Objectives: To compare the progression in the prevalence of stunting in a nationally-representative survey between that accumulated up to 59 months and what it was at the 24-months landmark, within a sample of children classified according provinces of Guatemala. This is justified insofar as Guatemala has the highest prevalence of chronic malnutrition, better termed “stunting” in the Western Hemisphere, with a composite prevalence across the nation of 46.5%.

Methods: The data of the prevalence of stunting in children provided by the ENSMI 2014-2015 (the DHS survey for Guatemala) were taken; and the sample was distributed according to the place of residence. For the first time in history and based on the concept of the “window-of-opportunity” of the first 1000 days of life, anthropometric status was reported both conventionally under-five prevalence and also through 0-24 months separately. The provinces were arranged in ascending order by the interim accumulation of stunting values. The difference of the
prevalence of stunting at both landmarks was calculated, province by province. Descriptive statistics and comparisons were generated.

Results: The respective descriptive statistics for prevalence of stunting from 0 to 24 mo and from 0 to 59 mo were 41±15%, (with a median of 38% and range of 19.1% to 70.3%) and 46±14%, (with a median of 46% and range of 25.3% to 70%), across the 22 provinces of Guatemala. The overall mean interim difference in stunting accumulation was 4.7±4.0 percentage points (median: 4.6 percentage points). However, when assessed by individual provinces, the values in the time-dependent difference between the 5-year accumulated rate in relation to the 2-year data range from -1.6 percentage points in the Guatemala Province, to a +13.2% percentage points in Alta Verapaz.

Conclusions: The difference in prevalence between the stunting accumulated up to 59 months and what it was at 24 months was variable among the 22 diverse provinces of Guatemala, and the environmental and contextual factors operating over the last 3 y of the under-five period emerge as a subject for future investigation.

Objective: To delineate the contribution of estimation of percent body fat (%BF) on the conventional weight-status classifications by body mass index (BMI) in adult women of the Province of Sololá in Guatemala. This is justified insofar as BMI, is simply a measure of relative weight in terms of the stature squared; is used as a useful measure of overweight and obesity. The relationship between BMI and %BF has been studied to estimate the capacity of BMI to predict adiposity. Although BMI is often considered an indicator of body fatness, it can be a variably-valid surrogate measure of body fat because it measures excessive relative weight rather than excess fat.

Methods: Height and clothing-adjusted body weight were determined in 111 women between the ages of 19-68 years, and their BMI calculated. The women were classified in three sub-groups of n=37 according to WHO BMI classifications. All the women were measured for %BF assessment with BIA, using the mBCA 525 BIA instrument (SECA, Hamburg, Germany), with tetrapolar placement of skin electrodes, with specific attention to the %BF value (BIA-%BF). %BF was concurrently estimated by two predictive equations developed by Deurenberg et al (1991) and Gallagher et al (2000).

Results: We measured all variables in 111 women (median: 35 years and 148 cm), equally divided and distributed into three BMI classes: “normal” (median BMI: 23 kg/m2 , n=37), “overweight” (median BMI: 27 kg/m2, n=37) and “obese” (median BMI: 31 kg/m2, n=37). Respective median %BF across the classification groups were 32%, 39% and 46%. The overall Pearson r value for BMI vs %BF (n=111) was 0.80. BIA-measured %BF was, on average, 3.2 percentage-points higher than the theoretical association from the Gallagher and Deurenberg predictive equations.

Conclusions: The median stature in the sample was only 3 cm above the cutoff criterion of 145 cm for “adult stunting” in women. The distinct methods for assessing %BF provide values that are highly inter-correlated. The approach based on BIA provided the highest %BF among the methods. In this population, women of normal weight are not necessarily “lean”; however, almost all women classified as obese by BMI have elevated adiposity by both universal and conditional body fatness criteria.

Objective: The aim of this study was to determine the geographical correspondence of the body-size height and weight indices of under-five year-old children from a nationally-representative sample of the Republic of Guatemala.

Methods: We reviewed Table 14b from the Key Indicators document of the 2014-15 Demographic and Health Survey (ENSMI in Spanish) and ordered the 22 provinces in descending order from that with the lowest to highest prevalence of stunting (<-2 Z scores, height-for-age) and underweight (<-2 Z scores, weight-for-age). Analysis of correspondence was made by assessing homology with the tertiles and by a Spearman rho coefficient of the ordinal rank for provincial prevalence of both deficits in body size.

