Abstract
The World Health Organization recommends assessing compliance with key clinical practices of the Baby-Friendly Hospital Initiative (BFHI; Steps 3–9) using birthing women’s self-reports. Globally, compliance is mainly assessed using health staff reports, and the use of women’s self-reports in selected countries has deviated from the Global Standards for the BFHI. Therefore, we aimed to provide insight into the appropriate method of incorporating women’s self-reports in assessing compliance with Steps 3–9 of the BFHI. We developed questions and coding algorithms for assessing compliance with Steps 3–9 based on Global Standards for BFHI compliance, and implemented them via a cross-sectional survey of 302 women who gave birth to a live baby in Sri Lankan hospitals. Compliance with specific practices within each of Steps 3–9 and overall compliance with each step were described as percentages. Compliance with specific practices and each BFHI Step ranged from 15.9%–100% and 7.0%–100%, respectively. Our findings particularly emphasize the potentially enhanced usefulness and robustness of assessing all specific practices within BFHI key clinical steps and not focusing only on one practice within a step, to derive more useful health service guidance globally for capturing BFHI compliance and its impact on breastfeeding outcomes. This method could be translated across multiple settings globally. It would enable more specific identification of care advancements required by health services to improve the effectiveness of breastfeeding support and address the prevailing undervaluing and under-use of women’s experiential data to evaluate and guide health service improvement.
Keywords
Key Messages
Globally, compliance with key clinical practices of the Baby-Friendly Hospital Initiative (BFHI) are assessed primarily based on reporting from health staff.
Transpositioning the Global Standards of BFHI compliance into women-centric compliance measures produces a method for assessing compliance using women’s self-reported care experiences.
Applying this method in health facilities may enable health services to review existing compliance monitoring practices and to adopt a reliable and less biased method for supplementing
Background
The Baby-Friendly Hospital Initiative (BFHI) is an internationally recognized guideline that comprises 10 steps to support short and long-term breastfeeding, including key clinical practices in Steps 3–10 (World Health Organization [WHO] & United Nations Children’s Fund [UNICEF], 2018). The WHO recommends that BFHI compliance is evaluated by assessors independent of health facilities, including interviews with health staff and birthing women (WHO & UNICEF, 2018). However, in most countries, BFHI compliance is primarily assessed using health staff self-report (International Baby Food Action Network [IBFAN]-Asia, 2019). Although there are some examples of using women’s self-reports of care received to assess BFHI compliance (Agbozo et al., 2020; Araújo et al., 2019; Çaylan et al., 2022; Spaeth et al., 2018), there has been no research to establish methods to measure this systematically and comprehensively according to existing recommendations. Without this, the implementation and effectiveness of BFHI for improving breastfeeding rates cannot be adequately assessed. In this paper, we provide insights for assessing BFHI compliance using women’s self-reported experiences guided by the Global Standards of BFHI compliance
We propose specific methods required for assessing compliance with BFHI steps and examining the BFHI policy more closely. We share the practicalities of trans-positioning questions from the hospital criteria for key clinical practices given in the Global Standards of BFHI compliance (WHO & UNICEF, 2018) to questions that can be asked to birthing people, and applying those methods to assess compliance in a Sri Lankan cohort.
The BFHI was launched in Sri Lanka in 1992 in collaboration with UNICEF. The IBFAN-Asia (2019) reported that 94% of government and private health facilities in Sri Lanka were designated as baby-friendly between 2012 and 2017, and also ranked Sri Lanka highly for the quality of BFHI implementation based on the outcome of the World Breastfeeding Trend Initiative (WBTi) assessment conducted in 2018. However, the same WBTi 2018 report indicated that monitoring of BFHI compliance is not standardized nationally, it is primarily implemented only at the hospital level, and it relies on motivated individual administrators (IBFAN-Asia, 2019).
