Effective interventions for the promotion of breastfeeding and healthy complementary feeding in the context of Primary Health Care

Abstract Objective: To develop a rapid review on effective actions for the promotion of breastfeeding and healthy complementary feeding in primary health care and to summarize a list of actions and their elements for implementation. Data source: The review included systematic reviews on the effectiveness of interventions to promote breastfeeding and/or healthy complementary feeding for mothers and other caregivers, and/or professionals who work with this population, in comparison with any usual approach or none. Data synthesis: A total of 32 systematic reviews were included in the evidence synthesis. Ten types of interventions were evaluated in systematic reviews on promotion of breastfeeding and four types of interventions on promotion of healthy complementary feeding. The synthesis allowed six aspects to be discussed, and these must be considered to increase the chances of interventions’ impact: type of intervention, target audience, timing of intervention, actors that can implement it, strategies and methods of intervention, and intensity of intervention. Conclusions: It was possible to assemble a list of actions whose effectiveness has already been demonstrated, providing elements for local adaptations. Evidence is expected to support and strengthen the implementation of programs aimed at promoting breastfeeding and complementary feeding in primary health care.


INTRODUCTION
Adequate nutrition in the first years of life can impact children's development and health.Increasing breastfeeding (BF) practices could save the lives of more than 800,000 children.It is recommended to breastfeed in the first hour of life, offer exclusive breastfeeding (EBF) for the first six months and continue to breastfeed the child until two years of age or more. 1,2From six months onwards, children should receive healthy complementary feeding (CF) to meet the nutritional needs for their development, which could prevent 100,000 deaths of children under five years old annually. 1,2e are still far from reaching infant feeding targets of the Global Agenda and its 2030 Sustainable Development Goals. 1,2In Brazil, several actions to promote healthy BF and CF are proposed; 3 within the scope of Primary Health Care (PHC), the implementation of the Brazilian Breastfeeding and Feeding Strategy (EAAB, acronym in Portuguese) is recommended for the promotion, protection and support of BF and healthy CF.Using critical-reflection methodology, the EAAB trains tutors to support Basic Health Units' (BHU) teams in organizing the work process to implement interventions aimed at promoting BF and healthy CF according to the local context. 4y 2019, more than 48 thousand PHC professionals were involved in EAAB workshops.However, EAAB monitoring data do not inform whether interventions being developed in BHUs are supported by evidence of positive impact on BF and healthy CF indicators. 5,6n 2019, the General Coordination of Food and Nutrition (CGAN, acronym in Portuguese) and the Coordination of Child Health and Breastfeeding (COCAM, acronym in Portuguese) of the Ministry of Health started a project to strengthen the implementation of EAAB based on Evidence-Informed Policies for health decisions.7 So, the objectives of this study were to develop a rapid review on effective interventions for the promotion of BF and healthy CF in PHC and to synthesize a list of actions and their elements for implementation, in order to support BHU teams.

METHOD
A rapid review was prepared based on a previous protocol, using methodological shortcuts to recommended steps for the elaboration of systematic reviews (SR).This type of review has the advantage of shortening time of evidence delivery for decision-making in health, maintaining the transparency, systematic process and reproducibility of the SR model. 8n this study, the system PICOS-Population; Intervention; Comparison; Outcomes; and Study Design 9 -(Table 1) was used as inclusion criteria and in the construction of the research question: "Which interventions are effective for the promotion of BF and healthy CF in PHC?".Publications in English, Spanish and Portuguese were included with no restriction regarding the year of publication.Interventions offering food supplements or whose sole outcome was the child's growth were not included.Scoping reviews, integrative reviews, synthesis of evidence for policies, health technology assessments, economic evaluations and primary studies were excluded.
The indexed databases MEDLINE (via PubMed), Regional Portal of the Virtual Health Library (VHL), Epistemonikos, Embase and Health Evidence were searched.The search strategies strengthen the implementation of programs aimed at promoting breastfeeding and complementary feeding in primary health care.

Intervention
Interventions to promote breastfeeding and/or healthy complementary feeding.

Comparator
Any usual approach or none.

Outcomes
Improvement in prevalence and/ or duration of breastfeeding and exclusive breastfeeding and healthy complementary feeding practices.

