The role of exclusive breastfeeding and sugar-sweetened beverage consumption on preschool children's weight gain.

BACKGROUND: Sugar-sweetened beverages (SSBs) and breastfeeding practices have been recognized as important factors linked to children's weight status. However, no other studies have simultaneously investigated the role of each factor on children's conditional weight gain (CWG). OBJECTIVE: To evaluate the role of exclusive breastfeeding (EB) and the SSBs consumption on CWG from birth to the survey date among Brazilian preschool children (24-59 months old). METHODS: A nationally represented cross-sectional survey with complex probability sampling (n = 2421) was conducted. The outcome variable - CWG - represents how much an individual has deviated from its expected weight gain, given his or her prior weight. The multivariate linear regression to analyse the effects of EB and the consumption of SSBs on CWG were adjusted for economic status and maternal variables. RESULTS: There was a significantly protective effect of EB duration during the first year of life on CWG from birth to the survey date (-0.02 [-0.03; 0.00 95% confidence interval]); however, the SSBs intake promoted an effect on the weight gain that was 2.5-fold higher (0.05 [0.02; 0.08 95% confidence interval]) than the EB. CONCLUSION: As hypothesized, the exposure variables acted in opposite directions, but the harmful effect of SSBs intake had greater magnitude than the beneficial effect of EB on children's CWG.

Risk factors for overweight and obesity in adolescents of a Brazilian University: a case-control study.

OBJECTIVE: To analyze the risk factors associated to overweight and obesity in freshmen of a public university in the Center-West region of Brazil. METHODS: A case-control study comprising 1,465 adolescents, identifying 106 cases (Body Mass Index (BMI)>or = 85th percentile of National Center for Health Statistics -NCHS) and 233 controls (BMI > 5th and < 85th percentile of NCHS). Interviews were made to collect information on socioeconomic data, eating habits, physical activity and health habits. Information on parents' height and weight were obtained over the telephone. RESULTS: Non-conditional multivariate and hierarchical logistic regression showed that overweight and obesity were positively associated to males (odds ratio (OR) 2.25, 95% confidence interval (CI) 1.37-3.69), mother's BMI (OR 3.45, 95% CI 2.16-5.8), overweight in childhood (OR 2.81, 95% CI 1.62-4.85), consumption of fruits less than once a day (OR 1.84, 95% CI 1.05-3.21) and the habit of weight-loss dieting (OR 6.33, 95% CI 2.68-14.94). CONCLUSION: Results emphasize the need for educational interventions at early ages involving the whole family to control the excess of weight.

Accuracy of obesity diagnosis in Brazilian adolescents: comparison of Cole et al and Must et al criteria with DXA percentage of fat mass.

OBJECTIVES: to assess the accuracy of the two most used anthropometric criteria: Must and Cole to diagnose obesity in adolescence comparing with percentage of fat mass determined by DXA. METHODOLOGY: cross-sectional study with 418 adolescents (52.4% males) attending a private school in São Paulo/Brazil. Anthropometric measures of height and weight were taken and BMI was calculated. Analysis of body composition was performed using the DXA to detect percentage of fat mass. Using the method proposed by Ellis & Wong (ERM) two sex-specific linear regression models of fat percentage for age in years were fitted. The comparison between the methods was carried out through the analyses of specificity and sensitivity with two residual percentiles as cutoff points (ERM85th and ERM95th) as standards. A logistic model was fitted to estimate the probability curves of obesity classification. RESULTS: the comparison of the two classic criteria for the diagnosis of obesity with the ERM85th and ERM 95th, yields for females the same sensitivities of 0.50 and 0.20 for both criteria. For males sensitivities for ERM 85th were 0.61 (Must) and 0.49 (Cole); while for ERM95th the sensitivities were 0.81 (Must) and 0.64 (Cole). Therefore, there are high probabilities that those criteria diagnose adolescents as obese, when actually they are not. CONCLUSION: the Must and Cole criteria were similar and present flaws for the diagnosis of obesity. In clinical practice and field studies anthropometric criteria should be evaluated as to the diagnostic accuracy along with other clinical parameters and, when feasible, the analysis of fatness percentage. However, the anthropometric criteria evaluated are efficient in the identification of non-obese adolescent in the two cutoff points considered.

Plasma total homocysteine in Brazilian overweight and non-overweight adolescents: a case-control study.

