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

Breast-feeding training for health professionals and resultant institutional changes M.F. Westphal,1 J.A.C. Taddei,2 S.I. Venancio,3 & C.M. Bogus4 Assessed is a breast-feeding training course 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 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 profession- als 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. Introduction Breast-feeding promotion is a basic strategy for controlling infant mortality and morbidity. Based on data from an evaluative study, Rodrigues-Garcia & Schaefer stated that an increase in the length of breast-feeding might save 1 million infant lives in developing countries (1). Other studies demonstrate that compared with exclusively breast-fed infants, non-breast-fed infants have a mean relative risk of morbidity from diar- rhoea of 3.5-4.9 during the first 6 months of life (2, 3). In Brazil, Victora et al. showed that the risk of dying from diarrhoea was 25 times greater during the first 6 months of life among infants who were not breast-fed (4). In Peru similar results were found when the effects of breast-feeding on the frequency of diarrhoea were evaluated (5). Despite such evidence, the incidence and dura- tion of breast-feeding have been decreasing, espe- cially in Latin American countries (1). Efforts are now being made to modify this tendency. I Associate Professor, School of Public Health; and Researcher, Epidemiological and Nutritional Research Center (NUPENS), University of Sao Paulo, Av. Dr. Arnaldo 715, 01246-904, Sao Paulo-SP, Brazil. Requests for reprints should be sent to this author. 2 Associate Professor, Paulista School of Medicine, Federal Uni- versity of Sao Paulo, Sao Paulo, Brazil. 3 Paediatrician, Guilherme Alvaro Hospital; and Researcher, ACA Project, NUPENS, University of Sao Paulo, Brazil. 4 Public Health Educator, Sao Paulo State Health Department; and Researcher, ACA Project, Sao Paulo, Brazil. Reprint No. 5620 In Brazil, the national breast-feeding programme was set up in 1981. Its activities include the creation of labour laws to allow working mothers to breast- feed their babies, implementation of mechanisms to control breast-milk substitutes sold in the market, diffusion of breast-feeding promotion messages through the media, in-service training for health pro- fessionals, use of educational material, and establish- ment of mothers' self-help groups. In 1987, the programme was evaluated in Sao Paulo (6) and Recife (7), where the average length of exclusive breast-feeding was found to be 43-67 days and 14-32 days, respectively. Also, health profes- sionals were found to have increased their under- standing about the theoretical basis for breast- feeding, although they were not able to solve practi- cal breast-feeding problems. These findings indicate that better training programmes on breast-feeding are needed for health staff, especially those involved in maternal and child health care, paying particular attention to the practical aspects.a In 1989, WHO supported the establishment of a centre for breast-feeding training in Brazil to enable health teams to promote breast-feeding activities. WHO's interest in supporting such programmes is mainly based on studies that have shown the impact breast-feeding can have on infant morbidity and mortality caused by diarrhoea (8). When the centre was set up a Brazilian health team consisting of a paediatrician, an obstetrician, a Rea MF, Toma TS. [Assessment of a national programme of incentives for breast-feeding: the practice of professionals]. Unpublished, Sao Paulo, 1988 (in Portuguese). Bulletin of the World Health Organization, 1995, 73 (4): 461-468 © World Health Organization 1995 461 M.F. Westphal et al. and a nurse were sent for training to the Wellstart Center, San Diego, CA, USA (9). Since completing training, the team has been operating at a public hos- pital (Guilherme Alvaro Hospital) in Santos, Sao Paulo. The hospital is linked to Santos Medical Sciences College, where breast-feeding promotion activities have been carried out for 18 years. These activities include the participation of paediatricians in antenatal care, encouragement of breast-feeding in delivery rooms, the practice of rooming-in, a special- ized outpatient clinic for the promotion of breast- feeding, as well as training activities for medical students. Recently the hospital has received the "Baby-friendly hospital" award from UNICEF. The Center for Health Professionals Training, Santos Lactation Center (SLC), is also based at this hospital. Training courses have been offered at the SLC since 1990 and feedback from the 120 health profes- sionals who have attended them suggest that they have introduced changes in the routine in their insti- tutions leading to an increase in the length of time mothers breast-feed their children. In view of the need to carry out a systematic evaluation of the training centre and its impact on breast-feeding promotion, we carried out a study that, inter alia, had the following objectives: - to identify the positive and negative characteris- tics of the various topics covered in the SLC course and hence improve its effectiveness; and - to quantify and qualify the structural changes that had occurred in institutions whose staff had attended the training course, by comparing them with similar institutions that had not been exposed to it. Theoretical framework The "ten steps for successful breast-feeding" recom- mended by WHO/UNICEF (1) (Table 1) were used to assess the contents and results of the breast- feeding promotion programmes. This approach was used because the "ten steps" have been widely adop- ted and UNICEF has identified them as a strategy for implementing programmes for the promotion of breast-feeding (12) The evaluation of institutional changes resulting from the participation of staff in the SLC course was carried out using the domain theory (10). Kouzes & Mico's theory of institutional domains states that institutions consist of the following distinct domains: the policy, management, and service domains, each of which operates according to different principles, success measures, structural arrangements, and work modes (or technologies). The simultaneous existence Table 1: Ten steps for successful breast-feeding Step number Step 1 Have a written breast-feeding policy that is routinely communicated to all health staff. 2 Train all health care staff in skills necessary to imple- ment this policy. 3 Inform all pregnant women about the benefits and man- agement of breast-feeding. 4 Help mothers initiate breast-feeding within half an hour of birth. 5 Show mothers how to breast-feed and how to maintain lactation even if they should be separated from their infant. 6 Give newborn infants no food or drink other than breast milk unless medically indicated. 7 Practise rooming-in - allow mothers and infants to remain together 24 hours a day. 8 Encourage breast-feeding on demand. 9 Do not give artificial teats or pacifiers to breast-feeding infants. 10 Foster the establishment of breast-feeding support groups and refer mothers to them upon discharge from hospital or clinic. of the three domains within an institution, with their different and often disharmonious interests, causes conflicts, since each one aims at the legitimation of its own standards, producing dissonant and contra- dictory interactions. Based on this framework, we devised an opera- tional model to evaluate the influence of the policy, management and service domains on the impact of the SLC course on the implementation of routines to enhance breast-feeding. Methodology The effects of the health professionals' training on their institutions, as well as on the average length of time infants born there were breast-fed, were asses- sed in eight health institutions, group-paired, and assigned at random to either the experimental group (exposed to the SLC course) or the control group. In order to achieve comparability, the eight institutions satisfied the following criteria: they were public or philanthropic; they were located near the city of Sao Paulo (within 100 km); they had not pre- viously been exposed to a similar course; they had professional staff available to attend the course on a full-time basis for a 3-week period; and they had at least two births per day in their maternity facilities. The course was evaluated, inter alia, in order to control for the intermediate (institutional) and final WHO Bulletin OMS. Vol 73 1995462 Breast-feeding training for health professionals results. The methodology used to assess the course and the institutional changes that followed are dis- cussed below. Evaluation of the SLC course Participant observation was used to become familiar with and to evaluate the teaching-learning process employed to train the teams that had been randomly assigned to the experimental group (13). Two re- searchers monitored and participated in all the activ- ities, alongside the teams of three professionals, one from each of the four institutions in the experi- mental group. In order to systematize the analysis of the teach- ing-learning process, each unit of the course was assessed separately by the researchers and by each of the participants, according to the following aspects: contents; sequence of activities; use of audiovisual resources; teaching performance; bibliographic mate- rial; work-load; classroom performance; students' participation; evaluation technique; and adequacy of the methodology in relation to the proposed goals. Each item was rated as very bad, bad, sufficient, good or very good. The course content was evaluated by applying WHO/UNICEF's recommendations for successful breast-feeding (11). Two aspects were taken into account in evaluat- ing the participants final results: improvements in trainees' knowledge (determined using pre- and post- course tests); and changes in their attitudes towards breast-feeding (determined using group dynamics) carried out at the beginning and at the end of the training. Evaluation of institutional changes Institutional changes were evaluated by