International BMI-for-age references underestimate thinness and overestimate overweight and obesity in Bolivian adolescents.

BACKGROUND: Since no growth standards for adolescents exist and a single reference applicable everywhere is still in debate, it is recognized that the best reference should be derived from the growth pattern of the healthy population that will use it. In 2007 a study developed references for body mass index for 12th to 18th y Bolivian school adolescent (BAP. OBJECTIVES: To compare nutritional status outcomes applying BMI references from the BAP, the Center for Disease Control and Prevention CDC 2000, the International Task Force (IOTF), and the 2007 WHO, to determine appropriateness of use in Bolivian adolescents. SUBJECTS/METHODS: References were applied in 3306 adolescents, 45.0% male, 55% female, 12th to 18th y selected from a nationally representative sample. RESULTS: Main findings reveal that the CDC and the 2007 WHO underestimate underweight (p<0.001) and the three international references overestimate overweight (p<0.001) with variation between ages and gender. CONCLUSIONS: Bolivian health providers are advised to replace CDC, OITF and 2007 WHO references for the use of BAP in Bolivia which reflects its healthy adolescent population growth pattern. International references may lead to incorrect conclusions when applied on Bolivian adolescents. They could deflect efforts from population which need prompt intervention and mislead treatments and budget to unnecessary ones. We recommend validation of international references where appropriate until a standard is released.

Anthropometry of height, weight, arm, wrist, abdominal circumference and body mass index, for Bolivian adolescents 12 to 18 years: Bolivian adolescent percentile values from the MESA study.

Anthropometry is important as clinical tool for individual follow-up as well as for planning and health policy-making at population level. Recent references of Bolivian Adolescents are not available. The aim of this cross sectional study was to provide age and sex specific centile values and charts of Body Mass Index, height, weight, arm, wrist and abdominal circumference from Bolivian Adolescents. Data from the MEtabolic Syndrome in Adolescents (MESA) study was used. Thirty-two Bolivian clusters from urban and rural areas were selected randomly considering population proportions, 3445 school going adolescents, 12 to 18 y, 45% males; 55% females underwent anthropometric evaluation by trained personnel using standardized protocols for all interviews and examinations. Weight, height, wrist, arm and abdominal circumference data were collected. Body Mass Index was calculated. Smoothed age- and gender specific 3rd, 5th, 10th, 25th, 50th, 75th, 85th, 90th, 95th and 97th Bolivian adolescent percentiles(BAP) and Charts(BAC) where derived using LMS regression. Percentile-based reference data for the antropometrics of for Bolivian Adolescents are presented for the first time.

[Development and testing of a semi-quantitative food frequency questionnaire for Bolivian adolescents]. / Desarrollo y evaluación de un cuestionario semi-cuantitativo de frecuencias alimenticias para adolescentes bolivianos.

INTRODUCTION: Since previous studies suggest that the dietary patterns and lifestyle in adolescence have an impact in health later in life, it is pertinent to monitor and document their dietary patterns. OBJECTIVE: To develop and test the reliability and validity of a semi-quantitative food frequency questionnaire for Bolivian adolescents. SETTING: La Paz, September - October 2004. METHODOLOGY: A convenience sample of 82 students from a secondary participated in the study. A semi quantitative food frequency questionnaire was applied twice with one month interval (Q1 and Q2). A 3 days dietary record (3R) was obtained from 32 of the Students as standard. RESULTS: Most of the questions (87%) reporting intake of each individual food item agreed moderately to substantially while only 2 individual food items fell into the category of slight agreement and 11 into the category of fair agreement using the kappa statistic. Quantities obtained from both applications of the questionnaire yielded similar results (P > 0.05). Nutrient estimates obtained from the Q1 and Q2 and the 3R were not statistically different (P > 0.05). CONCLUSION: The semi-quantitative food frequency questionnaire is apparently reliable and valid for the measurement of absolute food intake in Bolivian adolescents. Care will be taken in the future to prevent overestimation of intake measured by this particular questionnaire.

Female gender and wealth are associated to overweight among adolescents in La Paz, Bolivia.

