Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Nutrition surveillance: A planners' perspective

Soekirman and Darwin Karyadi

Why nutrition surveillance?

In the mid-1970s, when we in Indonesia looted for an instrument that could be used to predict the possible occurrence of periodic food shortages that affected nutrition status, we were unaware of any activity or programme that could meet our need. What we wore leaking for was an instrument or system to provide early warning information for prompt action before serious malnutrition became prevalent.

Why did we need an early warning system? Until the early 1980s periodic food shortages in several areas of same Indonesian provinces led to a high frequency of acute protein-energy malnutrition, known as hunger oedema in adults, particularly pregnant women and the elderly. The disorder was common under colonial governments and in the early stages of independence. However, after self-sufficiency in rice was achieved in 1985, the government could no longer tolerate such food crises without early and prompt action by local authorities to minimize their negative effects. In particular, policy makers and planners at the National Development Planning Agency (BAPPENAS) were uneasy whenever they read reports on victims of starvation. "This should not have happened. Something must have gone wrong," they complained.

In a staff meeting in the late 1970s, the lead author was asked whether any health or nutrition technique or system could be used to predict the possible occurrence of food shortage so that early action could be taken. In response to this challenge, efforts were made in 1979 to develop an early warning system for food availability and consumption at the household level in villages prone to food crisis.

An information system was developed with technical assistance from Cornell University, with the following criteria, established by policy makers and planners from BAPPENAS:

  1. Information should make it possible to prevent malnutrition epidemics during times of food crisis.
  2. The information should be easy to collect and process and should be available promptly to various levels of government administration and the local community for making policy decisions and developing action programmes.
  3. The information should not be developed as a completely new system but should attempt, as far as possible, to use data and information already available at the local level.
  4. The system should be designed so that the information could be collected and processed by the local government and the community. In other words, from the beginning the system should belong to the local authorities at the subdistrict level and to communities.
  5. Since the information should serve as a monitoring mechanism for higher level government administrators (province, central), it should be communicated to them promptly without creating an additional reporting mechanism.

These criteria represent a planners' perspective for an information system for Indonesia that is now called the nutrition surveillance system (NSS). It is worth noting that the initiative to develop this system in Indonesia came from policy makers and planners, because the system should facilitate prompt actions and these parties were very much concerned about such actions. An advantage of policy makers and planners being involved early in the development of an NSS is the assurance that funding will be available for promptly taking actions indicated by the system.


What is the nutrition surveillance system?

After considering what the policy makers and planners needed in such a system, we tried to define the scope of the NSS for Indonesia. First, it had to function as a timely warning and intervention system (TWISS), a system to link problem-prone areas (districts, subdistricts, villages) with higher authorities at provincial and central levels, provide indicators that act as early detection mechanisms for food crises, and guide prompt actions to cope with declining food availability and consumption, particularly among poor households.

As a TWISS, the NSS is aimed primarily at government officials who have the authority to mobilize resources to deal with declining household food availability that may arise because of higher food prices, crop failure, a decline in market supplies of food, especially rice and other staples, and high peaks in unemployment. In Indonesia, these government officials include the village head, who is elected by the people, the subdistrict head, who is an appointed government official at the lowest level, the district head, who is appointed by the central government but elected by the district parliament, the governor as head of the province, and certain ministers and ministries at the central level.

The definition of an NSS depends on the users.

For the village head, it is a first-hand information mechanism indicating the people's wellbeing. If data show an immediate problem, the village head may call for an urgent meeting with the village council. The council is a community organization representing the needs and concerns of the people as brought to the attention of the village head or others.

The subdistrict head cannot take action alone, having no authority to mobilize resources, but forwards NSS information to the district head with a proposal for solving problems through prompt action.

The district head may use NSS information to instruct various sectoral agencies to take action, with or without authority from the provincial or central levels. These agencies are primarily the local Rice Logistic Agency for local market operation to secure rice market supplies so that prices can be controlled in favour of both the producers (farmers) and consumers, the Office of Agriculture for agricultural inputs or information, the Office of Manpower Agency for labour-intensive operations, together with other agencies to provide short-term employment, and the Office of Health Service for appropriate health and nutrition services.

High-level authorities use TWISS information to monitor local food and nutrition problems. Such authorities can be found within the ministries of Internal Affairs, Agriculture, and Health (primarily the Division of Nutrition), as well as the National Rice Logistic Agency, which became a Ministry of Foods in the new cabinet (starting in 1993) with a coordinating function for supply-side (input) nutrition.