Study design
Systematic reviews and overviews, with or without meta-analysis.
included structured keywords based on the acronym PICOS, using MeSH and DeCS terms.The terms and synonyms used were: "Primary Health Care", "Breastfeeding" and "Physiological Phenomena of Infant Nutrition", and their translations into Portuguese and Spanish.A resource of filters available in the databases was used to select SR or Overview studies.
Exclusion of duplicates and study selection were based on titles and abstracts, by means of an application for bibliographic management, Rayyan QCRI. 10 Eligible studies were selected for full reading and data was input to an Excel ® spreadsheet.As this is a rapid review, data selection and extraction were not performed by peers, but individually by researchers experienced in conducting review studies and evidence syntheses.The country in which the intervention was carried out was considered in the analysis and interpretation of data of all articles.The methodological quality of the included studies was evaluated by peer review using AMSTAR-2-A Measurement Tool to Assess Systematic Reviews. 11

RESULTS
The search, carried out in October 2019, identified 700 references, of which 596 were screened considering title and abstract after the exclusion of duplicates.Sixty-four articles were selected for full reading, of which 32 were excluded  ; the justifications are presented in Table 2. Thirty-two SR-type studies were included in the qualitative synthesis, of which 26 evaluated interventions to promote BF;  four assessed interventions to promote healthy CF; [71][72][73][74] and two 75,76 addressed interventions for both (Figure 1).
5][66][67][68][69][70] Positive results were identified for the combination of trained professionals and lay people, 63,64,66,68,69 while an SR 64 reported that interventions by health professionals were more effective than when only by laypersons.Training to promote exclusive breastfeeding (EBF) and BF was important in low-income countries. 68,69Positive effects were seen for Panjwani et al. 23 Park et al. 24 2019

Interventions to promote breastfeeding
Sguassero et al. 25 2007 Not found for full reading Kim et al. 26 2017 Mushtaq et al. 27 Renfrew et al. 28 Witten et al. 29 2017

Did not address interventions
Blaney et al. 30,31 Garcia et al. 32

Interventions in specific contexts
5]68 Isolated offer of printed material, without counseling, or strategies with no or brief face-to-face interaction (e.g.leaflets delivered to mothers, telephone support, and non-continuous actions) had no results.Therefore, interventions in health systems and communities by trained professionals and lactation counselors are effective to promote EBF and BF. 2. Support by health professionals, laypersons or peers, including counseling: these interventions were identified in eight SRs and had positive effects when the actions were adapted to local needs, conducted by trained people (professionals, lay people and mothers), provided face-to-face and routinely, during pregnancy or in the postpartum period. 46,48,52,53Support in home visits was effective; the authors recommend performing at least one prenatal visit, one postpartum visit (days 1-3), and continuing visits beyond the first month of postpartum.Higher frequency of visits increased success, although a threshold appears to be reached with seven visits. 49,51,75n contrast, when peer support was provided at low intensity (fewer than five planned contacts), actions to promote BF were ineffective. 50,51,756][47][48][49][50] Prenatal interventions significantly increased the rate of short-term BF, while the combination of pre-and postnatal interventions increased the rates of medium-and long-term BF. 48Although there is little evidence for effectiveness of group activities, these actions seem to increase the rates of initiation and duration of BF in developed countries. 475][56] There was a significant effect on increasing breastfeeding confidence when strategies combined individual and group actions, face-to-face and overthe-phone sessions, in the pre-and postnatal periods. 55 program based on Bandura's social cognitive theory and adult learning principles increased self-confidence in BF within four weeks of delivery. 56In one SR, the interventions had no effect in the medium and long term. 54Evidence is limited on the positive effect of these interventions on BF initiation, BF duration, and BF attitudes and knowledge.The results come from studies conducted in developed regions with wide access to the internet or cell phone. 58,767. Father involvement: father involvement appears to be promising for BF promotion. 61,62The probability of EBF for six months and other outcomes such as reduced use of infant formula was doubled when the target was mothers and fathers. 62In two SRs, actions involving fathers included information-education-communication actions, such as streaming videos and discussing them, individual or group counseling, mass and electronic media, delivery of informative materials, and a combination of these.Such actions were conducted by health professionals (physicians, pediatricians and community health workers) in hospitals, PHC services, and households. 61,62Overall, counseling, education and information actions had an impact on several outcomes: timely introduction of CF, quantity, frequency and diversity of foods offered, and responsive feeding. 71An intervention based on eight home visits for pregnant women and monthly home visits in the first year of the infant's life was successful in decreasing anemia rates, promoting maternal knowledge about nutrition, improving feeding practices, BF rates and infant growth at 12 months of age. 72

