5
A Role for Nutrition Surveillance in Addressing the Global Food Crisis

The food price and economic crises have highlighted the need for collecting data in order to understand the effects of these phenomena on populations and make decisions to improve the situation. There are a variety of tactical measures and approaches to nutrition surveillance that will be explored in this chapter. Workshop presentations discussed an array of nutrition surveillance systems and lessons learned. This chapter considers what roles nutrition surveillance might be able to play in the future, including an investigation of the separate capacities of various agencies and specific projects. A number of presenters spoke of the need to aggregate data, compile it quickly using new technologies, and deliver it to the food security and nutrition community for decision making at the program and policy level. As described by moderator Keith West of Johns Hopkins Bloomberg School of Public Health, the following presentations helped to encourage a broad discussion of nutrition surveillance, including existing nutrition surveillance systems, their capacity to monitor food price fluctuations, and the gaps and needs for improved surveillance.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 75
5 A Role for Nutrition Surveillance in Addressing the Global Food Crisis T he food price and economic crises have highlighted the need for collecting data in order to understand the effects of these phenomena on populations and make decisions to improve the situation. There are a variety of tactical measures and approaches to nutrition surveillance that will be explored in this chap- ter. Workshop presentations discussed an array of nutrition surveillance systems and lessons learned. This chapter considers what roles nutrition surveillance might be able to play in the future, including an investigation of the separate capacities of various agencies and specific projects. A number of presenters spoke of the need to aggregate data, compile it quickly using new technologies, and deliver it to the food security and nutrition community for decision making at the program and policy level. As described by moderator Keith West of Johns Hopkins Bloomberg School of Public Health, the following presentations helped to encourage a broad discussion of nutrition surveillance, including existing nutrition surveillance sys - tems, their capacity to monitor food price fluctuations, and the gaps and needs for improved surveillance. 

OCR for page 75
 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS NUTRITION SURVEILLANCE IN RELATION TO THE FOOD PRICE AND ECONOMIC CRISES John Mason, Ph.D., Professor, Department of International Health and Deelopment School of Public Health and Tropical Medicine, Tulane University Nutrition Surveillance: Making Decisions to Improve and Protect Nutrition Nutrition surveillance means to watch over and make decisions that will lead to improvements in the nutrition of populations (FAO/WHO/UNICEF, 1976). Nutrition information—when appropriately linked to interventions, policies, and programs—can help mitigate malnutrition, particularly in developing countries (Figure 5-1). The current economic and food crises serve as stimuli for the world to be concerned about nutrition and perhaps can foster movement in the direction of positive change for nutrition surveillance. In public health, the “surveillance cycle” is well established and understood. One starts at the point of collecting data, the analysis leads to decisions on action, the action is implemented, the situation is followed, and new data can be used to monitor and evaluate that situation at the same time as detecting new problems (Figure 3-1). This surveillance process applies in the area of nutrition, usually at the population level but may also refer to individuals or households. Do We Really Know How Many People Are Hungry? (And What Does That Mean?) The official Food and Agriculture Organization of the United Nations (FAO) estimates from 2009 show that the number of the world’s “hungry” has hit 1 bil - lion (Food and Agriculture Organization of the United Nations, 2009). This esti- mated increase in the world’s hungry is similar to that seen around the year 1970, at the time of another world food crisis. Yet hunger is a very “lazy” indicator; its meaning is not very clear and not often explained, and its method of calculation is suitable only to inform that the hunger situation is getting worse. The true definition of this “hungry” indicator is the number of people who during the course of the past year did not on average get enough food energy to maintain moderate activity and body weight (UN ACC-SCN, 1993). The calculation of this indicator takes the estimated calorie availability at a national level (dietary energy supply [DES]) and sets that together with an estimate of the coefficient of variation (CV) of the distribution of consumption and estimates the number below a certain cut point. It does not take into account a number of 1 The following people contributed to Dr. Mason’s presentation: Megan Deitchler, M.P.H., FANTA; Marito Garcia, Ph.D., The World Bank; A. Sunil Rajkumar, Ph.D., The World Bank; and Roger Pearson, M.A., UNICEF.

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE Data Observation Detect New Problems Decide on Monitor & Action Evaluate Take Action/ Implement FIGURE 5-1 Surveillance cycle. factors, such as that most people have a very high degree of correlation between intake and requirement: most individuals do not change weight very much over R01545 the course of a year. The international nutrition community should consider Mason F-1 whether this hunger indicator is appropriate as the primary means for assessing global food deprivation and undernutrition. A newle of indicators or measure- vector editab set ments could be adopted to keep track of the global situation regarding undernutri- tion, hunger, and malnutrition. The ability to understand the number and percentage of households that experience hunger would be extremely useful. The methodology needs to evolve towards adding other measures to the existing FAO DES/CV approach to assess- ing the problem. Suitable methods were previously assessed (Food and Agri - culture Organization of the United Nations, 2002) (Table 5-1). One promising method (the fifth listed) involves qualitative data collection—three simple ques - tions about the experience of hunger in households, referred to as the “house - hold hunger scale”—which has now been tested in several countries and could be applied more widely (Deitchler et al., 2009). Other existing measures of the dimensions of hunger are more complicated and expensive to apply. The inter- national community has continued to rely on the first of these indicators; indeed, it can be argued that the current hunger indicator may actually be obstructie to progress in hunger assessment because other methods have not been developed as fast as they would have been, were there less reliance on this first (DES/CV) indicator.

