Rapid ethnographic assessment of breastfeeding practices in periurban Mexico City.

Rapid ethnographic assessment of breastfeeding practices in periurban Mexico City M. L. Guerrero, 1 R. C. Morrow, 2 J. J. Calva, 3 H. Ortega-Gallegos, 4 S. C. Weller, 5 G. M. Ruiz-Palacios, 6 & A. L. Morrow 7 Before carrying out a breastfeeding promotion programme in a periurban area of Mexico City, we conducted a rapid ethnographic study to determine the factors associated with absence of exclusive breastfeeding. The responses to pilot interviews were used to develop a standardized questionnaire regarding reasons for infant feeding choice, sources of advice, and barriers to breastfeeding. We interviewed a random sample of 150 mothers with a child <5 years of age; 136 (91%) of them had initiated breastfeeding; but only 2% exclusively breastfed up to 4 months. The mothers consistently stated that the child's nutrition, health, growth, and hygiene were the main reasons for the type of feeding selected; cost, comfort, and the husband's opinion were less important. Physicians were ranked as the most important source of advice. Reduction or cessation of breastfeeding occurred on the doctor's advice (68%); or when the mothers encountered local folk illnesses such as ``coraje'' (52%) or ``susto'' (54%), which are associatedwith anger or fright; or had ``not enough milk'' (62%) or ``bad milk'' (56%); or because of illness of the mother (56%) or child (43%). During childhood illnesses and conditions, breastfeedingwas reduced and the use of supplementary foodswas increased. This study emphasizes the importance of cultural values in infant feeding choices, defines specific barriers to breastfeeding, and provides a basis for interventions to promote exclusive breastfeeding in the study population. Voir page 328 le re sume en francËais. En la pa gina 329 figura un resumen en espanÄ ol. Introduction Exclusive breastfeeding during the first 6 months postpartum provides many nutritional, immunologi- cal and psychosocial benefits, including protection of the infant against infectious diseases, improved child spacing due to lactational amenorrhoea, and en- hanced maternal±infant bonding (1-3). WHO and other responsible international agencies have recom- mended to governments and health-care profes- sionals that mothers should breastfeed their children exclusively for 4±6 months from birth and to continue breastfeeding, supplemented by other appropriate foods, up to the second year of life or later (4, 6). Because breastfeeding has declined worldwide in recent years as a result of urbanization, marketing of infant formula, and maternal employ- ment outside the home (6±12), exclusive breastfeed- ing to 6 months is uncommon in many areas (8±10). For the past 15 years, we have conducted longitudinal studies among mothers and infants in San Pedro MaÂrtir, a periurban neighbourhood of Mexico City. The results indicate that breastfeeding enhances immunological mechanisms of protection against specific diarrhoeal diseases (13±18). The studies also identified serious problems with breastfeeding prac- tices, e.g. while breastfeeding was started by >90% of mothers in San PedroMaÂrtir, it rapidly declined during the first few months postpartum and exclusive breastfeeding was rarely practised. Water, tea and formula feedings were typically introduced during the first 2 months of life (18). Studies conducted in different parts of Mexico have shown that such supplementary practices are common (10±12), although the reasons for this are not well understood. Prior to initiating a community-based inter- vention project to promote exclusive breastfeeding in San PedroMaÂrtir, we conducted a rapid ethnographic assessment (REA) to identify major influences affecting infant feeding practices. Interventions that influence infant feeding depend on behavioural change and need to be well supported by demo- graphic, psychosocial, and cultural factors that affect the mother-infant pair. REA methods have been recommended as the basis of community-based intervention to reduce disease and promote commu- nity health (19±23). The specific aims of this 1 Epidemiologist, Department of Infectious Diseases, Instituto Nacional de la Nutricio n, Mexico City, Mexico. 2 Epidemiologist, US Navy Environmental Health Center, Norfolk, VA, USA. 3 Professor, Department of Infectious Diseases, Instituto Nacional de la Nutricio n, Departamento de InfectologõÂa, Vasco de Quiroga 15, Tlalpan 14000, Mexico, D. F., Mexico. Requests for reprints should be sent to Dr J. J. Calva at this address. 4 Social worker, Department of Infectious Diseases, Instituto Nacional de la Nutricio n, Mexico City, Mexico. 5 Anthropologist, Associate Professor, Department of Preventive Medicine and Community Health, University of Texas, Galveston, TX, USA. 6 Professor and Head, Department of Infectious Diseases, Instituto Nacional de la Nutricio n, Mexico City, Mexico. 