Assessment of PEPFAR’s Impact on Selected Health System Parameters in Sub-Saharan African Countries Presented by: Anya Shen Viviane D. Lima, Wendy Zhang, Carly Heung, Alexis Palmer, Julio Montaner, Robert Hogg, Nathan Ford, Edward Mills MOAE0101
Feb 08, 2016
Assessment of PEPFAR’s Impact on Selected Health System Parameters
in Sub-Saharan African Countries
Presented by: Anya ShenViviane D. Lima, Wendy Zhang, Carly Heung, Alexis Palmer, Julio Montaner,
Robert Hogg, Nathan Ford, Edward Mills
MOAE0101
Overview
• PEPFAR results by September 2008 Supported treatment for more than 2.1 million peopleOverall HIV prevention estimates unknown. Estimated
240,000 mother-to-child transmissions averted10.1 million people received care globally
• Other published studiesDecreased HIV-related deaths in PEPFAR focus countries in
Sub-Saharan Africa comparing to other countries in the same region1
• ‘The President’s Emergency Plan for AIDS Relief in Africa: An Evaluation of Outcomes’ Bendavid, et all. 2009
Research Objectives
• PEPFAR is a vertical program• There is a need to investigate the effect that disease-
specific vertical programs have on health systems and population health
• The objective:– Assessment of PEPFAR’s impact on selected population
health parameters in Sub-Saharan African countries
MethodsFocus countries: Botswana, Cote d’lvoire, Ethiopia, Kenya, Mozambique, Namibia, Nigeria, Rwanda, South Africa, Tanzania, Uganda, Zambia
Control countries: Angola, Benin, Burkina Faso, Burundi, Cameroon, Central African Republic, Chad, Congo, Democratic Republic of Congo, Djibouti, Eritrea, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Lesotho, Liberia, Madagascar, Malawi, Mali, Niger, Senegal, Sierra Leone, Somalia, Sudan, Swaziland, Togo, Zimbabwe
Focus
Control
Methods• Longitudinal data collected from WHO, UN, World
Bank, US Census Bureau• Time frame
1. Pre-PEPFAR era [1999-2002]2. PEPFAR era [2004-2007]
• Comparison of trends in focus and control countries before and after the initiation of PEPFAR programs
• Mixed-effects Models, General Linear Regression Models
Mean Change between Pre-PEPFAR era and PEPFAR era, CI
Comparison of Mean Change between Focus and Control, CI*
P-value
Focus -10.76 (-13.49, -8.03 )* -5.14(-0.63, -9.66)* P=.0028
Control -5.62(-7.40, -3.83)*
60
80
100
120
140
160
180
1999 2000 2001 2002 2003 2004 2005 2006 2007
Un
der
5 M
ort
ali
ty R
ate
(p
er
1,0
00
liv
e b
irth
s)
Year
Under 5 Mortality Rate
Focus
Control
40
50
60
70
80
90
100
110
1999 2000 2001 2002 2003 2004 2005 2006 2007
Infa
nt M
ort
ali
ty R
ate
(per
10
0,0
00
li
ve b
irth
s)
Year
I nfant Mortality Rate
Focus
Control
Average Change between Pre-PEPFAR era and PEPFAR era, CI
Comparison of Average Change between Focus and Control, CI
P-value
Focus -3.18 (-5.50, -0.86)* 1.61(-2.21, 5.42) P=.246
Control -4.79(-6.28, -3.29)*
400
500
600
700
800
900
1999 2000 2001 2002 2003 2004 2005 2006 2007
Mate
rnal
Mort
ality
Rate
(p
er
10
0,0
00
liv
e
bir
ths)
Year
Maternal Mortality Rate
Focus
Control
Average Change between Pre-PEPFAR era and PEPFAR era, CI
Comparison of Average Change between Focus and Control, CI
P-value
Focus -43.46 (-68.21, -18.71 )* -6.58(-47.26, 34.09) P=.653
Control -36.88 (-52.80, -20.95)*
47
48
49
50
51
52
53
1999 2000 2001 2002 2003 2004 2005 2006 2007
Lif
e E
xp
