Background Information

 

The linear programming module of Nutrisurvey: a new tool for designing low cost, nutritionally adequate diets in developing countries

 

 

The development of sound food based dietary guidelines consistent with FAO/WHO nutrient intake recommendations during the complementary feeding period can be facilitated by the use of linear programming.

 

Linear programming is a mathematical technique for optimizing a linear objective function subject to a system of linear inequalities called constraints. It is perfectly adapted to give clear answers to practical questions faced in the field by nutritionists working in developing countries:

 

 

The linear programming module of Nutrisurey was made to design diets compatible with local food habits and fulfilling different sets of nutritional recommendations at the lowest possible cost.

 

 

Nutritional recommendations

 

The nutritional recommendations incorporated in the Linear Programming Module of Nutrisurvey are defined for the following age groups: 6-8 months; 9-11 months; 12-23 months. The choice between these recommendations are made by choosing the files fao-who6-8m.lpr; fao-who9-11mo.lpr and fao-who12-23mo.lpr respectively in the Recommendation windows when the linear programming function is activated.

The energy content of the diets obtained for these age groups is derived from the PAHO/WHO Guiding Principles for Complementary Feeding of the Breastfed Child (available at: http://www.who.int/child-adolescent-health/New_Publications/NUTRITION/guiding_principles.pdf ; see p 18 ). i.e. 615 kcal for 6-8 months, 686 kcal for 9-11 months and 894 Kcal for 12-23 months.

The minimum protein content of the diet for the age group is set at  9.1g, 9.6g, 10.9g for the age groups of 6-8 months, 9-11 months and 12-23 mo. respectively. These figures were obtained from the report: Complementary feeding of young children in developing countries: a review of current scientific knowledge, 1998, WHO/NUT/98.1, Table 25 p93; Available at: http://www.who.int/child-adolescent-health/New_Publications/NUTRITION/WHO_NUT_98.1/chapter4.pdf

The minimum content of fat of the whole diet, including breast milk, is set at 30% of total energy to be consistent with PAHO/WHO Guiding Principles on Complementary feeding (reference above, p 23). This represents a  minimum fat intake of 20.5g/d, 22.8g/d and 29.8g/day for 6-8 months, 9-11 months and 12-23 months respectively .

Minimum contents of calcium, magnesium, zinc, iron, thiamine, riboflavin, niacin, vit. B6, pantothenic acid, folate, vit. B12, vit. C, vit. A included in the programme correspond to the FAO WHO 2002 recommended intake (available at: ftp://ftp.fao.org/es/esn/nutrition/Vitrni/pdf/APPENDIX.pdf ).

For zinc, chosen recommended values correspond to a bioavailability of dietary zinc of 30%, corresponding to the diets of moderate bioavailability. For iron, it corresponds to a bioavailability of 10%.

When the bioavailability of zinc and iron is thought to be different, users are invited to change the iron and zinc requirements according to the FAO/WHO recommendations for different levels of bioavailability (see: Tables below, derived from: ftp://ftp.fao.org/es/esn/nutrition/Vitrni/pdf/APPENDIX.pdf).

 

 

Recommended intake of zinc (mg/day) for different levels of bioavailability

 

Age groups (months)

High (50%)

Moderate (30%)

Low (15%)

 7-11

 12-23 

 2.5

 2.4

 4.1

 4.1

8.3

 8.4

 

 

Recommended intake of iron (mg/day) for different levels of bioavailability

 

Age groups (months)

High (15%)

Moderate (10%)

Low (5%)

 7-11

 12-23

 

 6

 4

9

6

 19

 12

 

Food data base:

 

The food data base used by this software is derived from the data bases used in the  FAO WORLDFOOD DIETARY ASSESSMENT SYSTEM. This programme is available at: http://www.fao.org/infoods/software_worldfood_en.stm. Breast milk has been added. A different carotene retinol conversion factor was used and the niacin content of different foods was modified to take into account the contribution of tryptophane (see below)..

Foods found in duplicate in the original data base were removed. Foods with identical names but different nutrition content were kept with the name of the country of the original data base added.

The composition of breast milk was taken from the mean value reported in: Complementary feeding of young children in developing countries: a review of current scientific knowledge, 1998, WHO/NUT/98.1, Table 22 p84; Available at: http://www.who.int/child-adolescent-health/New_Publications/NUTRITION/WHO_NUT_98.1/chapter4.pdf

The energy content of breast milk was assumed to be 65 Kcal/100 ml as mentioned p 45 of Chapter 3 of the same report. Available at: http://www.who.int/child-adolescent-health/New_Publications/NUTRITION/WHO_NUT_98.1/chapter3.pdf

            Users can add new foods in the data base using, in the "Food" menu the function "modify food data base"

 

Estimation of zinc, iron and calcium absorption

 

To guide the user to select the recommended intake of zinc and iron in the FAO-WHO requirements, the proportion of absorption was estimated for these two nutrients. The methods is based on rough assumption, and the results should be interpreted with caution.

