0.1

0.2

0.3

0.4

0.5

0-250 250-500

500-750

750-1000

1000-1250

1250-1500

1500-1750

1750-2000

Nyanza and Campaign Nyanza Province (KAIS)Campaign participants

No significant difference in the distributions

CD4+ cell count distribution in Nyanza Province (KAIS survey) and the Kisii campaign

hospital cohort

0.0

0.1

0.2

0.3

0.4

0.5

0-250 250-500

500-750

750-1000

1000-1250

1250-1500

1500-1750

1750-2000

Nyanza and ReferenceNyanza Province (KAIS)Kisii Hospital reference cohort

Hospital cohort significantly lower

CD4+ cell count distribution in Kisii hospital reference cohort and the Kisii campaign

0.0

0.1

0.2

0.3

0.4

0.5

0-250 250-500

500-750

750-1000

1000-1250

1250-1500

1500-1750

1750-2000

Reference and Campaign Kisii Hospital reference cohortCampaign participants

Campaign cohort significantly higher

0.00

0.20

0.40

0.60

0.80

1.00

0 500 1000 1500 2000CD4/micro-litre

Cum

ulat

ive

prob

abili

ty

ReferenceCampaignKAIS (Kissii)

Hospital referenceCampaignNyanza (KAIS)

CD4+ cell count frequency distribution (Hospital reference, Campaign, KAIS)

Modeled impact of campaign approach for Nyanza Province

Percent of HIV positive population

Number started on ART

EstimatedHIV transmissions averted

Deaths averteddirectly

TB cases averted per year

CD4< 200 8% 24,000 5,000 24,000 2,400

CD4 <350 26% 76,000 27,000 76,000 3,600

Immediate 90% 268,000 241,000 268,000 4,700

Conclusions

• Integrated campaigns have a significant potential to bring services to people where they live– Possible to bring CD4 technology to the community

• Improved access translates into earlier HIV diagnosis

• Early HIV diagnosis has significant prevention, care and treatment benefits

• To reach Universal Access and achieve MDGs we will need to expand access to HIV services in the community

People of Kisii town and surrounding areas Nicolas Muraguri (MoH Kenya)Alex Doyen (Vestergaard Frandsen)Navneet Garg (Vestergaard Frandsen)Brian Williams (SACEMA)MoH Kenya staffVestergaard Frandsen

Thank you

Public health is purchasable. Within a few natural and important limitations any community can determine its own health.

--Hermann M. Biggs(29 Sep 1859 - 28 Jun 1923)New York City's Public Health Officer and public health pioneer