OBJECTIVES OF THE REPORT
DATA PRESENTATION
THE INTERAGENCY FOOD AND NUTRITIONAL SURVEILLANCE PROGRAMME (IFNS)
A SHORT DESCRIPTION OF THE LAYOUT OF THE CASE-STUDY GRAPHICS
Long-term trends in nutrition in the developing countries were generally towards improvement, if slowly, in the two decades up to around 1980. These trends were described at regional level in the ACC/SCN's "First Report on the World Nutrition Situation"1. In that report, we stated as follows: " In most parts of the world, nutrition has improved over the last 25 years. Infant and child mortality rates, which summarize many factors, reflect this. ... The estimates of child nutritional status demonstrate that improvements of the 1970's ceased, on average, in the 1980's. Economic stress, as well as severe drought, have contributed to the overall picture of deterioration in Africa, and no significant improvement in South America, the latter contrasting with previous progress. However, proportions of the population "undernourished" are estimated to have fallen in South and South East Asia, Central America and the Near East."
1 ACC/SCN, 1987. "First Report on the World Nutrition Situation"; ACC/SCN, c/o WHO, Geneva.Economic recession and structural adjustment, made worse by drought in Africa, were thus thought to have contributed to increasing malnutrition in many countries in the early 1980's (ACC/SCN, 1986)2. Case-studies from ten countries compiled by UNICEF (1988)3 demonstrated this. But a need was felt, first, to update and extend the results given in the "First Report" to country level, for a substantial number of countries - both those with and without economic crises. Second, factors influencing changes in nutrition, particularly with reference to economic adjustment, should be included. The present "Update" report therefore gives more detail, at country level, than could be compiled for the first report, and includes as recent data as could be obtained - usually including 1987 and sometimes into 1988.
2 ACC/SCN, 1986. "Report of the Twelfth Session of the ACC Sub-Committee on Nutrition and its Advisory Group on Nutrition"; ACC/1986/PG10, Tokyo, 7-11 April, 1986. ACC/SCN, c/o WHO, Geneva.Timeliness requirements of the information to be included, and limited resources, have determined to a large degree the nature of the report. The trade-offs between how sure one can be of the trends, how much detail to cover, the need to put out information quickly on a wide range of countries, but use minimum resources, has given a report that reflects what we consider plausible judgements, not definitive answers. For example, the search for recent data was started in November 1987, and closed in early July 1988: in this time data from over 40 countries from diverse sources were compiled, evaluated and analyzed as needed - 33 data sets are included here (see Map on p5 and accompanying list)1. The limitations of the data and their interpretation - as discussed in the Methods Section, p178 - must be kept in mind.3 Cornia, G.A., Jolly, R., Stewart, F., 1988. "Adjustment with a Human Face, Volume 2 - Ten Country Case Studies"; UNICEF, New York.
1 Additional country data are known to exist, e.g. for the Near East and Latin America, however these were not available to us in time for incorporation in the present report.
The report aims to describe recent trends in nutrition and certain potential determining factors, country-by-country. The approach is unavoidably ad hoc, due to the variable availability of data. Selected indicators of likely determinants of year-to-year change are emphasized, especially related to economic stress. Such factors are thought to be having significant effects now, and protective measures are often needed. At the same time, structural adjustment programmes may provide important opportunities both for direct intervention for the vulnerable, and for moulding longer-term policies for sustained improvement. The information here, showing associations between economic stress and nutrition, is intended to be useful in considering measures to protect nutrition during adjustment and recovery.
One intended audience for the report is thus policy-makers and programme planners, in governments and international agencies. A major point it is intended to make to this audience is that nutrition does respond, negatively or positively, both when economic, food, and/or health conditions deteriorate; and when measures are successfully taken to improve these. The human suffering from economic recession and from some aspects of adjustment is real and measurable; and it could be prevented.
The users of this report should also include advocates for nutrition - national and from international agencies. In those countries for which data are given here, the country case-study may provide a starting point for discussion of trends in nutrition, their possible causes and policy implications. For countries not included here, a first step may be to make a search for similar information, possibly starting surveillance if no suitable data exist. For inter-country comparative purposes, perhaps one of the more striking observations here is the very consistency of the nutritional trends and associated indicators in country after country, highlighting the need to carefully examine policies for their nutritional effects. For other users of the report - particularly for technical people - one benefit may be an improved understanding, based on observation, of the ways in which nutrition changes: seasonally, associated often with food prices, related to drought, and so on. The graphics in the case-studies, with patterns becoming familiar as they repeat across countries, give a fresh perspective not seen in this way before.
The indicators are presented here - as in the First Report on the World Nutrition Situation - to facilitate viewing linkages between indicators within countries, and for comparing the same indicators between countries. As stressed throughout, trends rather than levels of indicators are important here. The choice of particular economic and food access indicators for inclusion in the country graphics is discussed in section 4 on Methods. A brief introduction to the layout of the two pages of graphics per report will be found in the box at the end of this Section (p7).
