Mitchell J. Besser, MDFounder and Medical Director
mothers2mothers
14 December 2010
mothers2mothers: Preventing Mother-to-Child HIV
Transmission in Africa Using New Paradigms in Health Care
Delivery
Population HIV Prevalence
0
10
20
30
Bo
tsw
an
a
Le
so
tho
So
uth
Afric
a
Za
mb
ia
Ma
law
i
Ke
ny
a
Ta
nza
nia
Ug
an
da
Rw
an
da
Eth
iop
ia
Co
te D
'Ivo
ire
Bu
rkin
a
Sie
rra L
eo
ne
Ca
me
roo
n
Gu
ine
a
Gh
an
a
Se
ne
ga
l
Ma
li
Ca
mb
od
ia
Ha
iti
DR
Pe
ru% Population
Zim
bab
we
So
uth
A
frica
Bo
tswan
a
Sen
egal M
ali
with high HIV prevalence:
ZimbabweSouth AfricaBotswana
with low HIV prevalence:MadagascarSenegalMali
Source: UN Department of Economic and Social Affairs (2001) World Population Prospects, the 2000 Revision.
30
35
40
45
50
55
60
65
Lif
e e
xp
ecta
ncy
(y
ears
)
1950– 1955
1955- 1960
1960-1965
1965-1970
1970-1975
1975-1980
1980-1985
1985-1990
1990-1995
1995-2000
2000-2005
Life Expectancy: 1950-2005
Grim RealityU
NA
IDS
: 20
10
FACTS:
• 2.6 million new HIV infections (2010)
• 1.8 million adults and children died of HIV/AIDS (2010)
• 1.2 million people started on treatment in 2010
Each year:►Twice as many people become infected with HIV as start on treatment;►Twice as many people die of AIDS as start on treatment.
Global HIV Prevalence
33.3 million living with HIV in 2010
Sub-Saharan Africa
22.4 million
90% of HIV-positive pregnant women are in Sub-Saharan Africa
1.4 million pregnant women in low- and middle-income countries are infected with HIV
Towards Universal Access, WHO, 2009
Grim Inequities
Prevalence in Pregnancy
Nigeria 2.4 - 4%
South Africa 29%
US and UK 0.6%
Adult Prevalence
Nigeria 3%
South Africa 18%
US and UK 0.3 - 0.6%
PMTCT Coverage
Nigeria 10%
South Africa 73%
US and UK > 95%
ARV Coverage (children)
Nigeria 12%
South Africa 61%
US and UK > 95%
Pediatric HIV infections in U.S.
80% decline
CDC
PACTG 076 – AZT treatment starts
Siripon Kanshana, 2007
Pediatric HIV infections in Thailand
80% decline
HIV testing PMTCT ARVs
National PMTCT program
UNAIDS estimates 2008
Pediatric HIV infections in World
1100 children infected per day in 2008
≈1000 in Africa
<1 in the US1 in Europe
100 in Asia & the Pacific
Annual pregnancies in HIV-positive women
United States (7000)
Rwanda (8000)
South Africa (300,000)
Baragwanath Hospital (8000)
(Soweto)
0
10
20
30
40
50
60
70
80
90
100
ANC clinic visits Accepting VCT Receive results HIV-positive NVP Mom NVP baby
PMTCT CascadePMTCT Cascade
Gap = Missed Opportunities
Reducing the Risk: Treatment
PMTCT TreatmentTransmission
Rate at 6 Weeks
No Treatment 25%
Single-dose Nevirapine (sdNVP) 12%
AZT (28 weeks) + sdNVP 3%
HAART 1%
Myth of the 80s
640
512
Enter into program
800
488 No Rx (25% MTCT):
122 infected
200
160
128
Missed - no PMTCT
Transmission rates:• NVP (12% MTCT): 61
infected• AZT+NVP (3% MTCT): 15 infected• HAART (1% MTCT): 5
infectedAdapted from P. Barker, IHI, WHO PMTCT Mtg Nov 2008, L Mofenson, 2009
Number of HIV+ babies:• NVP: 183• AZT+NVP: 137• HAART: 127
Attend ANC: 80%
Counseled and tested for HIV: 80%
Get ARVs: 80%
1000 positive mothers
Reality of the 80s
865
804
Enter into program
930
196 No Rx (25% MTCT):
49 infected
70
65
61
Missed - no PMTCT
Transmission rates:• NVP (12% MTCT): 96
infected• AZT+NVP (3% MTCT): 24 infected• HAART (1% MTCT): 8
infectedAdapted from P. Barker, IHI, WHO PMTCT Mtg Nov 2008, L Mofenson, 2009
Number of HIV+ babies:• NVP: 145• AZT+NVP: 73• HAART: 57
Attend ANC: 93%
