Design and evaluation of a lipid-based nutrient supplement to enrich local complementary foods: Results from Ghana Kathryn G. Dewey, PhD Program in International & Community Nutrition University of California, Davis
Dec 25, 2015
Design and evaluation of a lipid-based nutrient supplement to enrich local
complementary foods: Results from Ghana
Kathryn G. Dewey, PhDProgram in International & Community NutritionUniversity of California, Davis
Complementary feeding (6-24 mo): need for fortified products
Difficult to meet nutrient needs of infants (especially iron & zinc) without fortified foods
Commercially produced fortified complementary foods are being evaluated in several countries – key issue is cost
Home-fortification of complementary foods is a less costly alternative– Sprinkles (micronutrients only)– Lipid-based products (e.g. “Nutributter”)
Advantages of lipid-based products for home fortification
Stable, resistant to spoilage, prevent micronutrients from interacting
Provide some additional energy (and increase energy density of complementary foods)
Provide essential fatty acids May enhance absorption of fat-soluble vitamins Taste good; can mask taste of other nutrients such as
B vitamins Can be consumed alone or mixed with other foods Can easily divide the dose during the day Can be locally produced & stimulate local economy
Randomized controlled trial of home fortification of complementary foods
with 3 types of micronutrient supplements in Ghana
S Adu-Afarwuah1, A Lartey2, KH Brown1, A Briend3, S
Zlotkin4, KG Dewey1
1 Prog. Intl. Nutr., UC Davis2 Univ. of Ghana, Legon3 Inst. de Recherche pour le Développement, Paris, France 4 Hospital for Sick Children, Univ. of Toronto, Canada
Study design
At 6 mo, infants (n=313) randomly assigned to receive supplement daily, 6-12 mo
– Sprinkles (Fe, Zn, Vit A, Vit C, folate)– Nutritabs (all of the above plus 11 other micronutrients)– Nutributter (19 micronutrients plus fat; 108 kcal/d)
Anthropometric & biochemical assessment at 6 & 12 mo; diet & morbidity assessed weekly; motor development observed at 12 mo
Non-intervention group (n=96), eligible but not randomly selected for the intervention, assessed at 12 mo only
Sprinkles (SP) (per sachet/d)
Vitamin A (μg RE) 300 Vitamin C (mg) 50 Vitamin D3 (μg) 7.5 Folic acid (μg) 150 Iron (mg) 12.5 (microencapsulated, fumarate) Zinc (mg) 5 (gluconate)
Nutritabs (NT) (per tablet/d)
Vitamin A (μg RE) - 400 Vitamin C (mg) - 30 Folic acid (μg) - 80 Vitamin B1(mg) - 0.3 Vitamin B2(mg) - 0.4 Vitamin B3(mg) - 4.0 Pantothenic acid (mg)- 1.8 Vitamin B6(mg) - 0.3 Vitamin B12 (μg) - 0.5 Iron (mg) - 9.0 (sulfate) Zinc (mg) - 4.0 (oxide) Calcium (mg) - 100 (carbonate) Potassium (mg) - 152 (chloride) Copper (mg) - 0.2 (sulfate) Selenium (μg) - 10 (sodium selenite) Iodine (μg) - 90 (potassium iodate)
Nutributter (NB) (per 20 g dose/d)
Vitamin A (μg RE) - 400 Vitamin C (mg) - 30 Folic acid (μg) - 80 Vitamin B1 (mg) - 0.3 Vitamin B2 (mg) - 0.4 Vitamin B3 (mg) - 4.0 Pantothenic acid (mg) - 1.8 Vitamin B6 (mg) - 0.3 Vitamin B12 (μg) - 0.5 Iron (mg) - 9.0 (sulfate) Zinc (mg) - 4.0 (sulfate) Copper (mg) - 0.2 (sulfate) Selenium (μg) - 10 (sodium selenite) Iodine (μg) - 90 (potassium iodate) Calcium (mg) - 100 (phosphate) Potassium (mg) - 152 Total energy (kcal) - 108 Linoleic acid (g) - 1.29 Linolenic acid (g) - 0.29
Plus some P, Mg and Mn, mainly from the ingredients
Based on RNI, 6-12 mo
RNI minus amount from other sources
RNI minus amount from other sources
Study design
Potential participants: n=612
Approx. 75% randomly selected for intervention at 5 mo (n=442)
Not selectedat 5 mo (n=170)
SP: n=98 NT: n=102 NB: n=98 NI: n=96
SP: n=105 NT: n=105 NB: n=103
Randomized into 3 Intervention groups at 6 mo (n=313)
SP= Sprinkles; NT=Nutritabs; NB= Nutributter; NI=Non-intervention
Seth Adu-Afarwuah explains some details to study mother
Picture by Dr. K.H. Brown
A child at 9 mo being weighed in her home
Picture by Seth Adu-Afarwuah, UCD and Univ. of Ghana,
Measuring the head circumference of a nine-month-old boy in his home
Picture by Seth Adu-Afarwuah, UCD and Univ. of Ghana,
A field worker demonstrates to a study mother how she should administer Nutritabs to her child. We provided a plastic cup and a stainless steel spoon to the mother to help her measure the appropriate amount of food (about 3 spoonfuls) and mix it with the supplement.