comparing the situation before exposure to the course with that observed 6 months after it had been completed. For this purpose, the participating institutions - both the experimental and control - were visited twice, on each occasion over a 2-day period, the objectives of which are discussed below. Institutional evaluation questionnaire. This instru- ment is based on the Wellstart's questionnaire, and is designed to collect data that demonstrate whether institutions comply with the ten steps for successful breast-feeding (11). Basically it consists of conduct- ing interviews with professionals in the health team who hold managerial positions or who have direct contact with mothers and newborns; interviewing women who have just delivered; observing the vari- ous maternity areas (rooming-in, nursery, delivery- room, and antenatal clinic); and interviewing preg- nant women at the antenatal clinic.b Individual interviews. Individual interviews were held 6 months after the SLC course with directors in charge of the process of decision-making policies in the participating institutions; with administrative managers in charge of financial and human resources' allocation for attaining the above-men- tioned decisions; and with professionals in charge of antenatal, rooming-in, nursery and outpatient areas. The interviews, which were tape-recorded and then transcribed, followed previously established guide- lines and contained either open-ended or yes/no questions, and were designed to facilitate a detailed qualitative analysis of the data collected (3). Focus-group sessions. Structured discussions with 6-15 persons, directed by a moderator were held 6 months after the course with the staff involved in the institution's routine activities (14). The discussions were tape-recorded and analysed later. Our main concem was to identify how the professionals per- ceived the breast-feeding promotion activities. A total of 16 focus-group sessions were held, two in each of the institutions participating in the study; in each case one session was held with the front-line personnel, and the other with graduate-level profes- sional staff who were involved in the care of the newbom. Analysis of the data Institutional questionnaire data. Data gathered through application of the institutional question- naires were analysed in two ways, as described below. First, a qualitative analysis was made based on structured observations derived from the researcher's notes made after each visit. The notes were used to generate a chart showing an individual institution's compliance with each one of the "ten steps", taking into account a greater (5) or lesser (0) adherence to the required pattem. Second, a quantitative analysis was carried out, based on the total score (the "ten-step score"). This score was obtained from the variables in the institu- tional questionnaire, grouped according to each one of the ten steps. A weighting was assigned to each variable, depending on the degree of reliability of the source of information. For example, for the analysis b Venanclo Si. ACA project: operational and impact evaluation of a breast-feeding promotion course for health professionals. (Institutional Report No. 1). Sao Paulo, NUPENS/WHO, unpublished document, 1992. WHO Bulletin OMS. Vol 73 1995 463 M.F. Westphal et al. of information on breast-feeding (step 5), mother's reports were given a greater weighting than those of the staff. In contrast, to check whether newborns were receiving food other than mother's milk (step 6), a greater weighting was attached to observations made in the nurseries than to mothers' reports. The total score permitted a comparison of the situation before and 6 months after the course.C In order to identify the factors that might be influencing the implementation of the expected insti- tutional changes, we analysed qualitatively individu- al interviews and those with the focus groups, both tape-recorded and transcribed. The influence of the policy and management domains on the impact of the SLC course was assessed using the tape-recorded and transcribed interviews held individually with general, clinic, and/or administrative directors of the experimental and control institutions. Evaluation of the service domain was carried out using data gathered from individual interviews with the health professionals responsible for the various maternity areas, and through focus-group interviews with graduate profes- sionals and auxiliary personnel. Because of the great amount of data gathered for the service domain, we compared the various institu- tions using a comprehensive thematic analysis of the perceptions of the graduates and auxiliary staff about the following: the incentive to breast-feed in antenatal care; the incentive to breast-feed in delivery care; the incentive to breast-feed in nursery areas; the incentive to breast-feed in rooming-in areas; the incentive to breast-feed in outpatient clinics; human resources' influence, both positive and negative; the acceptance of the activities developed, expec- tations related