Behavioral and non-behavioral interventions: two SRs
presented these interventions to improve diet diversity and vegetable acceptance.Approaches included repeated and varied exposure to vegetables, introduction of healthy CF with vegetables, improvement of parents' eating practices, responsive feeding, parenting, and support for lactation and mothers' health. 73,74Introducing fruits and vegetables at the beginning of CF and promoting repeated exposure were successful strategies to improve acceptance of this food group.All studies that investigated the association between EBF duration and subsequent vegetable intake, reported significant positive associations. 73Home visits focused on improving eating practices of both the parents and their children had positive effects on fruit and vegetable intake. 74No effects were found for vegetable intake using the Baby-Led-Weaning (BLW) technique or taste learning compared to spoon-feeding. 73. Community workers: an SR identified the impact of home visits carried out by women in the community educated on BF practices and healthy CF.Visits took place after the child was born: once between the third and fifth day of life, then again between the seventh and tenth day of life, on the 21st day, at 1.5 months of age, and then once a month until 5.5 months of age.Healthy CF at six months was significantly higher in the intervention group. 75. Use of mass media with or without nutrition education: an SR evaluated interventions conducted with mothers of children up to 24 months, pregnant women and mothers who had given birth in the last five years.They were offered a variety of formats of information (printed material, television, voice messages and/or SMS, radio messages, megaphone/speaker messages, videos, social media posts and music/dramatization), and duration ranged from nine weeks to four years (mostly one year).The results showed an increase in the prevalence of healthy CF.When nutrition education was combined with mass media, the effects were more robust.76

Income Level of Study Countries
75,76 The income classification of countries where the primary studies were conducted, according to the World Bank list of economies from July 2018, is shown in Table 3.

