FEATURES

Highlights of the World Nutrition Situation - Updated trends from SCN report
Food Prices and Nutrition
Food Security and Nutrition, 1971-91 - Lessons Learned and Future Priorities
Long-Term Effects of Improved Childhood Nutrition

Highlights of the World Nutrition Situation - Updated trends from SCN report

This is a summary of the findings in the recently published ACC/SCN Second Report on the World Nutrition Situation - Volume I. Global and Regional Results. This feature provides an all round overview of the current situation to be followed by more details in the next issue of SCN News.

Malnutrition continues to affect large numbers of people, as summarized in the table below. An estimated 20% of the population has inadequate food consumption. Growth failure affects one-third of children, and over 40% of women are underweight and/or anaemic. At least one billion people worldwide are probably affected by one or more nutritional deficiency.

Famine in Africa

The period 1990-1992 marked one of the most severe times of food shortage and famine in Africa. Drought and war led to famine in Ethiopia, Somalia, and Sudan in the Horn of Africa, and in Mozambique and Liberia. These caused massive movements of refugees into neighbouring countries. However, the nutritional effects of the severe drought in many other parts of southern and eastern Africa were contained by government and external action.

Malnutrition in Developing Countries, 1975-1990

	Percent affected			Number (millions)
	1974-76		1988-90	1974-76		1988-90
GENERAL MALNUTRITION
1. Population (all ages) with energy intake (kcals/caput/day) on average below 1.54 BMR over one year	33%		20%	976		786
	1975		1990	1975		1990
2. Children (under five years) with weight below -2 S.D. of reference	42%		34%	168		184
		1980s			1980s
3. Women (15-49 years old) with weight below 45 kg		45%			400
MICRONUTRIENT MALNUTRITION
4. Anaemia: women (15-49 years old) haemoglobin <12 g/dl (non pregnant) or <11 g/dl (pregnant)		42%			370
5. Iodine deficiency disorders (IDD) Goiter (all ages)		5.6%			211
6. Vitamin A deficiency: children (under five years) with xerophthalmia		2.8%			13.8

Notes:

1.	Data on population with low energy intake (underfed) were calculated by FAO. The estimates are averages for 1974 to 1976 and 1988 to 1990.
2.	Underweight children results are estimated by ACC/SCN, for children aged 0 though 60 months, using a cut-off of -2 S.D. of the median NCHS reference.
3.	The estimate of underweight adult women is calculated from ACC/SCN's database on women's nutrition. The 45 kg cut-off is used as a basis for comparison, as (hat commonly reported. The prevalence estimates exclude pregnant and lactating women but these are included in the calculation of numbers.
4.	Anaemia estimates are based on ACC/SCN's database on women's nutrition. The cut-off points for anaemia use the WHO reference for pregnant and non-pregnant women.
5.	IDD estimates are based on WHO and ICCIDD data.
6.	Vitamin A deficiency estimates are based on WHO data.

(Source: ACC/SCN (1992) Table 1.1, p. 2.)

Overall, the percentage of children underweight fell in the 1980s, from around 38% in 1980 to 34% in 1990 (see figure 1). The improvement was more rapid from 1975 to 1980, from 42% to 38%. This rate of improvement in the late 1970s was just enough to reduce the total numbers of pre-school children underweight, but these are estimated to have risen again in the 1980s, from around 164 million in 1980 to 184 million in 1990.

Figure 1. Trends in prevalence of underweight children, 1975-1990

(Source: ACC/SCN (1992) Figure 1.2, p.10)

The proportion of the population in developing countries underfed - consuming dietary energy inadequate to sustain more than light activity on average - was estimated to have fallen substantially over the last 15 years, from around one in three people in 1975, to one in five in 1989. This implies a considerable reduction in the numbers so affected, from nearly 1,000 million to just below 800 million. These calculations from FAO use a new indicator of low consumption, and revised methods of estimation, now including China. By this calculation, less people today are underfed than at any time in the recent past.

This estimate is considerably influenced by the improving situation in China. Excluding China (and certain other East Asian countries, such as Korea and Vietnam), a slight increase in total numbers underfed is still estimated by FAO, from around 540 million in 1979/81 to about 580 million in 1988/90.

Sub-Saharan Africa

Nutritional trends have generally deteriorated or remained static in Sub-Saharan Africa during the 1980s, in contrast to all other regions. There are indications that some African countries with extensive community-level programmes may have achieved nutritional improvement, against the trend.

South Asia

South Asia is estimated to be improving slowly, according to recent results from India and elsewhere, at around a reduction of half a percentage point prevalence of underweight children per year. Nonetheless, the underweight prevalence in South Asia remains the highest in the world, and over half the world's underweight children are in this region. Indications are that calorie consumption remained low throughout the 1980s, with little change, although this may have improved slightly for some poorer groups such as the landless.

