Hostname: page-component-7c8c6479df-hgkh8 Total loading time: 0 Render date: 2024-03-27T21:10:46.029Z Has data issue: false hasContentIssue false

Parasite and maternal risk factors for malnutrition in preschool-age children in Belen, Peru using the new WHO Child Growth Standards

Published online by Cambridge University Press:  01 December 2007

Martin Casapía
Affiliation:
Asociación Civil Selva Amazónica, Urbanización Jardin 27, Iquitos, Peru
Serene A. Joseph
Affiliation:
Division of Clinical Epidemiology, McGill University Health Centre, Royal Victoria Hospital, V Building, 687 Pine Avenue West, Montréal, QuébecH3A 1A1, Canada
Carmen Núñez
Affiliation:
Asociación Civil Selva Amazónica, Urbanización Jardin 27, Iquitos, Peru
Elham Rahme
Affiliation:
Division of Clinical Epidemiology, McGill University Health Centre, Royal Victoria Hospital, V Building, 687 Pine Avenue West, Montréal, QuébecH3A 1A1, Canada Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Purvis Hall, 1020 Pine Avenue West, Montréal, Québec, H3A 1A2, Canada
Theresa W. Gyorkos*
Affiliation:
Division of Clinical Epidemiology, McGill University Health Centre, Royal Victoria Hospital, V Building, 687 Pine Avenue West, Montréal, QuébecH3A 1A1, Canada Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Purvis Hall, 1020 Pine Avenue West, Montréal, Québec, H3A 1A2, Canada
*
*Corresponding author: Dr Theresa W. Gyorkos, Division of Clinical Epidemiology, McGill University Health Centre, fax +1 514 934 8293, email theresa.gyorkos@mcgill.ca
Rights & Permissions [Opens in a new window]

Abstract

Child malnutrition, including wasting, underweight and stunting, is associated with infections, poor nutrient intake, and environmental and socio-demographic factors. Preschool-age children are especially vulnerable due to their high growth requirements. To target interventions for preschool-age children in a community of extreme poverty in Peru, we conducted a household survey between October 2005 and January 2006 to determine the prevalence of malnutrition and its risk factors. Of 252 children < 5 years old, the prevalence of wasting, underweight and stunting was 26·6, 28·6 and 32·1 %, respectively, based on the new WHO Child Growth Standards. Risk factors for wasting were: (1) moderate–high intensity Trichuris infection (OR 2·50; 95 % CI 1·06, 5·93); (2) hookworm infection (OR 6·67; 95 % CI 1·08, 41·05); (3) age (OR6-month 1·27; 95 % CI 1·11, 1·46); (4) maternal education (secondary incomplete) (OR 5·77; 95 % CI 2·38, 13·99); and (5) decreasing maternal BMI (OR1 kg/m2 1·12; 95 % CI 1·02, 1·23). Risk factors for underweight were: (1) moderate–high intensity Trichuris infection (OR 4·74; 95 % CI 1·99, 11·32); (2) age (OR6-month 1·22; 95 % CI 1·07, 1·38); (3) maternal education (secondary incomplete) (OR 2·92; 95 % CI 1·40, 6·12); and (4) decreasing maternal BMI (OR1 kg/m2 1·11; 95 % CI 1·02, 1·21). Risk factors for stunting were: (1) age (OR6-month 1·14; 95 % CI 1·02, 1·27) and (2) decreasing maternal height (OR1 cm 1·12; 95 % CI 1·06, 1·20). Overall, risk factors for malnutrition included both child and maternal determinants. Based on these data, locally appropriate and cost-effective dietary, de-worming and educational programmes should be targeted to mothers and preschool-age children.

Type
Full Papers
Copyright
Copyright © The Authors 2007

Child malnutrition is a serious concern in developing countries due to the high mortality and morbidity with which it is associated. Preschool-age children under the age of 5 are particularly susceptible to the adverse effects of malnutrition as they are in a vulnerable growth period with high growth requirementsReference Blossner and de Onis1. It is estimated that 54 % of deaths in children under 5 are associated with malnutritionReference Gordon, Mackay and Rehfuess2, and that approximately 5 million children die each year due to causes that are directly or indirectly related to malnutritionReference Gross and Webb3. Morbidity related to malnutrition can occur in infancy and childhood, as well as in adulthood, and includes adverse effects on health, cognition and behaviour4Reference Grantham-McGregor and Baker-Henningham7.