7 Associate Professor of Pediatrics, Center for Pediatric Research, Children's Hospital of the King's Daughters, Eastern Virginia Medical School, Norfolk, VA, USA. Reprint No. 5778 323Bulletin of the World Health Organization, 1999, 77 (4) # World Health Organization 1999 assessment were to identify the following: cultural values and beliefs regarding infant feeding practices; the major sources of infant feeding advice given to mothers; situation-specific supplementation prac- tices; and the barriers to breastfeeding as perceived by mothers. This project was designed to be conducted over a short period of time so that the findings could be used to plan and guide the development and implementation of a community-based peer-coun- selling project to promote exclusive breastfeeding. Population and methods Study area San Pedro MaÂrtir, located in the south-western outskirts of Mexico City, is a transitional neighbour- hood with a sociodemographically diverse popula- tion of approximately 30 000. In the older parts, the families live in houses with running potable water, tiled floors, central sewage and rubbish disposal systems, and have access to roads, clinics, schools, and public transportation. In the most recently settled area, families live in huts with earth floors and water from pipe outlets some distance away. Over half of these hut dwellers are families withmore than four persons occupying one room. Households consist of extended family members or nuclear families. Most families obtain their income from manual work with little variation in the amount of individual wages. Some 15% of men living in the area with young families are in the military and reside in San Pedro MaÂrtir while in training. Few girls and women work outside the home, although their employment level is increasing. There is a govern- ment primary health care centre, which offers antenatal care, immunization, and paediatric out- patient services. While most deliveries occur in nearby hospitals, some deliveries occur at home attended by traditional midwives. Study design From June to August 1994, a survey was conducted in San Pedro MaÂrtir using a combination of epidemiological and ethnographic interview techni- ques to identify and better understand maternal attitudes and behaviours, as well as specific psycho- social and health factors that may lead mothers away from exclusive breastfeeding. Study population Included were mothers residing in the San Pedro MaÂrtir area whose youngest child was < 5 years of age. The free and informed consent of the subjects was obtained. The study was approved by the institutional review board of the Instituto Nacional de la NutricioÂn. Instrument/questionnaire development During the preparation for the survey, we conducted free recall listing interviews (24). These consisted of open-ended questions inwhich five fieldworkers and 15 mothers, randomly selected from the community, were asked to ``list all the things you like about breastfeeding'', ``list all the things you don't like about breastfeeding'', ``list all the supplements that can be given to a breastfeeding baby'', and ``list the reasons why breastfeeding might be interrupted''. A set of 29 items salient to our study population was obtained from these interviews. Project investigators summar- ized the responses and selected 9 items for use in the systematic data collection phase. A 33-item, standardized questionnaire was developed. The questionnaire covered feeding practices relevant to the mother's experience with her youngest child and problems encountered by the mother that reduced the frequency of, or precluded, breastfeeding. In addition, respondentswere asked to rank-order the following items in terms of their importance or relevance to her experience: the reasons to feed a child breast milk or formula; perceived infant health status and infant feeding choice; and sources of infant feeding advice. Each interview required approximately 0.75-1 hour to complete. Survey sampling and data collection A total of 150 mothers were randomly selected from the three sections of San Pedro MaÂrtir. Face-to-face interviews were carried out at door-to-door house- hold visits, using a list of eligible subjects obtained from a recent census of the community. If a mother had moved, but another eligible mother was identified during a visit to the listed address, the newly arrived mother was interviewed instead. Data analysis Basic descriptive analyses of the data were performed using frequency distributions and correlation ma- trices. Responses to questions where respondents were asked to rank a series of items were analysed using cultural consensus methods. A formal math- ematical model and theoretical foundation are presented fully elsewhere (25, 26). The method evaluates the degree of agreement among respon- dents to determine whether their responses are homogeneous enough to warrant aggregation and description as a shared set of cultural beliefs. If goodness-of-fit criteria are met (eigenvalue ratio >3:1), the responses are considered to indicate a single set of shared beliefs, and themodel can provide estimates of culturally agreed upon ordering of items and cultural knowledge scores for each individual. Results Demographic characteristics