Doss the NSS work in Indonesia?

In certain areas, especially in some subdistricts on Lombok Island in West Nusa Tenggara province, the NSS had effectively been in operation as a TWISS for eight years (1981/1988) when rice cultivation was still based on a local traditional farming system with a low production capacity. During the third five-year plan, starting in 1978, a massive agricultural programme to increase production through a nationwide "green revolution" was implemented in the province. As a result of this programme, the people began to benefit by the mid-1980s. Since then, the province has completely changed from having an insufficient rice supply to exporting significant amounts of rice to other provinces while retaining rice self-sufficiency. As a result, periodic food shortages no longer occur, and there has been no report of acute malnutrition among the adult population.

Local government officials and the people then lost interest in the TWISS, largely because the food shortage was resolved. This also happened in other provinces, and interest rose only when the food situation became vulnerable once more or when an area was considered to have an inadequate supply of food, especially of rice. In such instances the objective of the NSS emphasizing timely warning and intervention was reassessed and shaped into a general nutrition-monitoring system.


The NSS revisited: The planners' perspective

From the planners' point of view, since Indonesia has achieved self-sufficiency in rice and no acute malnutrition has been reported, the continuation of the NSS has become debatable. This is contrary to the opinion of nutrition specialists, who believe that the NSS should be continued even in areas of food and rice surplus as a part of the process of nutrition programme development. We must continue to educate policy makers and planners on the significance of nutrition-improvement programmes as an integral part of national development. This is not easy, especially since we cannot rely on biomedical or traditional nutrition information to convince these officials. It requires what has been called a nutrition engineer to translate nutrition science into issues of economic, social, and political development (Alan Berg, personal communication, 1992).

It seems that nutrition has won the game in Indonesia. It was decided that the NSS should be continued but reworked. From 1988 to the present, the definition and scope of the NSS has been broadened to monitor the nutrition status of people in general. For the periodic monitoring of children's nutrition status on a national basis, the design, collection, and processing of data and information are integrated into a National Social and Economic Survey under the General Bureau of Statistics. With this integration, periodic data on weight and age are available for the national and regional levels. In addition, there are plans to monitor the height of primary-school entrants periodically. The pilot activity was completed in 1990, and the monitoring will be implemented nationwide in the sixth five-year development plan, starting in 1994. Consideration has also been given to using the growth-monitoring data and information collected monthly for children under five years of age in about 67,000 villages, covering almost every child, in the revised NSS.

The revised NSS is important primarily in that it provides data and information for better, more effective nutrition programmes. It also shows evidence of the positive impact of well-planned programmes.



From the perspective of the planners, the NSS has achieved what they once expected a nutrition programme to do: develop a system that could be put into operation by key sectoral agencies (agriculture, health, home affairs, etc.) to prevent acute malnutrition due to food crises.

If we want to develop the NSS beyond being a TWISS, special efforts must be made through more effective communication and education so that good nutrition can be presented in a more attractive way. It is important that the overall concept of the NSS be relevant to national development priorities. Thus, nutrition personnel who operate the NSS have to be trained beyond the basic science of nutrition. They must understand professional linkages between NSS data and other available information such as socioeconomic data from the National Social and Economic Survey in Indonesia. The objectives and the data-collection methodology should be reviewed and updated periodically to be consistently relevant to the needs of development programmes. This happened in the early stage of the NSS in Indonesia, in which a broader concept of an NSS was narrowed down to a TWISS and then was developed back into an NSS.


Nutrition surveillance: Use of information for programme planning and management

Ligia Rodriguez

The basic goal of nutrition surveillance is clearly stated in the theme of this symposium: to support "actions towards better nutrition." It is also central to the definition of nutrition surveillance itself: "watch[ing] over nutrition in order to make decisions which will lead to improvements of nutrition in populations" [1]. The mandate is clear. The implementation process is not.

Supporting actions towards better nutrition implies using information, which is a much more complex and uncertain matter than producing information. The production of information deals with data, whereas the use of information deals with persons, organizations, and society.


Conditions for the use of surveillance information

For nutrition surveillance to assist programme planning and management, it must develop in a supporting political and social context, have active and efficient managers, have a clearly defined scope, and communicate results in an appropriate fashion to decision makers. Ideally also it should produce the information near the decision makers.