Counseled and tested for HIV: 93%
Get ARVs: 93%
1000 positive mothers
HIV Infected Babies
Myth of the 80s
Reality of the 80s
NVP:
12%183 146
AZT+NVP:
3%137 73
HAART:
1%127 57
HIV Infected Babies
Myth of the 80s
Reality of the 80s
NVP:
12%183 146
AZT+NVP:
3%137 73
HAART:
1%127 57
Four Prongs of PMTCT
Prevention of unintended pregnancies among
HIV infected women
Preventing mother to child
transmission of HIV
Primary prevention of HIV infection in women
Provision of care and support for HIV
infected mothers, their infants, partners and families
Incident HIVIncident HIV
Challenges and Responses
Country Ratio Data Source
Ethiopia 6:1 DHS-05
Tanzania 3:1 AIS 03/04
Kenya 2:1 DHS-03
Uganda 2:1 AIS-04/5
Discordant/concordant
Couple Status – Discordance Predominates
Country …% Woman HIV + Data Source
Ethiopia 27% DHS-05
Tanzania 37% AIS 03/04
Uganda 55% AIS-04/5
Kenya 57% DHS-03
If man is HIV+ and in a couple…
Couple Status – Discordance Predominates
Incident HIV in pregnancy = new infections in women with a documented negative test in that pregnancy
Where effective interventions have reduced transmission in women identified as HIV-positive, new infections during pregnancy may be a major
source of MTCT.
MTCT rates associated with new infection to mother:• 70% during pregnancy and childbirth• 36% during breastfeeding
HIV Incidence in Pregnancy
T Creek, personal communication 2008
Impact of Incident HIV Infection in Pregnancy
A study in Botswana showed:
Among women testing negative in early pregnancy:
Botswana National PMTCT program transmission data show:
Extrapolating incident HIV to the national Botswana figures :
Incident HIV is thus estimated to account for 470/1090 (43%) of all infant infections in 2007
1.3% were infected in 17 weeks before
delivery
1.8% were infected in the first year after
delivery
13,900 HIV+ women infected estimated 620 infants (4.7%)
Estimate 950 women acquired HIV during pregnancy or first postpartum year, and infected 470 infants
Feeding the BabyFeeding the Baby
• Infant feeding is one of the most difficult and most emotive issues in HIV management in low-resource settings
• An estimated 300,000 children acquire infection through breast feeding each year
• HIV transmission in early childhood remains a challenge in places where infant formula cannot be safely provided
Infant feeding and HIV
Am J Epidem 1995
Timing of HIV Transmission (non Breastfeeding cohort)
Months
0
5
10
15
20
25
30
35
1 12 18
% H
IV
infe
cte
d
Transmission
Cumulative
4%4% Transmission of HIV for every Transmission of HIV for every 66 months of breast-feeding months of breast-feeding
6-3
Deli
very
Timing of HIV Transmission
Delivery
Infant Feeding and HIV Transmission
Infant Age
Coutsoudis,13th AIDS Conf, 2000
8% 7% 7%
19% 19%
26%
0
10
20
30
% T
ran
smis
sio
n
1 Day 6 Mos
Never Breastfed (N=157)Exclusive Breastfed (N=118)Mixed Feeding (N=276)
35.1% 35.1% 29.9%
18.0%
35.2%
11.9%
0%
20%
40%
60%
80%
100%
% E
xclu
sive
BF
Uganda (J Trop Ped
2004)
Kenya (JAIDS2004)
S Africa (AIDS 2001)
Botswana (CROI 2005)
Ivory Coast (AIDS Conf
2004)
Zimbabwe(AIDS Conf
2004)
Exclusive Breastfeeding in HIV+ Women in Clinical Trials in Resource-Poor Countries
Requirements for safe replacement feeding
Assessment should also include:
• Health service accessibility• Counselling and support
available
Is replacement feeding…
…for the mother and baby
Henderson, WHO, 2006
A cceptableF easibleA ffordableS ustainable S afe