OCR for page 75
 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS TABLE 5-1 Suitability of Different Measures for Trend Estimation Method Suitability of Trend Analysis Dimensions of Hunger Measured FAO DES/CV Only reflects DES change as Energy intake, with averaged CV held constant; only method adequacy at population level available for all countries and all years Household income and Potentially suitable when there Energy intake, with some expenditure survey are repeated comparable large- household adequacy; economic scale surveys aspects (e.g., employment, wages, food prices) also possible Food consumption and Repeated comparable large- Energy, better chance of relating individual intake scale surveys very rare and to requirement, hence adequacy expensive Anthropometry Suitable and widely used for Some aspects of health; changes trends, but does not measure often related to food access only (or even) food security changes Qualitative methods Probably very suitable within Suffering, behavior, and country, but cross-country economic activity may be comparisons need more work assessed SOURCE: Food and Agriculture Organization of the United Nations, 2002. Surveillance Inextricably Linked to Interventions Although data from surveillance systems may be interesting in its own right, clearly the ultimate goal (and inextricable partner) of surveillance data is the link to, and triggering of, appropriate interventions and successful implementa - tion of programs. The interventions to protect nutrition, referred to in a recent article by the UN’s Standing Committee on Nutrition (SCN), range from safety nets to a series of essential nutrition interventions, usually best implemented through community-based programs and primary health care. A number of well- established and evolving safety net interventions can mitigate the negative effects of the current food situation, particularly: • Conditional cash transfers, • Unconditional cash transfers, • Food and nutrition programs (school feeding, micronutrient and food supplement distribution), • Price subsidies for food or energy, • Public works employment (cash or food), and • Fee waivers. These programs, particularly the conditional cash transfers (CCTs), are wide- spread and are increasing rapidly. One estimate shows 137 CCT programs in 37

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE countries in sub-Saharan Africa (Social Safety Nets: Moing Forward for an Afri- can Agenda, 2009). CCTs have a dual or multipurpose rationale: one is to provide income support to needy people; the second is to encourage behaviors that could improve human capital, nutrition, and the education of children. In this sense, the conditionality is to ensure that children are immunized, go to school, or are cared for in ways regarded as beneficial. When public expenditures, infrastructure, and public services fail to reach the very poor, CCTs offer a method of targeting the poor and vulnerable. For these reasons, CCTs may be an important way to foster the development of human capital, as well as healthier, better developed, and more properly nourished children. How in Theory Would Better Nutrition Surveillance Work? What are the problems that improved nutrition surveillance could solve? Interventions are inadequate and untimely in preventing worsening malnutrition caused by rising food prices, increasing unemployment, and reduced public or private services. These crisis situations have a particular effect on certain vulner- able groups including pregnant women (with irreversible effects in utero to the unborn child) and the urban poor. The international community, then, needs to think about interventions that can take place quickly and immediately, at least for certain highly vulnerable groups. Then, resources must be allocated to organize and strengthen effective programs and to evaluate whether these programs are having the intended effect. Surveillance and data collection usually focus on the population level; increasingly, however, surveillance may apply to the individual level as well. There are three types of population-level data that are very pertinent to the pres- ent economic situation: 1. The real price of food—the ratio of the food price index changes to the general price index—does predict quite well when there is difficulty getting food (when there is food insecurity) and is likely correlated with malnutrition. 2. The “household hunger scale” is a second method of assessing food inse- curity or hunger. 3. The third method is tracking the prevalence of malnutrition, usually assessed by anthropometry (the study of human body measurements, especially on a comparative basis). Malnutrition is the result of a number of different factors. It is not the same thing as food insecurity, although it tends to track in the same direction. The “household hunger scale” could be included in many household surveys (e.g., Demographic and Health Surveys). It uses a short set of questions:

