Landscape of Nutrition: the 2008 Lancet
Series on Maternal and Child Nutrition
Reynaldo
Martorell
Hubert Department
of Global Health,
Rollins School of
Public Health
Emory University
Atlanta, GA. USA
The Causes of Malnutrition
(adapted from UNICEF)
Malnutrition and Death
Poor diets
Insufficient
access to
FOOD
Infection
Inadequate
child CARE
BASIC CAUSES
Poor environment
and inadequate
HEALTH services
Individual
Family
Society
About the Series
Paper 1: prevalence and short-term consequences
(deaths and disease burden)
Paper 2: long-term educational and economic effects
and associations with adult chronic diseases
Paper 3: evidence-based interventions to
significantly reduce the effects of undernutrition
Papers 4 & 5: scaling up interventions through
actions at national and global levels
Gestation and the first 2 years of life represent a
window of vulnerability for human development
• High nutritional
requirements
• Rapid growth and
development
• Greater susceptibility
to infections
• Fully dependent on
others for care
Dietary Recommendation for Iron
(mg/kg/day) from high quality diets+
1.5
1.20
1.0
0.5
0.12
0.0
Infant, 9 months
+ Dietary Reference Intakes. IOM, 2006.
Adult Male
Mean Z-scores for height-for-age relative to
the new WHO standards for Peru, 2000
Age (months)
Z
Mean Z-scores for height-for-age relative to
the new WHO standards for Peru, 2000
Age (months)
Z
Human brain development
Original: Thompson, 2001; taken from Grantham-McGregor, 2007
Brain development in early life
• Brain development early life is
dramatic and depends on the interaction
between genes and how the child
experiences the world.
• Nutrition, child care and stimulation in
early life effect directly the wiring of
the brain and this has a lasting effect on
learning capacity, behavior and the
ability to regulate emotions.
+
Reversing the Real Brain Drain. Early Years Study. April 1999.
High Prevalence of IUGR, Stunting and Severe
Wasting in Children Under 5 ( Black et al, 2008)
13 million babies are born each year
with intrauterine growth restriction
178 million children are stunted; 32%
of all children
19 million children are severely
wasted
PAPER 1
Mortality in children < 5 y attributable to
nutritional problems
Deaths
Stunting
1,491,188
14.5%
Wasting
449,160
4.4%
IUGR
337,047
3.3%
↓ Vitamin A
667,771
6.5%
↓ Zinc
453,207
4.4%
↓ Iron +++
20,854
0.2%
↓ Iodine
3,619
0.03%
Total++
+
% of deaths in children
<5y
2,800,000
Total+
2,184,973
21.4%
Total
1,145,451
11.1%
28%
Takes into account the joint distribution between stunting and wasting.
++
Takes into account the joint distribution among nutritional problems.
+++
Responsible for 115,000 maternal deaths per year or 20% of maternal mortality
Suboptimal Breastfeeding
Suboptimal
breastfeeding
accounts for
1.4 million
deaths
annually
PAPER 1
Mortality
statistics do not
capture long
term effects on
human capital
Maternal and child undernutrition
(paper 2)
Consequences for
adult health and human capital
Victora CG, Adair L, Fall CHD, Hallal PC,
Martorell R, Richter L, Sachdev HPS
• a review of published data linking
maternal and child undernutrition with
adult outcomes
• a combined analysis of new data from 5
developing countries where people have
been monitored from birth to adulthood
Delhi
Guatemala
Cebu
Soweto
Pelotas
Damage Suffered in Early Life Leads to
Permanent Impairment
Undernourished children are more likely to
become short adults and to give birth to
smaller babies
Evidence links stunting to cognitive
development, school performance and
educational achievement
Poor fetal growth or stunting in the first 2 years
of life leads to reduced economic productivity
in adulthood
Child’s height for age is best predictor of
human capital
PAPER 2
Rapid Weight Gain After Being
Undernourished Increases Chronic Disease
Risk as Adults
Children whose early growth is restricted and
gain weight rapidly later are more likely to
have high blood pressure, diabetes and both
cardiovascular and metabolic disease
No evidence that rapid weight (?) or length
gain in first 2 years increases risk of chronic
disease later in life
By supporting early nutrition and growth,
incidence of chronic disease could be
reduced
PAPER 2
Years of increased schooling associated with a standard deviation
shift in birth weight and conditional weight gain (0-24 and 24-48m)
in 5 cohort studies, after control for confounding
Adjusted for SES and maternal education
0.43
0.21
0.07 (N.S.)
