Building on Uganda’s Progress in Reducing Anemia: From

Evidence to Action

Jack Fiedler, Deepali Godha, Manisha Tharaney, Jolene Wun, Alex Mokori, Sarah Ngalombi

Presentation Outline

• Anemia Situation in Uganda

• Rationale for the Anemia Landscape

Analysis

• Methods

• Findings of the Study

• National Stakeholder Meeting

• Next Steps

2

Anemia in Uganda

• Major public health

issue for many years

• Earliest national survey in 2001:

– 71% of children <5

years

– 37% of women of

reproductive age

• 2002- National Anemia Policy

3

Women’s anemia rates have been about half the rate of children, and have fluctuated relatively more over the 10 year period.

72 73

50

3641

23

0

10

20

30

40

50

60

70

80

2000/01 2006 2011

Pre

vale

nce

of

An

em

ia (

%)

Trends in Anemia in UgandaChildren (6-59m) and Non-Pregnant Women (15-49y)

6-59m Women 15-49y

+1%

-32%

+14%

-44%

-32%

4

Rationale for the Analysis

• To assess the plausibility of key anemia

prevention and control programs that may have contributed to the decline among women and children

• To stimulate cross-program and multi-sectoral discussion around anemia

programing

• To develop consensus about which programs may have been the largest contributors

5

Data: Uganda Demographic and Health Surveys

• DATABASE: 3 Household surveys 2001,2006 and

2011 conducted by the Uganda Bureau of

Statistics (UBOS)

• SAMPLE: stratified, two-stage cluster – 2000/01: 7,885 households

– 2006: 8,807 households

– 2011: 10,086 households

• REPRESENTATIVENESS: Provides estimates of

population and health indicators: – Nationwide,

– By rural and urban areas

– By regions (definitions change over time)

6

Sample Analyzed UDHS Population 2001 2006 2011

Women 15-49 years

Not pregnant at the time of survey

7988 3505 3208

Had at least one birth at the time of the survey

7609 3320 2918

Children 6-59 months

Children 6-23 months 2122 2367 2218

Children 24-59 months 3561 4415 4340

7

72 81 78

63

73

52

62 54

31

50

0

10

20

30

40

50

60

70

80

90

Central Eastern Northern Western Nationwide

Pre

vale

nce

of

An

em

ia in

Ch

ildre

n 6

-59

m

Changes in the Prevalence of Childhood Anemia by Region

2006 2011

-20 ppt -19 ppt -24 ppt -32 ppt -23 ppt

The largest gains were in Western—where rates were already the lowest—and in Northern Region.

Note: ppt = percentage points 8

76 74 69

59

50 52 46

42

0

10

20

30

40

50

60

70

80

no education primary secondary higher

Pre

vale

nce

of

An

em

ia in

Ch

ildre

n 6

-59

m

Educational Attainment of Mother

Changes in the Prevalence of Childhood Anemia by Mothers' Educational Attainment

2006 2011

Differences in anemia prevalence by educational level narrowed. The uneducated had the largest reductions.

Note: ppt = percentage points 9

80 75 73 74

61 62 53 52

42 38

0

10

20

30

40

50

60

70

80

90

poorest poorer middle richer richest

Pre

vale

nce

of

An

em

ia in

Ch

ildre

n 6

-59

m

Household Wealth Quintile

Changes in the Prevalence of Childhood Anemia by Household Wealth Quintile

2006 2011

-21 ppt -32 ppt -23 ppt

Differences in the anemia prevalence rates by household wealth grew. The two richest quintiles’ rates fell to about 40%, others remained above 50%

Note: ppt = percentage points 10

63 65

42

0

10

20

30

40

50

60

70

80

90

100

2001 2006 2011

Per

cen

t

Children 24-59 months

85 87

63

0

10

20

30

40

50

60

70

80

90

100

2001 2006 2011

Pe

rce

nt

Children 6-23 months

Age Differences in Childhood Anemia

6-23month olds: 24 PPT reduction, 28% decrease. 24-59 month olds: 23 PPT reduction, 35% decrease.

Note: ppt = percentage points 11

WRA with any anemia, by location

26

38

28

43

21 23

0

5

10

15

20

25

30

35

40

45

Urban Rural

2001

2006

2011

Change in Prevalence by Wealth Quintiles Among Women 15-49 years, (2001-2011)

39 41

37 38

26

43 42 45

41

34

29

25

18

22 20

0

5

10

15

20

25

30

35

40

45

50

Poorest Poor Middle Rich Richest

2001

2006

2011

Why Analyze Program Participation and Behavior Changes?