South East Asia

Nutrition in South East Asia is improving rapidly in many countries, at around one percentage point of underweight prevalence reduction per year; this is in line both with considerable economic development, and with vigorous and widespread health and nutrition programmes at village level. Food consumption is relatively good, and has risen during the 1980s, along with marked success in food production to the point that a number of countries changed from net food importing to exporting.

Middle America/Caribbean/South America

Middle America and the Caribbean, and South America, saw some improvement in nutritional indicators during the 1980s, but little change in the latter pan of the decade. Reductions in underweight prevalence (e.g. in Brazil, Chile), helped by such factors as improved education and lowered birth rates, as well as direct nutrition interventions, seemed surprisingly resilient to economic stress.

Micronutrient Deficiencies

Deficiency of vitamin A affects at least 40 countries, and, out of an estimated 14 million with resulting eye damage, blinds up to half a million preschool children each year. Important recent research shows that improving vitamin A status in children in deficient populations reduces mortality among young children by around one quarter. Vitamin A supply in some parts of South Asia and Eastern and Southern Africa is so low that deficiency is almost inevitable. In most regions the trend in availability is upwards, except in East Africa. Trends in the deficiency itself cannot be assessed at present.

Iron deficiency, one result of which is anaemia, is the only nutrition problem showing a general deterioration in many parts of the world. Prevalence is especially high in South Asia, where more than 60% of women are anaemic. In general, trends in dietary iron supply are downwards, for example in South Asia due to reduced production of legumes with the green revolution, in line with the worsening of anaemia. The main exception is Near East and North Africa, where iron supply is up and anaemia is down.

Figure 2. Projections: Trends in prevalence of underweight children: Year 2000

Scenario A - pessimistic Scenario B - optimistic

(Source: ACC/SCN (1992) Figures 6.1 and 6.2, p. 66)

Outbreaks of scurvy, pellagra and beri-beri have occurred among populations of refugees and displaced people. These deficiency diseases (of vitamin C, niacin and thiamine respectively) had not been seen for decades. They afflict people in camps whose diets are severely restricted in variety.

Women's Nutritional Status

The extent of stunting, underweight, and wasting in women in developing countries was assessed for the first time. The results show that these problems are very extensive in developing countries, particularly low body weight and thinness in Asia. Malnutrition in women is generally in line with estimates of low birth weight, and the intergenerational effects, of malnourished women having small babies who grow up to be small mothers, can readily be seen. Indications are that anaemia prevalences, already high, may be rising in South Asia and Sub-Saharan Africa. Maternal mortality rates are also feared to be increasing in Sub-Saharan Africa, in contrast to other regions.

Diet-Related Chronic Diseases

Assessments of premature mortality rates from chronic diseases possibly linked to diet indicate these to be falling slightly in the developing countries for which data are available, and to be generally below rates for developed countries. Dietary trends show increases in fat consumption with rising incomes in many countries. In general, the differences in fat intake between countries are explained more by food habits than by changes in incomes. However, because morbidity has not been assessed in developing countries in transition, and in the light of widespread reports of increased incidences of, for example, cardiovascular disease, it is urgent to begin to monitor the situation more closely, and to consider steps to reverse undesirable dietary trends.

Projected Nutritional Trends

Projections of possible nutritional trends are shown in figure 2 (using the indicator of percent underweight children). These are based on the historical trends, the pessimistic scenario being a continuation of the worst 5 year trend in the last 15, the optimistic being the best of these. These indicate continued improvement in most regions except Africa; however, the rate of improvement would be generally far below that needed to meet internationally accepted goals, such as those for the World Summit for Children (1990), endorsed by the International Conference on Nutrition (1992). Nutrition in Africa may continue to deteriorate. For South Asia, where the underweight prevalence is by far the highest in the world, although improvement may continue, the rate projected is such that many decades would pass before the problem is solved. Southeast Asia, although improving more rapidly, would still need to have an accelerated rate of improvement to meet proposed goals. South America, if the rate returns to that of the late 1970s, might meet the goals by the year 2000.

Source: ACC/SCN (1992), Second Report on the World Nutrition Situation: Volume I. Global and Regional Results. ACC/SCN, Geneva.

Food Prices and Nutrition

Can food price be used to warn of a deteriorating nutrition situation?

Often in times of shortage the price of food rises, and rising food prices in turn lead to decreased access to food in poor societies. Could, then, the monitoring of food prices - which are already widely collected and reported as part of consumer price indices - be used as a means of giving timely warning of a deteriorating nutrition situation?

Solid evidence that such a relationship between food prices and nutrition does exist has come from national nutrition monitoring programmes in several African countries. An early example was given in SCN News No. 4 (1989) in which retrospective analysis showed similar trends in food prices and child underweight prevalence in Ghana over the time of the severe economic crisis in 1982-84. Moreover, the results appeared to show that food price changes preceded changes in underweight prevalences by approximately three months - indicating that in this case food prices could have provided early warning.