One standard deviation (z) of
birth weight =0.5 kg
weight gain (0-24m) = 0.7 kg
weight gain (24-48m) = 0.9 kg
Martorell et al., under review
Lancet Series on Maternal and
Child Undernutrition: paper 3
What works? Interventions to affect
maternal and child undernutrition
and survival globally.
Bhutta et al, The Lancet, 2008
Evidence-Based Interventions
Effective interventions are available to reduce stunting,
micronutrient deficiencies and child deaths and nutrition-related
disability
Interventions showing the most promise for reducing child deaths
and future disease burden include:
•
•
•
•
Breastfeeding promotion
Appropriate complementary feeding
Supplementation with vitamin A and zinc
Appropriate management of severe acute malnutrition
PAPER 3
Interventions with Sufficient Evidence to
Implement in All Countries
Maternal and Birth
Outcomes
• Iron folate
supplementation
• Maternal supplements of
multiple micronutrients
• Maternal iodine through
iodization of salt
• Maternal calcium
supplementation
• Interventions to reduce
tobacco consumption or
indoor air pollution
Newborn Babies
Infants and Children
• Promotion of
breastfeeding (individual
and group counseling)
• Promotion of
breastfeeding (individual
and group counseling)
• Behavior change
communication for
improved complementary
feeding
• Zinc supplementation
• Zinc in management of
diarrhea
• Vitamin A fortification or
supplementation
• Universal salt iodization
• Handwashing or hygiene
interventions
•Treatment of SAM
Interventions with Sufficient Evidence to
Implement in Specific Situational Contexts
Maternal and Birth
Outcomes
• Maternal supplements of
balanced energy and protein
• Maternal iodine
supplements
• Maternal deworming in
pregnancy
• Intermittent preventative
treatment for malaria
• Insecticide-treated bednets
Newborn Babies
Infants and Children
• Neonatal vitamin A
supplementation
• Delayed cord clamping
• Conditional cash transfer
programs (with nutritional
education)
• Deworming
• Iron fortification and
supplementation programs
• Insecticide-treated
bednets
Paper 3 Key Messages
If implemented at scale, these
interventions would:
• Reduce all child deaths by one-quarter in the short
term
• Reduce prevalence of stunting at 36 months by
one-third
PAPER 3
Meta-analyses of micronutrients
and birthweight (Martorell in press)
Composition
# of
studies
Unweighted
birthweight Unweighted
differences effect sizes
(g)
95% C.I.
P
0.011
Iron vs.
Placebo
7
110
0.24
0.06-0.43
MM vs. Iron
8
63
0.14
0.05-0.33 < 0.001
MM vs.
Placebo
0
?
?
?
?
Meta-analyses of the preventive impact of
zinc supplementation on morbidity and
mortality ( Bhutta et al, 2008)
Outcome
Risk ratios (95% CI)
Reduction
Mortality
0.91 (0.82 – 0.99)
9%
Diarrhea
0.86 (0.79 – 0.93)
14%
Severe diarrhea
/dysentery
0.85 (0.75 – 0.95)
15%
Persistent diarrhea
0.75 (0.57 – 0.98)
25%
Pneumonia
0.80 (0.70 – 0.92)
20%
Meta-analysis of the impact of zinc
supplementation on linear growth in children
(Bhutta et al, 2008)
• No new analyses about zinc and growth were carried out for
the Lancet series.
• The results published by Brown, Peerson, Rivera and Allen
(2002) were accepted for the series.
• These authors found a mean effect size on height changes of
0.35 Z (0.19 – 0.51) in an analyses that included children of
all ages .
• Zinc supplementation features as one of the most potent
measures to reduce stunting.