There is limited evidence to assess BFHI compliance within specific Sri Lankan health facilities. Existing evidence has exclusively been derived from health staff reporting (IBFAN-Asia, 2019) and provides inconsistent BFHI compliance information (Lokeesan et al., 2022). For example, compliance with immediate and uninterrupted skin-to-skin contact in labor rooms after giving birth (Step 4) ranges from 1.8%–83.3%; and postnatal wards’ provision of breastfeeding support (Step 5) ranges from 23.1%–90.0% across Sri Lankan health facilities (Bandara et al., 2012; Fernando & Prathapan, 2017). Monitoring of BFHI by hospital staff who have an interest in positive evaluation results (Rowel, 2020) may create self-report bias, which can lead to overestimation of compliance in some health facilities.
Contradictory international and national evidence regarding the implementation of and compliance with BFHI in Sri Lanka means that an accurate and consistent picture of BHFI in Sri Lanka remains elusive. Through this paper, we provide our innovative method as a model for measuring and reporting BFHI compliance using self-reports from birthing women consistently in Sri Lankan health settings, and to inform the potential wider application of assessing BFHI compliance based on recommended standards for assessment in global health settings.
The Clinical Innovation
We developed a tool to assess compliance with BFHI key clinical practices using women’s self-reported experiences of BFHI care received during their postpartum stay in a hospital. This period is critical for initiating breastfeeding and establishing exclusivity as soon as possible after giving birth, ensuring long-term exclusive breastfeeding. We included key clinical BFHI practices outlined in Steps 3–9 in our tool, as women and infants directly experience care across these practices up to and including their postpartum hospital stay. At that time of measurement, these steps are therefore appropriate for self-reported assessment from maternity service users. Also, breastfeeding support practices within Steps 3–9 of the BFHI are the responsibility of health services up to and immediately after birth, and are critical for establishing breastfeeding during the early postpartum hours in hospitals (Table 1).
Measuring Baby-Friendly Hospital Initiative (HI) Compliance (Steps 3–9) for a Self-Reported Care Assessment.
Note. BF = breastfeeding; compliant care = these were categorized as “Yes-received compliant care”; HI = hospital initiative; MI = medical indication (participants who had a medical reason for not receiving a particular type of care) SSC = skin-to-skin contact. This table describes the rules that were developed in this study based on Global Standards of HI compliance, to measure the compliance according to participants’ responses to each question related to specific practices of each of Seps 3–9 of HI.
This column represents the survey questions and responses given in Table S1 (see the online supplemental material). Question numbers given for compliance measures in this column represent the relevant questions from Table S1. Therefore, some of the measures were given more than one question number, that is, Q2–Q7 (for breastfeeding information) means Q2–Q7 from Table S1 collectively representing the compliance measure given in this column, and responses related to each question have been collectively reported under the survey items in this column.
Among the response options given in the column labeled “survey items,” responses other than given in the column labeled “criteria to determine compliant care” were categorized as “no compliance” with relevant HI practices.
Length of the time that the skin-to-skin contact lasted was not assessed to determine compliance with skin-to-skin contact.
Compliance criteria of Step 9 of HI were derived to assess compliance with that step within the health facilities. It is different from the measures in Global Standards of HI compliance which is more likely to be implemented after discharge from the hospital.
Whether participants received bottles from health staff or were recommended by health staff to use bottles was indirectly measured via the question of feeding supplement (if infant received supplements) with?
As described in Table 1, we converted the compliance criteria given in the Global Standards of BFHI compliance (WHO & UNICEF, 2018) to self-reported measures for individual women. For example, within Step 3, the compliance criteria of “at least 80% of women who received prenatal care also received prenatal counseling on breastfeeding” was converted to individual criteria of “received breastfeeding counseling” (Table 1). We developed questions and possible responses to address the self-reported BFHI care measures related to each of the specific practices within Steps 3–9. Some of the existing surveys and reports related to BFHI care and infant feeding (Begley et al., 2008; Sakala et al., 2018; Toronto Public Health Division, 2017; WHO, 1999) were used to guide question and response development with appropriate wording to reflect the BFHI practices. Compliance measures related to each of the specific BFHI practices within Steps 3–9 were addressed using single or multiple questions as given in Table 1.