Political and social contexts

Political and social contexts define the resources available to address problems and the legislative support required to take action. It is unlikely that information from nutrition surveillance will be of use to a government with little social commitment, or when conditions of uncertainty are present such as sudden economic contraction during a civil war [2]. The government's social commitment, its stability, and the continuity of technical teams despite political changes are all critical determinants of the use of information in programme planning and management.


Active management

Effective programme management requires the power to make decisions. It also implies personal and institutional involvement in those decisions [3]. Political and administrative support of the officials in charge and key personal characteristics such as leadership, creativity, involvement, and interest are necessary if programmes are to be adjusted or changed in accordance with the results of nutrition surveillance. An effective programme manager is critical.

Some information from nutrition surveillance may be considered threatening for the officials in charge of a programme [4, 5]-for example, when it is demonstrated that the cost-benefit ratio of the programme is very high. In addition, changes may imply extra effort, and particular officials may not be willing to risk change. To prevent such problems it is necessary to know who the information users (i.e., decision makers) will be and to make efforts to ensure their early involvement in the nutrition surveillance process [2].


The scope of nutrition surveillance

Various authors have stressed the need to focus surveillance around nutrition-related decisions [6, 7] to provide sufficient information for making a specific decision. The objective is to meet the minimal information requirements for a specific decision using existing resources [8]. It is also recommended that the decisions that need to be supported should be taken as a starting point from which to work backwards in order to determine what data are required.

Many nutrition-surveillance systems have been established to obtain data on the overall nutrition status of a population without making provision for information on the underlying causes of nutrition problems. An opposite multisectoral view of nutrition surveillance was clearly set forth by the World Food Conference of 1974, which defined the scope of surveillance as "a wide range of information on all factors which influence food consumption patterns and nutritional status." Given this point of view, nutrition surveillance will usually require data that go beyond those considered nutritional, to include economic, sociocultural, and biological determinants.

Neither approach meets the surveillance needs for action. In one case the information is too limited, and in the other important information can be lost in mountains of details. In either case, action can be paralysed. Nutrition surveillance should explicitly define its scope beforehand, to produce information clearly related to the specific nutrition or nutrition-related decisions that it is going to support. Assertive, focused information is usually less complex. Keeping it simple is critical [9].


Production of information near decision makers

Producing information near the place where decisions are made, that is, within the same sector or institution where specific nutrition or nutrition-related decisions are to be taken, favours the user's involvement, adaptation to the user's needs, timely information delivery, and closer knowledge of the organizational context [10].

In Costa Rica same of the problems of information use have been avoided by a strategy called SISVAN Minimo (Minimum Food and Nutrition Surveillance System), which focuses nutrition surveillance on marginal groups, selects specific decisions to be addressed in each step of the food chain, gives priority to sector-specific information needs, and tends to reinforce and use existing data sources of information systems, located in different sectors.


Communicating results

Information from surveillance is likely to be used by decision makers if it responds to their specific needs, is communicated on time, and is presented in a simple, short, direct way. The use of information is a matter of cross-cultural communication among persons with different perspectives: the rational scientific perspective on the information-generation side, and the pragmatic, political perspective on the users' side [10]. To establish a link between these two worlds and enhance information use, time and effort should be devoted to gaining insight on the decision-making processes and the decision makers themselves, as well as on the particular organizational contest.

Our experience indicates that nutrition surveillance, particularly for programme management, should be located within the institution in charge of the programme or programmes. Otherwise timely information is seldom produced, and potential users do not become involved, thus wasting time and effort. The development of technical capabilities for policy analysis and communication skills will also contribute to a more efficient dialogue with decision makers. Continued training of the producers of information is required in order to enhance the political and managerial value of surveillance results.


Decisions supported by nutrition surveillance

For programme planning and management, nutrition surveillance should provide information to support decisions on the focus and coverage of the programme, the types of services to be provided, organizational and process adjustments of the programme necessary to accomplish its objectives and goals, and adjustments of the programme itself, including its objectives and proposed goals.


Experience in Costa Rica

In Costa Rica, the socially oriented governments, the stability of the regimes, and the resources allocated to education and health have been key factors favouring information use. Laws to democratize education, regulate labour, and extend social security were all established in the late eighteenth and early nineteenth century. The army was disbanded in 1948, and the savings in military expenditures permitted resources to be channelled to infrastructure, education, health, and other programmes. As a result, the country reached a high level of development in most social indicators. Figure 1 shows the Costa Rican infant-mortality rate in comparison with that of selected other countries. The nutrition status of the population also improved (fewer than 3% of pre - school children have moderate or severe malnutrition). Despite all this, however, there are particular counties, districts, and population groups where the situation is not nearly so favourable and problems are much more prevalent [12,13].