OBJECTIVE: To test the hypothesis that overweight adolescents have higher plasma total homocysteine (tHcy) levels than non-overweight adolescents and to explore the association between plasma tHcy levels with folate, vitamin B12 and some risk factors for CVD in both groups. METHODS: A case-control study conductec with 239 adolescentes aged 15-19 years in the city of São Paulo, Brazil; 86 overweight and 153 non-overweight frequency matched by age, gender, pubertal and socioeconomic status. tHcy, folate, vitamin B12, lipid profile, glucose, insulin and insulin resistance were measured. RESULTS: No significant differences were found in tHcy, folate and vitamin B12 levels between overweight and non-overweight groups. The geometric means of tHcy were elevated in both groups (overweight: 11.8 micromol/L; non-overweight: 11.6 micromol/L) higher for boys than for girls (P < or = 0.001). Folate deficiency was identified in 68.6% of total studied population. Triacylglycerol, LDL cholesterol, insulin resistance were higher and HDL cholesterol was lower in overweight that non-overweight adolescents. In the multiple linear regression model, in overweight group, tHcy was independently associated with age (P = 0.041), sex (P = 0.004) and folate (P = 0.022) and in non-overweight group, with age (P = 0.049), sex (P < 0.001), folate (P = 0.018) and vitamin B12 (P = 0.030). CONCLUSIONS: Obesity was not a determinant factor of tHcy levels. Age, sex and folate were independent determinants of plasma tHcy levels. The high prevalence of folate deficiency may have been responsible for the elevated tHcy levels in these adolescents, increasing the risk for future development of CVD.

An application of bootstrap resampling method to obtain confidence interval for percentile fatness cutoff points in childhood and adolescence overweight diagnoses.

OBJECTIVE: To present a resampling approach to obtain confidence intervals (CIs) and the empirical distributions for the studentized regression residuals percentiles when used as cutoff points for overweight and obesity diagnosis in children and adolescents. METHOD: A tutorial for the nonparametric bootstrap with bias accelerating correction is presented. A classical method, the Binomial interpretation, is used as comparing criterion. SUBJECTS: A case study comprising 418 randomly selected subjects from a private secondary school (age: 10-17 y, boys: 52%). MEASUREMENTS: Body fat percentage (by), age (y) and Tanner criteria. RESULTS: The empirical distributions presented skewness suggesting that the CIs should not be symmetric. CIs obtained by the proposed approach were more realistic than the classical ones. CONCLUSIONS: We propose a simple and efficient way to obtain the interval estimates and the distribution properties of cutoff points for overweight and obese classification using a sample-based method that allows the comparison of cutoffs among many subpopulations.

Hábitos alimentares de escolares (1ª e 2ª séries) durante a merenda escolar em escolas públicas de São Paulo, SP / Alimentary habits of students during meal break time of public schools of São Paulo city

Objetivo:estudar o hábito alimentar de crianças e seus determinantes durante a merenda escolar, visando conhecer e intervir positivamente na formação dos hábitos alimentares. Metodologia: em amostra de 630 escolares, observou-se o consumo alimentar no horário destinado à merenda escolar, classificando-o em saudável e não saudável. Utilizaram-se tabelas de contingência para a descrição das frequências das variáveis. Para o estudo de efeitos individuais de cada variável independente, com controle dos efeitos das demais incluídas no modelo, foi utilizado modelo de regressão logística com estimativas de razões de chances (odds ratio), entre os grupos de escolares com hábito alimentar saudável e não saudável.Resultados: observou-se a alta prevalência de 77,5 por cento (n=488) do hábito alimentar não saudável. Obesidade e sobrepeso, sexo masculino e ir à escola a pé, que se constituíram em variáveis com significância de 0,10 na an´lise bivariada, deixaram de s

What can go wrong when you assume that correlated data are independent: an illustration from the evaluation of a childhood health intervention in Brazil.

The key analytical challenge presented by longitudinal data is that observations from one individual tend to be correlated. Although longitudinal data commonly occur in medicine and public health, the issue of correlation is sometimes ignored or avoided in the analysis. If longitudinal data are modelled using regression techniques that ignore correlation, biased estimates of regression parameter variances can occur. This bias can lead to invalid inferences regarding measures of effect such as odds ratios (OR) or risk ratios (RR). Using the example of a childhood health intervention in Brazil, we illustrate how ignoring correlation leads to incorrect conclusions about the effectiveness of the intervention.

Breastfeeding training for health professionals and resultant changes in breastfeeding duration.