Compliance with specific practices within BFHI steps was determined with the dichotomous measures of “Yes” and “No” based on the responses that indicated women received care related to that specific practice (Table 1). Responses that reflect that women received care or women received care/advice, but that they do not remember specifics about it were classified as “Yes, women received care.” When multiple responses in a question reflected the compliant care of a BFHI practice, compliance with that specific practice was determined if participants provided any one response among all applicable responses as given in Table 1. However, for some practices, such as receiving advice on latching, compliance was determined only if participants provided all responses that reflected the care compliant with that practice. Moreover, responses that indicated women did not receive care or women do not remember whether they received care or received any advice from health staff were classified as “No, women did not receive care.” Overall compliance with a BFHI clinical step (Steps 3–9) was defined as participants having received care that was compliant with every practice within that step.
We defined women as having received care that was compliant with Step 3—giving pregnant women information on breastfeeding benefits and management (WHO & UNICEF, 2018)—if they reported that they attended pregnancy care and received breastfeeding counseling regarding at least two of the following: the importance of breastfeeding, pros and cons of exclusive breastfeeding, risks of supplementation, risks of using bottles to feed babies and mother-baby skin-to-skin contact (Table 2).
Compliance Rates for Overall Steps and for Each Practice Within Each Step of the Baby-Friendly Hospital Initiative (BFHI; N = 302).
Note. BF = breastfeeding. Bolded numbers are overall BFHI compliance rates. Steps 3–9 of BFHI are coded according to the compliance criteria described in Table 1.
We defined compliance with Step 4—immediate postnatal care management (WHO & UNICEF, 2018)—if women reported having skin-to-skin contact with their baby within 5 minutes of birth, and placing the baby on the breast within 1 hour of birth (Table 2).
Compliance with Step 5—supporting women with instructions on how to breastfeed correctly and continuously (WHO & UNICEF, 2018)—was defined as women reporting: (1) that health staff supported them in managing difficulties while breastfeeding, (2) advised them on how to position and attach the baby to the breast for effective breastfeeding, and (3) advised on methods to increase the milk production for their infants (Table 2).
Women were defined as receiving care compliant with Step 6—not recommending breastmilk supplements unless medically indicated (WHO & UNICEF, 2018)—if they reported that their baby was not given any supplements or was given supplements recommended by health staff when medically indicated (Table 2).
We defined women as receiving care compliant with Step 7—mothers and infants should “room in,” that is, remain together 24 hours a day (WHO & UNICEF, 2018)—if they reported that they always stayed with their baby after giving birth and that health staff advised them about rooming in (Table 2).
Compliance with Step 8—support for women in giving their baby their own milk according to the baby’s need (on-demand; WHO & UNICEF, 2018)—was defined as women reporting that they were advised by health staff about feeding on demand (Table 2).
Compliance with Step 9—no artificial teats or pacifiers should be given to breastfeeding infants (WHO & UNICEF, 2018)—was defined as women reporting that their baby was not given a pacifier by health staff and no health staff recommended using bottles for feeding their baby in the hospital (see Table 2).
The overarching principles of our new measurement approach were that: (1) the measures needed to account for all specific practices within each BFHI clinical step; (2) we only aimed to measure the key clinical steps that are the responsibility of health services to implement up to and immediately after birth (Steps 3–9); (3) the measures needed to capture women’s experiences of care; and (4) the measures needed to be suitable for early postpartum administration in the hospital. Therefore, the compliance measures that we developed for a few BFHI steps were different from the criteria given in the Global Standards of BFHI compliance. For example, for Step 9, the compliance criteria for “advising women regarding the risks of using bottles, teats, and pacifiers” given in Global Standards is mostly applicable during the late postpartum period and, therefore, we developed new measures (see Table 1) that are significant for establishing breastfeeding within the early postpartum hours.