of respondents All 150 mothers selected (age range, 17±42 years, mean, 26.1+5.6 years) agreed to participate in the study and were interviewed. At the time of the Research 324 Bulletin of the World Health Organization, 1999, 77 (4) interview, none had formal employment. A total of 30 (20%) reported having a child aged <6months, 24 (16%) had a child aged 6±12 months, and the rest (64%) had children in the age range 12±48 months (median age, 24 months). No differences in responses were observed between mothers with a child aged <2 years and mothers with older children. Infant feeding patterns A total of 136 (91%) mothers reported having breastfed their infant (Table 1). Duration of breastfeeding ranged from 2 days to 3 years, with a median duration of 6 months. Only 25% of the mothers breastfed their infant to the end of the first year. Tea, water and bottled milk were introduced early by many mothers. Tea or water was given by 21%ofmothers to their infants on the first day of life, and 63%gave these fluids by day 20. Bottledmilk was introduced by 23% on day 1 postpartum, by 40% on day 14, and by 59% at 3 months. By the end of the third month, 63% of mothers had introduced solid foods. Source of infant feeding advice Doctorswere ranked as themost important source of advice (Table 1). Moreover, 42% of mothers indicated that at some time while they were breastfeeding a doctor had advised them to cease doing so; half of these mothers reported complying with the advice (Fig. 1). The doctors consulted by 50% of the mothers who did so had their offices within the area. Within the family, the mother of the new mother had the greatest influence on feeding practices, followed by the mother-in-law; other individuals were also mentioned by some mothers, but not as frequently. Cultural values and beliefs Cultural consensus analysis revealed that mothers consistently ranked the following reasons in order of importance when choosing an infant feeding meth- od: ``so that my child has good nutrition'', ``so the child is sick less'', ``so the child grows better'', and ``so that the child is cleaner, more hygienic''. Maternal comfort, maternal freedom, cost, the husband's opinion, and the mother's physical appearance were less important. This same ordering applied when the mothers were asked about the relative importance of reasons for choosing breastfeeding. The average level of shared cultural knowledge (0.82) indicated that beliefs were highly shared among these women and that a single set of beliefs existed (eigenvalue ratio, 10.1:1; Table 2). The only factor that mothers mentioned as a reason to bottle-feed was ``it gives me more freedom to do other things''; however, this was given a low ranking as a cultural value for how the mother chose to feed the baby. While infant growth appeared to be a major cultural value for choice of infant feeding mode, and the principal reason for breastfeeding, 37% of mothers agreed with the statement that a baby is fuller and grows better if, in addition to breast milk, the baby receives infant formula. Perceived barriers to breastfeeding Of 106 women interviewed who had an infant aged 46 months, 95% reported that, while breastfeeding, they had encountered one or more of the problems listed in Table 3, and 76% of such mothers indicated that at least one of these problems had caused them to reduce or cease breastfeeding. For example, 52 of these women had ``coraje'' and 26 had ``susto'' (local folk illnesses associated with anger or fright). As a result, 52% and 54% of these women reduced or precluded breastfeeding, respectively. Fig. 1 shows the frequency of the mother's specific behaviour according to the problem encountered. Table 1. Infant feeding practices and sources of infant feeding advice reported by 150 study mothers No. Breastfed youngest child Ever breastfed 136 (91) a 5 3 months 113 (75) 5 6 months 88 (59) Introduced by 3 months Bottled milk 89 (59) Water or tea 139 (93) Solid foods 95 (63) Source of infant feeding advice: Physician 76 (51) Mother 40 (27) Mother-in-law 13 (9) Sister 7 (5) Husband 4 (2) Neighbour or friend 4 (2) Midwife 1 (1) Other 5 (3) a Figures in parentheses are percentages. Table 2. Cultural consensus analysis of the relative importance of nine factors for the choice of infant feeding method among Mexican study mothers Rank order Item a 1 ``So that my child has good nutrition'' 2 ``So that my child is sick less often'' 3 ``So that my child grows better'' 4 ``So that it is cleaner, more hygienic'' 5 ``So that it is more comfortable'' 6 ``That gives me freedom to do other things'' 7 ``A way that is cheaper'' 8 ``So that it is preferred by my husband'' 9 ``So that my breasts are not deformed or hurt'' a Eigenvalue ratio >10.1:1; average competency from factor loadings = 0.82. Assessment of breastfeeding practices in Mexico City 325Bulletin of the World Health Organization, 1999, 77 (4) A total of 37% of mothers said that other reasons to cease breastfeeding would be if they were not well nourished or were not eating well. Some mothers (19%) indicated that they had stopped