![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Division of Nutritional Sciences, Cornell University, Ithaca, NY and * Department of Nutrition for Health and Development, World Health Organization, Geneva, Switzerland
2To whom correspondence should be addressed. E-mail: eaf1{at}cornell.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • stunting • malnutrition • growth • development
Stunting (i.e., low height-for-age) reflects poor linear growth accumulated during the pre- and postnatal periods because of poor nutrition and health. Stunting in early childhood is associated with detrimental effects on intelligence quotient, psychomotor development, fine motor skills, and neurosensory integration (1,2). Stunting also is related to mental capacity and school performance, even in mild-to-moderate cases, often leading to reduced work capacity in the adult years (2,3).
Recent efforts in combating malnutrition have resulted in a steady improvement since 1980 when the estimated global prevalence rate of stunting was nearly 47% (4). These improvements have been crucial to child well-being because they have been strongly related to improvements in child survival (5). Nevertheless, progress has remained slow, and most international goals set for improving child nutrition and health had not been met by 2000. Today, 164 million children worldwide are stunted, 
1 of every 3 children < 5 y old (6). Furthermore, improvements have differed across global regions. For example, although Southeast Asia has been experiencing the highest rate of improvement, at 1%/y, the prevalence of stunted children in sub-Saharan Africa is not improving. This trend, compounded with high population growth rates, translates into larger numbers of African children stunted each year (6).
The aim of the study was to understand why improvement over the past decades has been achieved in some countries but not in others. We addressed this by asking 2 questions. First, what was the relative importance of long-term development vs. that of specific interventions more directly aimed to improve child stunting? Second, which particular national factors were most strongly associated with improvement?
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
National factors were separated into 2 categories, in accordance with the UNICEF framework (Fig. 1): 1) social, political, and economic (i.e., basic) factors that capture long-term, developmental changes; and 2) food, care, and health (i.e., underlying) factors, that were likely influenced by both development and specific interventions (7). The basic factors influencing a child’s nutritional status operate at the national and community levels, whereas the underlying factors operate at the community and household levels. The most important of these factors were identified through variables, and the relative importance of variables of the 2 subsets was examined. For most factors, the variables available and used did not fully capture the conceptualized factor.
|
Data
The data used are described below. Sources of the data are listed in Table 1.
|
Care is "the provision in the household and the community of time, attention, and support to meet the physical, mental, and social needs of the growing child and other household members" (10). Female education is critical to the provision of adequate care because an educated woman is more knowledgeable about care practices, better able to use health care facilities, and more adept at keeping her environment clean (10). Care was measured by female literacy as a percentage of male literacy. Women’s status is also important because women with a higher status are more likely to allocate extra resources under their control to their children than are fathers (10,11). It was assessed by women as a percentage of the total labor force and the percentage of ministerial government positions occupied by women.
The health environment and services factor refers to the availability of safe water, sanitation, health care, and environmental safety, including shelter (12). It was measured by the average percentage of 1-y-olds immunized for tuberculosis, diphtheria, tetanus, pertussis, and polio, oral rehydration therapy use, and percentage of the population with access to safe water.
Basic factors. The social environment may have multifaceted effects on child growth. Urbanization, for example, directly affects children because it potentially leads to weaker informal safety nets, greater participation of women in the labor force with its consequences for child care, greater availability of public services (such as water, electricity, sewage and health care), and governance by a new set of property rights that may deter government and nongovernmental organizations from providing allotted benefits to children (4). Urbanization was measured as the percentage of the total population living in an urban environment. Fractionalization, or the lack of homogeneity in society, often results in political instability, conflict, and an increase in government consumption that deflects potential resources away from child welfare (13). The degree of heterogeneity of a society was measured by a fractionalization index, which evaluates the probability that 2 randomly selected people within a country will not belong to the same ethnolinguistic or religious group (13). HIV/AIDS diverts care away from children, reduces government expenditure on child health, decreases food security in the home, and decreases household income from agricultural goods (14). Data on the prevalence of HIV/AIDS were obtained from UNAIDS (15).