OCR for page 75
0 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS • In the past four weeks, was there ever no food to eat of any kind in your household because of lack of resources to get food? • In the past four weeks, did you or any household member go to sleep at night hungry because there was not enough food? • In the past four weeks, did you or any household member go a whole day and night without eating anything because there was not enough food? The results of such questions are sensible and easy to interpret; they appear rel- evant and easily communicated across different cultures (Deitchler et al., 2009). At the indiidual level, eligibility criteria for cash transfers and the numbers of eligible recipients give an indicator as to levels of food insecurity, malnutri - tion, and hunger. Data from screening children can give a real-time estimate of the number of children who are malnourished. There are other indicators, such as the number of people who participate in public works (paid in cash or food), which can serve as gauges of hunger. The eligibility for access to programs could provide a new set of information and needs to be built upon. Nutrition surveillance has traditionally been seen as having three major purposes (Table 5-2): long-term planning, program monitoring, and timely warn - ing (Mason et al., 1984). Timely warning of crises is the most relevant in this TABLE 5-2 Population Data Sources and Their Use Program Monitoring Timely Warning to Source Long-Term Planning and Evaluation Prevent Crises Repeated national Main use Possible use, but No use, as too surveys rare as process data infrequent and too much limited and design lag time not ideal Area-level surveys Not usually, but Possible use, but Main use, together with some potential with rare as process data other data (e.g., prices) further analysis limited, design not ideal, and external validity may be unclear Reporting systems Not usually; Potential use for Potential main use if lag considered less process monitoring time can be minimized reliable than if lag time can be repeated national reduced surveys Sentinel systems Potential use Potential use Potentially important for evaluation if use carefully designed

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE context, and the tool that is particularly useful in this context for timely warning is small-scale area-level surveys. More than 1,000 small-scale surveys took place in the past 6 or 7 years in the Horn of Africa alone (Mason et al., 2007). These are useful in allowing fairly rapid assessments of the current nutritional situation and should be continued in a consistent manner. The collective experience of reporting systems from clinics or nutrition programs in Africa is that they are not sustainable. If the reporting system is useful at the local level, then it is continued, but if people are collecting infor- mation just to pass up to other agencies, the reporting system is not continued. Again, these systems need to be part of ongoing programs and need to be linked to community-based programs that are probably the most effective way to deal with child malnutrition. Sentinel systems (facilitated reporting from selected sites, either of data routinely collected, or by household sample surveys) are appealing because they use information from a limited number of places with implications for predict - ing trends more widely. Where applied, such systems have been successful. For example, in Zimbabwe, with UNICEF support, 6 monthly estimates have been made over the past 2 or 3 years, and these estimates have given remarkable infor- mation about the extent of, and changes in, malnutrition. One outcome has been to show that the proportion of people who had one meal or less the previous day went from about 15 percent in 2007 to 34 percent in July 2008, and to 47 percent in November 2008 (Zimbabwe Combined Micro- nutrient and Nutrition Sureillance Surey: Summary of Main Findings, 2008; Zimbabwe Nutrition Sentinel Site Sureillance System: Summary of Main Find- ings, 2008). In practice, nutrition interventions are more effective and sustainable when based upon much larger programs. For example, the health extension program in Ethiopia, supported by The World Bank, is a massive reform that increased outreach of health services. This program provides an opportunity to build on nutrition-protecting programs and make them more effective. The role of nutrition information for safety net interventions, then, is to target populations; measure their needs, the levels, and particularly the changes in nutritional status; assess the outcomes (including in terms of evaluation); and conceivably contribute to individual eligibility assessment. Next Steps Intervention policies of a large scale are required if such policies are to reverse the increased hunger and malnutrition that undoubtedly resulted from the food price and economic crises starting in mid-2008, even if the extent of these problems are not yet measured in any detail. The international nutrition community needs to help develop systems to provide timely and disaggregated information to support these large-scale intervention policies. Nutrition surveil -

OCR for page 75
 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS lance should focus more on routine data collection (e.g., prices) and qualitative assessment (e.g., household hunger scale), as well as continuing to use repre - sentative surveys. There will likely be a convergence of the safety net eligibility at the individual and population levels, as well as a convergence between the response to such crises as droughts and economic stress and the response to other large crises such as HIV/AIDS. INSIGHTS FROM 25 YEARS OF HELEN KELLER INTERNATIONAL’S NUTRITION SURVEILLANCE IN BANGLADESH AND INDONESIA Andrew Thorne-Lyman, M.H.S., Department of Nutrition Harvard School of Public Health There are few articles in the peer-reviewed literature documenting the impact of rising prices on nutritional and health outcomes in developing countries. Helen Keller International (HKI)’s Nutrition Surveillance Project (NSP) functioning in Bangladesh from 1990 to 2005 and its Indonesian Nutrition Surveillance System (NSS), which collected data from 1995 to 2005, stand out as relatively unique examples of surveillance systems that have generated analytical insights related to understanding the effects of economic crises on such nutritional outcomes as child underweight, stunting, maternal underweight, and micronutrient deficiencies (Torlesse et al., 2003; Block et al., 2004; Campbell et al., 2009a,b). Unique Focus and Objectives of the NSP and the NSS In many ways, the NSP and the NSS were distinct from the dominant paradigm of nutrition surveillance that was in place as of 1990 when the NSP was started (Bloem et al., 2008). At the time, surveillance was thought of as something that should be simple and inexpensive, including only a minimal number of indicators. Surveillance systems were intended to generate information to be used primarily at the community level to influence programs. The NSP and the NSS diverged from this model. From inception, both systems were oriented towards generating infor- mation that would be used to inform decision making at the national level, by policy makers working in multiple sectors including nutrition, health, and agriculture. Design of the Systems Nutritional status is an outcome that is the result of many complex causes and interactions. Analysis of such interactions through conventional nutrition surveillance approaches is often limited by both the sample size of surveillance systems and the relatively narrow scope of other variables (which typically reflect the sector providing funding to the system). Both the NSP and the NSS had relatively large sample sizes that enabled both geographic disaggregation and