Results: The lowest rate of stunting of 25.3% was found in Guatemala Province, and the highest of 70.0% in Totonicapán. The lowest and highest rates of underweight, 6.2% and 21.4%, were seen in Izabal and  Huehuetenango, respectively. Only 2 provinces (Petén and Sacatepéquez) were not common to the 7 districts in the tertiles of lowest frequencies of stunting and underweight (71.4% homology). Only one province (Jalapa) was not common to the tertile of the highest frequency (85.6% homology). Consequently, 3 provinces were non correspondent among 8 in the middle tertile (62.5% homology). The Spearman rho correlation coefficient for rank- order numbers for provincial prevalence were 0.835 (p<0.001). The prevalence of deficits in weight-for-height ranged nationally from 0.0 to 1.6%.

Conclusions: A correspondence in body-size height and weight indices across geography is not unexpected, given the lack of any substantial wasting. Therefore, the lower than normal weight-for-age is likely a direct consequence of the children’s being appropriate and proportionate to their severely-retarded stature in Guatemala, without any contribution from energy deficits.

Objective: To determine the correspondence between anthropometric indicators HAZ and WAZ in children under 5 years of age with the food-security scale at the household level in the 22 departments of Guatemala.

Methods: We reviewed Table 14b from the Key Indicators document of the 2014-15 Demographic and Maternal and Child Health Survey (ENSMI in Spanish) and ordered the 22 provinces in descending order from that with the lowest to highest prevalence of stunting (<-2 Z scores, height-for-age) and underweight (<-2 Z scores, weight-for age). The food security assessment used a variant of the Food Security Scales for Latin America and the Caribbean (ELCSA) with data from the National Survey of Living Conditions (ENCOVI); it was analyzed as the percentage not insecure, and the latter plus the mildly-insecure tier (combined). Analysis of correspondence was made by a Spearman rho coefficient of the ordinal rank for provincial prevalence of each deficit in body size in ascending order with degree of food security, and by assessing homology with in the tertiles.

Results: The lowest rate of stunting of 25.3% was found in Guatemala Province and the highest of 70.0% in Totonicapán. The lowest and highest rates of underweight, 6.2% and 21.4%, were seen in Izabal and Huehuetenango, respectively. The corresponding rates of full food security, ranging from 6.2% to 21.4%, were seen in San Marcos and Alta Verapaz, respectively. The Spearman rho correlation coefficient for rank-order of the stunting-prevalence hierarchy was 0.156 (p=0.487) and that for underweight-prevalence was 0.363 (p=0.097). In within-tertile analysis, the percentage homology for provinces for full food security with HAZ rank was18.2%, and was 27.3% with WAZ. Using the combined food security tabulation, the respective Spearman rho values were -0.085 (p=0.706) and 0.089 (p=0.695) and the geographical homology values were 27.3% and 36.4%, respectively.

Conclusions: Not unexpected is that the associations were higher with weight status than height adequacy. In fact, the low and non-significant associations between indicators of body-size with food security is consistent with a scenario in which short stature is largely determined by factors other than food and diet, while body mass is appropriate to the lower statures.

Objective: To establish the variation of severe infant stunting rate, as defined as <-3 Z scores of height-for-age, through 2 and 5 years of age across the 22 provinces of Guatemala, using national survey data. This is justified insofar as, Guatemala has the highest prevalence of chronic undernutrition, better termed “stunting.” among under-five children in Latin America (46.5%); this. is attributed to high rates of food insecurity, poor hygienic environments and social structural problems of inequality and exclusion. The most vulnerable are the communities of Mayan descent, where stunting affects 61.2% of children under five. Beyond the total stunting rate at the <-2 Z-scores of HAZ, is a classification of “severe stunting” at the <-3 HAZ). Any interaction between these grades is unknown.

Methods: Data reported in the National Survey for Maternal and Infant Health (ENSMI 2014-2015) for total (< -2 HAZ) and severe (< -3 HAZ) stunting rates across two (0-24 months and 0-59 months) age-intervals were analyzed to explore any variance in stunting-grades over time. Spearman rank-order correlation, was applied to determine the strength of any association of the proportion of stunting in the severe category with the total stunting prevalence across Guatemala’s 22 provinces. Pearson correlation coefficient was calculated for the percentage of severe stunting across time-intervals among the provinces.

Results: The provinces of El Progreso and Huehuetenango reported the lowest and highest values through the 0-24 month interval, 3.1% and 32.3%, respectively; Izabal and Totonicapán with 4.8% and 34.0% were the corresponding opposites through the 0-59 month interval. Spearman rank-order correlations for total and severe stunting rates for the respective landmark intervals were highly significant at 0.967 (p<0.00001) and 0.920 (p<0.00001). Mean percentage of stunting of the severe degree rose slightly from 14.4±8.0% to 16.1±8.6 across provinces with extending time-interval. Inter-province Pearson correlation coefficient for severe stunting across the time-points was r=0.953 (p=0.00001).