A major obstacle to the investigation of whether the relationship held up consistently had been the scarcity of data showing rapid (e.g. monthly) changes in nutrition. Considerable data accumulated during the 1980s, particularly as a result of the Catholic Relief Services' Programmes in Africa, and some of these data were used in the SCN's “Update on the Nutrition Situation” (1989). Further analysis (by the SCN and ILO) of the data on food prices and nutrition again showed striking links between food prices and nutrition at national level. This was observed both for seasonal fluctuations, and when the seasonal effect was analytically removed (Kelly, 1989). The association tended to hold in Botswana, Madagascar, and Togo as well as Ghana (see table 1).

Since the earlier work in 1986-89, a number of other studies have been published investigating food prices in a deteriorating nutrition situation. Some have been at sub-national level and have found evidence of more complex relationships between food price and nutritional status amongst the most vulnerable groups, which may be masked in national level studies.

Despite fairly strong evidence of a link between food price and nutrition, there has yet to be extensive use of food-price data in nutrition surveillance systems. In this article it was felt worthwhile to pull together the results of various publications to clarify their implications for surveillance. An overall conclusion is that one should perhaps no longer be too timid in using the expected relationships of food price to nutrition as a means of early warning of possible nutrition deterioration. The message is that food price-based indicators could now be built into nutritional surveillance systems more widely.

Food Access Indicators and Nutritional Change

There are several indicators of access to food which may be able to provide early warning of short-term changes in nutritional status. But for an indicator to be of practical use for this purpose, it needs to be available at monthly intervals, be routinely or easily collected, and have a high degree of association with nutritional status. Fulfilling the first two requirements for most countries in practice would strongly favour using wage or price data, with the latter being generally more readily available and reliable. Indicators such as the number of hours work necessary to buy a 'food basket' - which incorporate both wage and price effects - have potential, but are less likely to be routinely available.

If food price data is to be used as an early warning indicator, an important consideration is how well it fulfils the third condition of having a high degree of association with nutritional status. In the majority of poor areas, the income of a household together with prevailing food prices largely determine access to food and to some extent the nutritional status of its members, although other factors will be influential. Food price data may then tell only part of the story - for example, if food prices rise at the same time as household income, access to food may be unchanged. Other factors affect the nature of the relationship. Importantly food prices have different effects depending on which side of the market the household lies - net food purchasers will be adversely affected by food price rises, while net suppliers may benefit. Urban households dependent on wages differ from rural subsistence farming households in their response to food price changes, and such differences will be masked at national level. Rural areas themselves differ in the degree to which households are dependent on the market for food. In addition, urban areas often benefit from price-controls, which can distort assessment of the price-nutrition link at national level - for instance, if the group sampled for nutrition data is from the price-controlled group whilst the price data comes from rural areas with changing prices. Nutritional data should be from areas to which the price data apply. Many cross-sectional studies have reflected these differences and shown differing degrees of association, or none at all. Ideally, analyses should refer to local levels, where they are likely to be clearer and more meaningful.

Table 1. Relationship Between Food Price-Based Indicators and Nutrition for Selected Sub-Saharan Countries

Country	Correlation Coefficient between Consumer Price Index for Food and Prevalence of Underweight Children	Correlation Coefficient between Real Food Price Indexa and Prevalence of Underweight Children	Lag Period between Food Price data and Nutrition data (months)
GHANA	0.21*	0.59**	3
TOGO	0.81**	0.85**	1
BURUNDI	0.13	-0.36*	3
BOTSWANA	0.68**	0.58**	0-6
MADAGASCAR	0.56**	0.60**	2

^a Real Food Price Index = Consumer Price Index for Food/General Consumer Price Index as %

* significant at 5% level ** significant at 1% level

Source: Tabatabai, H. (1989)

The SCN/ILO study investigated the relationship between different price-based indicators and child nutritional status in the following Sub-Saharan African countries: Ghana, Togo, Burundi, Botswana and Madagascar (Tabatabai, 1989). National prevalences of underweight children (defined as less than 80% of the NCHS/WHO reference weight-for-age) were tracked monthly over a period of 4-8 years during the 1980s; while data were clinic-based, trends were considered adequately representative (see ACC/SCN, 1989, pp. 181-184), despite non-defined samples of low coverage. Monthly price indices for food and all consumer goods were extracted from ILO's quarterly Bulletin of Labour Statistics. The results of the analysis are shown in table 1 and figure 1.

For most of the countries listed in the table, there was a positive and significant relationship between food prices and child nutritional status. Burundi was the exception even showing a weak negative correlation. This may be explained by the fact that Burundi is much more of a subsistence-based farming economy than the others, where increases in food prices may represent incentives to production, conferring income and nutritional gains later.