breastfeeding for a few days and then started again. Supplementation practices When asked what they would feed an infant aged <6 months, mothers' answers depended on the condi- tions (Table 4). For example, breast milk was reported by 95% of the mothers as something to give in the first few days postpartum. When the baby had diarrhoea, 96% of the mothers said they would feed the child with oral rehydration solutions . If the infant was hot/thirsty, 93% said they would give water to the child; if ``mal de esto  mago'' (upset stomach) occurred, 89%would feed the child tea; and in case of fever, 66% would make the infant drink water. Thus, it appeared to be a common practice to reduce breastfeeding and increase the use of supplementary foods during many common childhood illnesses and conditions. Discussions In San Pedro MaÂrtir, 91% of the women in the study chose to breastfeed their infants. Consistent with that behaviour, we found that mothers held a firm belief on the positive value of breastfeeding. They consistently ranked breastfeeding as the best nutri- tion for their infant, to be more hygienic, and resulting in less illness. However, the dominant infant feeding patternwasmixed breast- and bottle-feeding; formula, tea and water were introduced during the first day postpartum before breastmilk appeared, and solid foods well before 3 months of age. Mothers believed that supplementation helped the baby to grow better. Whilemothers in this community accepted that breastfeeding protected the infant, they did not have a strong belief in the value of exclusive breastfeeding. They were convinced that they should supplement neonates' feeding with clear fluids until the milk began to flow during the first days postpartum. In many circumstances, fluids and formulas were added to promote growth, treat colic or other ailments, and provide water, which was thought to be needed by the child. In Guinea-Bissau, Gunnlaugsson et al. found that negative cultural perceptions about colostrum were a major obstacle to early start of breastfeeding (27). This belief was not shared by the mothers in our study. Ladas reported that lack of information was related to all the reasons given for premature supplementation and discontinuation of breastfeeding (28). Why mothers in our study tended to delay the start of breastfeeding was related to their experience in the hospital. Research has demon- strated that successful breastfeeding is related both to the time of initiation (within 2 hours of birth) and to its frequency (at approximately 2-hour intervals) (29). However, the existing postpartum care policies and practices in hospitals do not always reflect the needs ofmothers and infants and aremajor determinants of delayed initiation of breastfeeding (30, 31). The processes of secretion of breast milk and its delivery from mother to infant are not simple. Ability to breastfeed and the adequacy of milk supply are not assumed to be automatic by these mothers, with many mothers terminating if there is ``perceived breast milk insufficiency''. ``Susto'' and ``coraje'' may decrease the milk supply and are also perceived to Table 3. Frequency of problems encountered by 106 breastfeeding women with an infant46 months, based on interviews Problem Frequency of occurrence a Infant's illness 59 (56) b ``Painful breast'' 56 (53) ``Maternal milk was insufficient'' 53 (50) ``Coraje ''c 52 (49) ``Bad quality of maternal milk'' 46 (43) Doctor said to stop breastfeeding 44 (42) ``Susto''c 26 (25) Mother taking medication 25 (24) Maternal illness 25 (24) Mother working outside the house 17 (16) ``Embarrassment of breastfeeding'' 11 (10) a A woman may have had more than one problem. b Figures in parentheses are percentages. c ``Coraje'' and ``Susto'' are terms for folk illnesses or conditions. Fig. 1. Distribution of mothers who changed their breastfeeding behaviour after they encountered a problem during lactation Research 326 Bulletin of the World Health Organization, 1999, 77 (4) affect the quality of the milk (leading to diarrhoea in the infant), so that breastfeeding is undesirable. Harrison et al. report that, among women in Egypt, ``grief milk'' or ``sadness milk'' is expressed manually and the child nursed only after the first milk is discarded (32); these beliefs are related to the ``hot/ humoral'' theory (33, 34), analogous to ``the heat caused by walking while being exposed to sunlight'', after which a mother should cool down before breastfeeding. ``Susto'' and ``coraje'', as well as other reported cultural practices and beliefs among Mexican women (e.g. keeping the shoulders covered to preventmilk fromdrying up;mothers should drink ``atole'' (a corn-based beverage) to producemoremilk, or emotions can be transferred via maternal milk and harm the infant (35)), need to be better understood in relation to breastfeeding. The methodology for assessment and intervention should consider and respect the cultural elements that influence breast- feeding. Mothers should be educated on the value of exclusive breastfeeding during illnesses. Maternal illness, intake of