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE TABLE 5-3 Comparison of the NSP in Bangladesh and the NSS in Indonesia NSP (Bangladesh) NSS (Indonesia) Implemented by HKI and GOB (IPHN); HKI network of 17 NGO partners Dates of surveillance 1990–2005 1995–2005 Frequency of data collection 6 times/year 4 times/year Sampling Multistage cluster Multistage cluster ≈10,490 rural ≈33,600 rural Sample size per round ≈1,300 urban ≈10,800 urban Statistically representative National, divisional levels Seven densely populated (rural) provinces (70% of rural population) Urban Slums in three cities Slums in four large cities NOTE: GOB = government of Bangladesh; HKI = Helen Keller International; IPHN = International Poverty and Health Network; NGO = nongovernmental organization. precise estimation of such “tip of the iceberg” indicators as child night blindness, as well as facilitating disaggregated analyses that might not have otherwise been possible (Table 5-3). The NSS in Indonesia began in Java and expanded when the economic crisis hit, tracking the impact of the crisis and progress made as the country emerged from that crisis. In contrast, the system in Bangladesh was more oriented to col - lecting data to track development programs and to track the impact of develop - ment on a national level, although it was also used to measure the impact of crises such as hurricanes, pests, and flooding that affect different parts of the country over time. Malnutrition is a seasonal phenomenon in many countries, and one of the benefits of an ongoing surveillance system that collects data at multiple points throughout the year is the ability to understand the normal seasonal patterns of malnutrition, whereas a stand-alone survey taken at one point in time could eas - ily reach the wrong conclusions about trends in malnutrition or might wrongly attribute a rise or fall in malnutrition to a particular event (which might just be a regular seasonal phenomenon). Another unique feature of the NSP and the NSS was the inclusion of urban slums in the samples in both Indonesia and Bangladesh, which enabled the sys - tem to provide information about a growing segment of urban poor populations in both countries. The use of modules in the design of the system enabled flexibility to collect information about emerging issues of programmatic relevance, such as reasons for lower coverage of the vitamin A program in Chittagong, cyclone

OCR for page 75
 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS preparedness, and variables to be used to help measure the impact of the growing national nutrition program. Quality Control Quality control was an essential part of both the NSS and the NSP, especially because information was collected from a network of more than 17 partner orga- nizations. In Bangladesh—a country prone to floods, cyclones, and droughts— having a network of different sites throughout the country was invaluable during times of crisis. One of the downsides of using many partners, however, is the variability in the quality of the data collected. In an effort to ensure good qual - ity across all of the partners, there was a quality control team that revisited 10 percent of the households, administered part of the questionnaire, and made sure the questions were answered in the same way as had been previously reported. Refresher trainings were also held prior to each round of data collection to ensure good quality data. Breadth of Information Collected Because of the breadth of information collected, the NSP and the NSS were able to not only show trends in malnutrition and stunting over time, but also to explain the factors associated with these changes. Although information about the prevalence of malnutrition is often collected every several years in many coun- tries through other surveys, one of the virtues of the NSP and the NSS was the inclusion of information not typically available through health-focused surveys, such as food prices, household expenditures, agricultural and cropping patterns, land ownership, and female decision making within the household. These enabled a wider exploration of factors associated with malnutrition than is normally undertaken, as well as the isolation of relationships between variables by enabling adjustments for confounding factors. In both countries, information on micronu - trient status was also collected, facilitating the understanding that changes in the quality of the diet resulting from food price changes may have adverse impacts on health, even if they do not influence the prevalence of child undernutrition in affected populations (Bloem et al., 2005). Weaknesses of the NSP and the NSS Systems Often, surveillance systems are expected to serve as early warning sys - tems. This was not an explicit goal of either the NSP or the NSS. The relatively large sample size required collection of data over a period ranging from 4 to 6 weeks—such turnaround time makes it difficult for information to be used as an early warning. One downside of most early warning systems is that their reliance