FIG. 1. Infant mortality rate in Costa Rica in comparison with selected other countries, 1992 (Source: ref. 11)

In 1975 a Family Allowance Law was issued that established that 20% of the sales tax and a 5% tax on salaries would provide a financial fund for social development [14]. It was clearly stated it was to benefit the low-income population. As Figueres stated, "It is a subsidy from all [the population] to those that are worst off, those whose work is necessary to society, like workers in food or cash crop production" [15].

With the implementation of the law, decision makers were faced with the need for desegregated data-local-level information that could locate and describe the population to be reached. The data available from samples of national representative surveys did not provide such information for the smaller administrative units (cantons and districts) for geographical targeting.

To cope with this need, the Nutrition Information System chose the height census of children entering elementary school as a means to obtain the required nutrition information. Height-for-age data are an indicator of children's nutrition history and an indirect way to measure the socio-economic and environmental conditions to which they have been exposed [16]. These data collected through the census proved to be highly valuable, permitting the detection of counties and districts at risk for malnutrition. Maps showing the worst-off areas were constructed [17], and computerized presentations of the data in graphical form were prepared.

On the basis of this information, the resources of the Family Allowance Law (approximately US$150 million per year) were channelled to the poor through programmes in such areas as supplementary feeding, water supply, latrines, primary health care, training of unskilled workers, and pensions to the elderly. Later it also supported a housing programme for the low-income population.

Figure 2 compares the prevalence of low height-for-age in the ten worst-off districts found in the first school census of 1979 with the situation in 1985. The improvements seen in figure 2 are also reflected in an overall positive trend for the period 1979-1989 (fig. 3). These positive results can certainly be associated with the geographical targeting of resources to previously identified priority counties and districts, which otherwise could have been left behind.

Since 1979, when the first school height census took place, information on the height-forage status of seven-year-old children (by itself or combined with other socio-economic indicators) has been used in Costa Rica to allocate human and financial support to priority districts; to define coverage according to the selected districts' populations and the higher goals that are supposed to be achieved, usually from 80% to 100% of the target population in these districts; and to determine the type of service to be offered, for example, providing a complete lunch in these communities rather than just a glass of milk with crackers.

FIG. 2. Prevalence of low height-for-age (< - 2 SD) in children entering first grade, 1979 and 1985, in the Costa Rican counties with the highest prevalences in 1979 (Source: ref. 17)

FIG. 3. Mean prevalence of low height-for-age ( < -2 SD) in first-grade children-Costa Rica, 1979-1989 (Source: ref. 18)

Recently the information has been used as part of a social development index (SDI) developed by the Ministry of Planning, based on eight indicators in the areas of health and nutrition, education, and housing, as follows [18]:

» health and nutrition

- percentage of children with low height-for age,
- percentage of the population not covered by the social security system,
- mean time to the nearest health centre;

» education

- percentage illiterate among the population 10 years old or older,
- percentage of the population 10 years old or older with elementary-school education or less;

» housing

- percentage of occupied houses without electricity,
- percentage of occupied houses without water,
- percentage of houses in fair or bad condition.

These indicators were selected considering the information available and its reliability, taking into account the need for a balance between different areas of social interest.

The districts were ranked in order from the most to the least badly off for each indicator and assigned scores according to their decile position. Thus the worst-off districts, those in the highest decile, for a given indicator received a score of 10, with lower scores, ranging down to 1, for those that were relatively better off. The index for each district then was calculated by adding the scores for the individual indicators and obtaining their mean. The index values fluctuate between 1 and 3 for the better-off districts and 7 and 10 for the worst-off ones.

The SDI is used to categorize the districts and counties according to their relative development level, with cut-off points set at the mean value for the metropolitan area (where the better-off districts are located), the mean value for the whole country, and the mean value for peripheral districts (near the frontiers and the ocean). The resulting categories are as follows:

» better development level, 1.00-3.03,
» median development level, 3.04-5.49,
» low development level, 5.50-7.11,
» very low development level, 7.12-10.00.

Figure 4 shows the distribution of districts according to the SDI in the Atlantic region.