Feeding Method - Issues
• Clean water – “tap”
• Clean bottles – “flame” or “electric”
• Source of formula– Free– Accessible
• Disclosure
HIV Infection through Age 7 Months is Higher in Breast Fed than Formula Fed Infants
5.0% 4.6% 5.6%
9.1%
0%
10%
20%
% H
IV
Tra
ns
mis
sio
n
1 Month 7 Months
Infant age
Formula Breast
p=0.04
Thior I et al. JAMA 2006;296:794-805
7 Months
Mortality through Age 7 Months is Higher in Formula Fed than Breast Fed Infants
4.3%1.5%
9.3%
4.9%
0%
10%
20%
% M
ort
ali
ty
1 Month 7 MonthsInfant age
Formula Breast
p=0.004
Predominant causes infant death:Diarrheal disease and pneumonia
7 Months
No Difference in 18-Month HIV-Free Survival
Between Formula Fed and Breast Fed Infants
8.9%6.1%
12.6% 12.9% 14.2%15.6%
0%
10%
20%
30%
% H
IV-F
ree
Su
rviv
al
1 Month 7 Months 18 Months
Infant age
Formula Breast
p=0.41
Thior I et al. JAMA 2006;296:794-805
18 Months
No Difference in 18-Month HIV-Free Survival
Between Formula Fed and Breast Fed Infants
8.9%6.1%
12.6% 12.9% 14.2%15.6%
0%
10%
20%
30%
% H
IV-F
ree
Su
rviv
al
1 Month 7 Months 18 Months
Infant age
Formula Breast
p=0.41
Thior I et al. JAMA 2006;296:794-805
18 Months
FF: 33 infected, 46 deathsBF: 54 infected, 34 deaths
BAN: Probability HIV positive or death by week 28 visit in infants uninfected at birth
Age (weeks)
Pro
babi
lity
HIV
pos
itive
or
deat
h
1 4 8 12 16 20 24 28
0.00
0.02
0.04
0.06
0.08 Control vs Maternal HAART: p= 0.03
7.6%
4.7%
2.9%
Control – No ARVs
Maternal HAART
Infant NVP
BAN Study: Probability of HIV+ or Death by 28 week visit in infants uninfected at birth
BAN: Probability HIV positive or death by week 28 visit in infants uninfected at birth
Age (weeks)
Pro
babi
lity
HIV
pos
itive
or
deat
h
1 4 8 12 16 20 24 28
0.00
0.02
0.04
0.06
0.08 Control vs Maternal HAART: p= 0.03
7.6%
4.7%
2.9%
Control – No ARVs
Maternal HAART
Infant NVP
BAN Study: Probability of HIV+ or Death by 28 week visit in infants uninfected at birth
7.6%
4.7%
BAN: Probability HIV positive or death by week 28 visit in infants uninfected at birth
Age (weeks)
Pro
babi
lity
HIV
pos
itive
or
deat
h
1 4 8 12 16 20 24 28
0.00
0.02
0.04
0.06
0.08 Control vs Infant NVP: p <0.0001
7.6%
4.7%
2.9%
Control – No ARVs
Maternal HAART
Infant NVP
BAN Study: Probability of HIV+ or Death by 28 week visit in infants uninfected at birth
7.6%
2.9%
BAN: Probability HIV positive or death by week 28 visit in infants uninfected at birth
Age (weeks)
Pro
babi
lity
HIV
pos
itive
or
deat
h
1 4 8 12 16 20 24 28
0.00
0.02
0.04
0.06
0.08 Maternal HAART vs Infant NVP: p= 0.07
7.6%
4.7%
2.9%
Control – No ARVs
Maternal HAART
Infant NVP
BAN Study: Probability of HIV+ or Death by 28 week visit in infants uninfected at birth
4.7%
2.9%
Integrating HIV into the push for MDGs 4 & 5
GOAL 4: REDUCE CHILD MORTALITY
GOAL 5: IMPROVE MATERNAL HEALTH
GOAL 6: COMBAT HIV/AIDS, MALARIA & OTHER DISEASES Target 1: Have halted by 2015 and begun to reverse the spread of HIV/AIDS
Target 2:Achieve, by 2010, universal access to treatment for HIV/AIDS for all those who need it
Target 1: Reduce by two thirds, between 1990 and 2015, the under-five mortality rate
Target 1: Reduce by three quarters the maternal mortality ratio
Target 2:Achieve universal access to reproductive health
H
IV /
AID
S
Magical Magical ThinkingThinking
Health Systems
Global HIV Prevalence
Doctors Working in the World
Sub-Saharan Africa:
25% of global disease burden
3% of world’s health workers
Health care workers per 100,000 population (2007)