to breast-feeding duration, and the reasons for weaning; and whether there was a written breast-feeding pro- gramme. Results Course evaluation Structure and organization of the course. The course consisted of 45 units that covered theoretical and practical aspects of breast-feeding over a 133- c Venancio Si. ACA project: operational and impact evaluation of a breast-feeding promotion course for health professionals. (Institutional Report No. 2). Sao Paulo, NUPENS/WHO, unpub- lished document, 1992. hour period (14 days, full-time schedule). A total of 66% of this time consisted of lectures, with the re- mainder being spent on practical activities, realiza- tion of these practices, visits to services that promote breast-feeding, and video-cassette sessions. The ma- jority of the instructors (57%) were SLC profession- al staff, with the remainder being invited from other institutions. The group of trainees consisted of seven physicians (6 paediatricians and one obstetrician) and five nurses. Process and results evaluation. Analysis of the course content in terms of the ten steps for success- ful breast-feeding indicated that all the important topics were covered, although the emphasis on par- ticular issues varied. Steps 5 and 9 received the most emphasis followed by steps 3 and 1; step 10 was the least emphasized (Fig. 1). For the teaching-learning process, both evalua- tion averages (the students' and the researchers') were close to the maximum score and there were no significant differences between the two evaluations for the majority of the activities. These data there- fore indicate that the course was efficiently run. The pre- and post-course test results showed that there was, in general, an improvement in the student's knowledge on breast-feeding (Fig. 2). The assessment of the change in the student's attitudes, based on analysis statements they made in two group discussions (one at the beginning and the other at the end of the course), indicated that they intended to change the routines practised in their institutions. Institutional evaluation Comparison of the findings of the structured obser- vations made at the first and second visits to the study institutions indicated that changes had occur- red in institutions whose staff had attended the SLC Fig. 1. The SLC course: number of hours spent on each of the "ten steps for successful breast-feeding". 60- 50- 48 40- 34 36 0 3030 28 26 24 24 6 22 z 20- 10- 1 1 2 3 4 5 6 7 8 9 10 Step 464 WHO Bulletin OMS. Vol 731995 Breast-feeding training for health professionals Fig. 2. Evaluation of the SLC course according to the number of correct answers given by students In pre- and post-course tests (average number of correct ans- wers: pre-course: 20.27 ± 7.41; post-course: 26.92 ± 2.10). c Pre-course 35 U Post-course 25 - 891 11 z Student course. Changes were noted particularly in relation to steps 2 and 10. No such changes were observed for the institutions in the control group (Fig. 3). Institutions in the exposed group had a higher ranking in the ten-step score test than those in the control group, with the teaching institutions achiev- ing the best results (Fig. 4). In general, lack of cohesion between the policy, management and service domains was a feature of all the institutions studied. In the service domain the possibilities for change were limited by inter-personal conflicts, which were aggravated by the relative autonomy of the health professionals in carrying out their functions. This characteristic was observed chiefly in exposed insti- Fig. 4. Institutional changes in the study hospitals, according to their score In the "ten steps for success- ful breast-feeding" (Exp. = exposed institution; Cont. = control institution). 3 Exp. S Cont. 6 Exp. 8 Cont. 9 Exp. 11 Cont. 10 Exp. 4 C Pair 1 Pair 2 Pair 3 Pair 4 tutions, where the changes introduced were imple- mented to a greater or lesser degree, and was more evident among graduate-level professionals. The great majority of the frontline personnel working in all the institutions that were studied lacked information about how to receive training on breast-feeding. The professional health staff expressed a desire to improve their technical qualifi- cations in order to enable them to perform routine Fig. 3. Institutional changes in the study hospitals, according to structured observations (exposed group = Institutions 3, 6, 9, and 10; control group = institutions 4, 5, 8, and 11): * = first evaluation visit; * = 2nd evaluation visit. Institution Stepa 10 04 09 11 03 05 06 08 1 * * * t 2 ** ** **+ ** 3 ** 4 ** *** * 5 *4 *4 *** 0* 0 0 *4t 0* 6 ***t* *0*00* t0** 0 *4 0* 00 7 *0*0* *0*0* *00* *0* *0* 0* *0 8 *** ****+ *0* *0* *0* 0* *0* 9 **** *0*0 *4* 0* 0* 0* 0* 10 ** *0* **** 0*§ *** ** ** a See Table 1 for details of steps. WHO Bulletin OMS. Vol 73 1995 465 M.F. Westphal et al. activities, including dealing with the mothers of newborns. It was the unanimous opinion in the experimen- tal institutions that it is easier to implement new breast-feeding incentive activities (e.g., the creation of