Assessment of the double burden of malnutrition in six case study countries

G. Kennedy, G. Nantel and P. Shetty, Nutrition Planning, Assessment and Evaluation Service, Food and Agriculture Organization of the United Nations

Introduction

The concepts of nutrition transition and the double burden of malnutrition have been introduced over the past decade. There is documentation of the occurrence of each in many developing countries that are in rapid economic transition (Shetty and Gopalan, 1998; Shetty and McPherson, 1997; Public Health Nutrition, 2002; Gillespie and Haddad, 2003) This paper draws on evidence from six countries (China, Egypt, India, Mexico, the Philippines and South Africa) to document the nutrition transition and the double burden by summarizing the trends in dietary changes and accompanying changes in nutritional status and disease burden experienced in the past 20 years. Many contributory factors have influenced these processes, including urbanization, demographic shifts, sedentary lifestyles and the liberalization of markets. In-depth discussion of these drivers has been reviewed extensively in recent literature (FAO, 2004; Development Policy Review, 2003).

Double burden of malnutrition

The double burden of malnutrition refers to the dual burden of under- and overnutrition occurring simultaneously within a population. Historically, undernutrition has been associated with higher prevalence of infectious diseases; as populations move into epidemiologic and demographic transition, increases in overweight and obesity begin to appear, while undernutrition and infectious disease become past problems. Today, the burden of disease and malnutrition does not fit neatly into the classic stages of transition, but reflects a modified pattern referred to as the protracted-polarized model, where infectious and chronic diseases coexist over long periods of time (Frenk et al., 1989 in Chopra, 2004a). Evidence of this has been documented in countries as diverse as China (Cook and Dummer, 2003) and South Africa (Chopra, 2004a).

The protracted-polarized model represents a change in the documented pattern of the epidemiologic transition that occurred in Europe and North America in the nineteenth century. The classic pattern of "epidemiologic transition" constitutes a shift from high mortality and fertility patterns to lower mortality followed by lower fertility. Improvements in water and sanitation, and more effective public health services such as immunization result in an associated shift in disease burden from high rates of infectious disease to increasing non-communicable disease (NCD). In tandem with this shift, life expectancy increases and the demographic profile shifts towards lower child-to-adult dependency ratios and greater numbers of elderly in the population, with NCD becoming more predominant as longevity increases.

Underweight and obesity are both among the top ten leading risk factors for the global burden of disease (WHO, 2002). The current double burden of malnutrition seen in many developing countries is brought about by a coupling of risk factors. Progress in improving water and sanitation systems has been slow and the development of sound public health systems weak, thwarting efforts to reduce undernutrition. At the same time, increasing urbanization and changing dietary patterns and lifestyles are contributing to a rapid rise in overweight and diet-related chronic diseases.

Although there seems to be clear evidence of a double burden of malnutrition and disease at the global level, it is not clear how critical the issue is at the national level and to what extent developing countries need to concern themselves with the seemingly incongruous problems of under- and overnutrition and infectious and chronic disease. Some countries, such as South Africa and the United Republic of Tanzania, report no decline in numbers of cases of infectious diseases including tuberculosis (TB), malaria and HIV/AIDS, while the incidences of coronary heart disease (CHD), diabetes and stroke are on the rise (Kitange, no date). The protracted-polarized model of epidemiologic transition has been documented in South Africa, with poor people suffering increased mortality from infectious, chronic and accidental/violent causes (Chopra, 2004b). Regarding malnutrition, there is increasing documentation of rising rates of overweight and obesity among children and adults, and slow progress in reducing undernutrition, particularly in children under five years of age. This paper attempts to summarize and evaluate the problem of the double burden of malnutrition and disease as reported in the six country case studies, and discusses potential options for addressing both sides of the problem.