Region/Country Physicians Nurses
United States 256 937
South Africa 77 408
Botswana 40 265
Zambia 12 174
Zimbabwe 16 72
Lesotho 5 62
Mozambique 3 21http://www.hst.org.za/uploads/files/cahp9_07.pdf
Staffing Ratios
Health care workers per 100,000 population (2007)
Region/Country Physicians Nurses
United States 256 937
South Africa 77 408
Botswana 40 265
Zambia 12 174
Zimbabwe 16 72
Lesotho 5 62
Mozambique 3 21http://www.hst.org.za/uploads/files/cahp9_07.pdf
Staffing Ratios
SA Population (2009): 49 million
population dependent on public health sector
health professionals in public sector
85% 44%
10%
vacant posts in public
health sector
Doctors
Nurses36%
34%30%
Has only minutes per
patient
1) Counsel for HIV test 2) Perform HIV test, explain results3) Dispense single dose nevirapine,4) Explain how to take5) Discuss infant feeding options6) Reinforce exclusive infant feeding7) Perform infant HIV test at 12-months, 8) Explain results
PMTCT Program Interventions:
In 2001…
Transmission Rates: 14-16%
Nurse:
Still has only minutes per
patient
1) Counsel for HIV test 2) Perform HIV test, explain results3) Perform CD4 test, get and explain results. Refer for HAART if CD4<3504) Dispense cotrimoxazole5) Discuss infant feeding options6) Dispense AZT (from 14 weeks), explain how to take7) Dispense HAART (if eligible), explain how to take8) Counsel on adherence to HAART
9) Screen for HAART related toxicity 10) Reinforce exclusive infant feeding 11) Where ARVS for breast feeding are
available, explain how to use 12) Perform infant HIV test at 6
weeks, 13) Explain results 14) Refer mother to follow-up care,
15) Encourage her to attend 16) RH/FP
PMTCT Program Interventions:
In 2010…
Transmission Rates: 2-5%
Nurse:
...
mothers2mothers
• Individual and group meetings
• Daily presence for education and support
• Mentor Mothers: professional members of health care team - paid for service
Mothers are a community’s single greatest resource
Mothers living with HIV (Mentor Mothers) educate and support HIV-positive pregnant women and new mothers in health facilities
Simple, scale-able model of careSimple, scale-able model of care
Goal 3. Empowerment
Goal 2. Healthy mothers
and infants
Threem2m goals:
Goal 1. PMTCT
Site Coordinators and Mentor Mothers
• Recruited locally
• Selection Criteria: Mothers HIV-positive Attended PMTCT Disclosed
• Basic numeracy and literacy skills
• Site Coordinators manage services
• Mentor Mothers engaged for up to two years
• Curriculum based education• 2 weeks - Mentor Mothers • 3 weeks - Site Coordinators• Periodic top-up training
Training
Training Cascade:National Trainer Site Coordinator / Mentor Mother Patients
Points of ServicePoints of Service
• Prenatal clinics• Labor and delivery• Postnatal programs• Targeted community outreach
m2m services
task-shifting
m2m Services
• Counsel for, or perform HIV testing• Provide medication• Distribute formula
• Support medical services that do
m2m Does:
m2m Does Not:
Site Management Plan
MM MM
MM
MM
SC
MM
SC
Tertiary Care
Hospital
Primary Health Center
Site Systems
Regional or District Program Manager
SC = Site CoordinatorMM = Mentor Mother
Community Outreach
Community Outreach Community Outreach
Satellite Health Centres
Hospital or Major HC
Site System
Research Questions
Does mothers2mothers:
• Increase HIV-positive women’s utilization of PMTCT services?
• Improve PMTCT outcomes and psychosocial well-being?