CONTEXT: Promotion of breastfeeding in Brazilian maternity hospitals. OBJECTIVE: To quantify changes in the breastfeeding duration among mothers served by hospitals exposed to the Wellstart-SLC course, comparing them with changes among mothers attended by institutions not exposed to this course. DESIGN: Randomized Institutional Trial. SETTING: The effects of training on breastfeeding duration was assessed in eight Brazilian hospitals assigned at random to either an exposed group (staff attending the Wellstart-SLC course) or a control group. SAMPLE: For each of the eight study hospitals, two cohorts of about 50 children were visited at home at one and six months after birth. The first cohort (n = 494) was composed of babies born in the month prior to exposure to the Wellstart-SLC course, and the second cohort (n = 476) was composed of babies born six months subsequent to this exposure. MAIN MEASUREMENTS: Kaplan-Meier curves were plotted to describe the weaning process and log-rank tests were used to assess statistical differences among survival curves. Hazard ratio (HR) estimates were calculated by fitting Cox proportional hazard regression models to the data. RESULTS: The increases in estimated, adjusted rates for children born in hospitals with trained personnel were 29% (HR = 0.71) and 20% (HR = 0.80) for exclusive and full breastfeeding, respectively. No changes were identified for total breastfeeding. CONCLUSION: This randomized trial supports a growing body of evidence that training hospital health professionals in breastfeeding promotion and protection results in an increase in breastfeeding duration.

Overweight and obesity in Brazilian adolescents.

OBJECTIVE: This study aimed to describe the prevalence of overweight and obesity (OW+O) among Brazilian adolescents and to identify risks for subpopulations defined according to the five country macro-regions and situation (urban-rural) of the domiciles, income, years of school attendance, age and sex. DESIGN: A nationwide home-based survey representative of the Brazilian civilian noninstitutionalized population, performed in 1989. METHODS: The sampling plans followed a stratified, multistage, probability cluster design in The National Research of Health and Nutrition sample, which collected anthropometric data of 14,455 domiciles. In all, 13,715 adolescents ranging from 10 to 19 y of age were studied. The OW+O was defined from a body mass index (BMI) equal or superior to the 85th percentile of the reference population of the NCHS. The prevalences in the different studied groups were compared using the adjusted odds ratio in logistic regression models. RESULTS: The prevalence of OW+O was of 7.7%, reaching 10.6% within the female group and 4.8% within the male group. A direct relation could be established between the socioeconomic level and OW+O. Adolescents of the most industrialized region of the country presented a risk of OW+O 1.86 (95% CI 1.51-2.30) times higher than that found in the least developed region. Male youngsters who lived in urban areas were more liable (OR=1.71, 95% CI 1.30-2.25) to overweight than their counterparts of rural areas. The occurrence of menarche increased two and a half times (OR=2.58, 95% CI 2.11-3.15) the risk of OW+O within the female group of adolescents. CONCLUSIONS: The results demonstrate a low prevalence of OW+O among Brazilian adolescents when compared with adolescents of more industrialized regions. The OW+O is twice as high within the female group, which represents a much greater difference than the one encountered in industrialized countries, probably owing to the muscular work carried out preponderantly by male adolescents of lower socioeconomic levels. Higher prevalences in subpopulations of higher socioeconomic level and of more industrialized regions show the great need for differentiated actions to control overweight and obesity in the country.

Influence of parental height and sociodemographic factors on adolescent height in Brazil.

OBJECTIVE: To analyze the influence of parental stature and environmental factors on the stature of adolescents from a national survey sample. METHODS: A nationwide survey was carried out in 1989 among a stratified, two-stage, probability cluster sample of 14,455 Brazilian households to provide estimates of anthropometric deficits for urban and rural populations from the five regions of the country. Stature was measured for 5681 boys and girls age 14-18 years, 78.9% of their fathers, and 93.8% of their mothers. Associations between explanatory variables and adolescent height in centimeters were assessed by fitting multiple linear models to the data. RESULTS: The predicted effects of parental stature and environmental conditions together sum to a total of 17 cm when comparing a boy born to parents with stature below the median and living in the underdeveloped rural Northeast region (1.56 m) with one born to parents with stature above the median and living in the partially industrialized urban South region (1.73 m). For girls, this estimated difference was 12 cm. For boys, the overall influence of parents' stature was 10 cm (R(2)= 0.40) and the sociodemographic factors had an overall influence of 7 cm (R(2) = 0.29). For girls, these values were 7 cm (R(2)= 0.35) for the parental influence and 5 cm (R(2) = 0.11) for the sociodemographic factors. CONCLUSIONS: Mother's stature had the same influence on adolescent's stature as father's stature. Independent of parental stature, environmental factors have a strong influence on adolescent stature, particularly among boys.