Political structure includes availability and recognition of political rights and civil liberties. Political rights indicate whether people can participate freely in political processes, including choosing leaders freely from among competing groups and individuals. Civil liberties give people freedom to act outside the control of government, including development of views, institutions, and personal autonomy (16). The political structure relates to child stunting through the degree to which the rights of women and children are protected by the law. It was assessed with a sum of measures of political rights and civil liberties (17).
Economic factors include major economic transactions within a country that affect economic output. The economic system determines how income, benefits, and assets are distributed within a society. An external debt will divert needed monetary resources away from social services that promote child health and development into paying off loans. A national debt will also discourage investment in the country resulting in even heavier economic losses; it was calculated as external debt as a percentage of the GNP. Government consumption, which consists mainly of public wage bills, is an indicator of consumers’ current benefits from government spending (18). Government consumption, as a percentage of gross domestic product, was used. A country’s health expenditures are theoretically representative of the input of national income into the improvement of public health and are an indicator of national priorities; it was measured by health services expenditures per capita. The percentage of gross domestic product attributed to agriculture was evaluated as the value added in agriculture as a percentage of gross domestic product; it is a reflection of the net output of a country’s agricultural sector (19). The value added in industry was highly correlated (–0.82) with value added in agriculture and was therefore not used. Income distribution was assessed by the percentage of national income held in the hands of the wealthiest top 20% of the population. Official development assistance was measured by development assistance as a percentage of gross domestic product.
Potential resources include the country’s environment, technology, and people that translate into real resources for food security, care, and health environments and services. Gross national product (GNP) per capita is considered an adequate measure to account for all of these overarching resources. Conflict is also included because it has influential effects on both the other basic and underlying factors. Conflict can lead to food insecurity and chronic underproduction; it is also associated with political and economic instability, loss of caregivers, and inadequacy of health environments and services for children. Also, expenditures on the military lower the national investment in health, education, agriculture, and environmental protection (20). Conflict was measured by dividing the months and years of conflict within a country by the total years that country was evaluated in this study. A measure of conflict intensity was also included. The following types of conflict were considered: ethnic wars, revolutionary wars, genocides, and politicides, and adverse regime change. A single composite measure was then created which was a sum of the numerical value for the proportion of analyzed years in war and the conflict intensity.
Stunting. Data on child growth were the prevalence of stunting among children < 5 y old in a country. Stunting was defined as low height-for-age at less than –2 SD of the median value of the National Center for Health Statistics/WHO (NCHS/WHO) international growth reference. Data on the prevalence of stunting were taken from the February 2003 version of the WHO Global Database on Child Growth and Malnutrition, which has standardized child growth data from population-based nutritional surveys conducted around the world since 1960 (21). A detailed description of the methodology and data quality control applied in the database is provided elsewhere (21).
Eighty-five developing countries in the WHO Global Database on Child Growth and Development were included in the study, 40 from Africa, 25 from Asia, and 20 from Latin America and the Caribbean. These countries had at least 2 national surveys for stunting that were more than 4 y apart and fit into the study’s time frame of 1970–2001 (21). Countries that were included and classified for this study as being from Africa were: Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Cape Verde, Central African Republic, Comoros, Côte D’Ivoire, Democratic Republic of the Congo, Djibouti, Egypt, Ethiopia, Gambia, Ghana, Guinea, Kenya, Lesotho, Madagascar, Malawi, Mali, Mauritania, Mauritius, Morocco, Niger, Nigeria, Rwanda, Sao Tome and Principe, Senegal, Sierra Leone, Sudan, Swaziland, Tanzania, Togo, Tunisia, Uganda, Zambia, and Zimbabwe. Countries classified as being from Asia were: Azerbaijan, Bahrain, Bangladesh, Bhutan, Cambodia, China, India, Jordan, Kazakhstan, Kuwait, Laos, Mongolia, Myanmar, Nepal, Oman, Pakistan, Philippines, Solomon Islands, Sri Lanka, Syria, Thailand, Turkey, Vietnam, Yemen, and Yugoslavia. Countries classified as being from Latin America and Caribbean were: Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, Peru, Trinidad and Tobago, Uruguay, and Venezuela.