Indicators of malnutrition include wasting, stunting and underweight, which represent different aspects or measurements of malnutrition8. Wasting as measured by low weight-for-height represents lower than expected body mass (tissue or fat) and is a good indicator of existing nutritional deficits. Stunting, or low height-for-age is an indicator of a reduced linear growth rate and represents a chronic state of malnutrition as it takes longer for impaired skeletal growth to become apparent. Underweight, which is measured by low weight-for-age, represents a combination of both wasting and stunting. A review of the literature lists infections, nutrition, child and family demographics, and household characteristics as risk factors for malnutrition in preschool-age childrenReference Hernandez-Diaz, Peterson, Dixit, Hernandez, Parra, Barquera, Sepulveda and Rivera9Reference Sakisaka, Wakai, Kuroiwa, Cuadra, Kai, Mercedes and Hanada14.

The reduction in the prevalence of underweight in children under the age of 5 is one of the indicators of the first Millennium Development Goal target which seeks to eradicate extreme poverty and hunger15. Although improvements have been made in reducing the worldwide prevalence of malnutrition, it is clear that concentrated efforts will be needed to reach the Millennium Development GoalReference de Onis, Blossner, Borghi, Morris and Frongillo16, especially in areas of extreme poverty where malnutrition is highest. The identification of populations at risk has recently been improved due to the newly released WHO Child Growth Standards by which accurate estimates of indicators of malnutrition can be calculated8. This new standard is based on a sample of children from six countries (Brazil, Ghana, India, Norway, Oman and the USA) and represents how children between 0 and 5 years of age should grow under healthy conditions. The US National Center for Health Statistics (NCHS) international growth reference was previously used as a comparison; however due to methodological issues (e.g. homogeneous sample of children from the USA only, insufficient frequency of measurements) and the fact that it is a reference and not a standard (i.e. represents how children grow, not how children should grow) its generalizability is limitedReference Garza and de Onis17.

The objective of the present study was to determine the prevalence of three indicators of malnutrition (wasting, underweight and stunting) using the new WHO Child Growth Standards and associated risk factors in a population of preschool-age children living in an area of extreme poverty in the Peruvian Amazon.

Methods

Survey area

A household survey was conducted in the community of Belen from October 2005 to January 2006. Belen is located outside of the capital city of Iquitos in the province of Loreto in the Peruvian Amazon. Due to the seasonal flooding of the Rio Itaya, most of the houses are built on stilts or floating on wooden platforms whereby access to potable water and proper sanitation is a challenge. Belen is geographically divided into zone ‘alta’, or low flooding area, and zone ‘baja’, or high flooding area. These zones also serve to divide the community into higher and lower socioeconomic status, respectively.

Survey planning

The area boundaries within Belen with associated estimates of size (number of households) were provided by the Instituto Nacional de Estadística e Informática in Iquitos which had completed a census of Belen in August 2005. The census information delineated groupings of household blocks (called ‘manzanas’) within zones ‘alta’ and ‘baja’. A strategy for choosing and sampling households was based on probability proportional to size with the household as the unit of observation and proceeded as follows: (1) blocks were listed according to zone ‘alta’ and zone ‘baja’; (2) blocks were grouped together geographically within ‘alta’ and ‘baja’ to form clusters of approximately the same size; (3) clusters were arranged in a list in geographical order and numbered; (4) based on the average number of households per cluster and the desired sample size, clusters to be surveyed were randomly chosen; (5) all households within each cluster were then visited.

Questionnaire

The questionnaire was developed in English using the Multiple Indicator Cluster Survey 2 to assist in the formulation of survey questions and then translated into Spanish. Questions were chosen for their relevance in obtaining socio-demographic indicators as well as other variables found to be important determinants of malnutrition in previous research.