The SDI is used not only for programme planning in the social sector but also to guide economic interventions, such as the establishment of tax-free zones in less-developed communities. This type of intervention aims to support economic development in marginal areas and to revert the tendency of migration to the urban central areas.

In conclusion, height-for-age information and the SDI are both useful for programme planning. It is important to point out that they wore developed in the context of a supporting social and political context, responded to specific information needs, and observed the keep-it-simple feature that should be kept in mind if truly useful information is to be gathered.

FIG. 4. District development levels, assessed by the social development index-Atlantic region, Costa Rica, 1991 (Source: ref. 19)



1. WHO. Methodology of nutritional surveillance. Report of a joint FAO/UNICEF/WHO expert committee. Technical Report Series, no. 593. Geneva: WHO, 1976.

2. Rodríguez L. A surveillance system on access to food in Costa Rica. MPS thesis, Cornell University, Ithaca, NY, USA, 1990.

3. Pichardo MA. Planificación y programación social. San José, Costa Rica: Editorial de la Universidad de Costa Rica, 1986.

4. Tucker K, Pelletier D, Rasmussen K et al. Advances in nutritional surveillance. CFNPP monograph 89-2. Ithaca, NY, USA: Cornell Food and Nutrition Policy Program, 1989.

5. Berg A. Sliding toward nutrition malpractice: time to reconsider and redeploy. Am J Clin Nutr 1993;57:3-7.

6. Mason J. Trowbridge F. Haga J. Defining nutritional data needs. Serie SISVAN-INTA 18/85. CIVAN 85 INTA-Cornell University. Santiago, Chile: INTA, 1985.

7. Rothe GE, Habicht J-P. Nutritional surveillance: state of the art. In: Latham MC, Bran T. eds. World food issues. 3rd ed. Ithaca, NY, USA: Center for the Analysis of World Food Issues, Program in International Agriculture, Cornell University, 1990:105-10.

8. Pelletier D. Definition and types of nutritional surveillance. Discussion paper. Ithaca, NY, USA: Cornell Food and Nutrition Policy Program, 1989.

9. Habicht J-P, Pinstrup-Andersen P. Principles of nutritional surveillance. Pew/Cornell Lecture Series on Food and Nutrition Policy. Ithaca, NY, USA: Cornell Food and Nutrition Policy Program, 1990.

10. Ewell PT. Information for decision: What's the use? In: Ewell PT, ed. Enhancing information use in decision making. New Directions for Institutional Research, no. 64. San Francisco, Calif, USA: Josey-Bass, 1989:7-19.

11. Sandiford P. Rojas Z. A desk study of the health sector in Central America and Mexico. Liverpool, England: School of Tropical Medicine, 1992.

12. Sistema de Información en Nutrición. Encuesta nacional de nutrición 1978: aspectos socioeconómicos de la nutrición. San José, Costa Rica: Oficina de Información, Presidencia de la República, 1980.

13. Sistema de Indicadores Sociales, MIDEPLAN. Costa Rica: diferencias geográficas en el nivel de desarrollo social. San José, Costa Rica: Ministerio de Planificación y Política Económica, 1987.

14. DESAF. Dirección General de Desarrollo Social y Asignaciones Familiares: su ley y su reglamento. San José, Costa Rica: DESAF, 1989.

15. Figueres J. La pobreza de las naciones. San José, Costa Rica: Imprenta National de Costa Rica, 1973.

16. Ruel M. El uso de datos antropométricos en los procesos de vigilancia alimentaria nutricional. In: Sibrían R. Aburto A, eds. Memorias, seminario-taller subregional: Análisis y uso de la información alimentario-nutricional que se genera en los países del istmo centroamericano. Guatemala: INCAP, 1992:39-58.

17. Abarca N. Araya N. Estado nutricional de la población escolar de primer grado, Costa Rica 1979-1985. Serie informes 87-2. San José, Costa Rica: Sistema de Información en Nutrición/Oficina de Control de Asignaciones Familiares, 1987.

18. Tacsan L, Ascencio M, López E, Ramos G. Análisis del estado nutricional de la población costarricense. San José, Costa Rica: Ministerio de Salad, 1992.

19. Ramirez, RE, Barahona M. Costa Rica: zones de mayor y manor desarrollo relativo. San José, Costa Rica: Ministerio de Planificación y Política Económica / Fondo de Población de las Naciones Unidas, 1991.

Contents - Previous - Next