medications, and having a sick child while breastfeeding are associated with reduced frequency or cessation of breastfeeding, so beliefs need to be addressed in any intervention. A major factor in the exclusiveness and overall duration of breastfeeding is the strong influence that doctors had on the study mothers. In this study, 42% ofmothers indicated that at some timewhile they had been breastfeeding, a doctor had advised them to stop doing so, and half of them complied. If exclusive breastfeeding is acknowledged to be the optimal way to nourish an infant (4, 5) physicians must be trained to promote this mode of feeding. The child is also an important influence on the breastfeeding process. A common practice among mothers is to reduce breastfeeding and increase the use of supplementary foods during common child- hood illnesses and conditions. A mother meets the needs of her child by optimizing the quality and quantity of her milk (through her diet, psychological well-being, and physical health) and by providing different fluids as ``good remedies''. Different work- ers have labelled these as traditional or natural practices and identified them in broad cross-cultural comparisons (36). Cultural beliefs and practices need to be well understood to provide effective counsel- ling to breastfeeding women. Breastfeeding is a behaviour that cannot be imposed, but must be chosen. Several factors facilitate or hamper breastfeeding and effective programmes must examine and account for this variety of influences. Most interventions require the collaborative efforts of clinical researchers, epide- miologists, and social scientists. Each discipline offers its unique research perspectives, data-gather- ing tools, analytical methodologies, and conceptua- lizations of data analysis and inference. We used the data from the ethnographic study to guide a peer counselling programme for the promotion of exclusive breastfeeding in the San Pedro MaÂrtir community. To address the negative influence of doctors on breastfeeding, we co-hosted a seminar for community physicians with the La Leche League of Mexico. Maternal concepts that were barriers to exclusive breastfeeding were speci- fically addressed as part of the training of peer counsellors (promotoras). Health conditions and situa- tions in which mothers were likely to reduce breastfeeding were addressed to encourage mothers to give breast milk only. Also the influence of family members was considered by encouraging the promotoras to include key family members in discus- sions about breastfeeding. In this way, the ethnographic study helped focus the community intervention project, whichwas initiated in March 1995. It is intended to publish a detailed description of the intervention study methods and the results elsewhere. Briefly, pregnant women were identified by community census and invited to participate. Enrolledwomenwere allocated randomly to three groups: no intervention (control groups), 3 visits, and 6 visits during pregnancy and early postpartum (experimental group). In these two latter groups maternal concepts that were barriers to exclusive breastfeeding, health conditions, and situa- tions leading mothers to reduce breastfeeding were addressed by the peer counsellors to encourage them to breastfeed only. A total of 130 women participated in the study; 52 in the 3-visit group, 44 in the 6-visit group, and 34 in the control group. At 3 months postpartum, exclusive breastfeeding was practised by only 12% of mothers in the control group, as opposed to 52% in the 3-visit group and 67% in the 6-visit group (P<0.001). Thus, the intervention study demonstrated an increase in exclusive breastfeeding in this urban community through well-designed maternal support including early intervention and repeated contact (37). Community intervention programmes are very difficult, and succeed only when they match the values and needs of the community. WHO has therefore advocated the use of focused ethnographic Table 4. Relationship between perceived infant health status and infant feeding choice % who would feed their <6-month-old infant with the item listed if the infant: Beverage Was a few Had Was thirsty/ Had an Had days old diarrhoea hot upset fever stomach Breast milk 95 69 28 33 46 Bottle milk 19 10 7 5 11 Water 21 49 93 17 66 Tea 62 60 51 89 53 Atole a 1 33 3 6 4 Rice water 3 64 12 24 7 Oral rehydration solution 1 96 9 9 25 Soda 0 3 12 1 3 a Acorn-based beverage commonly used in this community. Assessment of breastfeeding practices in Mexico City 327Bulletin of the World Health Organization, 1999, 77 (4) studies (FES) to provide essential data for designing communication messages, adapting communication training materials for health workers, and for other aspects of programme planning and problem solving (19). Also, WHO has developed data collection guidelines (e.g. HIV/AIDS Rapid Anthropological Assessment Procedures) to develop meaningful and culturally appropriate educational interventions for the prevention and treatment of acquired immuno- deficiency syndrome. We recommend rapid ethno- graphic