ect
an
cy
Year
Life Expectancy
Focus
Control
Average Change between Pre-PEPFAR era and PEPFAR era, CI
Comparison of Average Change between Focus and Control, CI
P-value
Focus 0.42 (-0.20, 1.04) -0.08 (-1.10, 0.94) P=.825
Control 0.50 (0.10, 0.90)*
Pre-PEPARMean Annual Change (1999-2002)
PEPFARMean Annual Change (2004-2007)
Difference in Mean Annual Change between Two Eras*
Focus -0.30(-0.65, -0.05)* 0.81(0.52, 1.10)* 1.11(0.57, 1.75)*
Control 0.18(0.04, 0.41)* 0.45(0.37, 0.54)* 0.27(-0.04, 0.50)
47
48
49
50
51
52
53
1999 2000 2001 2002 2003 2004 2005 2006 2007
Lif
e E
xp
ect
an
cy
Year
Life Expectancy
Focus
Control
PRE-PEPFARMean Annual Change (1999-2002)
PEPFARMean Annual Change (2004-2007)
Difference in Mean Annual Change between Two Eras
Focus 2.95(-3.70 to 9.59) -2.52(-5.01 to -0.02)* -5.47(-14.60 to 3.68)
Control -1.31(-6.41 to 3.80) 3.26(-5.75 to 12.26) 3.26(-5.75 to 12.26)
HIV/TB Related Death Rate
20
30
40
50
60
70
80
90
100
1999 2000 2001 2002 2003 2004 2005 2006 2007
Year
HIV
/TB
De
ath
Rat
e
FocusControl
60
70
80
90
100
110
120
130
140
1999 2000 2001 2002 2003 2004 2005 2006 2007
TB
Death
Rate
Year
TB Death Rate
Focus
Control
Pre-PEPFARMean Annual Change
(1999-2002)
PEPFARMean Annual
Change (2004-2007)
Difference in Mean Annual Change between Two Eras
Focus 5.52(-2.09 to 13.12) -3.41(-6.22 to -0.60)* -8.93(-19.34 to -1.49)*
Control -1.33(-7.21 to 4.55) 2.64(-1.74 to 7.03) 3.97(-2.81 to 14.24)
200
250
300
350
400
450
500
550
1999 2000 2001 2002 2003 2004 2005 2006 2007
TB
In
cid
en
ce R
ate
Year
TB I ncidence Rate
Focus
Control
Pre-PEPFARMean Annual Change
(1999-2002)
PEPFARMean Annual
Change (2004-2007)
Difference in Mean Annual rate of Change
between Two Eras
Focus 32.45(23.94, 40.97)* -13.21(-17.86, -8.57)* -45.66(-58.83, -32.51)*
Control 15.89(11.25, 20.53)* 2.44(-0.30, 5.17) -13.45(-20.83, -6.08)*
Mean Annual Change (1999-2002)
Mean Annual Change (2004-2007)
Difference in Mean Annual Change between Two Eras
Focus 18.84(11.01, 26.67)* -8.36(-11.81, -4.90)* -27.20(-38.48, -15.91)*
Control 8.29(4.51, 12.08)* -0.20(-2.04, 1.63)* -8.49(-2.88, 14.12)*
0
50
100
150
200
250
300
350
1999 2000 2001 2002 2003 2004 2005 2006 2007
HIV
/TB
Inci
denc
e R
ate
Year
HI V/ TB I ncidence Rate
Focus
Control
Limitations
• Other confounders - Global Fund• Countries not picked at random & significant
baseline differences.• Population-based survey estimates in resource
limited countries have the potential for bias and problems with results based on aggregated measures.
• Lack of data in key indicators.• Lack of continuous data.
Conclusion• This analysis demonstrates the impact of PEPFAR
funding on mortality rates & population health indicators in Sub-Saharan Africa
• A significantly improving trend is shown in U5MR and Life Expectancy in PEPFAR-focus countries when compared with control countries
• Positive trends are evident in maternal mortality (though non-sig) rates in focus countries from Pre-PEPFAR era to PEPFAR era.
Acknowledgement
• The author would like to thank Dr. Viviane D Lima and Wendy Zhang for their statistical support, co-authors and the rest of the BC Centre for Excellence staff for their support and encouragement.
• Special thanks to Dr. Edward Mills for his encouragement and guidance for without which this project would not have been started.