DISCUSSION
Several interventions have shown effectiveness in improving outcomes related to BF and healthy CF, and these findings may contribute to the planning of PHC actions.The synthesis of results allowed us to discuss six aspects that should be considered to increase the chances of interventions having an impact.5][56] These interventions presuppose, respectively: defining contents or themes that will be worked with the target audience in educational actions; focus on the needs of the target audience and use of a counseling approach to support interventions; behavior change techniques aimed at cognitive and behavioral aspects of BF practice and coping with difficulties in interventions of maternal self-efficacy theories.Although many studies have reported the importance of training the actors involved, there is no evidence that training alone is effective in promoting changes in BF and healthy CF.][49][50]57 Finally, the combination of diverse approaches (education and support) involving different scenarios (hospital and community) had a great potential to impact the outcomes analyzed.56,61,[64][65][66][67][68][69][70] The target audience is the second aspect to consider. In mot SRs, the interventions were aimed at women in the period of pregnancy, in the immediate postpartum period and taking care of children up to two years old.Two of them reported positive results with the involvement of the father, which puts the spotlight on the importance of their participation when implementing interventions.61,62 The third aspect is the moment of intervention.Educational actions performed only in prenatal care significantly increased the rate of BF in the short term, but the combination of actions in pre-and postnatal periods increased the rates of BF in the medium and long terms.Therefore, interventions that cover both the prenatal and the postnatal periods tend to be more effective.46,48,52,53,[63][64][65]68 The fourth aspect relates to the actors that can implement interventions.Health professionals, specialized consultants, trained lay people or "peer support" (generally mothers supporting other mothers) were mentioned.46,48,52,53,59,71,72 In the Brazilian context, we can consider that "peer support" is similar to the role of community health workers because they are members of the community in which they work.77 One SR showed effectiveness in promoting healthy BF and CF through an intervention practiced by community health workers in low-income and middle-income countries.75 In general, interventions that combine the work of health professionals and lay people tend to be more effective than isolated actions, and non-formal trained professionals can also support BF rate increase in low-middle and low-income countries.63,64,66,68,69 The fifth aspect refers to strategies and methods, that is, how interventions are applied.Educational actions can be carried out individually or in groups, but there are few in favor of group prenatal care and better results were identified when carried out in small groups.47,55 Interventions had limited effectiveness when: a single method of education was used; only written material on BF was offered; there was no direct contact with any health professional; a more formal and non-interactive method of providing health education was used.[68][69][70] Face-to-face support strategies are more likely to be successful when compared, for example, to over-the-phone sessions only.There is evidence that support-based intervention using home visits is effective.49,51,72,74,75 As a strategy, the results suggest that media-based interventions can improve BF initiation, EBF duration, and attitudes and knowledge about BF, but the results are based on studies conducted in developed regions, with internet or cell phone access, and may not be applicable to developing regions.58,76 Support-based interventions should be adapted to local needs and offered routinely by trained personnel.
Finally, the intensity of intervention is another relevant aspect for its success.Some findings deserve attention: educational interventions structured in a single contact between the postpartum woman and the professional along with low-intensity peer support seem to be ineffective for any type of BF . 51,75Home visits should take place at least once in the prenatal period, after birth, and in the postpartum period (days 1-3), along with visits after the first month of postpartum, in a high frequency of up to seven visits. 49,51,75t is also worth mentioning that, in order to achieve effectiveness, it is essential that the actions are implemented as planned.Often, the plan defines the actions, the target audience, the moments, the actors, the strategies and the intensity, but the flow is modified during implementation.In this sense, monitoring the fidelity of intervention is crucial to identify and overcome possible barriers. 78t should be borne in mind that this rapid review used previously defined methodological shortcuts, such as limiting the number of databases, restricting languages to Portuguese, English and Spanish, including only SRs and excluding primary studies, and data collection by a single reviewer.It all must be considered when interpreting the results.
Although these methodological shortcuts were adopted, the study served its purpose of responding in a timely manner to the Ministry of Health's demand to support the EAAB implementation.The time spent in searches, data collection and quality assessment of the articles was limited; however, a systematic and transparent process was adopted at all methodological stages.The SRs provided relevant information on the effectiveness of different categories of interventions aimed at BF and healthy CF, as well as brought about lessons to be learned for their implementation, even though most SRs were classified as critically-low quality.It is noteworthy that cultural and economic aspects related to different realities of studies must be considered.Therefore, additional caution was taken when interpreting the results of SRs, identifying and presenting the context of the countries where the primary studies were conducted, so we could better understand the potential for application of interventions in the context of PHC in Brazil.

CONCLUSION
Rapid reviews can be an agile and consistent way to respond to management's demands for evidence-informed policy implementation.Education and support interventions based on scientific evidence and adapted to local contexts are key to promoting optimal BF and healthy CF practices.It is hoped that the evidence presented here can support the implementation of programs for BF and healthy CF in PHC, offering professionals a list of effective actions and providing elements for local adaptations.

Table 1 .
Systematic review inclusion criteria.

Table 2 .
List of references of excluded studies and reasons for exclusions.
, the results were considerably lower, suggesting that only printed information is not as effective as individualized contact. 46,496. Use of media and digital technologies: two SRs included interventions by means of web technologies, CD-ROM, virtual interactive approach, e-prompt, television, printed material, voice and/or SMS messages, radio, megaphones/ speakers, videos, social media, and music/ dramatization.
5. Distribution of written materials: two SRs depicted interventions using materials such as a list of key points, pamphlets and leaflets with or without personal prenatal support.When interventions were made only by delivering written printed materials, without the support of BF: breastfeeding; CF: complementary feeding.Figure 1.Study selection flowchart.Rev Paul Pediatr. 2023;41:e2021362 professionals

Board of Lactation Consultant Examiners
® (IBLCE ® ): these interventions were reviewed by an SR; consultants made at least two prenatal contacts, one contact at the hospital, and three to nine postpartum telephone calls.Of 16 primary studies, 10 included interventions in PHC setting (prenatal and postpartum).1.Information and/or education and/or counseling:two SRs showed interventions carried out by counselors and/ or trained nutritional educators, as well as in peer support groups.They included training in nutrition counseling for professionals and laypersons (peer support) and counseling on healthy CF and responsive feeding directly to mothers.One intervention provided a set of basic meal preparation tools (bowl, spoon and an illustrated card).The frequency of interventions ranged from monthly, weekly, three or six sessions.

Table 3 .
Income level of countries of primary studies in systematic reviews.