Conclusions: When overall stunting is low, the severe form is proportionally low, with the converse relationship in the higher echelons of stunting prevalence. Within-province, the prevalence of severe stunting tends to be constant throughout the 5-year period of public health interest.

Objective: To examine the relationships between body mass index (BMI) cut-off groups and the associated percent body fat (%BF) values, produced using bioelectrical impedance analysis (BIA) and predictive equations methods, among a sample of women living at the Sololá province in the western highlands of Guatemala. This is justified insofar as the epidemic of obesity is expanding around the world, body fat composition (%BF) has become a valuable indicator of health criteria among diverse demographics. WHO BMI cut-off values are widely use among populations to determine normal, overweight and obese status, but fell short in the assessment of health if %BF is not linked. BMI cut-offs and %BF values vary for different ethnic groups hence the importance of exploring the linkages between cut-off values for BMI and the corresponding %BF among different populations, in order to make better estimations of risk factors in health status.

Methods: Height, and clothing-adjusted body weight were measured in 111 women aged 19-68 years, BMI was calculated and women were equally grouped (n=37) among BMI cut-off classifications, %BF values were produced using the measurements taken with the mBCA 525 BIA SECA instrument and the calculated values of the Deurenberg (1991) and Gallagher (2000) equations. Pearson correlations were applied for the three %BF estimation methods vs BMI, itself.

Results: Pearson r values showed higher correlation with the predictive estimation equations Gallagher (r=0.97) and Deurenberg (r=0.94), than for BIA (r=0.80). Measured with BIA 50% of women showed elevated %BF, of >30%, although only 33.3% were classified by BMI as obese.

Conclusions: Estimated by BIA, evidence of elevated adiposity can be found across all three BMI classifications. Higher Pearson r values for Gallagher and Deurenberg equations, are due to the inherit nature of including BMI in their calculations.

Objectives: The aim of this study was to determine the variation between lactating stages and the breast-milk content of docosahexaenoic acid (DHA, 22:6n-3) estimated by dried milk spot analysis in nursing, urban women in the Western Highlands of Guatemala.

Methods: 40 lactating women, 10 each at 40, 80, 120 and 160 days postpartum with a from the urban area of Quetzaltenango, Guatemala, were recruited to this study by convenience sampling. Participation was voluntary with written consent. Women with premature neonates were excluded from the study. At least 30 mL were collected breastmilk from each participant using manual breast-pumps by trained researchers. All samples were prepared in the laboratory using the dried milk spot preservation and then mailed at ambient temperature to Lipid Technologies LLC, Austin, Minnessota, USA. Data was evaluated with SPSS software version 22.0. ANOVA test was used to evaluate the differences in DHA composition among lactating stages.

Results: The mean maternal age and postpartum days was 24.6 ± 5.1 years old (range 17-37 y) and 99.6 ± 47.0 days (range 37-172 days), respectively. DHA content ranged from 0.05-0.58% of total fatty acids, with mean (±SD) content of 0.15±0.09%. The median DHA content by lactation-stage groups studied was 0.14%, 0.14%, 0.09%, 0.14% respectively. No significant differences between groups were observed (p=0.08) but reflecting a near-significant trend. DHA content of breastmilk only differed significantly between 80 and 120 days postpartum (p=0.02), based on pair-wise comparisons.

Conclusion: In general, the DHA content in this setting of the Western Highlands is low in terms of normative international values. The significant decline in breast milk DHA content, at least through the 40- to 120-day interval is consistent with literature reports of progressive decrements over time. There was a rebound at the point, i.e. at 6 months, at which milk begins to form a lesser contribution to infant nutrition.

Objectives: The aim of this study was to assess the magnitude of the association between human milk fat content (MF) with docosahexaenoic acid (DHA, 22:6n-3) estimated by dried milk spot analysis among lactating urban women in the Western Highlands of Guatemala.

Methods: 40 lactating women with infants aged between 37-172 days from the urban area of Quetzaltenango, Guatemala, were recruited to this study by convenience sampling. Participation was voluntary with written consent. Women with premature neonates were excluded from the study. At least 30 mL breastmilk from each participant was collected using manual milk pumps by trained researchers. MF (g/L) was determined from the average of three creamatocrit measurements for each sample using Lucas’s equation (1978). To determine DHA breast milk concentration, all samples were prepared in the laboratory using the dried milk spot analysis on PerkinElmer 226 five spot RUO cards and then mailed at ambient temperature to Lipid Technologies LLC, Minnesota, USA. Data were  evaluated with SPSS software version 22.0. Pearson correlation coefficient (r) was used to evaluate the association between total MF and DHA composition.