Picture by Dr. K H Brown, UCD
Assessment of motor development at 12 mo at the laboratory. The little girl holding the table is going through the “standing with assistance” procedure, whilst the other boy has already been through all the assessment procedures.
Picture by Seth Adu-Afarwuah, UCD and Univ. of Ghana,
Iron status: Ferritin < 12 ug/L at 12 mo
0
10
20
30
40
50
60
SP NT NB Control
% lowa
aa
b
Anemia: Hemoglobin < 100 g/L at 12 mo
0
10
20
30
40
50
60
SP NT NB Control
% lowaa
a
b
Growth and energy intake from CF, 6-12 mo
SP (n=96)
NT (n=101)
NB (n=97)
p
Wt gain (kg) 1.39 ± 0.50ab 1.35 ± 0.48a 1.57 ± 0.73b 0.02
Ln gain (cm) 7.9 ± 1.4ab 7.8 ± 1.4a 8.3 ± 1.4b 0.04
Energy from CF(kcal/d, 7-12 mo)
139 ± 76a 133 ± 69a 224 ± 71b <0.0001
SP = Sprinkles; NT= Nutritabs; NB = Nutributter; CF = Complementary foods.
Length-for-age z-scores of intervention (SP, NT, NB) and non-intervention (NI) groups
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
6 12
Age (mo)
LA
Z
NB
NT
SP
NI
SP=Sprinkles; NT=Nutritabs; NB=Nutributter; NI=Non-intervention
SP, NI
NB
NT
Motor development: % walking at 12 mo
0
10
20
30
40
50
60
SP NT NB Control
%
aa
a
b
What is responsible for improved outcomes in the Nutributter group?
Possible explanations: Increased energy intake Favorable ratio of essential fatty acids Inclusion of milk
Path analysis of observed effects
Exogenous variable:
Group assignment
Dependent variables:
-Weight gain
-Length gain
Potential intermediary variables:
-Energy from CF -FA status
Does increased energy from CFs explain effect of Nutributter on weight and length gain?
MeanDiff. from NB
Mean Diff.from NB, with CF energy in model
% Diff. due to CF energy
Mean p Mean p %
Weight gain, kg1
SP group NT group
-0.18-0.20
0.040.02
-0.10-0.11
0.280.25
4346
Length gain, cm1
SP group NT group
-0.39-0.44
0.040.02
-0.46-0.50
0.030.02
N/A
Controlling for baseline values, child sex and maternal height.SP=Sprinkles, NT=Nutritabs, NB=Nutributter.
Answer: Yes (partially) for weight gain; No for length gain
Plasma fatty acid analyses
Individual fatty acids and fatty acid classes– 14:0, 16:0, 18:0, 20:0, 22:0 and 24:0 – 16:1n7, 18:1n7, 18:1n9, 20:1n9 and 24:1n9– 18:2n6, 18:3n6, 20:2n6, 20:3n6 and 20:4n6 – 18:3n3, 20:5n3, 22:5n3 and 22:6n3
Total fatty acids– Saturated fatty acids (SFA)– Monounsaturated fatty acids (MUFA)– n-6 fatty acids– n-3 fatty acids– Polyunsaturated fatty acids (PUFA)
Ratios– PUFA:SFA– n-6:n-3 PUFA
Plasma α-linolenic acid (mg/L) at 12 mo
1 Mean ± SD adjusted for baseline values, mean BF frequency at 6-12 mo and sex; 2 Mean ± SD adjusted for BF frequency in the week prior to 12 mo and sex. ** Different from NB, p< 0.05; * Different from NB, p <0.08.
NI SP NT NB p
3-group comparison1
n/a 5.5 ± 4.9**(n=70)
5.2 ± 2.9**(n=74)
7.3 ± 5.0(n=70)
0.03
4-group comparison2
6.2 ± 5.5*(n=72)
5.6 ± 5.5**(n=76)
5.6 ± 5.5**(n=78)
7.9 ± 5.6(n=71)
0.05
Plasma DHA (mg/L) at 12 mo
1 Mean ± SD adjusted for baseline values, mean BF frequency at 6-12 mo and sex; 2 Mean ± SD adjusted for BF frequency in the week prior to 12 mo and sex.
NI SP NT NB p
3-group comparison1
n/a 117 ± 53(n=70)
120 ± 53(n=75)
131 ± 54(n=70)
0.24
4-group comparison2
128 ± 61(n=72)
122 ± 61(n=76)
127 ± 61(n=78)
136 ± 62(n=71)
NS
Plasma saturated fatty acids (% total) at 12 mo
1 Mean ± SD adjusted for baseline values, mean BF frequency at 6-12 mo and sex; 2 Mean ± SD adjusted for BF frequency in the week prior to 12 mo and sex. ** Different from NB (p< 0.05).