Field survey

Field teams were provided with a map of each cluster to be surveyed. Each household was visited on two consecutive days, with interviews being conducted with the female head of the household. During the first visit, households were informed of the purpose of the survey, informed consent for the participation of the household was obtained, and a container for stool specimens was given. On the second visit, the questionnaire was completed. Height and weight measurements and blood and stool samples were also collected from the interviewee and one child under 5. If two or more children under the age of 5 were present in the household, one was randomly chosen for inclusion in the study. The weight of mothers and of children able to stand alone, in light clothing and without shoes, on a digital portable scale, was measured in pounds, and converted into kilograms. For children unable to stand alone, the difference between the mother's weight and that of the mother and the child, on the digital scale, was similarly ascertained. Height, without shoes, was measured to the nearest millimetre using a measuring stick against a wall. Length for infants under 2 years of age was measured on a flat surface using a measuring tape, to the nearest millimetre. Blood samples were collected using fingerprick blood. The second drop of fingerprick blood was drawn into a microcuvette for Hb determination (g/dl) using a HemoCue® machine (HemoCue Inc., CA, USA). Fingerprick blood was also tested for malaria infection using the Giemsa stained thick and thin smear technique. The Kato-Katz technique was used to determine the presence and intensity of intestinal helminth eggs in the stool specimens.

Statistical analyses

Gender-specific anthropometric z-scores for weight-for-height, weight-for-age and height-for-age were calculated using the new WHO Child Growth Standards18. Moderate to severe wasting, underweight and stunting were defined as z-scores < − 2 sd from the WHO Standards for weight-for-height, weight-for-age and height-for-age, respectively. Estimates of severe wasting, underweight and stunting were calculated using z-scores < − 3 sd from the standards. Children who had potentially incorrect or out-of-range z-scores based on height or weight measurements were flagged by the software and excluded from analyses. For comparison, anthropometric z-scores were also calculated using the NCHS international growth reference18. Anaemia was defined as a Hb value < 11·0 g/dl for children and < 12·0 g/dl for mothers19.

Means and frequency procedures were used to describe the individual and household characteristics of the study population. Logistic regression was used to determine risk factors for moderate to severe wasting, underweight and stunting, adjusting for potential confounders. All potential variables to be included in model building were first examined for correlations with one another. Although socioeconomic and other household information was gathered, there was a high correlation of these variables with one another and with more proximate indicators of malnutrition such as infection prevalence and other individual and maternal characteristics. The variables chosen for inclusion in model building were those that better explained the outcome in univariate analysis, or those previously determined to be more relevant from the literature. Univariate logistic regression was used to determine potential variables to be included in multivariate analyses at P < 0·20. Variables with P < 0·05 in the final multivariate logistic regression model were considered independent risk factors for moderate to severe wasting, underweight and stunting. Models were re-run using the generalized estimating equation to take into account the effect of clustering. All statistical analyses were performed using the Statistical Analysis Systems statistical software package version 9.1 (SAS Institute, Cary, NC, USA).

Ethics approval

Ethics approval was obtained from the Ethics Review Board of the Research Institute of the McGill University Health Centre in Montreal, Canada, and the Ministerio de Salud in Iquitos, Peru.

Results

Final population sample

Households were visited between October 2005 and January 2006. Zone ‘baja’ was surveyed first. It was also imperative to survey zone ‘baja’ before the rainy season (December to April) due to the high potential for flooding in this area. In zone ‘baja’, there were a total of thirty-eight clusters of households. Anticipating an average of twenty households per cluster and a total desired sample size of 220 households, a random sample of eleven clusters was selected. All eleven clusters (100 %) were sampled, with 213 households (96·8 %) participating in the survey. After completing the surveys in zone ‘baja’, surveys were then conducted in households in zone ‘alta’. As it was apparent that there was not enough time to survey all of the anticipated households in zone ‘alta’, a decision was made to do systematic sampling of households based on geographical order until the available time and financial resources had been expended. In zone ‘alta’, there was a total of eighty-six clusters of households. Anticipating an average of thirty households per cluster and a total desired sample size of 780 households, a random sample of twenty-six clusters was selected. A total of fifteen clusters (58 %) were sampled, with 274 households (35·1 %) participating in the survey.

Of 487 participating households in zone ‘baja’ and ‘alta’, a child under 5 was present in 288 of the households. Thirty-six children were excluded from subsequent analyses for the following reasons: missing anthropometric and Hb information (n 1), no stool specimen obtained (n 21) and flagged anthropometric values (n 14). Therefore, our final sample size of children with complete data was 252 children.