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE on small sample sizes leads to a lack of precision. Differences in the objectives of surveillance systems should influence their design. The issue of cost and sustainability is another criticism that has been made of these systems. Although cost information is not readily available, these systems were more expensive than most. But in the context of the present food crisis, many rapid assessments were done at significant cost, because they required the establishment of new infrastructure to collect information. Most lacked baseline information from the time period prior to the food price rises that could have been used to better understand the potential impacts of the crisis itself. The value of having ongoing systems in place to collect information is often not appreciated until after a crisis hits. FAMINE EARLY WARNING SYSTEMS NETWORK, NUTRITION SURVEILLANCE, AND EARLY WARNING Chris Hillbruner, M.S., Food Security Warning Specialist Chemonics Introduction to FEWS NET The goal of the Famine Early Warning Systems Network (FEWS NET) is to provide an early warning of an impending food crisis. While FEWS NET is not a direct implementer of nutrition surveillance systems, it has a role to play in collecting nutrition information that could provide an early warning of food insecurity and other crises that are detrimental to the nutritional status of sur- veilled populations. FEWS NET is at work in 20 countries and has broad experience look- ing at the type of information available in different places and applying that data where it can be successfully used. FEWS NET is a partnership that pulls together a variety of different information flows from a core group of interna - tional partners, including academic institutions, consulting companies, and U.S. government institutions. FEWS NET also collaborates with a wide network of formal and informal partners. FEWS NET has nine internal partners: Chemonics International, Michigan State University, FEG Consulting, WebFirst, Inc., Intana, U.S. Geological Survey, National Oceanic and Atmospheric Administration, U.S. Department of Agriculture, and National Aeronautics and Space Administration. Members and collaborators of FEWS NET include UN World Food Programme (WFP); FAO; several national ministries of agriculture, rural development, and health; various price and market information systems; meteorological centers; nongovernmental organizations (NGOs); and other UN agencies (UNICEF, UN High Commissioner for Refugees [UNHCR]). FEWS NET focuses primarily on early warning with a 4- to 6-month out- look, providing information that is credible and actionable for decision makers.

OCR for page 75
 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS than regular surveillance, it is more difficult to look at trend analysis and use that information to project into the future. Moving Forward There is a need to start advocating for a bigger role for nutrition and health data in early warning systems. Additionally, more nutrition surveillance systems must be developed. These should be timely and ongoing, focus on vulnerable areas and populations, have some analytical capacity (not just a data collection function), and leverage partnerships for contextual data collection. A system should recognize who is currently collecting information and focus on where the gaps are, as opposed to creating a heavy and duplicative system. LISTENING POSTS PROJECT: A CONCEPT FOR A REAL-TIME SURVEILLANCE SYSTEM NESTED WITHIN A PROGRAM Anna Taylor, Head of Hunger Reduction Save the Children UK Introduction to Listening Posts Save the Children and Action Against Hunger are collaborating in an effort to integrate surveillance into routine program work. During the food price increases of 2008, Save the Children realized it was ill-prepared to determine the effects of the crisis in the communities where it had programs. In an effort to rectify this gap, the Listening Posts project was developed; it has a threefold purpose: • Listen and pick up shocks and their nutritional effects at a local level. • Respond and be able to use information to inform program work and adjust programming according to the findings of the surveillance. • Inform others about what is happening in the vulnerable communities under the project’s surveillance, and link what is happening at the local level to global shocks as they occur, in hopes of implementing a larger response. The following six principles guide the design of the Listening Posts system: 1. The system should be as light as possible so as to be easily integrated into program work with very little additional cost. The proposed level of investment is one team of two people working for 5 days every 3 months gathering data.

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE 2. The system should operate as rapidly as possible. The project is currently investigating new technologies for data management, such as mobile technology, in order to transmit data quickly from local points to capital entities and eventually to the rest of the globe. 3. The system should be local and global. The project is attempting to strike a balance between information that is locally useful and context specific, as well as information that allows discussion with some authority about what is happening to vulnerable communities in multiple locations for global policy application. 4. The system should be as widely endorsed as possible to ensure the find- ings are quickly accepted and acted upon. . 5. The system should be linked into national early warning systems. Ide- Ide- ally, the data would be useful for FEWS NET and used in the food secu - rity classification processes that are happening increasingly in various countries. 6. The system should be as replicable as possible so that, if successful, it can be easily used by other programs. Indicators to Be Collected The Listening Posts project would use a very small indicator set. It would collect data on the prices of staples and the ratio between those prices and labor rates as a very broad measure of changes in the economic situation. Dietary diversity and feeding frequency among children aged 6 to 24 months would also be collected using the standard indicators, along with mean weight gain, MUAC, and edema for children aged 6 to 24 months. Some optional indicators that country programs may choose to use include details around coping strategies and talking to children about their experiences as well as such impact-level assessments as feeding center admissions, changes in child labor, marriage, and reduced school attendance. Such data would be more relevant, for example, when an education program is running in the same area. How Would It Work? Country programs would be required to implement the Listening Post strat - egy in a minimum of one livelihood zone. Each livelihood zone would be divided into six quadrants, selecting the community closest to the center of the quadrant for the nutritional data; that community would be the listening post. There would be six listening posts per livelihood zone. At each listening post, 16 children would be selected to survey. When a child exceeded 24 months of age, a new child would be selected. According to this design, there would be 96 children per livelihood zone.