The new data in this report concerns trends in prevalences of underweight children, or other measures of child nutritional status. All other indicators were directly compiled from data already available through the UN agencies.
Three types of common indicators were sought: economic, food, and child nutritional status. Other health indicators were also investigated1. For economic indicators, measures of debt, and of economic performance (GNP) were taken from available sources. Food indicators were chosen to reflect production changes, and food available for consumption. Prices, as the general price index, and of food, are included. Most results are only available at national level. In particular, urban and rural populations, which may be differentially affected by economic recession and adjustment, cannot usually be distinguished.
1 While it is fully appreciated that infectious and other diseases do influence child nutritional status, to obtain an equivalent set of health indicators - consistent across countries - would be a major undertaking and was considered infeasible for this report. As an indication of their potential usefulness, some annual health statistics have been included for certain countries.Limitations of child nutrition data are important to bear in mind in viewing the report. The cautions are given in detail in Section 4. To summarize (with particular reference to the monthly data derived from health centres) these are as follows.
The nutritional data derived from the health system are from an unknown sample; more crucial for estimating trends, possible changing bias in this sample cannot be assessed. Thus, the comparability of month-to-month samples is unknown, so that no statistical assessment of the likely validity of observed trends can be made. The reasons for thinking that the trends are meaningful are given in detail in Section 42.
2 In brief, these include the consistency of the patterns of changing prevalence, at different levels of aggregation; the strong association with likely causal factors; and the exclusion analytically of at least some sources of confounding.The indicators chosen and their presentation monitor factors in complicated situations. A number of associations appear to exist, for which there is good theoretical reason to expect that they may be causal. However, the caveats in interpreting these associations need to be stressed at the outset.
On the one hand, with such complex systems involving whole societies, great caution should be exercised in concluding that one factor caused another to change. Confounding effects must be borne in mind. For example, although price changes often seem to precede child underweight prevalence changes, in the expected direction, these could both result from a common cause: adjustment policy changes might raise food prices and reduce health expenditure; and depleted health services might be the real cause of increasing numbers of malnourished children. Such scenarios cannot be excluded. Further, whether causal or not, historical associations do not necessarily predict future changes.
One way, it is suggested, to look at the data is as a description of outcome - child nutritional status - under recent conditions in a variety of countries; and of likely intermediate factors. Common sense indicates that economic stress and drought are likely to lead to - to cause -hardship; and it looks as if they do. At an absolute minimum, child malnutrition should be prevented from rising under these circumstances.
On the other hand, these and other data (e.g. in the "First Report on the World Nutrition Situation") do show very strong associations at the aggregate level. Despite the uncertainties, at the population level the relationships in fact seem stable and indeed predictable.
We could cautiously go further, given that in reality most decisions have to be made based on inadequate information. It does seem sufficiently plausible that major economic changes are actually causing significant changes in nutritional status - and that this in turn reflects changes in other aspects of human welfare and health - that consideration of these serious, and fairly immediate, consequences should feature prominently when deciding on major economic policies. These consequences were suspected earlier, and "Adjustment with a Human Face" documented some; now they are becoming more firmly established, and cannot be ignored. Finally, they should now be much more widely monitored.
Map and regional listing of countries included in the report.
|
Africa |
Asia |
Latin America & the Caribbean |
|
Benin |
Bangladesh |
Bolivia |
|
Botswana |
China |
Chile |
|
Burkina Faso |
Indonesia |
Colombia |
|
Chad |
Philippines |
Costa Rica |
|
Ethiopia |
Sri Lanka |
Cuba |
|
Gambia |
Thailand |
Guatemala |
|
Ghana |
|
Jamaica |
|
Lesotho |
|
Nicaragua |
|
Madagascar |
|
Peru |
|
Mali |
|
Uruguay |
|
Mauritania |
|
Venezuela |
|
Niger |
|
|
|
Rwanda |
|
|
|
Sudan |
|
|
|
Tanzania |
|
|
|
Togo |
|
|
The IFNS programme was started to provide assistance to developing countries in establishing or strengthening surveillance systems. A brief introduction to the programme is given in Appendix A. There are a number of linkages with this report to which it may be worth drawing attention.
Part of the data used here derives from nutritional surveillance systems. The report thus illustrates possible information outputs from such systems. It must be strongly emphasized at the same time that the key characteristic of surveillance - in nutrition as in other fields - is the use of information. A surveillance system should include the mechanism to respond to information, and only the information needed for response should be produced. Thus, while the data here illustrate outputs possibly useful at national or sub-national level, the more important aspect will be the response mechanism. The illustration is only partial therefore: this is one type of data that may be useful, being tapped here for monitoring purposes.
Only a minority of countries, albeit a considerable number, have data on trends in nutrition, so that this report is by no means comprehensive geographically. Nonetheless, for those countries included, it is hoped that the information on which the report is based may be directly useful; often further analysis may be required. Support from the IFNS programme could be drawn upon both to generate information within countries lacking data, and to improve on and use that already available.