Child characteristics

The mean age of the children was 27·5 (sd 16·2) months with a range between 0·62 and 59·6 months; 49 % were female; 12 % (n 30) attended preschool, all of whom were over the age of 36 months. The prevalence of anaemia was 56·0 %, and 29·4 % had had diarrhoea in the last week. For those over the age of 12 months, the recommended age by which children should be vaccinated against measles, the vaccination rate was 91·2 %. The prevalence of hookworm, Trichuris and Ascaris infections was 3·2, 38·9 and 32·1 %, respectively; 48 % of the children were infected with at least one infection of hookworm, Trichuris or Ascaris. Only one child was positive for malaria (Plasmodium vivax).

Maternal and household characteristics

The mean age of mothers who had children included in the present study was 28·7 (sd 7·7) years; 38 % had completed secondary education and 42·1 % were employed outside of the home. The mean BMI of mothers was 24·8 (sd 4·4) kg/m2, which is considered in the normal range. For their last birth: 78·5 % had given birth in a hospital or health centre; the mean duration of breastfeeding (exclusive or predominant) was 15·3 (sd 7·8) months (for the 147 women not currently breastfeeding); 78·4 % had a doctor or trained obstetrical nurse to assist in delivery; and 84·1 % attended the recommended six or more prenatal visits. The characteristics of mothers and households in relation to wasting, underweight and stunting are given in Table 1.

Table 1 Prevalence of wasting, underweight and stunting with respect to household and maternal characteristics of 252 preschool-age children, Belen, Peru, October 2005 to January 2006

* Totals do not sum to 252 due to missing answers on the questionnaire.

Anaemia: < 12 g Hb/dl = yes; ≥ 12 g Hb/dl = no.

Underweight = BMI < 18·5; normal = BMI 18·5–24·9; overweight/obese = BMI ≥ 25·0.

Risk factors for malnutrition

It was found that 50 % of the children suffered from either wasting, underweight or stunting, and 27·4 % had at least two of these indicators of malnutrition (Table 2). The prevalence of moderate to severe wasting, underweight and stunting in the population of preschool-age children was 26·6, 28·6 and 32·1 %, respectively. The prevalence of severe wasting, underweight and stunting in children was 17·9, 18·3 and 11·1 %, respectively.

Table 2 Comorbidity of wasting, underweight and stunting in a population of 252 preschool-age children in Belen, Peru, October 2005 to January 2006

* Wasting, underweight and stunting are defined as weight-for-height, weight-for-age and height-for-age < − 2 sd from WHO Child Growth Standards18.

Risk factors for moderate to severe wasting were: (1) moderate–high intensity of Trichuris infection (OR 2·50; 95 % CI 1·06, 5·93); (2) hookworm infection (OR 6·67; 95 % CI 1·08, 41·05); (3) increasing age of child (OR6-month 1·27; 95 % CI 1·11, 1·46); (4) maternal education level (secondary incomplete v. secondary complete) (OR 5·77; 95 % CI 2·38, 13·99); and (5) decreasing maternal BMI ( 1·12; 95 % CI 1·02, 1·23) (Table 3).

Table 3 Risk factors for wasting, underweight and stunting in a population of 252 preschool-age children in Belen, Peru, October 2005 to January 2006

* All variables with a P < 0·20 were included in the multivariable model.

All models control for child gender (male or female) and maternal age.

Wasting, underweight and stunting are defined as weight-for-height, weight-for-age and height-for-age < − 2 sd from WHO Child Growth Standards18.

Risk factors for moderate to severe underweight were: (1) moderate–high intensity of Trichuris infection (OR 4·74; 95 % CI 1·99, 11·32); (2) increasing age of child (OR6-month 1·22; 95 % CI 1·07, 1·38); (3) maternal education level (secondary incomplete v. secondary complete) (OR 2·92; 95 % CI 1·40, 6·12); and (4) decreasing maternal BMI ( 1·11; 95 % CI 1·02, 1·21).

Risk factors for moderate to severe stunting were: (1) increasing age of child (OR6-month 1·14; 95 % CI 1·02, 1·27) and (2) decreasing maternal height (OR1 cm 1·12; 95 % CI 1·06, 1·20).

There were negligible differences in risk factors for wasting, underweight and stunting when using the generalized estimating equation method to take clustering into account (results not shown).