Pilot Administration and Outcome
We conducted face validation of questions and responses by administering the tool to five Sri Lankan women who had birthed a live baby between September and December 2021. We then conducted a pilot administration of this tool in a cohort of postpartum women in Sri Lanka to assess the applicability of this tool and to test the method for presenting compliance estimates. We asked 302 postpartum women who had live, single, or multiple births, did not have a neonatal death, and were sufficiently physically and mentally stable to communicate and participate in this assessment. These women were from four randomly selected government hospitals in Sri Lanka. In Sri Lanka, 99% of women receive pregnancy care at government health facilities and 94% of births occur in government hospitals (Department of Census and Statistics [DCS]-Sri Lanka, 2017, 2019). Women were approached in the postnatal wards of the selected hospitals by the principal investigator of this assessment who is a researcher and independent from Sri Land Health Services. We approached potential participants at least 12 hours after giving birth and before discharge, so participants had an opportunity to most easily and correctly recall their recent experiences of care received in the hospital. Participants received a thorough explanation of the purpose of the assessment and then written consent was obtained before asking them about their experiences with the breastfeeding support care they received up to and immediately after birth.
To implement the compliance measuring criteria, we assessed whether every participant received care across the specific practices of BFHI Steps 3–9. We derived the total number and percentage of participants who received overall care compliant with each key clinical step and received care compliant with specific practices within the clinical steps (Table 2). The percentage that reflects the overall compliance rate with each key clinical step (bolded in Table 2) was derived from the proportion of participants who received care with every specific BFHI practice within a key clinical step.
Implications
From the pilot administration, we observed widely varying compliance across specific practices within each of Steps 3–9 of BFHI. However, we believe that assessing compliance across key clinical practices contributed meaningfully to the accuracy of overall compliance estimates compared to previously used assessment criteria. For example, in Ghana (Agbozo et al., 2020), the compliance rate of Step 4 was measured only with the criteria of “mother-baby skin-to-skin within five minutes of birth.” In Taiwan (Chien et al., 2007), the reported compliance rate with Step 4 was measured using only the criteria “initiating breastfeeding within 30 minutes of birth.” In Brazil (Araújo et al., 2019) and Turkey (Çaylan et al., 2022) compliance with Step 5 had been assessed with a single criterion of “breastfeeding support and management of difficulties.” Overall compliance estimates derived in these studies may be affected by misclassification bias through assessment of a single popular practice within a BFHI step and inconsistency with Global Standards of BFHI compliance.
Further, assessment of specific practices within each BFHI step can assist health services to better identify areas for improvement that affect overall compliance. For example, within Step 4 of BFHI, the compliance rate with mother-baby skin-to-skin contact was 15.9% while compliance with placing the baby on the mother’s breasts within 1 hour of giving birth was 70.5%, and the overall compliance was 11.9%. According to our method of overall compliance assessment, variations in compliance across the specific practices within each step have significantly contributed to the overall compliance rate of that step. In this case, health facilities can easily identify the practice that needs to be improved to enhance overall compliance with each BFHI step.
In Sri Lanka and globally, it is worth conducting a comprehensive assessment of BFHI compliance using women’s self-reported care experiences as health staff may over-report their practice with self-monitoring to produce more favorable results (Araújo et al., 2019). Evidence from Brazil (Araújo et al., 2019) demonstrated that reported compliance across Steps 3–9 of BFHI was greater than 80% when self-monitored by health staff. When assessed with an external evaluation which included women’s self-report (Araújo et al., 2019), the compliance rate was less than 70% for most of Steps 3–9 of BFHI, with a difference of 20%–30% from estimates based on health staff reports (Araújo et al., 2019). Women’s self-reported maternity care experiences and outcomes are at least as reliable as medical records completed by health staff, and play a vital role in making decisions on care delivery (Chen et al., 2022; McCarthy et al., 2023). Therefore, comparing our estimates of BFHI compliance against compliance estimates derived from health staff’s monitoring in the same health facilities may be valuable for health services to understand the need and value of supplementing health staff’s self-monitoring with women’s self-reported care experiences.