OBJECTIVE: To describe the prevalence of overweight, obesity and being thin in adolescents of La Paz City, Bolivia. DESIGN: Cross-sectional study, clustered sampling. SETTING: La Paz, Bolivia, August-September 2003. SUBJECTS: Height and weight of 525 adolescents (mean age 16 y; s.d.+/-1.3) attending public and private schools were measured; sociodemographic characteristics were assessed with a short questionnaire. RESULTS: The present study reveals that 9.3% (+/-2.5) of the sample are thin adolescents; 19.8% (+/-3.4) is the overall prevalence of overweight and 2.3% (+/-1.3) the overall prevalence of obesity. The odds of being thin were higher among those working (P=0.03) and among those of higher age (P<0.001). The odds of being overweight were higher among female subjects (P<0.001). The odds of being obese were higher among younger adolescents (P=0.046) and among the more wealthy ones (P=0.044). CONCLUSIONS: A large prevalence of 22% overweight/obesity in adolescents from La Paz City was observed. The findings support the need of enhanced nutritional surveillance and interventions particularly in female adolescents.

Changes in the nutritional status of Bolivian women 1994-1998: demographic and social predictors.

INTRODUCTION: Bolivia, as one of the poorest Latin American countries, has dealt with the problems of undernutrition for the last 50 y. Little importance has been given to the increase in overweight and obesity among the population, despite the scientific evidence linking overweight and obesity with mortality and morbidity. OBJECTIVE: To describe the social and demographic determinants of the nutritional status among women in Bolivia between 1989 and 1998 to gain a better understanding of the nutrition transition phenomena and to identify urgent research needs. METHODOLOGY: Secondary analysis of the raw data of the Bolivian National Demographic and Health Surveys of 1994 and 1998. Changes in the prevalence of underweight, obesity and overweight are described by sociodemographic characteristics of Bolivian women. Social and demographic determinants of nutritional status have been fitted into a logistic model. RESULTS: The prevalence of overweight (defined as 25< or =BMI<30 kg/m2) among women of reproductive age (20-44 y) increased by 9 percentage points between 1994 and 1998 (P<0.001), while the prevalence of normal BMI decreased by 10 percentage points (P<0.001). The decrease in the prevalence of underweight (defined as BMI <18.5 kg/m2) from 2.4% in 1994 to less than 1% in 1998 was statistically significant (P<0.001). Obesity (defined as BMI> or =30 kg/m2) was positively associated with geographical region (P=0.001), educational level (P<0.001), age (P=0.003) and total number of children (P=0.001) and negatively associated to rural locality (P=0.001) and native languages (P<0.001). Overweight was inversely associated with rural locality (P=0.013) and with Quechua language (P=0.04), while the total number of children (P<0.001) and year of survey (P<0.001) were positively associated. Underweight decreased dramatically (P<0.001), being positively associated with the region of residence (P=0.04) and inversely associated with the total number of children (P=0.006). CONCLUSION: The present study suggests that the population of Bolivia is in a transitional stage, with overweight becoming as much of a problem as undernutrition.

Hair zinc and copper in Indonesian infants.

The objective of this study was to determine the zinc and copper status in hair of a group of Indonesian infants aged 0-5 months, a period when growth faltering in this population is known to occur, and to determine the daily zinc and copper availability in the habitual diet. A mixed cross-sectional longitudinal design was used. Infants 0-3 months of age were recruited in two villages on the south coast of the island of Madura, Indonesia and followed up to the age of 5 months. All newborns during the study period were included. Hair samples were collected between the ages of 0 and 5 months at monthly intervals. Zinc and copper concentrations were determined by flame atomic absorption spectrophotometry and compared with a sample of Belgian controls, recruited cross-sectionally. Zinc and copper content of the habitual diet was calculated on the results of a food intake study previously performed in the same community. For the 42 Madurese infants recruited, 107 hair zinc and 96 hair copper concentrations were determined. Belgian infants (15 boys, 15 girls) served as controls. One Madurese infant died during the study and six moved from the area. Hair zinc concentrations were found to decrease with age in both populations, while the zinc and copper values did not differ from the Belgian controls. Hair zinc values were not correlated with growth performance. The boys had lower hair zinc values than did the girls. Copper values among the Indonesian infants did not show a trend over time; however, the Belgian children showed an increase towards the age of 12 months, although this was not significant. The mean daily zinc and copper availability in the habitual diet was less than half of the recommended daily allowance for adult women. The situation was much worse for lactating women given that the availability of these elements increased very little. The hair zinc and copper values indicate that they are not responsible for the early onset of linear growth retardation. The lower zinc values in boys might be an indication of a marginal deficiency. The very low zinc content of the diet consumed in this population could be an indication of a zinc and copper deficit in the Madurese population, although this needs to be confirmed.

Pattern of linear growth velocities of infants from birth to 12 months in Madura, Indonesia.