The highest degree of association emerged with the real food price indices. This indicator is likely to be available and particularly useful in urban areas, whereas in rural areas a suitable substitute could be based on the retail price of one or two major staples in a moderately-sized local market.

In Madagascar, underweight prevalences showed marked seasonal variations, whilst prices were relatively static, although the correlation was significant. In this case, the data needed de-seasonalizing before being interpreted. Seasonally was less of a concern for the analyses in Ghana, Togo and Botswana.

In another study in Sudan, between 1981-86, Teklu et al (1991) found that nutritional status showed a similar lagged correlation with changes in the relative prices of cereal and livestock. Poor market integration results in livestock prices dropping as cereal prices rise during drought and famine situations. Such unfavourable market changes strongly affected the poor, and children in particular.

Figure 1. Food Price and Child Underweight Prevalence in Madagascar, Togo, and Ghana Jan 1980 - Dec 1987

Notes.

1.	FPI/CPI = Food Price Index/Consumer Price Index
2.	MADAGASCAR: Deseasonalized prevalence of underweight (<80% Wt/Age) children aged under 5 years. Health Centre Data. TOGO: Deseasonalized prevalence of underweight (<80% Wt/Age) children aged 6-60 months. Health Centre Data. GHANA: Deseasonalized prevalence of underweight (<80% Wt/Age) children aged 7-42 months. Health Centre Data.

Source: ACC/SCN (1989)

While at national level trends in food prices may show a strong lagged association with trends in child nutritional status, at district or community level, there may be other important relationships between price, food consumption and child anthropometry. Reductions in child nutritional status have conventionally been viewed as late indicators of acute food insecurity and famine. This is because it is assumed that people only reduce their food intake late on in the process of impoverishment and/or there is in any case a lag period between reduced food intake and reduced anthropometric status. However, studies at community level have shown patterns of behaviour such that changes in child nutritional status in vulnerable groups could provide early indicators of food problems in the larger population. Here are some examples.

A study from 1981 to 1986 in Niamey, Niger, examined relations between various price-derived indicators, anthropometric indices, and measures of “food crisis” (Khan et al, 1992). Results showed that in this case, again, changes in millet prices proceeded changes in wasting prevalences by around three months. They also indicated that wasting prevalence increases (deseasonalized) predicted generalized food crisis, as occurred in 1984/5 in Niger. One conclusion was that wasting prevalences “among the most vulnerable start increasing at an early stage, again about three months before the initial crisis turns into a community syndrome”. For the poorest households, anthropometry is a later indicator, but if they can be monitored, such changes may predict community level crisis. Indeed, this is often the purpose of using sentinel sites, where these are deliberately aimed at detecting changes among the most vulnerable.

In Wollo, Ethiopia, in 1982-88, monthly changes in child mean weight-for-length against monthly changes in grain price were again observed to show a more or less linear association. In this case, prices and nutritional status seemed to change simultaneously. Moreover, in 1984, an unusually steep drop in mean weight-for-length actually preceded the substantial grain price rise characteristic of severe acute food insecurity. Again, the poorest groups in Wollo were the most vulnerable to factors causing food price increases, for example, drought. One of the first effects of a drought is a reduction in need for agricultural wage labour. Therefore, for some, household income and access to food were drastically cut whilst increases in prices followed later when stocks began to be depleted. In this way, changes in child nutritional status could have preceded changes in prices. An additional explanation may have been that in Wollo, households decided on dietary restriction in response to perceived food insecurity - choosing to go hungry (reflected in nutritional outcomes) in order to improve their chances of future survival. Similar sequences of response to this have been well documented elsewhere, for example in Darfur, Sudan (de Waal 1989, Young and Jaspars 1992). Undernutrition here may not have represented a failure in coping, but a cost of coping (Corbett, 1988). Where such a coping strategy is commonly adopted, anthropometric change in both children and adults would themselves provide early indicators of food crises.

Conclusions

At the national level, trends in food price data have been seen in several African countries to be associated with trends in child underweight prevalences, price changes usually preceding underweight changes with a lag period of between 1-3 months. This is particularly useful as price data are usually readily available. The price-nutrition relationship also appears to be strongest where change is most extreme e.g. during drought, as seen in Ghana, Sudan, Ethiopia and Niger.

At community level, other important inter-relationships may occur between food prices, household energy consumption and child anthropometry, which will usually differ in their strength of association and periodicity between socio-economic groups. While food price change is still predictive of nutritional problems among children, the lag period between the two may vary for the most vulnerable socio-economic groups - so much so that changes in nutritional status of children in these at risk groups could be an earlier indicator of overall food crises in the wider population. Thus, if the need is to predict community-level food crises, both food price data and child anthropometric data in the most vulnerable socio-economic groups may provide useful early warnings.