The validity of our assessment method requires further research. Some caution is recommended in reporting the estimates derived for BFHI compliance by applying our method of BFHI compliance assessment in different health facilities before further validation. There is a need for health services to decide who will assess women’s exposure to BFHI care because we have designed the tool to be administered in the wards during the early postpartum period. Social desirability bias may arise if women are asked about the care received by a healthcare worker administering the tool during this vulnerable time. To manage this, assessors should be independent of the health services, replicating the processes applied in our pilot administration. During the pilot administration, we approached potential participants between 12 hours of giving birth and discharge, as we believed that obtaining information from women during the early postpartum hours might minimize recall bias. However, it is also possible that health service providers may not have had an opportunity to comply with some of the practices being assessed during the early postpartum hours, for example, practices within Steps 5, 7, and 8. Future research and applications should consider appropriate timing for collecting adequate information from women regarding their experiences with all practices within Steps 3–9 of BFHI to maximize validity and minimize recall bias.
Implementing our method of assessing compliance across all specific BFHI practices using women’s self-reported care experiences will be a new approach in Sri Lanka and requires significant policy change. The new BFHI policy in Sri Lanka would require three aspects: resources to capture self-reported experiences of BFHI care, commitment to measuring compliance with specific practices within each BFHI step, and targeted training of health staff to improve BFHI compliance in response to comprehensive patient-reported data. However, adopting such a policy in Sri Lanka may be difficult given the economic crisis experienced by Sri Lanka since late 2021 (Dayal, 2023) and the continual dependency on international non-government organizations to fund maternal and child health initiatives at the provincial and district levels (Ministry of Health [MoH] & Family Health Bureau [FHB]–Sri Lanka, 2018). Currently, in Sri Lanka, BFHI compliance monitoring is primarily implemented only at the hospital level, and it relies on motivated and individual administrators with no national coordination (IBFAN-Asia, 2019). Inadequate BFHI training for health staff is reported to be a result of inadequate funding, irregular planning, and scheduling of ongoing training for health professionals on BFHI compliance practice, and no training coordinator positions in hospitals (MoH & FHB–Sri Lanka, 2012). Therefore, without dedicated national policy change, or long-term BFHI funding commitments from international non-government organizations, it will be difficult to raise the quality of BFHI compliance monitoring in Sri Lanka.
Conclusion
Development and pilot administration of the tool that we developed for comprehensively assessing BFHI compliance provides insight into the importance of measuring compliance with individual and multiple practices within the key clinical steps (Steps 3–9) of BFHI and also considering women’s self-reported care experiences as a valuable resource in compliance monitoring. Such insights enable health services to review existing BFHI monitoring practices and consider adopting assessment systems for supplementing health staff monitoring of BFHI compliance. Including maternal self-report would allow health services to better identify the specific practices with low levels of compliance and the circumstances that influence the delivery of compliant care. More useful BFHI compliance monitoring can better guide health staff education and training for maximizing compliance with BFHI practices and providing adequate support for every woman breastfeeding in health facilities.
Footnotes
Acknowledgements
We would like to thank all the women who participated in this study. We also acknowledge the Director/Medical Superintendents of selected hospitals for permitting us to conduct the study; Chief Nursing Officers of selected hospitals, Consultant Obstetricians and Gynecologists, Nurses in-charge, nursing officers, and the doctors and midwives of the postnatal wards of selected hospitals for supporting data collection; and the Faculty of Health, the Queensland University of Technology, Australia for funding travel costs for the principal researcher to collect data.
Author Contributions
Availability of Data and Material
Access to this dataset is outlined in the Queensland University of Technology repository Research Data Finder (QUT - Research Data Finder). The dataset generated and/or analyzed during the current study is available upon request.
Disclosures and Conflicts of Interest
The authors disclosed no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The faculty of Health, Queensland University of Technology [QUT], Australia funded Laavanya Lokeesan to conduct data collection in Sri Lanka. Open access publication was facilitated by SAGE-Queensland University of Technology (QUT) publishing agreement via the Council of Australian University Librarians.
Ethics Approval
Ethics approval to conduct this study in selected hospitals was obtained from the University Human Research Ethics Committee of Queensland University of Technology (Approval number – 2000001001) on the 28th of February 2021, Ethics Review Committee of Faculty of Health-Care Sciences, Eastern University, Sri Lanka (Approval number – E/2021/05) on August 3, 2021, and the Ethics Review Committee of Department of Health Services, North-Western province, Sri Lanka (Approval number – ERC/NWP/2021/11) on December, 7 2021.