Comparison between National Center for Health Statistics and WHO estimates

Using the NCHS reference to calculate the prevalence of malnutrition in the present study, the estimates of moderate to severe wasting and underweight (25·4 and 32·5 %, respectively) were similar to those obtained using the new WHO Child Growth Standards (26·6 and 28·6 %, respectively), while a lower prevalence was found for moderate to severe stunting (23·8 and 32·1 %, respectively). For the ‘severe’ categories, the NCHS and WHO estimates were similar for the category of severe underweight (17·9 and 18·3 %, respectively), but the WHO estimates were higher for severe stunting (11·1 and 7·1 %, respectively) and for severe wasting (17·9 and 11·9 %, respectively).

Discussion

Although progress has been made in reducing malnutrition on a national basis in Peru20, it is clear that not all areas have benefited from this decline. The prevalence of moderate to severe underweight in the present study population was 28·6 % which is almost three times higher than the national average in Peru from 200020. The prevalence of severe underweight was more than 20 times higher than the reported national prevalence of 0·8 %. These numbers show the importance of targeting high-risk areas, such as Belen, which are not accurately reflected in national statistics. Compared to other studies of preschool-age children in Latin American countries (Table 4)Reference Hernandez-Diaz, Peterson, Dixit, Hernandez, Parra, Barquera, Sepulveda and Rivera9Reference Marins and Almeida11, Reference Sakisaka, Wakai, Kuroiwa, Cuadra, Kai, Mercedes and Hanada14, Reference Ferrari, Solymos, Castillo and Sigulem21, Reference Bronte-Tinkew and DeJong22, our estimates of malnutrition, particularly underweight and wasting, are considerably higher. Previous studies, as well as the national statistics from Peru, used the NCHS international growth reference to estimate the prevalence of malnutrition which limits direct comparability with the present results. However, regardless of the reference population chosen, the prevalences of both moderate to severe and severe malnutrition are exceptionally high in the present study population.

Table 4 Prevalence of wasting, underweight and stunting in previous cross-sectional studies in preschool-age children in Latin America and the Caribbean*

* All prevalences of malnutrition are based on the US National Center for Health Statistics reference population.

Blank cells indicate outcomes that were not measured.

The present data also show that Trichuris and hookworm infections are important risk factors for underweight and wasting in children under 5. This is consistent with previous researchReference Muniz, Ferreira, Ferreira, Conde and Monteiro23. Mechanisms by which helminth infection can lead to malnutrition in children include decreased appetite and food intake, depletion and impaired absorption of micronutrients, and anaemiaReference Crompton and Nesheim24. In the short term, de-worming with a recommended broad-spectrum anthelminthic, such as single-dose albendazole or mebendazole, is necessary to cure or decrease the worm burden in preschool-age children. Previous studies have shown growth improvements in children of this age group after anthelminthic treatmentReference Awasthi, Pande and Fletcher25, Reference Stoltzfus, Chway, Montresor, Tielsch, Jape, Albonico and Savioli26. School-based de-worming programmes are recommended by WHO in endemic areas where helminth prevalence and intensity exceed certain thresholds27. The present data lend support for similar interventions at the preschool-age level.

The increasing risk of malnutrition with age shows that malnourished children continue to deviate from expected growth standards as they get older. Previous studies demonstrated that preschool-age children are in a vulnerable growth periodReference Kwena, Terlouw and De Vlas6, Reference Hernandez-Diaz, Peterson, Dixit, Hernandez, Parra, Barquera, Sepulveda and Rivera9, Reference Marins and Almeida11, Reference Bloss, Wainaina and Bailey12, Reference Espo, Kulmala, Maleta, Cullinan, Salin and Ashorn28Reference Maleta, Virtanen, Espo, Kulmala and Ashorn30 and that interventions are needed to improve their nutritional state. With the increasing risk of malnutrition with age, these children should be targeted before entering school so that they do not face the longer-term educational, behavioural and physical consequences of adverse health in school age.