assessment as a potential and effective foundation for culturally appropriate community interventions to promote breastfeeding. n Acknowledgements We are grateful to the mothers of San Pedro MaÂrtir who participated in this study, as well as Genoveva Figueroa, Rosalba Martõ  nez, Refugio Martõ  nez, Yolanda Martõ  nez and Patricia HernaÂndez for their invaluable help in making this research possible, and Mrs Jane Bravo and La Leche League of Mexico for training and assistance. The study was supported by Wellstart Inter- national/USAID through its Expanded Promotion of Breastfeeding (EPB) Program. Re sume Evaluation ethnographique rapide des habitudes d'allaitement maternel dans la zone pe riurbaine de Mexico L'allaitement maternel exclusif au cours des six premiers mois suivant la naissance pre sente bien des avantages pour la meÁ re et l'enfant sur le plan nutritionnel, immunologique et psychosocial, mais reste une pratique peu fre quente dans de nombreuses re gions du monde. Avant de mettre en úuvre un programme de promotion de l'allaitement au sein dans une zone pe riurbaine de Mexico, on a effectue une enqueà te associant des techniques utilise es en e pideÂmiologie et en ethnographie afin d'identifier et de mieux comprendre les attitudes et les comportements maternels, ainsi que certains facteurs sanitaires et psychosociaux particuliers pouvant dissua- der les meÁ res d'allaiter compleÁ tement leur enfant. Les re ponses obtenues lors d'interrogatoires pilotes ont e te employe es pour mettre au point un questionnaire normalise portant sur les raisons du choix du mode d'alimentation des nourrissons, l'origine des conseils ayant motive ce choix et les obstacles aÁ l'allaitement maternel. On a interroge un e chantillon ale atoire de 150 meÁ res ayant un enfant de moins de 5 ans. Cent trente-six d'entre elles (91%) avaient commence par allaiter leur be be ; mais seules 2% ont continue aÁ l'allaiter compleÁ tement jusqu'aÁ l'aà ge de quatre mois. L'analyse du consensus culturel sur la question a re ve le que lesmeÁ res affirmaient re gulieÁ rement que l'e quilibre nutritionnel, la sante , la croissance et l'hygieÁ ne e taient ce qui avait motive le choix du mode d'alimentation de leur enfant ; les questions de couà t, de commodite et l'opinion du mari avaient moins d'impor- tance (niveau moyen des connaissances culturelles partage es 0,82%; rapport des valeurs propres 10/1). On a obtenu le meÃme ordre de priorite lorsqu'on a interroge les meÁ res sur l'importance relative des diverses raisons ayant motive le choix de l'allaitement au sein. La seule raison invoque e par les meÁ res pour l'allaitement artificiel a e te «il me donne davantage de liberte pour faire d'autres choses»; toutefois, ce facteur n'est pas apparu comme une valeur culturelle importante ayant dicte le choix de l'alimentation du be be . Les me decins ont e te range s parmi les personnes dont les conseils e taient les plus e coute s. Cependant, 42% des meÁ res ont indique qu'aÁ unmoment ou aÁ un autre c'e tait unme decin qui leur avait conseille d'arreà ter l'allaitement; la moitie d'entre elles ont suivi le conseil. Dans la famille, c'e tait la meÁ re de la jeune accouche e qui avait le plus d'influence. On a e galement constate un espacement ou une interruption de l'allaitement au sein lorsque la meÁ re souffrait de ce que la me decine populaire de signe sous le nom de «coraje» (coleÁ re, 52%) ou de «susto» (frayeur, 54%), ou lorsqu'elle a cru ne pas avoir suffisamment de lait (62%) ou du lait de mauvaise qualite (56%), ou encore parce qu'elle (56%) ou l'enfant (43%) sont tombe s malades; 37% des meÁ res ont de clare que si elles n'e taient pas bien nourries, ce serait e galement une raison pour arreà ter l'allaitement au sein. En cas de maladie de l'enfant l'allaitement a e galement e te espace et on a observe une augmentation de l'utilisation des compleÂments alimentaires. En effet, beaucoup de femmes consideÁ rent que le lait maternel est un e leÂment aÁ donner dans les premiers jours suivant la naissance (95%). Lorsque les be be s ont la diarrhe e, les meÁ res les nourrissent avec des solutions de re hydratation orale (96%). Si l'enfant est chaud/a soif, 93% d'entre elles lui donnent de l'eau; en cas d'indigestion, 89% lui donnent du the et, en cas de fieÁ vre, 66% lui donnent de l'eau. L'allaitement au sein est un comportement qui ne peut eà tre impose mais doit eà tre choisi. Les programmes d'intervention communautaires sont treÁ s difficiles aÁ mettre en úuvre et n'obtiennent des succeÁ s que lorsqu'ils correspondent aux valeurs et aux besoins de la communaute . Cette e tude souligne l'importance des valeurs culturelles pour le choix de l'alimentation des nourrissons, indique quels sont les obstacles particuliers aÁ l'allaitement au sein, et fournit une base aÁ partir de laquelle e laborer des interventions visant aÁ promouvoir l'allaitement au sein exclusif dans la population d'e tude. Research 328 Bulletin of the World Health Organization, 1999, 77 (4) Resumen Evaluacio n etnogra fica ra pida de la pra ctica de la lactancia natural en una