Penny
Umeta (I)
Umeta (II
Gardner
Kikafunda
Lira (I)
Lira (II)
Rivera
Rosado (I)
Rosado (II)
Ruz
Ninh
Sempertegui
Castillo-Duran
Dirren
Bates
Hong
Walravens
Walravens
Walravens
Walravens
EFFECT SIZE
Wuehler (I)
Wuehler (II)
Wuehler (III)
Brown (I)
Brown (II)
Fahmida
Berger (I)
Berger (II)
Heinig
Olney (I)
Olney (II)
Silva
Wasantwisut (I)
Wasantwisut
Brooks
Gardner
Alarcon
Black
Black (I)
Black (II)
Lind (I)
Lind (II)
Diaz-Gomez
Muller
Zlokin
Osendarp
Yang (I)
Yang (II)
Castillo-Duran
Dijkhuizen (1)
Dijkhuizen (2)
Hershkovitz
EFFECT SIZE
-2
Effect of zinc on growth in height
in children less than 5 y
(Ramakrishnan, Nguyen and
Martorell, 2009)
Studies included by Brown
et al ( 2002)
Effect size 0.26 Z
(95% CI: 0.08, 0.43)
Studies not included by
Brown et al ( 2002)
Effect size 0.02 Z
(95% CI: -0.07, 0.11)
All available studies
0.07 Z (95% CI: -0.03, 0.17)
-1.5
-1
-0.5
0
0.5
1
1.5
2
Stratified analyses of zinc effects on
growth in height in children less than 5 y
(Ramakrishnan, Nguyen and Martorell, 2009)
1
0.5
0.35
0
0.04
-0.5
-1
Population not stunted at
baseline; mean HAZ ≥ -2
Population stunted at
baseline; mean HAZ ≤ -2
Interventions to improve
complementary feeding (Bhutta et al, 2008)
Intervention
Number
Type of
of studies population
Effect size
Height/age (Z)+
Education/communication,
without food or $ transfers
3
Food
secure++
0.25 (0.01-0.49)
Food or $ transfers,
with or without
education/communication
7
Food
insecure
0.41 (0.05-0.76)
Education /communication,
without food or $ transfers
0
Food
insecure
??
+ Weighted mean difference and 95% C.I.
++ More than 1 US$ per day per capita
From Bhutta et al, 2008
Poverty
Stunting
Y Chen and LA Zhou, The long-term health and
economic consequences of the 1959–1961 famine in
China. J Health Econ, 26: 659-681, 2007.
• 20 million people died in the 1959-62 famine in
China (Great Leap Forward)
• Survivors born in 1959-62 (exposed early in life)
were compared to cohorts born before or after the
famine.
• Famine exposure in early life was associated with
a reduction in:
– Adult height , 3 cm
– Annual per capita income, 33% lower (my estimate)
INCAP longitudinal study, 1969-77
•
•
Community-randomized supplementation trial
(2 large and 2 small villages).
Two villages (1 large, 1 small) received Atole,
a nutritious supplement made from
Incaparina, milk and sugar, and two (1 large,
1 small) received Fresco, a less nutritive
drink.
Martorell R, Habicht J-P, Rivera JA. History and design of the INCAP longitudinal study (1969-77) and its
follow-up (1988-89). Journal of Nutrition 125(Suppl. 4S):1027S-1041S, 1995.
Impact on total nutrient
intakes and on growth
•
•
Total diets of young children from Atole
villages were greater by 9 g of protein, 100
kcal/day and in micronutrients when
compared to diets of children from Fresco
villages.
Length was increased by 2.5 cm in atole vs.
fresco but only in the first three years of life.
Difference in length at 3 years of age between children (n = 453)
exposed to supplement during their lives and those measured at
baseline: village level analysis
t-test = 24.50
p < .005 (2-tail, df = 2)
cm
Large
Large
Small
Atole
Villages
Small
Fresco
Econometric analysis of exposure to improved
nutrition from 0-3 years of age on education (n = 1469)+
• Schooling: Effects found in women only
-- Improved by 1.2 years
• Cognition:
Effects found in men and women
– Improved Raven scores by 8% ++
• Reading:
Effects found in men and women
– Improved scores by 17% +++
+
Maluccio, Hoddinott, Behrman, Martorell, Quisumbing & Stein. The Impact of Nutrition during Early Childhood on
Education among Guatemalan Adults. The Economic Journal, 2009
++ Raven Progressive Matrices
+++ Inter-American Reading Series
Long terms effects of a nutrition intervention
carried out in Guatemala in 1969-77
Improved nutrition before,
but not after 3 years of
age, increased wages in
men by 46% and annual
incomes by $914
_______________________________
Hoddinott, Maluccio, Behrman, Flores
and Martorell. Effect of a nutrition
intervention during early childhood on
economic productivity in Guatemalan
adults (The Lancet, 2008).
The Lancet
series did not
consider
implementation
issues in detail
nor costs of
scaling up
programs
Letter to the editor of Lancet entitled
“Delivery Sciences in Nutrition”:
Meera Shekar y 17 others (May, 2008)
• Seeks to understand the implementation and cost
effectiveness at scale
• The letter proposes that we know more about “Discovery”
than “Delivery Sciences” and analyzes the causes of lack of
capacity in Delivery Sciences
– Lack of explicit training in leading universities
– Near absence of research on delivery sciences , including
operational research
– What little is done is of poor quality
– Scarce funding for research on delivery sciences
– Reluctance of journals to publish such research