Maternal factors, such as educational achievement, BMI and height were also found to be important risk factors for wasting, underweight and stunting in children under 5 years of age. This highlights the importance of targeting families in addition to individual children in order to have a greater reduction in child malnutrition. Maternal education has been previously found to be an important risk factor for child malnutritionReference Hernandez-Diaz, Peterson, Dixit, Hernandez, Parra, Barquera, Sepulveda and Rivera9Reference Marins and Almeida11, Reference Sakisaka, Wakai, Kuroiwa, Cuadra, Kai, Mercedes and Hanada14, Reference Shah, Selwyn, Luby, Merchant and Bano31. There are various direct and indirect ways by which maternal education can affect malnutrition in children. Mothers who are more educated may have better knowledge of proper health and nutrition behaviours for their family, including breastfeedingReference Wachs, Creed-Kanashiro, Cueto and Jacoby32. They may also be more likely to attend prenatal care and therefore be exposed to healthy pre- and post-pregnancy behaviours for themselves and their newborns. This would be an appropriate time-point for the health services to provide targeted information on healthy nutrition for both mother and child, which could be especially important for mothers with a lower level of formal education. In addition to educational level, maternal BMI and height were found to be risk factors for wasting, underweight and stunting. This is consistent with the mechanism of malnutrition in which mothers who are malnourished are more likely to have children of low birth weight, who are themselves more likely to suffer from wasting, underweight and stunting in childhood and adulthood. This finding also corroborates previous research in this areaReference Hernandez-Diaz, Peterson, Dixit, Hernandez, Parra, Barquera, Sepulveda and Rivera9.

Strengths of the present study include: (1) the use of a recent census of Belen for sampling methodology; (2) the collection of a wide range of potential confounding variables such as socio-demographic information and other family and maternal characteristics; (3) the use of the new WHO Child Growth Standards18 to give accurate estimates of the prevalence of malnutrition in this population; and (4) calculation of risk factors for wasting which has been difficult in previous studies due to low sample sizes. Limitations include: (1) the use of the survey design which limits the determination of the direction of cause and effect of the outcome and potential risk factors; (2) not having information on birth weight which has been shown to be a strong predictor of growth in childhoodReference Marins and Almeida11, Reference Maleta, Virtanen, Espo, Kulmala and Ashorn30; and (3) not surveying all of the clusters in zone ‘alta’ which may have limited the representativeness of the final study population.

Overall, the results show that both individual and family characteristics are related to the occurrence of malnutrition in preschool-age children. The literature highlights the potential for reducing or reversing malnutrition through individual and macro-level interventionsReference Stoltzfus, Chway, Montresor, Tielsch, Jape, Albonico and Savioli26, Reference Penny, Creed-Kanashiro, Robert, Narro, Caulfield and Black33Reference Alderman, Hoogeveen and Rossi36. In the short term, health and nutrition educational interventions targeted to mothers and children, and de-worming and dietary programmes for preschool-age children are needed. As these children would not yet be attending primary school and most do not attend preschool, it is important to target these children and their families through alternate means. For example, interventions for mothers could occur during prenatal and postnatal care to ensure that they are properly educated about good health and nutrition behaviours for themselves and their families. A combination of interventions may be most effectiveReference Alderman, Hoogeveen and Rossi36. It should be noted, however, that nutrient intake was not rigorously assessed in the present study and that future research should include a more comprehensive analysis of nutritional risk factors for wasting, underweight and stunting.

If the Millennium Development Goals are to be reached, extra efforts are needed to target vulnerable populations, such as those in areas of poverty or rural areas, or who otherwise do not have easy access to existing interventions or health services. Otherwise, continued health inequalities within countries will persist and high-risk individuals, even in countries with favourable Millennium Development Goal indicators, will be neglected. Long-term and sustainable solutions for combating child malnutrition require a strong commitment from government, non-governmental organizations, community groups and families. Future research should determine the cost-effectiveness of both short- and long-term interventions for child malnutrition, especially in vulnerable populations such as those living in areas of extreme poverty.

Conclusions

The community of Belen has long been neglected in terms of research attention and programmes to improve health and reduce poverty. It is clear that statistics of malnutrition at the national level in Peru are not representative of this community and likely other similar communities of extreme poverty. The present study provides a recent and accurate picture of the prevalence of wasting, underweight and stunting in preschool-age children in Belen and highlights the importance of both individual and family-based risk factors associated with malnutrition. This information can be used by the health authorities in the area to target appropriate health and education interventions, as a minimum, to pregnant women, new mothers and infants. The results of the present study will contribute not only to ensuring attention for the community of Belen, but also other communities of extreme poverty in Peru and elsewhere.

Acknowledgements

We are grateful to the Instituto Nacional de Estadística e Informática in Iquitos, Peru, for providing us with recent census information for Belen. We acknowledge the input provided by participants at the April 2005 multidisciplinary workshop in Iquitos, Peru and the research team of the Proyecto Canadá-Perú-Belén. This project was funded by the Canadian Institutes of Health Research (CIHR; grant number GLP 67 580).