[Childhood and adolescent obesity] / Obesidade exógena na infância e na adolescência.

OBJECTIVE: To review the medical literature in the last 5 years regarding obesity in children as well as its treatment. METHODS: Literature review. RESULTS: Obesity is a chronic disease that presents high percentages of therapeutic failure and recurrence, leading to serious physical and psychological consequences, especially in its most severe forms. Health care professionals and families usually neglect the treatment of obesity in the hope of a spontaneous reversal of the condition during adolescence. There is a great probability that obese children and adolescents will continue to be obese in adulthood, thus increasing morbidity and mortality of several diseases. CONCLUSION: Pediatricians are supposed to early identify children at greater risk for obesity, taking effective control measures in order to achieve a more favorable prognosis.

[Effects of survival bias on the prevalence of malnutrition in six-year-old children in Brazil, based on the national survey on health and nutrition, 1989]. / Efeito dos vieses de sobrevivência nas prevalências da desnutrição em crianças no sexto ano de vida. Brasil - PNSN, 1989.

Child growth as measured by anthropometric indicators is an important tool for assessing children's nutritional status and society's developmental stage. This study uses the height-for-age indicator with the cutoff point at -2 Z to estimate prevalence of malnutrition in a population of six-year-old children included in the Brazilian National Survey on Health and Nutrition (PNSN). Prevalence variability was analyzed according to gender, trimestral age range, per capita family income, and region of residence. Based on estimates of mortality rates for children under five whose deaths could be ascribed to malnutrition, the survival bias correction was performed using the Boerma methodology. This correction, in turn, was more conspicuous within the low-income and less-developed segment of the population. There was an increase in the number of malnourished children in relation to those surviving at the time of investigation. When comparing less and more developed areas of the country (the Northeast and Southeast, respectively), we observed that malnutrition prevalence rates within the six-year-old group showed no change in the Southeast region, while in the Northeast they increased from 26% to 34%, thus representing a 31% increase in the malnutrition rate. Therefore, in absolute figures, these rates account for the addition of 90,100 children to the malnourished group.

Overweight and obesity among mothers of malnourished children--Brazil--PNSN--1989.

OBJECTIVE: To evaluate the relationship between the nutritional status of the youngest child under 48 months of age (in families with the biological mother present) and their mothers among 3906 children selected from a sample of a national survey in 1989 (PNSN). RESULTS: Malnutrition was present in 5.8% of the children. From these, 21.8%, 60.9% and 17.3% had overweight/obese, eutrophic and malnourished mothers, respectively. Stratified analyses taking into account the regions, situation, income distribution and mother's educational level demonstrated that a lower proportion of malnourished children was concurrent with a higher proportion of overweight/obese mothers. The Kappa test evidenced a poor agreement between the nutritional conditions of the child-mother pairs (K < = 0.048). CONCLUSIONS: When the proportion of malnourished children decreased within the analyzed groups, the proportion of overweight/obese mothers increased. Such an epidemiological pattern indicates that within groups in which malnutrition is less prevalent, the proportion of children for whom a lack of food in the household is the main determinant factor for malnutrition is lower.

[Acute diarrhea: stool water loss in hospitalized infants and its correlation with etiologic agents and lactose content in the diet]. / Diarría aguda: perdas hídricas fecais em lactentes hospitalizados e sua correlação com agentes etiológicos e a presenca de lactose na dieta.