NI SP NT NB p
3-group comparison1
n/a 38.3 ± 2.1**( n=70)
38.4 ± 2.1**(n=72)
37.5 ± 2.1(n=68)
0.02
4-group comparison2
38.5 ± 2.2**(n=72)
38.2 ± 2.2(n=76)
38.6 ± 2.2**(n=77)
37.7 ± 2.2(n=71)
0.09
Correlations of plasma fatty acids at 12 mo with weight and length gain 6-12 mo
1 Pearson correlation coefficient for all children in intervention groups. Adjusted for baseline FA values, mean BF frequency at 6-12 mo of age and child sex
Fatty acids Length gain Weight gain
18:3n3 (mg/L) 0.20[p = 0.003]
0.03[p = 0.61]
SFA (% of total) -0.13[p = 0.05]
-0.10[p = 0.45]
Controlling for baseline values, child sex and maternal height.SP=Sprinkles, NT=Nutritab, NB=Nutributter
Mean Diff. from NB
Mean Diff. from NB with ALA in
model
% Diff. due to ALA
Mean Diff. from NB with SFA in model
% Diff. due to SFA
p p % p %
Length gain,cm1
SP group NT group
-0.39-0.44
0.040.02
-0.12-0.29
0.600.19
6934
-0.16-0.35
0.490.12
5921
Weight gain,kg1
SP group NT group
-0.18-0.20
0.040.02
-0.19-0.24
0.060.02
N/A -0.20-0.27
0.050.01
N/A
Do changes in fatty acid status explain the effect of Nutributter on growth 6-12 mo?
Answer: Yes (partially) for length gain; No for weight gain
Summary of potential explanations
Nutributter increased energy intake from comp. foods, increased blood PUFA levels (+33-40% in ALA) and decreased blood SFA levels.
Path analysis suggests that– Increased energy intake from CFs explained (part of)
the impact on weight gain, but not length gain– The shift in plasma FA explained (part of) the impact on
length gain, but not weight gain
However, growth effect could also be due to milk content
Acceptability
Supplement consumption (% of days, 6-12 mo): SP 86%, NT 88%, NB 88%
Child accepted food “well” when mixed with supplement: SP 90%, NT 78%, NB 86%
> 97% of mothers in all 3 groups a) liked giving the supplement, b) thought it helped child’s health & c) wanted to purchase in future
How much willing to pay? (US cents/day)
SP NT NB
~5 cents/d 93% 89% 99%*
~8 cents/d 69% 53% 81%**
Mean (SD) 8 (5) 8 (5) 10 (7)**
* p < 0.05; ** p < 0.01
Summary of results
All 3 supplements were well accepted All 3 supplements improved iron status compared to
the Non-Intervention group Only Nutributter improved growth Motor development was improved by all 3
supplements, but the effect was largest with Nutributter Providing a larger set of micronutrients in a lipid-based
product (with milk powder?) confers benefits beyond those of providing just a few selected key micronutrients
Lipid-based Nutrient Supplements (LNS): Convergence of approaches for treatment (SAM/mod malnut) & primary prevention?
RUTF (LNS in large doses) effective & feasible for treatment of SAM in the community
LNS (moderate dose) more effective than cereal-legume blends for supplementary feeding of moderately malnourished children
LNS for home fortification (small daily dose) more effective than preparations with micronutrients only
Daily ration of LNS can be chosen based on needs of target population & cost constraints
Goal: integrate CTC with programs for prevention of malnutrition
LNS Network
Development and evaluation of lipid-based nutrient supplements (LNS) for prevention of malnutrition: an innovative food-based approach
University of California, Davis, USA University of Tampere, Finland University of Malawi University of Ghana Many advisors and observers
Research agenda, LNS Network
Development and testing of new LNS formulations and doses for children 6-24 mo– EFA content, zinc content, milk + / -, ration/day
Supplementation of pregnant and lactating women
Large-scale effectiveness trials (primary outcome: stunting)
Socio-economic studies; cost-benefit evaluation Partnerships and scaling up LNS interventions
Publications (Ghana study)
Adu-Afarwuah, S., Lartey, A., Brown, K. H., Zlotkin, S., Briend, A. and Dewey, K. G. (2007), Randomized comparison of 3 types of micronutrient supplements for home fortification of complementary foods in Ghana: effects on growth and motor development, Am J Clin Nutr, 86: 412-20.
Adu-Afarwuah, S., Lartey, A., Brown, K. H., Zlotkin, S., Briend, A. and Dewey, K. G. (2008), Home fortification of complementary foods with micronutrient supplements is well accepted and has positive effects on infant iron status in Ghana, Am J Clin Nutr, 87: in press.
Acknowledgments (Ghana study)
Nestle Foundation and the Human Nutrition Institute of the International Life Sciences Institute (via USAID) for funding
Regional and municipal health administration, and laboratory staff in Koforidua, Ghana
Study team, nurses, mothers in Ghana
Laboratory of Dr. Bruce German at UC Davis for fatty acid analyses
Diane Vandepeute for administrative support at UC Davis
Jan Peerson for statistical support at UC Davis
Project staff in Koforidua, Ghana
Picture by Unknown