References

1Blossner, M & de Onis, M (2005) Malnutrition: Quantifying the Health Impact at National and Local Levels. WHO Report no. 12. Geneva: World Health Organization.Google Scholar
2Gordon, B, Mackay, R & Rehfuess, E (2004) Inheriting the World: the Atlas of Children's Health and the Environment. Geneva: World Health Organization.Google Scholar
3Gross, R & Webb, P (2006) Wasting time for wasted children: severe child undernutrition must be resolved in non-emergency settings. Lancet 367, 12091211.CrossRefGoogle ScholarPubMed
4The Partnership for Child Development (1999) Short stature and the age of enrolment in primary school: studies in two African countries. Soc Sci Med 48, 675682.CrossRefGoogle Scholar
5Norgan, NG (2000) Long-term physiological and economic consequences of growth retardation in children and adolescents. Proc Nutr Soc 59, 245256.Google Scholar
6Kwena, AM, Terlouw, DJ, De Vlas, SJ, et al. (2003) Prevalence and severity of malnutrition in pre-school children in a rural area of western Kenya. Am J Trop Med Hyg 68, 9499.CrossRefGoogle Scholar
7Grantham-McGregor, S & Baker-Henningham, H (2005) Review of the evidence linking protein and energy to mental development. Public Health Nutr 8, 11911201.Google Scholar
8WHO Multicentre Growth Reference Study Group (2006) WHO Child Growth Standards based on length/height, weight and age. Acta Paediatr Suppl 450, 7685.Google Scholar
9Hernandez-Diaz, S, Peterson, KE, Dixit, S, Hernandez, B, Parra, S, Barquera, S, Sepulveda, J & Rivera, JA (1999) Association of maternal short stature with stunting in Mexican children: common genes vs common environment. Eur J Clin Nutr 53, 938945.Google Scholar
10Nestel, P, Melara, A, Rosado, J & Mora, JO (1999) Undernutrition among Honduran children 12–71 months old. Rev Panam Salud Publica 6, 256265.Google Scholar
11Marins, VM & Almeida, RM (2002) Undernutrition prevalence and social determinants in children aged 0-59 months, Niteroi, Brazil. Ann Hum Biol 29, 609618.Google Scholar
12Bloss, E, Wainaina, F & Bailey, RC (2004) Prevalence and predictors of underweight, stunting, and wasting among children aged 5 and under in western Kenya. J Trop Pediatr 50, 260270.Google Scholar
13Aerts, D, Drachler, ML & Giugliani, ER (2004) Determinants of growth retardation in Southern Brazil. Cad Saude Publica 20, 11821190.CrossRefGoogle ScholarPubMed
14Sakisaka, K, Wakai, S, Kuroiwa, C, Cuadra, FL, Kai, I, Mercedes, AM & Hanada, K (2006) Nutritional status and associated factors in children aged 0–23 months in Granada, Nicaragua. Public Health 120, 400411.Google Scholar
16de Onis, M, Blossner, M, Borghi, E, Morris, R & Frongillo, EA (2004) Methodology for estimating regional and global trends of child malnutrition. Int J Epidemiol 33, 12601270.Google Scholar
17Garza, C & de Onis, M (2004) Rationale for developing a new international growth reference. Food Nutr Bull 25, S5S14.CrossRefGoogle ScholarPubMed
18World Health Organization (2006) WHO Anthro 2005, Beta Version: Software for Assessing Growth and Development of the World's Children. Geneva: WHO.Google Scholar
19World Health Organization (2001) Iron Deficiency Anaemia: Assessment, Prevention and Control. A Guide for Programme Managers. WHO/NHD/01.3. Geneva: WHO.Google Scholar
20United Nations (2006) Millennium Development Goals Indicators. Country Level Data. http://millenniumindicators.un.org/unsd/mdg/Data.aspx (accessed November 2006).Google Scholar
21Ferrari, AA, Solymos, GM, Castillo, RM & Sigulem, DM (1998) Risk factors for protein-energy malnutrition in pre-school shantytown children in Sao Paulo, Brazil. Sao Paulo Med J 116, 16541660.CrossRefGoogle ScholarPubMed
22Bronte-Tinkew, J & DeJong, G (2004) Children's nutrition in Jamaica: do household structure and household economic resources matter? Soc Sci Med 58, 499514.CrossRefGoogle ScholarPubMed