zona periurbana de la Ciudad de Me xico La lactancia natural exclusiva durante los seis primeros meses tras el parto tiene numerosos efectos beneficiosos de õÂndole nutricional, inmunolo gica y psicosocial tanto para la madre como para el ninÄ o, pero sigue siendo una pra ctica infrecuente en muchas zonas del mundo. Antes de llevar a cabo un programa de promocio n de la lactancia materna en una zona periurbana de la Ciudad deMe xico, se realizo una encuesta a base de entrevistas, combinando te cnicas epidemiolo gicas y etnogra ficas, para identificar y comprender mejor las actitudes y los comportamientos de las madres, asõ como los factores psicosociales y sanitarios especõÂficos que pueden disuadir a las madres de practicar la lactancia natural exclusiva. Se utilizaron las respuestas obtenidas en entre- vistas piloto para elaborar un cuestionario normalizado sobre las razones que habõÂan llevado a elegir la opcio n empleada para alimentar al lactante, sobre las fuentes de asesoramiento y sobre los obsta culos a la lactancia materna. Se entrevisto a una muestra aleatoria de 150 madres con ninÄ os menores de cinco anÄ os. En total, 136 madres (91%) habõÂan empezado a dar el pecho, pero so lo un 2% habõÂa seguido amamantando como forma de alimentacio n exclusiva hasta los cuatro meses. El ana lisis del grado de consenso cultural puso de manifiesto que las madres mencionaban sistema ticamente la nutricio n, la salud, el crecimiento y la higiene como razones principales del tipo de alimentacio n elegida; las razones econo micas, la comodidad y la opinio n del marido eran factores menos importantes (nivel promedio de los conocimientos culturales compartidos: 0,82; raõÂz caracterõÂstica: 10:1). Se obtuvo el mismo orden al pedir a las madres que indicaran la importancia relativa de las razones de la eleccio n de la lactancia natural. La u nica razo n que adujeron para usar el bibero n fue que asõ tenõÂan ma s libertad para hacer otras cosas; sin embargo, ese factor quedo clasificado en baja posicio n en el ordenamiento de los valores culturales que influõÂan en la eleccio n de la manera de alimentar al ninÄ o. Los me dicos quedaron clasificados como la fuente ma s importante de asesoramiento. Sin embargo, el 42% de las madres senÄ alaron que en algu n momento del periodo de amamantamiento un me dico les habõÂa aconsejado abandonar esa pra ctica, y la mitad de esas madres declararon que habõÂan seguido el consejo. En el seno de la familia, y la madre de la nueva madre constituõÂa la influencia ma s importante. Algunas madres tambie n reducõÂan o interrumpõÂan la lactancia natural cuando padecõÂan lo que en la medicina popular local se denomina «coraje» (enojo, 52%) o «susto» (54%), o cuando tenõÂan «poca leche» (62%) o «leche mala» (56%); o bien a causa de enfermedades de la madre (56%) o del ninÄ o (43%); el 37% de las madres consideraban que el hecho de no estar bien alimentadas o de no comer bien tambie n era un motivo para dejar de amamantar. Durante las enfermedades y afecciones propias de la infancia se reducõÂa la lactancia natural y aumentaba el uso de alimentos suplementarios. AsõÂ, por ejemplo, se decõÂa de la leche materna que era algo que debõÂa darse en los primeros dõÂas tras el parto (95%). Cuando el ninÄ o tenõÂa diarrea las madres lo alimentaban con soluciones de rehidratacio n oral (96%); cuando estaba caliente o sediento, el 93% le daban agua; el 89% administraba a los ninÄ os una infusio n cuando tenõÂan problemas de esto mago, y en caso de fiebre el 66% obligaba al ninÄ o a beber agua. La lactancia natural es un comportamiento que no puede imponerse, ha de ser voluntario. Los programas de intervencio n comunitaria revisten muchas dificultades y so lo prosperan cuando se ajustan a los valores y necesidades de la comunidad. El presente estudio subraya la importancia de los valores culturales en la eleccio n de las opciones de alimentacio n del lactante, identifica trabas concretas a la lactancia natural y permite fundamentar las intervenciones orientadas a promover la lactancia natural exclusiva en la poblacio n estudiada. References 1. Cunningham AS, Jelliffe DB, Jelliffe EFP. Breastfeeding and health in the 1980s: a global epidemiologic review. Journal of pediatrics, 1981, 118: 659±665. 2. Dolan SA, Boesman-Finkelstin M, Finkelstin RA. Anti- microbial activity of human milk against pediatric pathogens. Journal of infectious diseases, 1986, 154: 722±725. 3. Kennedy KI, Visness CM. Contraceptive efficacy of lactational amenorrhoea. Lancet, 1992, 339: 227±230. 4. World Health Organization/UNICEF. Innocenti Declaration: on the protection, promotion and support of breastfeeding. Florence, Italy, August 1990. 5. UNICEF. Children and development in the 1990s. World Summit for Children, New York, 29±30 September, 1990. 6. Emery JL, Scholey S, Taylor EM. Decline in breastfeeding. Archives of diseases in childhood, 1990, 65: 369±372. 7. King J, Ashworth A. Historical review of the changing patterns of infant feeding in developing countries: the case of Malaysia, the Caribbean, Nigeria and Zaire. Social science and medicine, 1987, 25: 1307. 