outpatient clinics for promotion of breast-feeding) than to promote changes in previously established routines (e.g., breast-feeding in the delivery-room, the non-maintenance of the breast-feeding of normal newborns in the nursery, the non-provision of food other than mother's milk to newborns, and the non- use of artificial teats). Our findings indicate that there were some diffi- culties experienced by the health professionals who had attended the SLC course in gaining support for their proposed changes. These difficulties may have arisen because the professionals concerned had insufficient authority, they had to perform adminis- trative tasks that limited the time needed to prepare and develop training courses, they lacked the support necessary for such initiatives, or they had encoun- tered obstacles in obtaining the necessary financial and human resources. All the health professionals stressed the impor- tance of promoting breast-feeding during antenatal care; none the less, few hospitals provided such care. The lack of antenatal care in all institutions and/or the lack of commitment to counsel mothers on the benefits of breast-feeding and to teach breast-feeding practice are major factors that hinder the promotion of such feeding. Discussion and conclusions The course implemented at SLC has, in general, the same structure, organization, and content as that run by the Wellstart Center in San Diego (15). However, the selection of participants who attend the SLC course differs from that run by the Wellstart Center, in that they do not necessarily work in teaching insti- tutions. Although we consider that this difference does not hinder the implementation of breast-feeding promotion programmes in health service institutions, it is probably easier for teaching institutions to attain the goals proposed by the SLC course. Choosing the right people to participate in the training course is important. The absence of obstetri- cians and the participation of health professionals who had a heavy workload, and consequently were not able to devote themselves to implementing the topics covered in the course, limited the attainment of the desired institutional changes. Since the total scores attained by the institutions in the exposed group showed that they had had some difficulty in implementing programmes for the pro- motion of breast-feeding, we believe that the SLC course should, in future, stress the importance of initiating feasible projects that can be realized within the prevailing conditions in each institution (16). Although the health teams found it difficult to devote 3 weeks to attend the training course, this period is required to develop all the aspects related to breast-feeding and to allow changes to occur in trainees' attitudes towards breast-feeding (17). The application of pre- and post-course tests to gauge improvements in trainees' knowledge has been used by other training courses for the promo- tion of breast-feeding (17); this reinforces the ad- equacy of the instrument to assess the results of the SLC course. Armstrong stresses the importance of following up trainee teams (17). Therefore, the SLC could carry out a systematic evaluation of the institutions that send staff to attend its course in order to guaran- tee the implementation of breast-feeding promotion programmes. Since the exposed institutions had introduced changes that encouraged breast-feeding, but had not fully attained all the "ten steps", we searched for hypotheses to explain our findings. For this purpose, individual interviews and focus-group sessions were analysed using the framework proposed by Kouzes & Mico (10). In this way, we identified that the lack of cohesion between the various institutional domains can hinder the implementation of breast- feeding promotion programmes. According to Stok- amer, the absence of an institutional policy of action and administrative support makes the success of such programmes almost impossible (18). Our data indicated that the service domain played a determinant role in governing the feasibility or not of the breast-feeding programme. Since both the policy and the management domains could have either positive or negative influences on the pro- gramme, it would appear that the policy domain, whose function is to define institutional guidelines, carries greater weight in programme implementation. Implementation of the breast-feeding program- me was not influenced directly by the administrative sector, but was indirectly - through its distribution policies, including human resources. The impor- tance of a good salary for the health professionals involved in activities related to breast-feeding and the availability of resources for training courses, publications, and research have been emphasized by Armstrong (17). The internal lack of cohesion in the service domain could, on its own, make the implementation of any programme impossible. In the study, the diffi- culties experienced with the graduate-level profes- sionals, especially the physicians, could have arisen because of their