Forty weaned male infants were studied during their first year of life, all hospitalized with acute diarrhea in the Gastroenterology and Metabolism Unit of the Hospital "Umberto I", São Paulo, SP, Brazil. We evaluated and quantified water fecal losses, employing the metabolic bed technique, relating the feeding formula employed with the different causal enteropathogenic agents. 67.5% of the studied infants were under six months and 40% under three months of age. Two groups were randomly assembled to receive, lactose or lactose free feeding formulae. Twenty one patients received a lactose-containing formula (Ninho 10%) and the other 19 children were fed caseine (Portagen) formulae. According to coproculture results and identification of enteropathogenic agents, we divided the studied infants relating feeding formula with the presence or absence of the enteropathogenic Escherichia coli (EPEC): I-13 with positive coproculture for EPEC and diets which included lactose--(L/EPEC+); II--eight with negative coproculture for EPEC and diets which included lactose--(L/EPEC-); III--seven with positive coproculture for EPEC and lactose free diets--(G/EPEC+); IV--12 with negative coproculture for EPEC and lactose free diets. (G/EPEC-). The most frequently isolated agent at coproculture was EPEC, in 20 of the cases (50%), followed by Campylobacter (7.5%). It was also possible to observe that the frequencies of EIEC, Salmonella and Rotavirus were all equal (2.5%). Mixed infections occurred only between EPEC and EIEC, registering a frequency of 5%. The EIEC samples, associated to EPEC 0111 were serotyped as 0 28 ac: H- and 0 152:H-. The use of metabolic bed made the evaluation of fecal volumes possible by a simple and quick technique, thus allowing a closer clinical monitoring, as well as a more reliable evaluation of the patients hospitalized with acute diarrhea. Average acceptance volumes of the formulae--either with or without lactose--were always below the amount recommended by FAO/WHO (100 kcal/day) which shows the impact of acute diarrhea on the decrease of food intake. The average volumes of watery fecal losses found among any of the studied subgroups may be considered quite relevant when compared to standard values. Especially within the L/EPEC+ group fecal losses, both on the first day (83.56 ml/kg/day) and, mainly, on the second (119.44 ml/kg/day) reached exceedingly high levels indicating a disastrous association between the presence of EPEC in the small intestine and lactose offer in the diet. Thus, the results show that there exists a positive and significant association between poor lactose absorption and the presence of EPEC in the feces. WHO's recommendation proposing the use of diluted cow milk, in universal and indiscriminate administration, in the two first days of the disease, may represent a risk factor, not only for malnutrition, but also for the survival rates of children with severe diarrhea, especially those under six months of age and hospitalized with EPEC enteroinfection.

Efeitos dos viéses de sobrevivência nas prevalências da desnutriçäo em crianças no segundo ano de vida - Brasil, PNSN, 1989 / Effects of survival bias on the prevalence of malnutrition among two-year old children

A análise do crescimento da criança, mensurado por indicadores antropométricos, constitui-se em instrumento importante para avaliaçäo da condiçäo nutricional e do grau de desenvolvimento de uma sociedade. Neste artigo, utiliza-se o indicador P/E, com ponto de corte em - 2 Z, para estimar a prevalência da desnutriçäo em uma populaçäo de crianças no segundo ano de vida. Os dados da Pesquisa Nacional de Saúde e Nutriçäo - PNSN-1989 - foram utilizados para análise da variabilidade da prevalência de acordo com o gênero, renda familiar per capita, regiäo domiciliar e faixa etária trimestral. Baseando-se em estimativas relacionadas às taxas de mortalidade de crianças no primeiro ano de vida, cujas mortes poderiam ser atribuídas à desnutriçäo, aplicou-se a metodologia proposta por Boerma (1992) para corrigir os viéses de sobrevivência. Observou-se que, na faixa de menor renda, as taxas de prevalência de desnutriçäo, no segundo ano de vida passam de 5,5 por cento para 12,9 por cento, correspondendo a aumento de 117.159 desnutridos. Na faixa de maior renda, a variaçäo é de 1,8 por cento para 2,5 por cento, aumento de 6.760 crianças. Já, em relaçäo às regiöes geográficas do Brasil, constatou-se que na regiäo Nordeste a variaçäo foi de 5,2 por cento para 11,7 por cento, correspondendo a um incremento de 82.612 desnutridos. Na regiÒo Sudeste, a variaçäo foi de 1,9 por cento para 3,4 por cento, correspondendo a um aumento de 17.755 crianças desnutridas no segundo ano de vida.

Capacitacion de profesionales de salud en lactancia natural y reformas institucionales subsiguientes / Breast-feeding training for health professionals and resultant institutional changes

Se evalua un curso de capacitacion sobre la lactancia natural al que asistieron varios profesionales de la salud en el Centro de Lactancia de Santos (CLS), Estado de Sao Paulo, Brasil, y se examina su efecto en la ejecucion de programas para la promocion de la lactancia natural en hospitales de maternidad. Ocho hospitales de este tipo se incluyeron en el estudio. De ellos, cuatro fueron asignados aleatoriamente al grupo experimental y enviaron a tres profesionales de salud a un curso de 18 dias en el CLS; las cuatro instituciones restantes formaron el grupo testigo. Para determinar en que grado se adheria cada hospital a las "Diez medidas para el exito de la lactancia materna" establecidas por la OMS y el UNICEF se calculo un puntaje antes del curso y otro seis meses despues de concluirlo. Las instituciones del grupo experimental tuvieron un mejor puntaje que las del grupo testigo. El curso del CLS fue eficiente porque permitio a los participantes promovieran la practica de la lactancia natural. Sin embargo, para ejecutar con exito los programas correspondientes, los profesionales de salud tambien tienen que aprender a aplicar los conocimientos adquiridos en el curso y lograr que todo el equipo del servicio de maternidad participe en las actividades correspondientes (AU)