Analysis
Because of high positive skewness, government consumption, official development assistance, health expenditures, HIV/AIDS prevalence, and per capita GNP were transformed using the natural logarithm. External debt was transformed using the square root.
Multiple linear regression analyses were implemented in Amos 4.0 software, which uses full-information maximum likelihood for estimation while accounting for missing data among the independent variables under the assumption of missing at random and without having to do imputation (22). Under this generally reasonable assumption, full-information maximum likelihood is unbiased and efficient, and is superior to the other most commonly used methods of handling missing data, i.e., pair-wise and list-wise deletion, and mean imputation (22).
The dependent variable was the change in prevalence of stunting over the studied time period for each country, expressed as the change in the percentage of stunting per year. That is, the prevalence of stunting was regressed against survey year for each country, using all available data points. The resultant slopes in units of prevalence per year were used as the dependent variable. As a robustness check, we also calculated the change in stunting between only the initial and final survey years, and found essentially identical results.
The analytic data set with the independent variables was constructed around the initial and final surveys available for the prevalence of stunting. Data for the independent variables were matched as closely as possible by year for each country to the stunting data.
In the regression analyses, both initial and final independent variables, corresponding in time to the initial and final stunting surveys, were examined if available. If the initial independent variable was included, then the interpretation of the coefficient for the corresponding final independent variable was the effect of a change in that variable on the change in stunting.
Nine multiple linear regression models were run. All models controlled for the initial rate of stunting, year of the first stunting data, and year of the last stunting data for each country. A null model was run with no other variables. Model 1 contained the underlying factors, Model 2 the social factors, Model 3 the political factors, Model 4 the economic variables, and Model 5 the potential resources. Model 6 contained all of the significant basic factors, as determined in Models 2–5. Model 7 combined all of the significant variables from Models 1 and 6. From Model 7, 2 techniques were used to create a more parsimonious Model 8, which included only the significant variables from Model 7. One represented a top-down technique in which the starting point was the underlying factors. Significant social factors, then political, then economic, and then potential resources were added into the model successively to determine which had remained significant even after accounting for the underlying variables. The second technique utilized a bottom-up approach in which the starting point was potential resources, and the variables going up the hierarchy of the conceptual framework were added successively until only the significant variables remained. The 2 techniques gave the same results for Model 8.
Analyses of scatter plots, leverage values, and residual plots, with and without inclusion of variables with many missing values (implemented in SPSS version 10), indicated that the data were adequate to fit the reported models and that there were no observations with undue influence. Adding region of the globe to Model 8 did not explain additional variation in changes in stunting. Similarly, the squares of y 1 and 2 were not significant.
The proportion of variability explained for each of the models was calculated as follows:
 |
This calculated the variance explained by the factors as a proportion of the variance remaining to be explained after accounting for that explained by the initial rate of stunting and the survey years (i.e., the null model).
Two sensitivity analyses were performed to gauge the robustness of the results. One analysis set more stringent criteria by excluding all countries with surveys that were <5 y apart; this excluded 6% of the sample. Another analysis excluded all countries with data before 1975; this excluded 8.5% of the sample. Both analyses yielded results essentially identical to those reported here.
Although it is possible that the analyzed independent variables interact with each other, such analysis was not conducted because it was beyond the scope of this study and would have been difficult to implement with the data available. Future research should examine such interactions when more data become available.
Results were estimated for Africa alone, and were essentially identical to the results reported here for the 3 regions combined. This comparison cannot be considered a test for whether there are regional differences in relations of independent variables to changes in stunting because Africa accounts for about half of the countries in the data set and sufficient data for the other 2 regions do not exist.