academic training, which empha- WHO Bulletin OMS. Vol 73 1995466 Breast-feeding training for health professionals sized treatment rather than prevention. In this sense, the physicians placed a low emphasis on events such as birth and breast-feeding (19). The presence of only one obstetrician among the group of professionals in the institutions exposed to the training course reinforces the notion that little attention was paid to breast-feeding by obstetricians. Many studies indicate that breast-feeding receives little attention in antenatal care (19, 20). Therefore, in the exposed institutions that we studied there was no institutional policy that gave priority to the breast-feeding programme and allocated human and physical resources to it. Also, many of the profes- sionals who participated in the SLC course did not hold any decision-making power and consequently could not increase the number of professionals engaged in the breast-feeding programme. These various factors illustrate the lack of cohesion among the different domains, which do not favour adoption of innovations and changes in working routines. Steps ought to be taken to change the content of the courses taken by health professionals during their academic training, especially to sensitize obstetri- cians to breast-feeding and make them more recep- tive to its promotion. The auxiliary health personnel were the most homogeneous group and also the most receptive to implementing the breast-feeding programme; how- ever, in the study institutions they were not involved at the planning level or prepared to accept new atti- tudes and functions. Instead, they were merely as- signed responsibilities to perform tasks related to the programme. Greater emphasis should be placed on training the auxiliary personnel; studies have demon- strated the favourable impact of such training on hospital policies and rules, leading to an increase in the prevalence of exclusive breast-feeding (16). The creation of outpatient clinics in the exposed institutions to promote breast-feeding probably arose because of the emphasis given by SLC to this activ- ity. We consider this to have been a positive step since the postnatal care services tend to approach breast-feeding inadequately. One of the advantages of this new activity is that it can be developed by a single professional, and does not require the involve- ment of the whole team. On the other hand, some difficulties were clearly experienced by the institutions in changing some of the maternity routines, such as providing a newborn with water and tea, no breast-feeding in the delivery- room, and observing the newborn in the nursery for 6 hours after birth. This could have arisen because these routines are widely used (1_49)d and because d See footnote a, p. 461. changes in routine presuppose the involvement of all the health professionals in contact with mothers and newboms. The negligence showed by the exposed institu- tions in failing to establish written breast-feeding programmes and in fostering breast-feeding support groups within the community could be a cultural problem. Though all the exposed institutions have outpatient clinics for follow-up, there is difficulty in meeting the demand, which is also true for the ante- natal care. These findings indicate that breast-feeding pro- motion programmes should be extended to the com- munity as a whole. Another point to be considered is whether the second evaluation of the exposed institu- tions should have been carried out a longer period after the first, since the health professionals might have displayed some resistance to implementing new routines favouring breast-feeding. In summary, it can be concluded that the SLC course has been very efficient in providing technical know-how on breast-feeding; however, it could benefit from introducing more topics on strategies for implementing breast-feeding programmes, includ- ing a critical analysis of the institutional changes required and the need for better cohesion among the institutional domains. Health departments, as well as the federal, state and municipal governments in Brazil should contrib- ute their political support to help health institutions achieve better results from training courses, by pro- viding hospitals with favourable operational condi- tions, particularly those to do with health personnel, bearing in mind the fundamental role that such indi- viduals play in implementing programmes. Acknowledgements We are deeply indebted to the WHO Diarrhoeal Disease Control Programme for supporting the project (Grant No. 91064). We also acknowledge the cooperation of the Santos Lactation Center, and the hospital directors and the health professionals of the eight participating hospi- tals, who contributed to the success of the project. Resume Formation des professionnels de sant6 en matiere d'allaitement maternel et modifications des pratiques au niveau des maternit6s Cette etude 6value un cours de formation en matiere d'allaitement maternel dispens6 aux pro- fessionnels de sant6 par le Santos Lactation