Characteristics of the nutrition transition

The nutrition transition refers to changes in the composition of the diet, usually accompanied by changes in physical activity levels. Popkin (2003) has characterized the nutrition transition into three stages: receding famine, degenerative disease, and behavioural change. In the first stage, diets are primarily derived from plant-based food sources, tend to be monotonous and are based more on home food production that requires high levels of physical activity related to planting, harvesting and processing. The second stage encompasses dietary changes that generally include more animal source foods, higher intakes of fat - both vegetable oils and saturated fat from animal products - increased use of sugar and other sweeteners, and higher reliance on food produced and processed outside the home or immediate community. Mandatory physical activity to produce food and procure water and fuelwood, including agriculture-based labour and household labour, is often also reduced. The final stage involves a shift to a diet with less saturated fat and decreasing reliance on processed foods. Typically, this stage encompasses increased intakes of whole grains, fruits and vegetables and decreased consumption of saturated fat, with a preference for animal source foods with lower saturated fat content (fish and poultry). Intensive physical labour related to agricultural production is not reintroduced, but non-obligatory physical activity is increased.

In which populations is the nutrition transition occurring?

The diets of most of the world’s population lie somewhere between the first and second stages of nutrition transition, while subsections of populations in North America and northern and southern Europe may be moving into the third stage. In most of the case study countries there is evidence of a rapid movement from primarily plant-based diets to diets with greater proportions of energy derived from meat, milk products, animal fats and vegetable oils.

Urban populations are typically the first to begin incorporating more fats, animal source foods and processed products into the diet. Dietary changes are not however limited to urban areas, nor to wealthier population groups. Research by Mendez, Du and Popkin (2004) on dietary transition in China used a scale of "urbanicity", which considered access to health care, housing, communications and transport in urban and rural areas. They found increasing intakes of animal source foods and edible oils in low urbanicity urban areas and more urbanized rural areas. In low-income areas of Brazil, processed bakery products, processed meat products, sweets and soft drinks were among the most commonly consumed foods (Sawaya, Martins and Martins, 2004). Falling prices are another stimulus for dietary changes. A high-fat diet is much more affordable today than it was 30 years ago (Popkin, 2002). In China, over a period of six years in the 1990s, the relative prices of fish, pork and oil all decreased (Mendez, Du and Popkin, 2004).

Key development indicators and linkages to the double burden of malnutrition in the case study countries

Economic, health and social indicators for each of the case study countries are presented in Table 1. Rapid urbanization has been linked to dietary change and obesity in developing countries (Mendez and Popkin, 2004). There is a wide range in the proportions of urban population among the case study countries, with Mexico being the most urbanized and India the least. Low birth weight (less than 2 500 g) has been identified as a risk factor for developing NCDs in later life (Barker, 2004). Among the case study countries, India has the highest percentage of infants born with low birth weight, followed by the Philippines and South Africa. The demographic transition, particularly the ageing of the population and longer life expectancy, can also contribute to increased incidence of NCD. Of the case study countries, China appears to be ageing the fastest, with the largest percentage of adults aged 65 years and older and the smallest of children up to 14 years. High adult literacy rates and improvements in water and sanitation contribute to decreasing undernutrition, particularly among children under five years of age. Mexico, the Philippines and China have adult literacy rates of more than 90 percent, infant mortality rates of less than 35 per 1 000 live births, and life expectancy of more than 70 years. HIV prevalence is a grave public health concern in South Africa and is reflected in this country having the lowest life expectancies for both men and women.

TABLE 1
Economic, health and social development indicators

¹ Access to safe sanitation is defined as access to adequate excreta disposal facilities such as a connection to a sewer or septic tank system, a pour-flush latrine, simple pit latrine or ventilated improved pit latrine. An excreta disposal system is considered adequate if it is private or shared and if it can prevent human, animal and insect contact with excreta.

² Access to safe water is defined as reasonable access to any of the following water supplies used for drinking: household connection, public standpipe, borehole, protected well, protected spring and rainwater collection.

Source: UNDP, 2004.