23Muniz, PT, Ferreira, MU, Ferreira, CS, Conde, WL & Monteiro, CA (2002) Intestinal parasitic infections in young children in Sao Paulo, Brazil: prevalences, temporal trends and associations with physical growth. Ann Trop Med Parasitol 96, 503512.Google Scholar
24Crompton, DWT & Nesheim, MC (2002) Nutritional impact of intestinal helminthiasis during the human life cycle. Ann Rev Nutr 22, 3559.Google Scholar
25Awasthi, S, Pande, VK & Fletcher, RH (2000) Effectiveness and cost-effectiveness of albendazole in improving nutritional status of pre-school children in urban slums. Indian Pediatr 37, 1929.Google ScholarPubMed
26Stoltzfus, RJ, Chway, HM, Montresor, A, Tielsch, JM, Jape, JK, Albonico, M & Savioli, L (2004) Low dose daily iron supplementation improves iron status and appetite but not anemia, whereas quarterly anthelminthic treatment improves growth, appetite and anemia in Zanzibari preschool children. J Nutr 134, 348356.Google Scholar
27World Health Organization (1998) Guidelines for the Evaluation of Soil-transmitted Helminthiasis and Schistosomiasis at Community Level. A Guide for Managers of Control Programmes. WHO/CTC/SIP/98.1. Geneva: WHO.Google Scholar
28Espo, M, Kulmala, T, Maleta, K, Cullinan, T, Salin, ML & Ashorn, P (2002) Determinants of linear growth and predictors of severe stunting during infancy in rural Malawi. Acta Paediatr 91, 13641370.Google Scholar
29Dangour, AD, Hill, HL & Ismail, SJ (2002) Height, weight and haemoglobin status of 6 to 59-month-old Kazakh children living in Kzyl-Orda region, Kazakhstan. Eur J Clin Nutr 56, 10301038.Google Scholar
30Maleta, K, Virtanen, SM, Espo, M, Kulmala, T & Ashorn, P (2003) Childhood malnutrition and its predictors in rural Malawi. Paediatr Perinat Epidemiol 17, 384390.Google Scholar
31Shah, SM, Selwyn, BJ, Luby, S, Merchant, A & Bano, R (2003) Prevalence and correlates of stunting among children in rural Pakistan. Pediatr Int 45, 4953.Google Scholar
32Wachs, TD, Creed-Kanashiro, H, Cueto, S & Jacoby, E (2005) Maternal education and intelligence predict offspring diet and nutritional status. J Nutr 135, 21792186.Google Scholar
33Penny, ME, Creed-Kanashiro, HM, Robert, RC, Narro, MR, Caulfield, LE & Black, RE (2005) Effectiveness of an educational intervention delivered through the health services to improve nutrition in young children: a cluster-randomised controlled trial. Lancet 365, 18631872.CrossRefGoogle ScholarPubMed
34Milman, A, Frongillo, EA, de Onis, M & Hwang, JY (2005) Differential improvement among countries in child stunting is associated with long-term development and specific interventions. J Nutr 135, 14151422.Google Scholar
35Roy, SK, Fuchs, GJ, Mahmud, Z, Ara, G, Islam, S, Shafique, S, Akter, SS & Chakraborty, B (2005) Intensive nutrition education with or without supplementary feeding improves the nutritional status of moderately-malnourished children in Bangladesh. J Health Popul Nutr 23, 320330.Google ScholarPubMed
36Alderman, H, Hoogeveen, H & Rossi, M (2006) Reducing child malnutrition in Tanzania. Combined effects of income growth and program interventions. Econ Hum Biol 4, 123.Google Scholar
Figure 0

Table 1 Prevalence of wasting, underweight and stunting with respect to household and maternal characteristics of 252 preschool-age children, Belen, Peru, October 2005 to January 2006

Figure 1

Table 2 Comorbidity of wasting, underweight and stunting in a population of 252 preschool-age children in Belen, Peru, October 2005 to January 2006

Figure 2

Table 3 Risk factors for wasting, underweight and stunting in a population of 252 preschool-age children in Belen, Peru, October 2005 to January 2006

Figure 3

Table 4 Prevalence of wasting, underweight and stunting in previous cross-sectional studies in preschool-age children in Latin America and the Caribbean*