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
|
|
|
|
|
Model 8 contained the underlying and basic variables from Model 7 with P-values < 0.1 (Table 8). The regression coefficient for initial immunization rate was negative, meaning that higher initial immunization rate was related to greater improvement in stunting. Similar relations were found for initial safe water and female literacy, meaning that countries that were better initially improved more. Countries with a greater change in immunization rate also improved more. Countries with a higher value in agriculture initially improved less, and countries with higher initial stunting rates improved more. Countries with greater change in safe water did not improve in stunting as much as countries with less change in safe water, although this variable explained <1% of the variation after accounting for other variables in the model. Countries with lower initial government consumption, more equitable initial income distribution, and lower initial rate of economy devoted to agriculture improved more. These variables together explained 65.5% of the variance of the change in stunting.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The underlying variables important to explaining changes in stunting included an increase in immunization rates for children < 5 y old and high initial rates of female literacy, which were associated with improvement. Neither of these findings is surprising, given previous emphasis on adequate health care and maternal education (23–26). The change in daily energy supply was significant when included with other underlying factors, but not when income distribution was included. Income may directly affect individuals’ consumption of commodities, which consequently affects diet (27). Food insecurity might thus be reflected in income inequality as well as in daily energy supply.
Unexpectedly, improvements in safe water were associated with increases in rates of stunting. A possible explanation is that safe water is a proxy variable for a healthy environment. Once a relatively healthy environment is reached, any further improvements in environment would contribute negligibly to child health if other underlying factors such as maternal education remained stagnant. This conjecture is supported by the fact that countries that started out with high rates of safe water improved more than did countries that had low initial rates for safe water.
Of the social factors, the initial prevalence of HIV/AIDS had a negative relation with stunting. That it was not found to be significant in Model 8 suggests that HIV/AIDS acts primarily through the underlying factors because it decreases food security, decreases caring potential of the family, and limits availability of household resources (28). Moreover, these findings may be understating the larger role that the current HIV/AIDS epidemic plays in the increased absolute number of stunted children in sub-Saharan Africa.
Neither heterogeneity nor the level of democracy was found to be significant. Of the economic factors, an increase in government consumption was inversely related to improvements in stunting. There are at least 3 possible explanations for this finding. First, on a macro-level, privatization of social services, although curbing government spending, may have been associated with improved children’s well-being. This assumes that government consumption measured the amount of money the government spent on social services including health care and education. Second, if, however, this money was not used for social services, but rather a disproportionately large amount of it was used to pay government officials, then government consumption may reflect a diversion of funds away from the welfare of children. Third, a government with a high consumption rate must impose a high tax on households, thus possibly reducing a household’s available income for food purchases, heath care, and education (29).
Countries that devoted more resources to agriculture tended to have children that were worse off, even after accounting for underlying factors and increases in per capita GNP. This suggests that, other things being equal, countries with greater resources in agriculture did not improve stunting rates as much as countries with a more diversified or developed economy.
Not surprisingly, inequitable income distribution was also associated with less improvement in stunting. Income inequality tends to be related to social conflict, political instability, a lower level of democracy, and a higher probability of revolution, and also tends to coexist with underinvestment in human capital, which translates into lower long-run economic growth (27). All of these factors would negatively affect child well-being.
This study found that countries with a higher initial GNP and with increases in per capita GNP had greater improvement in stunting, consistent with findings from other studies (12,23–25,30,31). Per capita GNP became insignificant, however, when income distribution was considered, supporting the idea that rapid improvement in nutrition will not necessarily be a direct result of increases in economic growth. That is, when income distribution is very unequal, benefits in increased national income may not reach the malnourished (28).