Results: The mean maternal age was 24.65 ± 5.13 years old (range 17-37 y) and postpartum days averaged 99.58 ± 47 days, respectively. The mean (±SD) MF was 32.1 ± 16.5 g/L (range 5.5-69.7 g/L) with a median of 33.1 g/L. The mean (±SD) DHA concentration was 0.15±0.09% (range 0.05-0.58%) with a median of 0.13% of total fatty acids. There was no association between MF and DHA (p=-0.081).

Conclusion: In urban lactating women of the Guatemalan Western Highlands, breast milk DHA content was low compared to the worldwide literature experience (Brenna JT et al, 2007). Human milk fat concentration in a sample is independent of DHA content in the milk fat, itself, thus justifying and validating this approach to nutritional assessment.

Objective: To estimate the distance from the place of residence of the households to the locations of purchase of unprocessed foods.

Methods: To date, 45 adult women in charge of buying food have been interviewed at their homes. A semi quantitative interview was conducted using the remainder from the previous day as a basis, asking about the places where each food was purchased, type of food purchased (processed or unprocessed), distance traveled for the purchase of food.and the locomotion or transportation vehicle used to travel. The distance was asked in meters or kilometer, there also were cases that did not provide this information, and it was necessary to use Google Maps to make a rough estimate of the distance traveled from home to the shopping sites. The type of food purchased was classified: unprocessed = foods those of vegetable or animal origin without added substances (e.g., sugar or sweeteners, salt, fat and oils, additives / preservatives), anf processed foods =products altered by the addition of the aforementioned substances not originally part of the original foods.

Results: When it comes to the venue for purchase of food items, the findings show that local markets (47.6%) are the main place to purchase food, followed by supermarkets (11.8%), neighborhood stores (10.5%), street vendors in the street (t 8.9%) and at the front doorstep (8.1%). Sixty-one percent of the food purchased was unprocessed and 39% processed. Walking (39.8%) and being transported by public bus (17.2%) are the most common forms of mobilization to acquire food, comprising almost 60% overall. The mean distances traveled specifically to buy unprocessed food was 1,200 meters.

Conclusions: We document a 3:2 predominance of consumption of unprocessed items compared to processed; most – but not all – of the former type of food is purchased in local markets. The degree to which distance to purchases sites dictates food selection cannot be assessed from the present data.

Objective: To determine the purchasing patterns of the foods as derived from from the consumption of a qualitative recall of the previous day in women of the urban area of Quetzaltenango.

Methods: To date, 45 adult women (median of 39 years) in charge of buying food for their households were interviewed. A semi-quantitative interview was conducted using the recall from the previous day as a basis, inquiring about the place or places of the purchase of food, the time of day or date on which food was purchased, to derive the time-interval between shopping and consumption of individual food items. Purchase-pattern time was classified as: daily (≤ 1 day); weekly (2-7 days). fortnightly (>1week <1 month). or monthly (> 1 month) foods in terms of the interval from when a food had been acquired. In order to classify the sites or establishments of purchase, a registry was created listing the places by categories depending on the type of products obtained. (Reported food items were classified in any of the 17 food outlet categories based on type of food expended.)

Results: Forty-six percent of items consumed had been consumed within the week (weekly), followed by 26% during the prior day (daily), with 24% having been brought into the home at a longer interval. The foods most frequently purchased (with the shortest interval to their consumption) were tortillas, bread, onions, drinking water, vegetables and, eggs; assorted other items fell into the “daily” category. In descending order the most frequent venues for food purchase were local markets, supermarkets and neighborhood stores. Only one woman among those interviewed in this urban, convenience sample had her own production of food for self-consumption in the household; among the products produced were: vegetables, herbs, corn, chicken meat, cow’s milk.

Conclusions: This study in a small convience sample of urban primary food purchasers confirms that foods that are perishable, consumed in large quantities or both need to be replenished on practically a daily basis. have a degree of flexibility in their behavioral response to distinct food expenditures, which might help those choosing healthier foods. This flexibility is however, dependent on the availability of foods in local markets and time shelter for the foods.

Objective: To obtain preliminary information about food environments on the outskirts of Quetzaltenango, Guatemala, to inform the design or modification of research protocols for study of these peripheral zones.

Methods: This exploratory study was conducted on the periphery of the municipality of Quetzaltenango, the first urban growing node other than Guatemala City to be investigated. A convenience sample of 5 peri-urban sectors in which CeSSIAM had previously been involved was enrolled. A Key-Informant in each sector was interviewed after oral consent was provided. Open-ended questions were asked regarding practices, challenges, household strategies and changes over time to guarantee access to food. Dialogues were digital recorded, field notes were and conventional content assessment approach used for findings and their analysis.