Review of trends in food available for consumption in the six case study countries

Trends in food availability using FAOSTAT data

The Food and Agriculture Organization of the United Nations (FAO) maintains a comprehensive database of food production from 1960 to the present. Country-specific food balance sheets provide information on the supply and utilization of many different commodities. Factors accounting for food supply include production, imports, stock changes and exports. The per capita supply of energy, protein and fats for many food commodities can be calculated by extrapolating from these data.

When analysing FAO food supply statistics it is important to consider the application of the per capita measurements. These figures are based on population totals for all ages and represent average, not actual, per capita availability. Actual food availability may vary by region, socio-economic level and season. Certain difficulties are encountered when estimating trade, production and stock changes on an annual scale. In order to reduce these errors, three-year averages should be calculated. This paper uses three-year averages for 1970-1972, 1980-1982, 1990-1992 and 2000-2002.

Trends in availability of dietary energy

Between 1970 and the present, per capita dietary energy supply increased in all the case study countries (Figure 1), although rates of growth were different. The largest absolute increase in caloric availability was in Egypt, and the largest percentage increase over the period from 1970-1972 to 2000-2002 occurred in China (49 percent). Over the same period, Egypt and the Philippines experienced increases of 41 and 30 percent, respectively. The slowest growth in per capita dietary energy supply over the past 30 years was in South Africa. However, of the six countries analysed, South Africa started with the highest per capita dietary energy supply and its still remains higher than India’s and the Philippines’.

FIGURE 1
Trends in dietary energy availability, 1970 to 2000

Commodity trends

Food availability and the percentages of dietary energy derived from basic food groups were calculated for 1970-1972 and 2000-2002 (Tables 2 to 4). Per capita supply of cereals and starchy staples has increased in all but one of the countries, but their percentage contribution to total energy supply has generally declined. Legumes, pulses and nuts have mainly remained stable or declined in terms of both quantity and percentage of dietary energy supplied. Oils, fats and animal products have increased in all case study countries, with the exception of fats/oils in Egypt. Fruits and vegetables have increased in most countries, as has percentage of energy from fruits and vegetables. The World Cancer Research Fund (1997) recommends that at least 7 percent of dietary energy be supplied from fruits and vegetables and, based on food balance sheet data, this goal would be achievable (assuming equitable distribution) in three countries. Sugar and sweeteners increased in all countries except South Africa, but the proportion of energy derived from sugar has not increased as dramatically as those from animal source foods and oil.

TABLE 2
Trends in food supply of different commodities (kg/capita/year), 1970-1972 and 2000-2002

TABLE 3
Percentage of dietary energy supply from major food groups, 1970-1972 and 2000-2002

¹ In China, the majority of the "other" category represents alcoholic beverages.

In terms of qualitative changes to the diet, China, India and Mexico exhibit the same pattern of declining per capita intakes of cereals and legumes and pulses and nuts, and increasing intakes of all the other food groups (Table 4). South Africa exhibits the most widely divergent pattern of increasing cereal and pulse intakes and decreasing intakes of meat, fish and poultry and sugars and sweeteners. Per capita supply of oils and fats has risen in all the countries except Egypt.

TABLE 4

Direction of the shift in percentage dietary energy from food groups, 1970-2000

Shaded arrows highlight declining trend

FIGURE 2
Percentages of daily energy supply from fat, sugar and alcohol, 2000-2002

In Mexico and South Africa, more than one-quarter of the dietary energy available per capita is assigned to sugar, fat or alcohol (Figure 2). In three of the six countries analysed, sugars and sweeteners account for more than 10 percent of daily energy supply,^[1] and in South Africa alcohol accounts for 5 percent of total dietary energy supply. WHO/FAO set a population nutrient intake goal of 15 to 30 percent of energy from fat (WHO/FAO, 2003). Extrapolating from FAOSTAT data, consumption of fat remains below the recommendation of 15 to 30 percent of total dietary energy intake in the case study countries.