A strength of this study is the quality of the data on changes in stunting. These data undergo a thorough process of quality control before their inclusion in the WHO database (21). Also, the study focused on the prevalence of stunting as a measure of long-term growth faltering resulting from chronic undernutrition, a better measure of changes over time than the prevalence of underweight (8,32). Another strength is that this study assessed factors that were not previously analyzed quantitatively (12,25,26) including HIV/AIDS, conflict, civil liberties, political rights, external debt, degree of ethnic/religious homogeneity within a country, degree of the GNP devoted to agriculture, official development assistance, and government consumption. There were limitations in some of these data, however. Many countries did not have complete statistics on national factors. Although a strong attempt was made to make use of reliable, nationally representative data, this was not always possible. The measures used were limited by data availability and were not necessarily ideal. For example, although female literacy and education have been used to measure maternal and child care, other more specific measures of child caring practices have not been available (33). Conflict was not associated with changes in stunting, presumably because of an inadequate measure. Also, in some cases, data on the same variable came from different agencies, each of which may have had its own quality standards and measurement criteria. In addition, each country might have set its own criteria in defining a variable, such as urbanization. Finally, because most conceptualized factors were not fully captured by the available variables, it is uncertain whether these variables are the most appropriate to change through specific interventions to prevent stunting.
Although data on stunting were available at subnational levels for most countries in the WHO database (5,21), this study was done at the national level because data on all of the independent variables were not available at subnational levels. These results may or may not be transferable to the subnational level (12) because characteristics of provinces and districts, for example, may vary considerably from the broad national characteristics assessed in the study, especially for large heterogeneous countries. The findings of Pelletier and Frongillo (5), however, were consistent at both levels.
This study provides information relevant for determining courses of action to be taken at the macrolevel to improve child health. Our analysis identified key national underlying and basic factors that have been most related to the reduction of stunting rates observed in the past 20–30 y. Improvements in stunting were related to differences in the underlying factors of food security, maternal and child care, health services and environment of the child, and also to differences in basic social and economic factors. Although the specific factors that were most associated with improvements in stunting in the past may not necessarily be the ones that are important in the future, one possible implication of these results is that substantial progress can best be made in reducing the current high prevalence of stunting by investing in both long-term development and in specific interventions. Such interventions, when aimed at factors amenable to change, should achieve important results in the short- to mid-term, but this requires further investigation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Manuscript received 20 June 2004. Initial review completed 3 August 2004. Revision accepted 1 March 2005.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Grantham-McGregor, S. M. & Fernand, L. C. (1997) Commission of Nutritional Challenges of the 21st Century 1997 United Nations ACC/SCN Geneva, Switzerland.
2. United Nations Children’s Fund (1998) The State of the World’s Children 1998 UNICEF New York, NY.
3. Jukes, M., McGuire, J., Method, F. & Sternberg, R. (2002) Nutrition and Education. Nutrition: A Foundation for Development 2002 ACC/SCN, Geneva Switzerland.
4. United Nations, Administrative Committee on Coordination, Sub-Committee on Nutrition (2000) Fourth Report on the World Nutrition Situation 2000 ACC/SCN in collaboration with IFPRI Geneva, Switzerland.
5. Pelletier, D. L. & Frongillo, E. A., Jr (2003) Changes in child survival are strongly associated with changes in malnutrition in developing countries. J. Nutr. 133:2302-2309.
6. de Onis, M., Blössner, M., Borghi, E., Morris, R. & Frongillo, E. A. (2004) Methodology for estimating regional and global trends of child malnutrition. Int. J. Epidemiol. 33:1260-1270.
7. United Nation’s Children’s Fund (1990) Strategy for Improved Nutrition of Children and Women in Developing Countries. Policy Review Paper 1990 UNICEF New York, NY.
8. Food and Agriculture Organization of the United Nations (2001) The State of Food Insecurity in the World 2001 FAO Rome, Italy.
9. Food and Agriculture Organization of the United Nations (2001) FAOSTAT: FAO Statistical Database, Rome, Italy 2001.
10. Engle, P., Menon, P. & Haddad, L. (1999) Care and nutrition: concepts and measurements. World Dev. 27:1309-1337.
11. Oniang’o, R. & Mukudi, E. (2002) Nutrition and Gender. Nutrition: A Foundation for Development 2002 ACC/SCN Geneva, Switzerland.