In reviewing the evidence, this article draws on findings from several countries in Sub-Saharan Africa - Ghana, Togo, Burundi, Botswana, Madagascar, Sudan, Niger and Ethiopia. In summary, these results lead to the following conclusions:

· food prices can provide early indicators of nutritional problems at national level, and this relationship is particularly useful as food price data are usually readily available;

· these relationships are more complex at local level;

· anthropometric changes in children in the most vulnerable groups may sometimes actually precede food problems and price changes among the wider community, thus still providing an early warning of community-wide food problems;

· in this case, the 'warning' that is given is more useful (because earlier) if the poorest groups in the community are monitored.

- S.R.G./J.B.M./V.E.

Bibliography

ACC/SCN (1989) Update on the Nutrition Situation. ACC/SCN, Geneva.

Corbett, J. (1988) Famine and Household Coping Strategies. World Development, 16:9, 1092-1112.

de Waal, A. (1989) Famine that Kills: Darfur, Sudan 1984-85. Clarendon Press, Oxford.

Kelly. A. (1989) Forecasting the Prevalence of Underweight Children. Working Paper for the ACC/SCN, Geneva.

Kelly, M. (1991) Anthropometry as an Indicator of Access to Food: Wollo Region, Ethiopia, 1982-88. Report to the Overseas Development Administration (UK), March 1991.

Khan, M.M., Mock, N.B., Bertrand, W.B. (1992) Composite Indicators for Famine Early Warning Systems. Disasters, 16 (3), pp. 195-206.

Tabatabai, H. (1989) Economic Indicators of Access to Food and Nutritional Status. Working Paper for the ACC/SCN. ILO, Geneva.

Teklu, T., von Braun, J. & Zaki, E. (1991) Drought and Famine Relationships in Sudan: Policy Implications. IFPRI Research Report No. 88. IFPRI, Washington D.C.

Young, H. & Jaspars, S. (1992). Nutritional Surveillance: Help or Hindrance in Times of Famine. Paper for discussion at IDS workshop, 8 May 1992, draft.

Food Security and Nutrition, 1971-91 - Lessons Learned and Future Priorities

by Eileen Kennedy and Lawrence Haddad, International Food Policy Research Institute
1200 17th Street, N.W., Washington DC, 20036

The following article is adapted from a testimony given by the authors to the US Congress in October 1991.

The current world food situation is dramatically different to that in the early to mid-1970s. Given the massive food shortages of that period, policies aimed at increasing agricultural production were stressed in most developing countries. Many of these improved agricultural technologies, often collectively dubbed “the Green Revolution” were successful in reducing hunger.¹ Their effects have been direct, through increased agricultural employment, as well as indirect, through lower food prices and increased non-agricultural employment stimulated by increased farmers' incomes.

Household Food Security²

The technological change in agriculture did help improve country level food security. However, even countries which achieved national food security did not banish food insecurity at the household and individual levels.

We now know that national food self-sufficiency is a poor proxy for household food security. It is common to have 20-30% of the population consuming less than 80% of caloric requirements even when the per capita supply of food within the country is at or above 100% of need.³ It is the household's ability to obtain food that is critical in ensuring household food security. Of course the ability to obtain food is related to the household's purchasing power, which in turn is related to the household's income.

Higher income households are more likely to be food secure. However, it is also important to emphasize that it is not simply the absolute levels, sources and flows of income that are important in influencing household food security. For example, research by IFPRI and others finds that income controlled by women, particularly in Africa, is more likely to be spent on food. At similar levels of income, households with more women-controlled income are more likely to be food secure. There is also evidence to suggest that “lump-sum” sources of income, such as large payments for cash crop remittances, are less likely to be spent on improving household food security.

Thus, new policies that shift control of income from women to men and/or result in “lumpier” income flows might have negative impacts on the food security of the household, despite increases in its overall income level.

Female-headed Households and Food/Nutrition Security

Policies can clearly affect household income, household food security and, in turn, nutritional status. But, recent evidence also suggests that non-income mediated pathways may be very important in the production of good health and nutrition. Work in Kenya and Malawi⁴ finds that at very low levels of income some types of female-headed households have lower levels of preschooler malnutrition than higher income male-headed households. This is accomplished by women allocating proportionately more of their incomes to food and then allocating more of the total calories to their children. In addition, other types of time-intensive nurturing behavior - for example, feeding children more frequently - are more common in certain types of female-headed households. These successful behaviors are important to understand. Whereas poverty alleviation through income generation is likely to be a long-term process, interventions that promote appropriate nurturing behaviour or that exploit incentives for households to invest in their children may be quite effective in providing short-term gains in child health and nutrition.

Diet Quality

The issue of diet quality will become more important in the food security debate throughout the 1990s. Much of the early work on consumption effects of policies concentrated on the effects of income on calories, and to a lesser extent on protein. The assumption was that as the household or individual caloric consumption increased, the consumption of other micronutrients (vitamins and minerals) would also increase. We now know that this does not always happen. Some recent work at IFPRI from Kenya shows that for children, as their food intake increases, the consumption of vitamin-A-rich foods decreases. Adequate levels of vitamin A consumption are generally thought to be linked with lower rates of morbidity and mortality. One reason that vitamin A consumption is decreasing is that many of these vitamin-A-rich foods are seen as low-prestige foods.