OCR for page 75
0 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS What Is the Cost-of-the-Diet (CoD) Assessment? The cost-of-the-diet (CoD) assessment developed by Save the Children over the past few years is a method for calculating the lowest cost diet that meets the is nutritional requirements of a whole family. It builds on the discussion about data quality and the international nutrition community’s inability to talk about the affordability of a quality diet. Food security is often inappropriately equated with energy security; the CoD assessment is an attempt to take the discussion a step further toward understanding affordability of a quality diet. CoD is an Excel-based tool that originates from the linear programming work that has been done at the World Health Organization (WHO). The system requires a list of locally available foods, their prices by season, and the size of a typical family. That data are entered into the linear programming tool, which then calculates the lowest cost nutritionally adequate diet for that location (Figure 5-2). This unique assessment approach takes seasonality into account and determines locally specific estimates of cost. It is also possible to build constraints into the program to limit the portion sizes of very cheap foods, which may not be possible to eat in large quantities, and to model the effects of introducing micronutrient supplementation or home fortification. List of foods available Additional constraints Fixed breast locally on contribution of milk intakes food types to diets for under-2s LOWEST COST, NUTRITIONALLY ADEQUATE DIET Size of a Food prices typical family by season Source of income data FIGURE 5-2 Schematic of the cost-of-the-diet assessment. R01545 Taylor F-1

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE Country Example From 2005 to 2006, the CoD system was used in Bangladesh. It found that in one community, 194 households, or 50 percent of families, could not afford a diet that met all nutritional requirements. (This scenario was estimated even before non- food items that the households may want to buy were considered.) About 15 percent of households could not even afford an energy-only diet (meeting only basic energy requirements and not other nutrient needs). In 2007–2008, a repeat assessment looked at the nutritional effects of food price increases. The system found that when considering both rising food prices and the harvest failure, there was not much difference in the proportion of people who could afford a quality diet. In fact, this second assessment showed that those who could afford a quality diet slightly increased, although the proportion who were pushed further into poverty and were unable to even afford an energy-only diet mas- siely increased. Save the Children UK will use CoD data as a key element of the Listening Posts surveillance system and will launch the software and guidelines in Rome in October 2009. Goals for the Future Ideally, the Listening Posts project is anticipated to involve quarterly data collection for the entire Listening Post system. One or two livelihood zones per country would be covered in this quarterly reporting, and all of the data would be fed into national early warning systems. Ultimately, an annual global report, capturing findings from a number of livelihood zones over a number of countries, could be produced on a quarterly basis. FOOD SECURITY, NUTRITION MONITORING, AND THE GLOBAL FOOD PRICE CRISIS: USAID/FFP TITLE II PROGRAMS Ellen Mathys, M.P.H., Senior Food Security Early Warning and Response Specialist Food and Nutrition Technical Assistance Project II (FANTA-2) Academy for Educational Development FANTA-2 and Food Security Programs The U.S. Agency for International Development (USAID)-funded Food and Nutrition Technical Assistance Project II (FANTA-2) undertakes activities designed to strengthen the capacity of awardees that are implementing multiyear development projects funded with USAID Office of Food for Peace (FFP) Title II

OCR for page 75
 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS resources (MYAPs2). FANTA-2 support to these projects includes technical sup- port, support for monitoring and evaluation, and in some cases support to nutrition surveillance activities. These awardees collect and report on program monitoring and evaluation data to the FFP, and the opportunity exists to use these MYAP data to monitor the food security and nutrition effects of the global economic and food price crises. These data that are already being collected for annual program monitoring and impact evaluation may be used to understand how chronically food-insecure communities are affected by these price shocks. FFP Requirements: Impact Indicators FFP Title II awardees are required to report on the following impact-level indicators for program evaluation, which allows the FFP and FANTA-2 to look at the following population-level outcomes in program areas: • Average number of months of adequate household food provisioning • Average household dietary diversity score • Percentage of children aged 0–59 months who are underweight • Percentage of children aged 6–59 months who are stunted A typical MYAP is 5 years long, and impact-level indicators are measured during the baseline study and final evaluation. These impact-level indicators are fixed (i.e., the indicator definitions and data collection techniques are stan - dardized). This limits how the data can be used to understand such things as seasonality because it is only assessed at baseline and final evaluation. However, the standardization of the indicators allows comparability when compiling data globally. FFP Requirements: Monitoring Indicators Title II awardees are also required to collect data on four program monitor- ing indicators annually. First, an anthropometric indicator of the MYAP’s choice must be used to regularly monitor the maintenance or improvement in nutritional status of beneficiaries. Such an indicator must reflect anthropometric measure - ments of child or adult growth, or graduation based on anthropometric measures. Acceptable indicators include the prevalence of stunting, underweight, wasting, low body mass index (BMI), or low MUAC; trends (weight gain, growth falter- ing); and graduation or exiting—the proportion of children or adults recuperating, according to defined anthropometric cutoffs. Second, the percentage of benefi- 2 Throughout this presentation, these Title II–supported multiyear assistance programs will be referred to as MYAPs.