Future updates on trends in nutrition should be more timely and comprehensive, as nutritional surveillance develops within more countries. Ultimately, this could lead to nutritional indicators becoming as widely used as the more familiar indicators (e.g. GNP, exchange rates) are today.
The indicators are presented for each country case-study on two pages. The first page shows (see overleaf) virtually the same data for each country, where available. The second page presents, at the top, sources of cereals and use of IMF credit. The Food Price Index is to be found here for the Africa section, but has been placed on the first page of graphics for the sections on Asia and Latin America & Caribbean (LAC) as 'Unfavourable Crop Conditions' and 'Food Shortages' have not been reported here for these regions.
The scales on the vertical axes vary between countries to some extent, to fit as much detail as possible into the space available and therefore due caution must be exercised if inter-country comparisons are attempted. Horizontal axes are for 1980 through 1987. Annual data are plotted mid-year. A fuller definition of the indicators and their source is provided in the Appendix.
PAGE 1 GRAPHIC
Top Panel: Each page 1 graphic is headed by a number of standard population statistics, e.g. population size, density and growth rate. Key country indicators, as used by UNICEF and the World Bank, are also given, viz. Infant Mortality Rate (IMR), Under 5 Mortality Rate (U5MR), and the per capita Gross National Product (GNP). These figures are calculated for 1987 apart from that for GNP which refers to 1986, and Population Growth which is based on 1980-'86. In addition, an estimated level of prevalence for underweight children (i.e. less than minus 2 standard deviations of the WHO/NCHS weight-for-age reference), based on a statistical model (see Glossary for details), is provided as a 10-point range. This estimate provides a general indication of the expected national level of underweight children for the mid-1980's. This is primarily intended to supplement the available prevalence estimates provided by health-centres or nutritional surveillance sources, which are unlikely to be nationally representative. The latter are presented at the bottom of the same page as "Trends in Underweight Children ".
Economic Indicators: Four indicators are plotted - the country's debt ratio, external debt (outstanding and disbursed), gross national product (expressed in local currency at constant 1980 prices, where available), and the monthly movement in the currency exchange rate with the US$.
Food Indicators: These are plotted in six panels: the annual food production index (per caput, base year 1979/'81 = 100), national annual cereal availability (after production, imports and food aid have been taken into account), monthly reports of unfavourable crop conditions and food shortages. (NB, as noted above, these are only reproduced here for African countries; country reports for Asia and LAC show the monthly food price index in this panel). The annual figures for nationally available Kcals per capita per day, the monthly general consumer price index (CPI, base year 1980 = 100) and, as a measure of the relative price of food, the ratio of the food price index (FPI) to that of the consumer price index, are also shown. An arbitrary guide line has been placed at the level of 2,000 Kcals to assist in inter-country comparisons. Both the Food Production Index and the Ratio of FPI to CPI have reference lines at the base value of 100.
Trends in Underweight Children: Available trend data on the prevalence of underweight children appear in this panel. If monthly health-centre data are available, e.g. from the Catholic Relief Service centres in a number of African countries, these have been used. The latter usually relate to the percentage of under 5's of less than 80% reference weight-for-age. A de-seasonalized plot will normally accompany the monthly trend data. For the majority of countries, this panel generally attempts to show changing prevalences during recent years. (NB: Particular attention must be paid to the figure titles and to the text which specify the measures and cut-points used. The representativeness of the data in terms of age, area or other demographic group should be noted and borne in mind. Cross-country comparisons of these figures are therefore difficult and due caution must be used.) Where possible, a guide line has been added to show the expected prevalence level corresponding to the given cut-point. For example, if the cut-point is set at <-2 Standard Deviations below the median of the reference curve, then the expected percentage would be 2.3% for a normal or Gaussian curve.
PAGE 2 GRAPHIC
Additional Food & Economic Indicators: The top of page 2 records the national production, imports, food aid, and availability of cereals on an annual basis. The use of IMF credit is also shown; should no IMF credit be used then this is indicated by a line centred on 0.0. Below this figure (for Africa only) the monthly change in the food price index is graphed.
Additional Indicators: The balance of page 2 is used to plot other relevant data, e.g. if regional trends in underweight are available they appear here. Also, for a number of countries, other nutrition and health-related statistics are given.
COUNTRY
|
POPULATION: (for 1987) |
IMR: (for 1987) |
|
POPULATION DENSITY: (for 1987) |
U5MR: (for 1987) |
|
POP. GROWTH RATE: (average 1980-'86) |
GNP (PER CAPITA): (for 1986) |
|
PERCENTAGE URBAN POP.: (for 1987) |
ESTIMATED PREVALENCE LEVEL UNDERWEIGHT CHILDREN: (for
mid-1980's) |
|
Estimated as % <-2 S.D.'s weight for age and given as a
10-point range. NB: this may differ from cut point used in presentation of data
in bottom panel. |