For iron, total absorbed iron was calculated, assuming a coefficient of absorption of  6% from plant source foods and 11% from animal source foods (including milk, which has lower iron bioavailability than meats) (ref: Dewey KG, Cohen RJ, Rollins NC; Informal Working Group on Feeding of Nonbreastfed Children. WHO technical background paper: feeding of nonbreastfed children from 6 to 24 months of age in developing countries. Food Nutr Bull. 2004 Dec;25(4):377-402). Iron absorption from breast milk will be assumed to be 20% (ref: Davidsson L, Kastenmayer P, Yuen M, Lonnerdal B, Hurrell RF. Influence of lactoferrin on iron absorption from human milk in infants. Pediatr Res. 1994 Jan;35(1):117-24 ; Abrams SA, Wen J, Stuff JE. Absorption of calcium, zinc, and iron from breast milk by five- to seven-month-old infants. Pediatr Res. 1997 Mar;41(3):384-90).  The ratio of absorbed iron over the total iron contained in the diet, expressed as a percentage, is calculated and displayed on the result sheet.

For Zinc, the percentage of absorption is calculated for each food, assuming the following values: 50% for all foods with a molar phytate:zinc ratio (PZR) less than 5;  30% for  PZR > 5 and PZR < 15; 15% for PZR >15 (World Health Organization (WHO), Food and Agriculture Organization (FAO), International Atomic Energy Association (IAEA). Trace elements in human health and nutrition. Geneva: World Health Organization, 1996).

Although the recommended intake of calcium in the FAO/WHO recommendations is not adjusted according to estimated absorption, the percentage of absorbed calcium was estimated to inform the user. Calcium absorption was assumed to be 25% for legumes, roots/tubers and grains, 5% for foods with high oxalate content (e.g. spinach), 45% for other fruits and vegetables and 32% for all other foods (including dairy products) (Weaver CM, Proulx WR, Heaney R.  Choices for achieving adequate dietary calcium with a vegetarian diet. Am J Clin Nutr. 1999 Sep;70(3 Suppl):543S-548S).

 

Estimation of retinol equivalents and niacin equivalents

 

A 12:1 conversion factor from carotene to retinol was used to calculate the retinol equivalent content of vegetable foods. A conversion factor of 2:1 was used for palm oil (see: IVACG, 2004. Conversion factors for vitamin A and carotenoids. Available at: http://ivacg.ilsi.org/file/webBookmark.pdf ).

            The World Food Dietary system assessment gives in its food composition table the content of preformed niacin. Instead, niacin equivalent is used in this programme, adding to the  preformed quantity of niacin the quantity which can be derived from tryptophan in a ratio of 1 mg of niacin equivalent for 60 mg of tryptophan. Breast milk will be assumed to contain 17 mg of tryptophan per g of protein (WHO Technical report 552, Geneva, 1973, p 58).

 

Breast milk intake

 

In absence of locally available data on breast milk intake, it is advised to use figures from the table below when choosing the amount of milk to enter in the models.

 

Table: Centiles of breast milk intakes for different age groups.

 

 

Age groups (months)

Centiles

6-8

9-11

12-23




3

390

293

197

10

480

396

309

25

572

500

423

50

674

616

549

75

776

732

675

90

868

836

789

97

958

939

901

 

These values are taken from the reference: Complementary feeding of young children in developing countries: a review of current scientific knowledge, 1998, WHO/NUT/98.1, Table 7 p47; Available at: http://www.who.int/child-adolescent-health/New_Publications/NUTRITION/WHO_NUT_98.1/chapter3.pdf . Centiles were calculated assuming a normal distribution around the mean: They were calculated using the Excel NORMINV function.

 

Price data

Price data should be entered in local currency. They should be entered for each food as price per 100g as eaten. This may require the introduction of some conversion factor for foods with a weight varying during cooking such are rice. Price data are incorporated in the food data base and can be saved when closing the linear programming function.

Minimum and maximum intake of each food

The range of intake should correspond to the average daily intake. A food eaten with a maximum amount of X g  twice per week should have for limit Xg x 2/7. The acceptable range of intake for foods selected from the food data base for each of the age groups can be saved in a separated file. This allows the user to save different sets of data corresponding to age groups or different seasons with different food availability scenarios or to rural or urban settings.

Food groups constraints

In order to avoid unrealistic diets, an upper limit of average daily intake can be set for the following food groups: 1)Grains and tuber; 2) Legumes/Nuts; 3) Dairy products; 4) )Flesh foods; 5) Eggs; 6) Vitamin A rich fruits or vegetable; 7) Other fruits and vegetables.

FAO and WHO (2002) recommended nutrient intakes included in the recommendation files of Nutrisurvey
 

 

6-8 months

9-11 months

12-23 months

energy (kcal)

616

686

894

protein ( g)

9,1

9,6

10,9

fat ( g)

20,5

22,9

29,8

Ret. equiv. (ug  )

400

400

400

Vit. B1 (mg)

0,3

0,3

0,5

Vit. B2 (mg)

0,4

0,4

0,5

Niacin equiv. (mg  )

4

4

6

pantoth. acid (mg)

1,8

1,8

2

Vit. B6 (mg)

0,3

0,3

0,5

folic acid eq. (µg)

80

80

160

Vit. B12 (µg)

0,5

0,5

0,5

Vit. C (mg)

30

30

30

calcium (mg)

400

400

400

magnesium (mg)

54

54

60

iron (mg)

9

9

6

zinc (mg)

4,1

4,1

4,1