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE ciaries adopting improved health, nutrition, or hygiene behaviors is measured. Third, the percentage of beneficiaries (in this case, farmers) using a project- defined minimum number of sustainable agricultural technologies is recorded. Fourth, the MYAP must determine the number of project-assisted communities with (due to the MYAP) improved physical infrastructure to mitigate the impact of price shocks, disaster early warning and response systems in place, safety nets to address the needs of the most vulnerable members, and improved community capacity. FFP Guidance: Trigger Indicators Trigger indicators refer to early warning indicators that MYAPs are encour- aged, but not required, to collect. Trigger indicators are used to determine the threshold at which MYAPs need to shift activities or when additional resources for new activities are required in response to a slow-onset shock. Trigger indica - tors typically include such variables as: • Climate or rainfall, • Key food and cash crop production, • Staple prices, • Livestock prices, • Coping strategies, • Remittances and debt, and • Nutrition trends (if available). MYAPs identify trigger indicators in advance and can collect them in an ongoing manner either through secondary data sources or primary data collec - tion activities. On the basis of trigger indicator information, MYAPs can request emergency resources for use in their MYAP area. This serves as an administrative convenience in a way that allows them to protect the gains of the MYAP while they respond to a deterioration of food security conditions in the area. Many NGOs do not feel comfortable performing what they see as early warn- ing activities; however, the benefits are savings in time, money, and the ability to respond quickly on the basis of localized information without the procedural requirements previously associated with emergency response (i.e., through devel- opment of a single-year assistance program proposal). Availability of Population-Level Outcome Data in MYAP Areas Although the FFP implements food security programs in many of the world’s most chronically food-insecure communities, the opportunity to use data col - lected on these populations—especially nutritional data—is not being used to full

OCR for page 75
 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS advantage. In 2007 for example, only 31 of the 79 MYAPs were in priority FFP areas and reported underweight data. Of these 31 MYAPs, only 7 were report- ing population-representative prevalence rates (the data were collected via the MYAP’s baseline or final evaluation). Thus only a minority of programs are pro- viding population-representative outcome data useful for nutrition surveillance. This is unfortunate because these MYAPs are typically in the most chronically food-insecure areas in their countries. Conclusions and Key Opportunities There are both positive trends and challenges that must be noted in a discus- sion of Title II awardees and their MYAPs. In terms of positive trends, there has been an increasing availability and quality of food security and nutrition data collected by MYAPs and their key national and international partners. Addition - ally, there has been strong support by USAID for monitoring and evaluation (including trigger indicators) in Title II programs. In terms of challenges, there are not enough resources to increase monitoring and evaluation responsibilities for MYAPs; yet the quality, scale, and frequency of data collection—particularly of population-representative data—are insufficient. It is difficult to compare the anthropometric monitoring data among MYAPs because of the variation in indicators used, the methodological constraints to attributing causality, and the complicated task of defining and harmonizing trigger indicators. Clearly, there is a role for nutrition surveillance systems to monitor impacts of price shocks on food security and nutrition, and this will complement, not duplicate, data col - lected through MYAPs. There are some key opportunities to improve current data collection and use of the data that are collected through MYAPs. In terms of data collection, several program approaches on the horizon would expand nutrition surveillance coverage, including the prevention of malnutrition in children under 2 years and mass treatment. New food security indicators are being developed that can be easily integrated into household surveys and monitoring, including the household hunger scale. Efforts are underway to strengthen and harmonize trigger indicators for food security monitoring and early warning in MYAPs. Mobile phone tech - nologies are being piloted for real-time transmission of food security early warn - ing data. NGOs increasingly collaborate in national food security and nutrition monitoring and early warning networks. In terms of data analysis and data use, new technologies are being developed to link multiple, agency-specific data sets together to enable triangulation and foster coordination. Food security scenario development must occur (including contingency and response planning) and a link must be forged between national and international alert systems.

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE DISCUSSION This discussion section encompasses the question-and-answer sessions that followed the presentations summarized in this chapter. Workshop participants’ questions and comments have been consolidated under general headings. Which Indicators? Several workshop participants argued that surveillance systems should be collecting very simple and interpretable indicators, like the cost of a local food basket or number of hours worked. The latter tracks the wage rate, a very easily understood indicator that is not very accurate, but says something that people understand. Such interpretable indicators track both wages and food prices, which are both extremely useful in this context. On the other side of the spectrum, such indicators as, “How many times a week do you eat a vitamin A-containing food?” is too complicated. Not only is the answer to such a question difficult to estimate, but the results are equally complex to interpret. This session on nutrition surveillance illustrated the range of data collection and systems that often function in the same country, for different organizations, using different types of monitoring systems. One participant suggested that when these data are brought back to counterparts at the national level, it can be confus - ing and less than helpful. It is urgent that the international nutrition community agree upon a set of indicators and approaches for certain conditions and purposes. It would do a great deal of good to have a sense of unity around interventions and indicators for nutrition. Because at the moment there is a greater sense of agreement on a set of interventions that have proven to be effective for nutrition, the nutrition community should be vocal about what interventions it supports and what indicators it wishes to collect. It was also posited that while coming up with a set of common indicators may be useful, it should also be noted that there are places where no data are being collected at all. In such countries, the need to start collecting some level of information may be more important than agreeing upon a certain set of data. The Time to Intervene One participant urged the international nutrition community to use the food basket (that corresponds to the culture and the price of that food basket relative to income) as the indicator that determines whether people have sufficient food quantity or quality. There needs to be more thinking in terms of region and ethnic group because indicators change according to these variables. WHO should discard set cut-points for when agencies react. For example, if a country shows 14.8 per- cent wasting (with the WHO cut-point at 15 percent); is it logical to say that only if wasting increases another .2 percent will there be a crisis? Additionally, the