12. Smith, L. C. & Haddad, L. (1999) Exploring Child Malnutrition in Developing Countries: A Cross-Country Analysis. Research Report 111 1999 IFPRI Washington, DC.
13. Annett, A. (2001) Social fractionalization, political instability, and the size of government. IMF Staff Papers 48: 561–592 2001 International Monetary Fund Washington, DC.
14. Gillespie, S, Haddad, L. & Jackson, R. (2001) HIV/AIDS, Food and Nutrition Security: Impacts and Actions. Nutrition Policy Paper No. 20 2001 ACC/SCN Geneva, Switzerland.
15. UNAIDS (2002) Epi fact sheets by country—2002 updates 2002 Accessed at www.unaids.org/hivaidsinfo/statistics/fact_sheets/index_en.htm on June 25, 2003.
16. United Nations, Administrative Committee on Coordination, Sub-Committee on Nutrition (1996) How Nutrition Improves 1996 ACC/SCN Geneva, Switzerland.
17. Freedom House (2002) Freedom House Country Ratings 2002 Accessed at www.freedomhouse.org/ratings/index.htm on June 25, 2003.
18. World Bank (1997) World Development Report 1997 Oxford University Press New York, NY.
19. World Bank (2000/1) World Development Report 2000/1 Oxford University Press New York, NY.
20. Messer, E., Cohen, M. J. & D’Costa, J. (1998) Food from Peace: Breaking the Links Between Conflict and Hunger. 2020 Vision Discussion Paper 24 1998 IFPRI Washington, DC.
21. de Onis, M. & Blössner, M. (2003) The World Health Organization Global Database on Child Growth and Malnutrition: methodology and applications. Int. J. Epidemiol. 32:518-526.
22. Arbuckle, J. L. (1996) Full information estimation in the presence of incomplete data. Marcoulides, G. A. Schumacker, R. E. eds. Advanced Structural Equation Modeling: Issues and Techniques 1996 Lawrence Erlbaum Associates Mahwah, NJ. .
23. United Nations, Administrative Committee on Coordination, Sub-Committee on Nutrition (1993) Second Report on the World Nutrition Situation. Volume II 1993 ACC/SCN Geneva, Switzerland.
24. Frongillo, E. A. & Hanson, K.M.P. (1995) Determinants of variability among nations in child growth. Ann. Hum. Biol. 22:395-411.[Medline]
25. Frongillo, E. A., de Onis, M. & Hanson, K.M.P. (1997) Socioeconomic and demographic factors are associated with worldwide patterns of stunting and wasting of children. J. Nutr. 127:2302-2309.
26. Smith, L. C. & Haddad, L. (2001) How important is food availability for reducing child malnutrition in developing countries?. Agric. Econ. 26:191-204.
27. Thorbecke, E. & Charumilind, C. (2002) Economic inequality and its socioeconomic impact. World Dev. 30:1477-1495.
28. Commission on the Nutrition Challenges of the 21st Century (2000) Ending Malnutrition by 2020: An Agenda for Change in the Millennium. Final Report to the ACC/SCN 2000 Accessed at www.unsystem.org/SCN/publications/UN_report.pdf on June 25, 2003.
29. Marsden, K. (1993) Employment strategy in Papua New Guinea: an international perspective. Int. Lab. Rev. 132:226-245.
30. United Nations, Administrative Committee on Coordination, Sub-Committee on Nutrition (1994) Update on the Nutrition Situation, 1994 1994 ACC/SCN Geneva, Switzerland.
31. Haddad, L., Alderman, H., Appleton, S., Song, L. & Yohannes, Y. (2003) Reducing child malnutrition: how far does income growth take us?. World Bank Econ. Rev. 17:107-131.[Abstract]
32. United Nations, Administrative Committee on Coordination, Sub-Committee on Nutrition (1997) Third Report on the World Nutrition Situation 1997 ACC/SCN Geneva, Switzerland.