Linkages between Agriculture and Nutrition

Despite breakthroughs in improving food security, it is estimated that approximately 700 million people in the developing world survive on grossly inadequate diets.⁵ Since many of the world's most food insecure and malnourished people live in rural areas where agriculture is the dominant sector, agricultural policies and programmes will continue to be important for achieving food security objectives. Agricultural policies that are labor intensive and thus generate employment for the rural landless offer an effective means for reaching vulnerable non-agricultural households.

We know that an increase in income at the household level is a necessary (but not sufficient) condition for improving the nutrition of women and children. But we need to explore innovative ways to improve the food security and nutrition impacts of income-generating policies themselves. This needs to be done in a way that does not create an expectation that the agricultural sector will become a substitute for primary health care delivery. Clearly it will not. However, given what appears to be the continuing problem with health care service delivery in rural areas, coordinating nutrition activities with agricultural programmes and projects already in place seems to be an attractive alternative for improving nutrition. One example of a type of agriculture/nutrition activity that offers the potential to improve food security and nutritional status includes credit with education programs targeted to rural women in Ghana and Mali. The assumption underlying this approach is that both income (via credit) and information are needed in order to bring about sustained improvements in nutritional status of household members. Other examples include agricultural extension services directed towards women with some food security messages integrated into the program such as is being done in Zimbabwe.

Two Important Goals for the 1990s

We believe two issues in particular should receive greater attention in this decade. The first is the impact of policies and programmes on women's food security and nutrition. Almost all the attention given to women has focused upon their roles as mothers. However, given women's key role as economic actors in food production it is important to understand the impact of policies on women's nutrition. Some IFPRI work suggests that women who are better nourished are more productive in agriculture.⁶ It is quite plausible that the major positive effect of policies and programmes on overall household food security and nutrition will come about through an improvement in women's nutritional status.

The second point is with respect to identification of the food insecure. Not only do we need to broaden our definition of food and nutrition security to include micronutrients, but we believe that we need to rethink our concept of cut-off points. Typically some absolute cut-off definition of food insecurity is used, such as households or individuals below 80% of requirements. This definition assumes that households just above that cut-off line are food secure. In many developing countries households just above the cut-off constitute a large percentage of the population. Rather than thinking about these households as food secure, it is more appropriate to classify them as very vulnerable to food insecurity. This issue will take on added importance if governments begin to dismantle programmes that have provided some protection to these vulnerable households. For these households the traditional coping mechanisms may not work.

We have learned much over the past 20 years about how to improve food security and nutrition. If this information can be translated into public policy during the current decade, more progress can take place.

Notes

1. J. Mellor, “Ending hunger: an implementable program for self-reliant growth”, in J.I. Hans Bakker, ed, The World Food Crisis: Food Security in Comparative Perspective, Canadian Scholars' press, Toronto, 1990.

2. Food security, as used in this report means the availability of sufficient food, at all times, for all people, in order to ensure active and health lives.

3. Food and Agricultural Organization, The Fifth World Food Survey, FAO, Rome, 1987.

4. E. Kennedy and P. Peters, (1992) “Household Food Security and Child Nutrition: The Interaction of Income and Gender of Household Head”. World Development, 20, 1077-1085.

5. Mellor, op cit, Ref 1.

6. E. Kennedy with contributions from Pauline Peters, Howarth Bouis and Ellen Payongayong, “Agriculture/nutrition linkages”, International Food Policy Research Institute, Washington DC, January 1991.

Long-Term Effects of Improved Childhood Nutrition

Enhancing Human Potential in Guatemalan Adults Through Improved Nutrition in Early Childhood¹

¹ Edited for SCN News by Viki Elliot

Results of research in Guatemala have provided evidence that improved nutrition during early childhood has longer-term payoffs than previously documented. The following is an edited version of an article based on material presented by Dr Reynaldo Martorell and colleagues at a symposium held at the 1992 Federation of American Societies for Experimental Biology (FASEB) meetings in Anaheim.

Poor nutrition is a powerful constraint to realizing human potential in poor societies. Operating in synergism with diarrhoeal, respiratory and other infections, poor diets in early childhood lead to growth failure, delayed motor and mental development, impaired immunocompetence and increased risk of complications and death from infections (Waterlow, 1992). Children who grow up in environments of poverty and malnutrition in developing countries have a diminished capacity for learning and are not able to take full advantage of even the limited educational opportunities to which they have access (Pollitt, 1990).