OCR for page 75
 IMPACTS OF THE GLOBAL FOOD PRICE CRISIS wasting indicator is inappropriate. For example, there was no increased wasting at all in sub-Saharan Africa during the HIV/AIDS crisis in the early 2000s, but there was a huge increase in underweight. These set indicators and fixed cut-points can easily be misinterpreted. A workshop participant urged that, instead, surveillance systems should collect situation-specific data and undertake a more sophisticated interpretation than merely stating that 15 percent wasting signifies a crisis. Ability of Surveillance Systems to Accurately Predict One workshop participant expressed concern that the trigger that exists in surveillance systems presumes that there is a high level of predictive value. Given that crises are rare (even though epidemiologically high sensitivity and specific - ity in terms of separating crisis from noncrisis is possible), the ability to predict crises with high levels of certainty using data from surveillance systems may be weak. Dr. Mason responded that there are enough crises to make predictions; even historical data sets can be used. For example, the FAO early warning system accurately predicts food crises. But it is remarkable how seldom surveillance systems are used for early warning of food and nutrition crises. REFERENCES Block, S. A., L. Kiess, P. Webb, S. Kosen, R. Moench-Pfanner, M. W. Bloem, and C. P. Timmer. 2004. Macro shocks and micro outcomes: Child nutrition during Indonesia’s crisis. Economics and Human Biology 21(1):21-44. Bloem, M. W., S. de Pee, and I. Darnton-Hill. 2005. Micronutrient deficiencies and maternal thin - ness: First link in the chain of nutritional and health events in economic crises. In Primary and Secondary Nutrition 2nd ed. Edited by A. Bendich and R. J. Deckelbaum. Totowa, NJ: Humana Press. Pp. 357-373. Bloem, M. W., S. de Pee, and R. D. Semba. 2008. How much do data influence programs for health and nutrition? Experience from health and nutrition surveillance systems. In Nutrition and Health in Deeloping Countries. 2nd ed. Edited by R. D. Semba and M. W. Bloem. Totowa, NJ: Humana Press. Pp. 831-858. Campbell, A. A., de Pee S., Sun K., et al. 2009a. Relationship of household food insecurity to neona - tal, infant, and under-five child mortality among families in rural Indonesia. Food and Nutrition Bulletin 30(2):112-119. Campbell, A. A., A. Thorne-Lyman, K. Sun, S. de Pee, K. Kraemer, R. Moench-Pfanner, M. Sari, N. Akhter, M. W. Bloem, and R. D. Semba. 2009b. Indonesian women of childbearing age are at greater risk of clinical vitamin A deficiency in families that spend more on rice and less on fruits/vegetables and animal-based foods. Nutrition Research 29(2):75-81. Deitchler, M., T. Ballard, A. Swindale, and J. Coates. 2009. HFIAS Validation Study: Identifying an Experience-Based Measure of Household Hunger for Cross-cultural Use. Washington, DC: Academy for Educational Development. FAO/WHO/UNICEF. 1976. Methodology of Nutritional Sureillance. Geneva: World Health Organization. Food and Agriculture Organization of the United Nations. 2002. Proceedings of International Sci- entific Conference: Measuring and Assessment of Food Depriation and Undernutrition, 00. Rome: Food and Agriculture Organization.

OCR for page 75
 A ROLE FOR NUTRITION SURVEILLANCE ———. 2009. More People Than Eer Are Victims of Hunger. Rome: Food and Agriculture Organization. Mason, J., J.-P. Habicht, H. Tabatabai, and V. Valverde. 1984. Nutritional Sureillance. Geneva. Mason, J., et al. 2007 (unpublished). The Impact of Drought and HIV on Child Nutrition in Eastern and Southern Africa. Tulane University. Social Safety Nets: Moing Forward for an African Agenda. 2009 (unpublished). The World Bank. Torlesse, H., L. Kiess, and M. W. Bloem. 2003. Association of household rice expenditure with child nutritional status indicates a role for macroeconomic food policy in combating malnutrition. Journal of Nutrition 133(5):1320-1325. UN ACC-SCN. 1993. Second report on the world nutrition situation. UN-SCN II:111-114. Zimbabwe Combined Micronutrient and Nutrition Sureillance Surey: Summary of Main Findings. 2008. GoZ Food and Nutrition Council and UNICEF. Zimbabwe Nutrition Sentinel Site Sureillance System: Summary of Main Findings. 2008. GoZ Food and Nutrition Council and UNICEF.

OCR for page 75