Population Council: Horizons Study (2007)
• PMTCT– 95% of mothers received nevirapine– 88% of babies received nevirapine
• Care– 79% had CD4 counts– 88% knew CD4 count results
• Infant Feeding– 89% chose exclusive infant feeding method
• Family Planning– 70% using contraception
• Disclosure – 97% disclosed (4.4x non-participants)
Population Council: Horizons Study (2007)
• Pregnant women felt they could:– Do things to help themselves – Cope with taking care of baby– Live positively
• Postpartum women felt less:– Alone in the world– Overwhelmed by problems– Hopeless about future
Program Participants Report Better Psychosocial Well-Being
Population Council: Horizons Study (2007)
m2m Outcomes – Lesotho (2010)
• Program started 2007– 58 program sites
• 64% m2m country coverage • 77% disclosure rates among clients attending
>3 times• 79% early infant diagnosis
CD4 and HAART uptake among m2m antenatal clients (N=1246)
Antenatal m2m clients
CD4 Tests
CD4 Results
CD4 <350
HAART
CD4 and HAART uptake among m2m antenatal clients by number of visits (N=1246)
“Standard” intervention
Clinic based care only
•Improved education on importance of early infant HIV testing•Use of m2m wheel to estimate date of return visit•Client information sheet – date and location of baby HIV test
“Enhanced” intervention
Clinic based care + active client follow-up
•8 week telephone call •10 week call + home visit if needed
Early Baby Testing Study
Baby HIV Testing Tool
Baby HIV Test Card
Resources Developed for StudyResources Developed for Study
“Standard” intervention
Clinic based care only
•Improved education on importance of early infant HIV testing•Use of m2m wheel to estimate date of return visit•Client information sheet – date and location of baby HIV test
“Enhanced” intervention
Clinic based care + active client follow-up
•8 week telephone call •10 week call + home visit if needed
Early Baby Testing Study
Cellphone Subscribers - 2009
North America: 276m
Africa: 358m
Non-ACFU Sitesn = 204
ACFU Sitesn=214
Eligible 8 week callYes =179 (84%), No =35 (16%)
Eligible 10 week call & home visit Yes =53 (32%), No =114 (68%)
* Of 167 consenting for home visit
10 week home visitReached =24 (45%)
Not Reached = 22 (42%)Excluded/too far = 7 (13%)
Early Infant Diagnosis StudyEarly Infant Diagnosis Study
Non-ACFU Sitesn = 204
ACFU Sitesn=214
Eligible 8 week callYes =179 (84%), No =35 (16%)
Eligible 10 week call & home visit Yes =53 (32%), No =114 (68%)
* Of 167 consenting for home visit
10 week home visitReached =24 (45%)
Not Reached = 22 (42%)Excluded/too far = 7 (13%)
Early Infant Diagnosis StudyEarly Infant Diagnosis Study
60% reached by phone (108)
Non-ACFU Sitesn = 204
ACFU Sitesn=214
Eligible 8 week callYes =179 (84%), No =35 (16%)
Eligible 10 week call & home visit Yes =53 (32%), No =114 (68%)
* Of 167 consenting for home visit
10 week home visitReached =24 (45%)
Not Reached = 22 (42%)Excluded/too far = 7 (13%)
Early Infant Diagnosis StudyEarly Infant Diagnosis Study
Non-ACFU Sitesn = 204
ACFU Sitesn=214
Eligible 8 week callYes =179 (84%), No =35 (16%)
Eligible 10 week call & home visit Yes =53 (32%), No =114 (68%)
* Of 167 consenting for home visit
10 week home visitReached =24 (45%)
Not Reached = 22 (42%)Excluded/too far = 7 (13%)
Early Infant Diagnosis StudyEarly Infant Diagnosis Study
Non-ACFU Sitesn = 204
ACFU Sitesn=214
Eligible 8 week callYes =179 (84%), No =35 (16%)
Eligible 10 week call & home visit Yes =53 (32%), No =114 (68%)
* Of 167 consenting for home visit
10 week home visitReached =24 (45%)
Not Reached = 22 (42%)Excluded/too far = 7 (13%)
Early Infant Diagnosis StudyEarly Infant Diagnosis Study
11% of total (214) reached by home visit
Early Infant Diagnosis StudyEarly Infant Diagnosis Study
(N = 55,13) (N = 115,167)
p < 0.01
Early Infant Diagnosis StudyEarly Infant Diagnosis Study
(N = 55,13) (N = 115,167)
p < 0.01
Early Infant Diagnosis StudyEarly Infant Diagnosis Study
(N = 55,13) (N = 115,167)
* Of those with known test results
mothers & babies
communities
healthcare systems
Demand and Supply Side:
South Africa
M2M2B - 2001M2M2B - 2001
South Africa
Malawi
Ethiopia
Kenya
Rwanda
Zambia
Swaziland
Lesotho
Botswana
Uganda
Mozambique ???
Tanzania
m2m 2010
84
mm22mm Today Today
Total HIV-positive pregnant women enrolled:
20% of the global disease burden
mm22mm Activities 2010 Activities 2010
Current Date Nov-10Sites 703Field Staff 1766Patient Encounters Per Month 267,103
New HIV-positive Women Per Month 22,111
Helping MothersSaving Babies