Research from around the world has documented that nutrition programs aimed at malnourished mothers and children lead to improved health and wellbeing. Typically, indicators used to evaluate the outcomes of these programmes have been measures taken concurrently with, or immediately following the intervention: among those usually included are birthweight, physical growth, mental development, and infant and child mortality rates. Until recent studies in Guatemala no one had attempted to assess the functional consequences of early childhood interventions in the adult.

In this article results are summarized of one such study, carried out to find out whether improved child nutrition leads to enhanced human potential in the adolescent and adult.

This study constituted the second phase of a research project which took place over nearly 20 years. In the first phase, the Institute of Nutrition of Central America and Panama (INCAP) carried out a longitudinal nutrition intervention study in four rural villages in Guatemala from 1969-1977. The second phase was a cross-sectional follow-up assessment carried out in 1988-89 on subjects of the earlier study and was a collaborative effort between Dr Juan Rivera at INCAP, Dr Ernesto Pollitt at the University of California at Davis, Drs Jere Haas and Jean-Pierre Habicht at Cornell, and Dr Reynaldo Martorell as principal investigator, at the time at Stanford University.

Objectives of the 1969-1977 Study

The INCAP Longitudinal Study (1969-1977) was designed to assess the impact of intra-uterine and preschool malnutrition on behaviour.

The design of the INCAP longitudinal study was heavily influenced by the belief held in the sixties that protein was the principal nutritional deficiency in Guatemala and in other poor countries. Thus, the study called for the provision of a supplement containing high quality protein during the critical first few years of life followed by comparison of physical and mental development in treated and untreated groups.

Design of the INCAP Longitudinal Study (1969-77)

Two liquid drinks were prepared and distributed in the villages. One, the treatment, was a hot gruel made of Incaparina (a vegetable-protein mixture developed by INCAP), dry skim milk, sugar and flavouring. Per cup (180 ml), the Atole, as it was called, provided 163 kcal and 11.5 g of protein; it was also a rich source of many vitamins and minerals. The Atole was distributed and consumed in a centrally-located feeding hall in two of the villages twice a day, every day, during mid-mornings and mid-afternoons. All inhabitants were free to consume as much as wanted but consumption, both attendance and volume ingested, was recorded for only the target subjects of the study: pregnant women; mothers of young children (<2 yrs); and children up to 7 years of age.

A cool liquid called Fresco was selected for the comparison villages. Initially, the object was to make it non-nutritional. Cyclamates were considered as a sweetener until news about their potential carcinogenicity led to the formulation of a drink containing sugar and flavouring, providing 59 kcal per cup (180 ml). In 1971, vitamins and minerals were added to the Fresco in similar concentrations as in the Atole.

Data Collection in Mothers and Children

A wide array of information was collected longitudinally for children in their first seven years of life. This included dietary surveys, continuous monitoring of morbidity, physical growth, and psychological testing. Maternal diet, morbidity and anthropometry were also monitored during pregnancy and in the first 24 months after delivery. Data about household composition, the occupations and education of family members, household income and wealth and other data were collected for all families. The INCAP Longitudinal Study is one of the most complete sources of growth and development data available in developing countries.

Guatemalan Follow-Up Study (1987-88)

The follow-up study is unique. There have been some follow-up studies in developing countries but these have focused on very specific groups such as survivors of severe malnutrition. As far as is known long-term follow-up studies of nutrition interventions have never been carried out. A distinguishing feature of the follow-up study is its comprehensive nature: no study in developing countries has included the range of measures of human function which were included in the follow-up.

The central question of the follow-up study was: does better nutrition during early childhood lead to adults with a greater potential for leading healthy, productive lives?

The use of the word “potential” was deliberate. It was realized that productivity in an economic sense could not be adequately assessed in a young population (subjects were aged 11-26 years at the time of the follow-up). This could only come later when the subjects formed independent households and settled into their occupations. Similarly, other functions, such as parenting, could only be measured later when the young subjects of the study formed families.

Data Collection for Follow-Up

The subjects of study were all former participants of the INCAP Longitudinal Study (born 1962 to 1977). Migrants were included, but only those that moved to Guatemala City and to the provincial capital of the study area. The target sample in the four study villages was the nearly 2,000 subjects of the original study and data was collected from 82% of these. If migrants are excluded from the target sample, the coverage rate was 90% - in other words, only 10% of those it was hoped to re-identify were missed.

The data collected included body size and composition, skeletal age, physical health, strength, work capacity and physical activity, fertility, school attendance and migration histories, intelligence, reading, numeracy and other functional performance tests.

Overview of Key Results from the Follow-Up Study

In this brief overview of results three aspects are emphasized: body size and composition, work capacity, and intellectual performance. All three areas are very important and improvements in one or all would be seen as contributing to human capital formation.