Cen- WHO Bulletin OMS. Vol 73 1995 467 M.F. Westphal et al. ter (SLC), Santos, Sao Paulo, Br6sil, et son impact sur la mise en ceuvre des programmes d'a- laitement au sein dans les maternit6s. Ces pro- grammes sont bas6s sur les "Dix conditions pour le succes de l'allaitement maternel" recomman- dees par l'OMS et l'UNICEF. Huit maternit6s ont particip6 a l'6tude: chacune des quatre maternit6s incluses par tirage au sort dans le groupe exp6ri- mental a envoy6 3 professionnels de sante au SLC pour y suivre un cours de 18 jours, et les quatre autres ont servi de groupe t6moin. L'6tude avait pour objectif d'6valuer le respect des "dix conditions" pr6cit6es avant et 6 mois apres le cours, en utilisant une 6chelle de notation. Les 6tablissements du groupe exp6rimental ont enre- gistr6 une augmentation de leur note, contraire- ment aux 6tablissements du groupe temoin. Des entretiens et des reunions par petits groupes ont egalement eu lieu avec les professionnels de sant6 de 1'ensemble des maternit6s participant a l'etude. Les r6sultats ont ete analyses selon la th6orie des domaines institutionnels de Kouzes & Mico, qui a permis d'identifier les difficult6s ren- contr6es dans la promotion des changements en matiere de pratiques hospitalieres. II a 6t6 conclu que le cours 6tait efficient en ce qu'il permettait aux professionnels de sant6 d'etre capables de promouvoir les pratiques d'allaitement maternel. Toutefois, pour r6ussir a r6ellement mettre en ceuvre des programmes d'allaitement maternel, ils devront egalement d6velopper leurs competences pour pouvoir mettre en pratique les connaissances acquises pendant le cours et faire participer 1'en- semble du personnel de la maternit6 aux activit6s. References 1. Rodriguez-Garcia R, Schaefer LA. [Promotion of breast-feeding and the participation of nursing per- sonnel]. Boletin de la Oficina Sanitaria Panamerica- na, 1991, 111: 1-15 (in Spanish). 2. Feachem RG, Koblinsky MA. Interventions for the control of diarrhoeal diseases among young chil- dren: promotion of breast-feeding. Bulletin of the World Health Organization, 1984, 62: 271-291. 3. Feachem RG. Preventing diarrhoea: What are the policy options? Health policy and planning, 1986, 1: 109-117. 4. Victora CG, Vaughan JP, Lombardi C. Evidence for protection by breast-feeding against infant deaths from infectious diseases in Brazil. Lancet, 1987, 2: 319-321. 5. Brown KH et al. Effect of continued oral feeding on clinical and nutritional outcomes of acute diarrhoea in children. Journal of pediatrics, 1988, 112: 191-200. 6. Rea MF, Berqu6 ES. Impact of the Brazilian nation- al breast-feeding programme on mothers in Greater Sao Paulo. Bulletin of the World Health Organiza- tion, 1990, 68: 365-371. 7. Rea MF. The Brazilian national breast-feeding pro- gram: a success story. International journal of gyne- cology and obstetrics, 1990, 31 (suppl. 1): 79-82. 8. Hogan R, Martines J. Breast-feeding as an inter- vention within diarrheal diseases control programs: WHO/CDD activities. International journal of gyne- cology and obstetrics, 1990, 31 (suppl. 1): 1 1 5-119. 9. Wellstart: the San Diego Lactation Program. Statement of corporate capabilities. San Diego, CA, Wellstart: the San Diego Lactation Program, 1990. 10. Kouzes JM, Mico PR. Domain theory: an introduc- tion to organizational behavior in human service organizations. Journal of applied behavioural sci- ence, 1979, 15: 449-469. 11. Nakajima H, Grant JP. Appendix 3: Protecting, pro- moting and supporting breast-feeding: the special role of maternity services; a joint WHO/UNICEF statement. International journal of gynecology and obstetrics, 1990, 31(suppl. 1): 171-183. 12. Ellis DJ. Supporting breast-feeding: How to imple- ment agency change. NAACOGS clinical issues, 1992, 3: 560-564. 13. Scrimshaw SCM, Hurtado E. Rapid assessment procedures for nutrition and primary health care. United Nations University, Tokyo; UNICEF/United Nations Children's Fund; Latin American Center, UCLA, CA, 1987. 14. Ramirez AG, Shepperd J. The use of focus groups in health research. Scandinavian journal of primary health care, supplement, 1988, 1: 81-90. 15. Nailor A. Professional education and training for trainers. International journal of gynecology and obstetrics, 1990, 31 (suppl. 1): 25-27. 16. Rodriguez-Garcia R, Shaffer L, Yunes J. Lactation education for health professionals. Washington, DC, Pan American Health Organization, 1990. 17. Armstrong HC. Breast-feeding promotion: training of mid-level and outreach health workers. Internation- al journal of gynecology and obstetrics, 1990, 31(suppl. 1): 91-103. 18. Stokamer CL. Breast-feeding promotion efforts: why some do not work. International journal of gynecol- ogy, 1990, 31(suppl. 1): 61-65. 19. Ellis DJ. The impact of agency policies and proto- cols on breastfeeding. NAACOGS clinical issues, 1992, 3: 553-559. 20. Breast-feeding promotion in maternity care services: hospital programs as a catalyst to policy change. International journal of gynecology and obstetrics, 1990, 31(suppl. 1): 43-45. 468 WHO Bulletin OMS. Vol 73 1995