Capacitacion de profesionales de salud en lactancia natural y reformas institucionales subsiguientes / Breast-feeding training for health professionals and resultant institutional changes

Se evalua un curso de capacitacion sobre la lactancia natural al que asistieron varios profesionales de la salud en el Centro de Lactancia de Santos (CLS), Estado de Sao Paulo, Brasil, y se examina su efecto en la ejecucion de programas para la promocion de la lactancia natural en hospitales de maternidad. Ocho hospitales de este tipo se incluyeron en el estudio. De ellos, cuatro fueron asignados aleatoriamente al grupo experimental y enviaron a tres profesionales de salud a un curso de 18 dias en el CLS; las cuatro instituciones restantes formaron el grupo testigo. Para determinar en que grado se adheria cada hospital a las "Diez medidas para el exito de la lactancia materna" establecidas por la OMS y el UNICEF se calculo un puntaje antes del curso y otro seis meses despues de concluirlo. Las instituciones del grupo experimental tuvieron un mejor puntaje que las del grupo testigo. El curso del CLS fue eficiente porque permitio a los participantes promovieran la practica de la lactancia natural. Sin embargo, para ejecutar con exito los programas correspondientes, los profesionales de salud tambien tienen que aprender a aplicar los conocimientos adquiridos en el curso y lograr que todo el equipo del servicio de maternidad participe en las actividades correspondientes (AU)

Se publico en ingles en el Bull. WHO. Vol. 73(4), 1995

Breast-feeding training for health professionals and resultant institutional changes.

Assessed is a breast-feeding training course that was attended by health professionals at the Santos Lactation Center (SLC), Santos, São Paulo, Brazil, as well as its impact on the implementation of breast-feeding programmes in maternity hospitals. Eight maternity hospitals were studied--four were randomly allocated to the experimental group and sent three health professionals to attend an 18-day course at SLC; the remaining four institutions constituted the control group. The compliance of all eight hospitals with WHO/UNICEF's "Ten steps for successful breast-feeding" was determined using scores obtained before and 6 months after the training course. Institutions in the experimental group had an improved score, but those in the control group did not. The SLC training course was efficient since it enabled the participants to promote breast-feeding practices. However, in order to succeed in implementing breast-feeding programmes, health professionals require also to develop skills to apply the knowledge they acquire in the course, as well as to involve the whole maternity unit team in the activities.

PIP: A breast-feeding training course was assessed that was attended by health professionals at the Santos Lactation Center (SLC), Santos, Sao Paulo, Brazil, as well as its impact on the implementation of breast-feeding programs in maternity hospitals. Eight maternity hospitals were studied, 4 were randomly allocated to the experimental group with 3 health professionals to attend an 18-day course at SLC; the remaining 4 institutions constituted the control group. The course consisted of 45 units that covered theoretical and practical aspects of breast-feeding over a 133-hour period. The compliance of all 8 hospitals with the World Health Organization (WHO)/UNICEF's 10 steps for successful breast-feeding was determined using scores obtained before and 6 months after the training course. Individual interviews were held 6 months after the SLC course with directors in participating institutions, with administrative managers, and with professionals in charge of antenatal, nursery, and outpatient services. 16 focus group sessions were also held with 6-15 persons of the staff 6 months after the course. Pre- and postcourse test results showed that in general there was improvement in the students' knowledge on breast-feeding (the average number of correct answers were 20.27 precourse and 26.92 postcourse). Institutions in the experimental group had an improved score, but those in the control group did not. Changes were more particularly in relation to steps 2 (train all health care staff in skills necessary to implement this policy) and 10 (foster the establishment of breast-feeding support groups and refer mother to them upon discharge from hospital). Among graduate-level professionals interpersonal conflicts limited the possibilities for change. The SLC training course enabled the participants to promote breast-feeding practices. However, in order to succeed in implementing breast-feeding programs, health professionals also have to apply the knowledge they acquire in the course as well as involve the whole maternity unit team in the activities.