Two aspects of the body size and composition results stand out. First, adolescents who had received the Atole (high energy supplement) during the first three years of life were taller and had greater fat-free masses (FFM) than those who received Fresco. And second, the anthropometric effects were greatest in females. The cut-off point of less than 149 cm, equivalent to a height of 4 ft 11 in., is often used as a criterion of obstetric risk in women. In females older than 16 years of age who were exposed to the supplements from birth to 3 years of age, 49% of Fresco subjects had very short stature compared to 34% of Atole women. Differences in FFM also stand out. Females from Atole villages had on average 2.1 kg more in FFM than females from Fresco villages. These differences are equivalent to an effect size of about 0.5, that is, equal to a positive shift of half a standard deviation. (Effect size is a commonly used measure of “magnitude”. For example, the difference between the mean FFM [Atole minus Fresco] is expressed as a % of the standard deviation of the pooled distribution of FFM in Atole and Fresco subjects.)

Work capacity was significantly improved in subjects exposed to the supplements in their first three years of life, but only in males. Atole males had maximal oxygen consumptions (VO₂ max) that were 0.3 liters per minute greater than those of Fresco males. The difference is equivalent to about 0.7 standard deviation units. Another interesting finding is that the larger working capacity of Atole males could not be explained by differences in fat-free mass (i.e. VO₂ max per kg of fat-free mass was still greater in Atole villages). The nature of these qualitative tissue differences between Atole and Fresco subjects is unclear.

A feature of all analyses carried out to date with respect to measures of intellectual performance has been that they control for schooling variables because the villages had differed in patterns of school attendance since before the study began. The Atole-Fresco differences found in children between the ages of 3 and 7 years were less than 0.2 of a standard deviation compared to differences of around 0.6 of a standard deviation found in adolescence using a summary variable of intellectual performance (i.e., a factor score that combines literacy, numeracy, general knowledge, Raven's Progressive Matrices, reading and vocabulary). In other words, the positive effects of childhood Atole supplementation appeared to become more marked as the children got older.

It is interesting to note that a further review of the Guatemalan follow-up study data (Martorell et al, 1992) has taken an alternative approach, looking at whether the level of stunting at 3 yrs showed any correlation with a number of outcome variables in early adulthood, including intellectual achievement. The study found striking relationships between early stunting and five indicators of later intellectual achievement. Figure 1 shows the deviation from the mean literacy score for adults > 18 yrs who had shown either severe, moderate, or no stunting at 3 yrs of age.

Potential Significance of the Results of the Follow-up Study

The greater body size and increased fat-free mass found in females would be expected to impact positively on reproductive fitness. Short stature is a risk factor for cephalopelvic disproportion, delivery complications and maternal obstetric mortality. Also, greater FFM has been found to lead to larger birthweights.

Figure 1. Lagged effects of early childhood malnutrition on adolescent intellectual performance. Deviations from the sex-specific mean scores on the literacy test (78.8% in males and 76.4% in females) at 18 years in relation to early childhood stunting at 3 years.

Note: Using the WHO reference population, severe stunting was defined as length-for-age values <-3 SDs below the reference median. Moderate stunting was defined as values between -3.0 and -2.0 SDs below the reference median, and the absence of stunting was defined as values within 2 SDs of the reference median.

Source: Martorell et al, 1992

Finally, sharp minds are valued by all societies and by parents everywhere in recognition that improvements in intellectual performance are bound to improve the capacity of individuals to function in a variety of settings. Consider two suggestions. One is that such improvements might lead to better employment opportunities and greater earnings. Another is that better intellectually endowed adults will be better parents, by virtue of being better providers as well as by being able to meet the developmental needs of their children.

Concluding Remarks

Previous and recently conducted analyses of the INCAP longitudinal data have established that a nutrition intervention to mothers and children results in improved health and nutrition in the vulnerable phase of pregnancy and the first three years of life. It should be stressed that the intervention - distribution of supplements at a center for direct consumption - was selected because it facilitated verification and measurement of individual consumption.

Actual programs would undoubtedly follow very different strategies for improving diets, including approaches other than the direct distribution of foods.

The Guatemala follow-up study has established that the contributions of improved nutrition during early childhood to health and nutrition are measurable in the adolescent and adult. Its hypothesis, that such interventions result in improved human capital, is supported by the results to date.

Dr Martorell is Professor of Nutritional Sciences at Cornell University and Chairman of the Advisory Group on Nutrition of the ACC/SCN. His current research focuses on the long-term functional significance of childhood malnutrition.

References

Martorell, R., Rivera, J., Kaplowitz, H. & Pollitt, E. (1992) Long-term consequences of growth retardation during early childhood. In: Human growth: Basic and clinical aspects (Hernandez, M. & Argente, J. eds). Elsevier.

Pollitt, E. (1990) Malnutrition and Infection in the Classroom. United Nations Educational, Scientific and Cultural Organization (UNESCO), Belgium.

Waterlow, J.C. (1992) Protein-Energy Malnutrition. Edward Arnold, London.