33. Frongillo, E. A., Sywulka, S. M. & Kariger, P. (2003) UNICEF Psychosocial Care Indicators Project. Final Report to UNICEF. Division of Nutritional Sciences, December 31, 2003 2003 Cornell University Ithaca, NY.
34. World Health Organization (2003) The WHO Global Database on Child Growth and Malnutrition 2003 Accessed at www.who.int/nutgrowthdb/ on February 27, 2003.
35. Food and Agriculture Organization of the United Nations (2002) The State of Food Insecurity in the World 2002 FAO Rome, Italy.
36. UNDP (United Nations Development Programme) (1990) Human Development Report 1990 Oxford University Press New York, NY.
37. UNDP (United Nations Development Programme) (1991) Human Development Report 1991 Oxford University Press New York, NY.
38. UNDP (United Nations Development Programme) (1993) Human Development Report 1993 Oxford University Press New York, NY.
39. UNDP (United Nations Development Programme) (1995) Human Development Report 1995 Oxford University Press New York, NY.
40. UNDP (United Nations Development Programme) (1998) Human Development Report 1998 Oxford University Press New York, NY.
41. UNDP (United Nations Development Programme) (1999) Human Development Report 1999 Oxford University Press New York, NY.
42. UNDP (United Nations Development Programme) (2001) Human Development Report 2001 Oxford University Press New York, NY.
43. UNDP (United Nations Development Programme) (2002) Human Development Report 2002 Oxford University Press New York, NY.
44. United Nations Children’s Fund (1984–2003) The State of the World’s Children 1984–2003 UNICEF New York, NY.
45. World Bank/The International Bank for Reconstruction and Development (1999) World Development Indicators 1999 Washington, DC.
46. World Bank/The International Bank for Reconstruction and Development (2001) World Development Indicators 2001 Washington, DC.
47. Country Indicators for Foreign Policy 2003 Accessed at www.carleton.ca/cgi-bin/cifp/ on June 25, 2003.
48. World Bank (1994) World Tables 1994 Washington, DC.
49. World Bank (1979) World Development Report 1979 Oxford University Press New York, NY.
50. World Bank (1986) World Development Report 1986 Oxford University Press New York, NY.
51. World Bank (1993) World Development Report 1993 Oxford University Press New York, NY.
52. Marshall, M. G. & Jaggers, K. (2001) Polity IV Country Reports. Center for International Development and Conflict Management. Integrated Network for Societal Conflict Research 2001 Accessed at www.cidcm.umd.edu/inscr/polity/ on June 25, 2003.
53. Marshall, M. G., Gurr, T. R. & Harff, B. (2001) State Failure Problem Set: Internal Wars and Failures of Government, 1955–2001. Integrated Network for Societal Conflict Research and Center for International Development and Conflict Management 2001 University of Maryland Accessed at www.cidcm.umd.edu/inscr/stfail/sfdata.htm on June 25, 2003.
This article has been cited by other articles:
|
|
 |
|
  S. Rasheed, E. A. Frongillo, C. M. Devine, D. S. Alam, and K. M. Rasmussen Maternal, Infant, and Household Factors Are Associated with Breast-Feeding Trajectories during Infants' First 6 Months of Life in Matlab, Bangladesh J. Nutr., August 1, 2009; 139(8): 1582 - 1587. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. T. Mosby, R. D. Barr, and P. B. Pencharz Nutritional Assessment of Children With Cancer Journal of Pediatric Oncology Nursing, July 1, 2009; 26(4): 186 - 197. [Abstract] [PDF]
|
|
|
|
 |
|
 J. C. Jenkins, S. J. Scanlan, and L. Peterson Military Famine, Human Rights, and Child Hunger: A Cross-National Analysis, 1990-2000 Journal of Conflict Resolution, December 1, 2007; 51(6): 823 - 847. [Abstract] [PDF]
|
|
|
|
 |
|
 P. Svedberg Declining child malnutrition: a reassessment Int. J. Epidemiol., October 1, 2006; 35(5): 1336 - 1346. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