Results: According to the key-informants, the periphery sectors studied had gained “urban” status, although extension of municipal services, support of infrastructure (formal food markets) to guarantee access to food are not in accordance to this classification. An additional increment in real estate demand has led to displacement, isolation and reduction of land for cropping; these have limited the access to staple foods such as corn and beans. Incremented insecurity and crime had reduced inhabitants’ mobility and had compromised access to fresh-food markets, making small grocery stores “tiendas” their main food sources. “Tiendas” represent access to more expensive and less nutritious food in these peripheral outskirt spaces.

Conclusions: A revision of the connotations of urban, peri-urban and rural categories is needed in emerging urban nodes such as Quetzaltenango in order to understand the underlying determinants of access to food. In studies aimed to understand nutrition environment ecology determinants more thoroughly, questions regarding social dynamics need to be asked.

Objective: To compare hydration-status estimates between bioelectrical impedance analysis (BIA) and urinary osmolality (Uosm) in lactating mothers of Guatemala in the Western Highlands.

Methods: Forty women, 10 in each of 4 stages of lactation – 40 days, 80 days, 120 days and 180 days – attending the public health center of Quetzaltenango were enrolled. Hydration status was estimated by two approaches: First by urinary osmolality (Uosm) on a urine sample collected at the beginning of the assessment and measured on an Osmometer (Vogel Löser model 8158) expressed in mOsm/kg; and second using bioimpedance data from application of a SECA mBCA 545 BIA instrument (Fat-Free Mass/Total Body Water expressed as percentage hydration). The comparison between methods was conducted by means of the non-parametric Wilcoxon test and by Chi-square.

Results: Evaluation of the 40 women yielded the following results (expressed in descriptive statistics of the mean±SD and min-max) for the osmolality was 599±227 mOsm/kg, (range: 69-980 mOsm/kg), with 24 women (60%) having euhydration status (defined as <700 mOsm/kg) and 16 others (40%) having hypohydration (defined as ≥700 mOsm/kg). With respect to assessment from BIA data, the mean was 75±6% (range: 55-85%), with hypohydration (defined as FFM/TBW X 100 ≥73.2%) in 13 women (32.5%) and 27 others (67.5%) with euhydration status (defined as <73.2%).

Conclusion: Although there is not a strong within-individual association of diagnosis as normally-hydrated or under-hydrated in assessment by the two approaches for evaluation, the tabulated estimation on a group basis is relatively consistent across methods. Most lactating women show euhydration, but it is a serious public health concern that such a high proportion, 32.5-40.0%, have hypohydration with potential implications for the infants’ nutrition as well.

OBJECTIVE: To describe the results found in CeSSIAM studies with respect to the hydration status in Guatemalan women and to identify the most important findings derived from the osmometer procedures to urine and breast milk in relation to hydration.

METHODS: Information was collected from the studies from 2013 to the present date, performed according to protocols approved by the CeSSIAM Human Studies Committee and after informed consent of the participants. A Vogel Löser model 8158 Osmometer was used in the osmolality laboratory, with values expressed in mOsm/kg. All studies were conducted into lactating women’s hydration, variously involving measurements on breast milk and/or urine, with and without hydration interventions or through freeze-fhaw cycles. Assorted statistical testing was used, appropriate to the design and the research questions. In a subset of urines, alternatives of specific gravity, by refractometry and urine color in relation to a color-chart were tested.

RESULTS: Baseline morning breast milk osmolality showed stability in all the studies, with a median value of 308 mOsm/kg (min-max; 282-316 mOsm/kg, n=46) with no detectable increase after ingestion of 1 L or not presenting difference even after the intake of water (p=0.325, n=23). Regarding urine, median osmolality was 627 mOsm/kg (min-max: 93-1073 mOsm/kg, n=46), without any significant association with the corresponding milk sample (Spearman rho= 0.214, p=0.153). The stability of the osmolality of the milk is not affected even by thawing or the application of termperature in the thawing process. When comparing methods of hydration evaluation, urine specific gravity (density) had a high specificity in relation to osmolality (r=0.888. p<0.001).

CONCLUSION: Our experience has shown that human milk osmolality is generally stable with values confined into a range that would not threaten the internal milieu of the digestive tract of a nursing infant. Urinary osmolality is not correlated to that of milk, but there is a variable maternal hydration status revealed by the osmometer and specific gravity. Milk can be frozen-stored without distortion of the true values.