The paper analyses growth velocity data of infants aged 0-11 months from Madura, Indonesia, with the aim of identifying the time of onset of linear growth retardation. Velocities were calculated as average velocities from mid-point intervals, so that they can serve as comparative information for further studies. Secondly, the relation between weight and length velocities is tested. Thirdly, growth velocities are related to birthweight, length at birth and ponderal index (PI = weight/height3 x 100 in g/cm3. The anthropometric information of infants is taken from two large longitudinal studies, East Java Pregnancy Studies Phase I and Phase II (EJPS I and II). These were conducted from August 1981 to December 1985 and January 1987 to December 1989, respectively, in two villages in Madura, Indonesia. The results support the following hypotheses: linear growth in the first year in Madurese infants shows two periods of deceleration. The early phase starts in the first month and is related to intra-uterine growth. It lasts up to about 4-6 months. Children with normal birthweight but with a low PI grow slowest in length after birth. The second period is towards the second half of the first year, when differences in linear velocity decrease with the references and velocity distribution change. Differences in weight velocity increase during this period. Factors outside the intricate fetal mother-child relationship could start to play a role.

Morbidity and growth performance of infants in Madura, Indonesia.

Between January 1987 and July 1988, all children born in two villages on the island of Madura, Indonesia were visited weekly by a field worker trained to interview mothers on disease symptoms. The maximum recall period was 1 week. All infants were measured (weight and height) at monthly intervals. Information on growth and morbidity is analysed for infants from birth until the age of 11 months. Morbidity, defined as acute respiratory tract infection (ARI), diarrhoea, fever and other diseases, is analysed over 4-week periods and related to growth performance. In total, there are 1373 4-week reporting periods with morbidity information from birth to the age of 12 months. Of all the diseases recorded (1021), 47% were ARI, 13% diarrhoea, 14% fever and 26% other diseases. The average (SD) duration of diarrhoea was 7 (11) days, ARI 14 (9) days, fever 6 (4) days and 16 (10) days for other diseases. The most striking results in this analysis are: (i) the lack of a relationship between morbidity and growth (either linear or weight) during the 1st 6 months of life; (ii) the existence of a relationship between illness and weight increment for which only ARI showed significant influence for infants of 6 months and more; (iii) the lack of a relationship between morbidity and linear growth performance at all ages; and (iv) the fact that no cumulative effect of disease on growth performance was found to explain the observations.

PIP: To collect data on infant morbidity and growth, all children born in an 18-month period in 1987-88 in two villages on the island of Madura, Indonesia, were visited weekly by a field worker. Infants were weighed and measured monthly. In total, 1373 4-week reporting periods covering birth through 12 months were available for analysis. Disease was present in 51% of these periods, and 22% of infants had more than one illness episode in a 4-week period. Of the 1021 disease episodes recorded, the most prevalent were acute respiratory infection (47%), fever (14%), and diarrhea (13%). The average durations of these three diseases were 14 +or- 9 days, 6 +or- 4 days, and 7 +or- 11 days, respectively. Extrapolation of these findings suggests the average infant in Mandura will have 4 episodes of acute respiratory infection, 1-2 episodes of diarrhea and fever, and 2-3 episodes of other diseases in the first year of life. There was no association between morbidity and growth in the first 6 months of life. However, between 6 and 11 months of age, acute respiratory infection (but not diarrhea) was significantly associated with incremental weight loss. Height was not affected by morbidity at any age. There was no cumulative effect of disease on growth performance. These findings suggest that if infants are breast fed, well cared for, and provided access to curative care in the first year of life, as the infants in Mandura were, intercurrent illness has no midterm adverse effect on growth.

Growth faltering in Madura, Indonesia: a comparison with the NCHS reference and data from Kasongo, Zaire.

Weight and height data of Madurese children from two longitudinal studies, the East Java Pregnancy Studies I and II, are analyzed cross-sectionally and compared with the NCHS reference and a rural population in Kasongo, Zaire. A total of 1945 children are included in the analysis, giving 25,202 weight and 17,196 height measurements. Madurese children start to accumulate a weight deficit compared with NCHS data at the age of 4 months. A linear deficit is already noticeable at 1 month, increasing rapidly in the 1st year of life. This period is responsible for most of the linear deficit accumulated by the age of 5 years. Madurese children are relatively fatter than NCHS children up to the age of 12 months. Comparing the Madurese children with those in Kasongo reveals that the accumulation of their linear deficit occurs at a period when weight increments follow those of Kasongo. Since in the studied population food intake and breastfeeding could not explain the growth dynamics in the 1st year of life, it is suggested that the explanation may lie in micro-nutrient deficiency and intra-uterine development.