8. Burkhalter BR, Marin PS. A demonstration of increased exclusive breastfeeding in Chile. International journal of gynecology and obstetrics, 1991, 34: 353±359. 9. Ryan AS et al. Recent declines in breastfeeding in the United States, 1984 through 1989. Pediatrics, 1991, 88: 719±727. 10. Ruiz FJ, Cravioto A. [Factors which affect the duration of breastfeeding in a cohort of urban mothers]. BoletõÂn Medico del Hospital Infantil de Mexico, 1989, 46(11): 705±708 (in Spanish). Assessment of breastfeeding practices in Mexico City 329Bulletin of the World Health Organization, 1999, 77 (4) 11. Gomez-Dantes H et al. Lactation and weaning patterns in Mexico. 1986 National Health Survey. Salud publica de Mexico, 1990, 32: 725±734. 12. Rodriguez-Garcia R, Aumack KJ, Ramos A. A community- based approach to the promotion of breastfeeding in Mexico. Journal of obstetric, gynecologic and neonatal nursing, 1990, 19: 431±438. 13. Ashkenazi S, Newburg DS, Cleary TG. The effect of human milk on the adherence of enterohemorragic E. coli to rabbit intestinal cells. Advances in experimental medicine and biology, 1991, 331: 173±177. 14. Cervantes LE et al. Campylobacter jejuni receptor analogs present in human milk. Microbiology, ecology and health disease, 1991, 4: S27. 15. Hayani K et al. Concentration of milk secretory immunoglobulin A against Shigella virulence plasmid-associated antigens as a predictor of symptom status in Shigella-infected breast-fed infants. Journal of pediatrics, 1992, 121: 852±856. 16. Ruiz-Palacios G et al. Protection of breastfed infants against Campylobacter diarrhea by antibodies in human milk. Journal of pediatrics, 1990, 116: 707±713. 17. Pickering LK, Morrow AL. Factors in human milk that protect against diarrheal disease. Infection, 1993, 21: 355±357. 18. Morrow AL et al. Protection against infection with Giardia lamblia by breastfeeding in a cohort of Mexican infants. Journal of pediatrics, 1992, 121: 363±370. 19. Gove S, Pelto GH. Focused ethnographic studies in the WHO programme for the control of acute respiratory infections. Medical anthropology, 1994, 15: 409±424. 20. Stanton B et al. Theory-driven behavioural intervention research for the control of diarrheal diseases. Social science and medicine, 1992, 35: 1405±1420. 21. Lewando Hundt GA, Forman MR. Interfacing anthropology and epidemiology: the Bedouin Arab infant feeding study. Social science and Medicine, 1993, 36: 957±964. 22. OcËampo P et al. Needs assessment for reducing infant mortality in Baltimore City: the Healthy Start Program. Southern medical journal, 1993, 86: 1342±1349. 23. Scrimshaw SCM et al. The AIDS rapid anthropological assessment procedures: a tool for health education planning and evaluation. Health education quarterly, 1991, 18: 111±123. 24. Weller SC, Romney AK. Systematic data collection. Qualitative research methods. Vol. 10. Newburg Park, CA, Sage Publications Inc., 1988. 25. Romney AK, Weller SC, Batchelder WH. Culture as consensus: a theory of culture and informant accuracy. American anthropology, 1986, 88: 313±338. 26. Bogartti S. Anthropac Software 3.02. Columbia, Analytic Technologies, 1990. 27. Gunnlaugsson G, Da SilvaMA, Smedman L. Determinants of delayed initiation of breastfeeding: a community and hospital study from Guinea-Bissau. International journal of epidemiology, 1992, 21: 935±940. 28. Ladas AK. Information and social support as factors in the outcome of breastfeeding. Journal of applied behavioural science, 1972, 8: 110±114. 29. Salariya E, Easton P, Cater J. Duration of breastfeeding after early initiation and frequent feeding. Lancet, 1978, 2: 1141±1143. 30. Winikoff B et al. Dynamics of infant feeding: mothers, professionals, and the institutional context in a large urban hospital. Pediatrics, 1986, 77: 357±365. 31. HoustonMJR, Field PA. Practices and policies in the initiation of breastfeeding. Journal of obstetric, gynecologic and neonatal nursing, 1988, 418±424. 32. Harrison GG et al. Breastfeeding and weaning in a poor urban neighborhood in Cairo, Egypt: Maternal beliefs and perceptions. Social science and medicine, 1993, 36: 1063±1069. 33. Gran P. Medical pluralism in Arab and Egyptian history: an overview of class structures and philosophies of the main phases. Social science and medicine, 1979, 13B: 339. 34. ReisslandN, Burghart R. The quality of a mother's milk and the health of her child: beliefs and practices of the women of Mithila. Social science and medicine, 1988, 27: 461. 35. Skeel LS, GoodME. Mexican cultural beliefs and breastfeeding: a model for assessment and intervention. Journal of human lactation, 1988, 4: 160±163. 36. Quandt SA. Patterns of variation in breast-feeding behaviours. Social science and medicine, 1986, 23: 445±453. 37. Morrow AL et al. A randomized trial of the effectiveness of peer counselling to promote exclusive breastfeeding among mothers in Mexico City. In: Proceedings of the International Society for Research on Human Milk and Lactation, 25±29 October, 1997, Plymouth, MA (in press). Research 330 Bulletin of the World Health Organization, 1999, 77 (4)