1. To understand how they changed (One of UNAP’s

objectives is to increase nutrition services delivery)

2. To explore the role they may have played in

reducing anemia

3. To identify opportunities for:

– Further improving the coverage of these services

– Promoting cross-program referrals/coordination in the delivery of other programs (provider side)

– Promoting the use of more services to more comprehensively address the many causes of anemia (consumer side)

14

10%15% 16%

4%11%

35% 35%30%

17%

29%

66% 68%74%

59%66%

0%

10%

20%

30%

40%

50%

60%

70%

80%

Central Eastern Northern Western Total

Pe

rce

nt

Changes in the Use of Bednets by All 6-23m Olds(Includes bednet owners and non-owners)

2001

2006

2011

9%17%

12%

3%10%

31% 30%25%

17%

25%

63% 60%

69%

59% 62%

0%

10%

20%

30%

40%

50%

60%

70%

80%

Central Eastern Northern Western Total

Pe

rce

nt

Changes in the Use of Bednets by All 24-59m Olds(Includes bednet owners and non-owners)

2001

2006

2011

15

Bed-net Ownership and Use among Non-Pregnant Women 15-49 years (2001-2011)

38% 36% 31%

62% 64% 69%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2001 2006 2011

Did not Sleep Under Slept Under

Proportion of non-pregnant women who owned a bed net and slept under it the previous night, 2001-2011. (Right) Among women who owned a bed net, percentage who slept under it, 2001-2011.

Increasing Coverage of Childhood Anemia-Prevention Programs

6-23 month olds 24-59 month olds

38

NA

10

67

3629

61

44

66

0

10

20

30

40

50

60

70

Vit

. A

Dew

orm

ing

Be

dn

ets

Per

cen

t

2001 2006 2011

40

NA

10

48 46

25

56 5562

0

10

20

30

40

50

60

70

Vit

. A

Dew

orm

ing

Be

dnet

s

Pe

rce

nt

2001 2006 2011

17

Program Participation/Behaviors by 6-59m olds: Number per Child

Includes: vitamin A supplementation, deworming and bednets. In 2001, deworming was not tracked, so the maximum number of programs was 2.

5736

12

39

35

29

4

23

33

0 625

0%

20%

40%

60%

80%

100%

2001 2006 2011

None One Two Three

5636

11

39

34

30

5

25

34

0 625

0%

20%

40%

60%

80%

100%

2001 2006 2011

None One Two Three

6-23 month olds 24-59 month olds

18

Continued Breastfeeding of Children at 1 and 2 Years of Age

93

52

88

44

0

10

20

30

40

50

60

70

80

90

100

At 1 Year At 2 Years

Pe

rce

nt

2006

2011

19

ANC Coverage (at least one visit), 2011 UDHS

< 65%

65-75 %

75-85 %

85-95 %

> 95 %

West Nile

Western

Southwest

North

Karamoja

Eastern

East Central

Central 1

Central 2

Kampala

Proportion of Women Receiving Various Anemia-Related Interventions during Pregnancy, 2001-2011

49

2

28

61

35

50

75

50

0

10

20

30

40

50

60

70

80

Deworming Fe Supplement Antimalarials

2001

2006

2011

Iron-Folic Acid Malaria prophylaxis

* Deworming not asked in 2001

IFA Coverage According to 2011 UDHS

< 65%

65-75 %

75-85 %

85-95 %

> 95 %

West Nile

Western

Southwest

North

Karamoja

Eastern

East Central

Central 1

Central 2

Kampala

IFA Falter Point Schematic Had at least

one ANC visit?

No

Yes

IFA tablets

received or

purchased?

Falter Point 1

No

Falter Point 2

Yes

IFA tablets

taken?

Falter Point 3

No

Yes

Took ≥ 180

tablets?

Yes:

SUCCESS No

Falter Point 4

Significance of Each Falter Point in Uganda

4%

22%

7%

65%

1% 0%

10%

20%

30%

40%

50%

60%

70%

80%

Falter Pt 1: Did not have ≥ 1 ANC visit

Falter Pt 2:Did not receive IFA

Falter Pt 3:Received but

did not take IFA

Falter Pt 4: Did not take ≥ 180 IFA

Did not falter: Took ≥ 180 IFA

Pe

rce

nt o

f W

om

en

UDHS limitations

• Three cross-sectional surveys, not panel data

• Data from different years are data on different individuals

• Not an analysis of causality, rather looks at correlations that are hypothesized to be causes, but can at most can only be regarded as plausible explanations

25

Dissemination of Findings • National Stakeholder

Meeting Oct 2-3, Uganda

• To develop consensus on

the status of the current

programs and their delivery

systems

• To begin development of

an action plan on “key

intervention” areas based

on available evidence and

local knowledge and

experiences

• 90 stakeholders from

national and sub-national

levels

Multi Sectoral Action Plan Developed

27

Thank You!

28