Influence of Perceived Maternal Self-Efficacy on Exclusive Breastfeeding Initiation and Consolidation: A Systematic Review

1. Introduction

Various international organizations support exclusive breastfeeding (EBF) during the newborn's first six months of life, as well as to subsequently complement it with food options up to the age of two years or more [1,2,3].

Both for newborns and for breastfeeding (BF) mothers, the benefits of EBF are many and are widely described in the literature [4,5,6]. In addition, it can be pointed out that BF is a key determining factor to promote public health and reduce inequalities in health [7]. Despite all the above, globally only 38% of the newborns receive EBF during their first six months of life [4]. The BF rates decrease rapidly during the first weeks postpartum, and it is in the first month that the most noticeable change is found in the number of women who interrupt BF [8].

Demographic, physiological, and psychological factors can interfere both positively and negatively in BF interruption or maintenance [9,10,11,12,13]. Among the psychological factors is the mothers' perception of supplying insufficient milk to their newborns. Although this perception is usually called hypogalactia, it is not always a true case of this condition (non-production of milk resulting from organic factors) but is often because of scarce or non-existent production related to an inadequate BF technique and to the technical aspects of BF (such as latch, position of the newborn and of the mother, etc.). This perception of BF non-efficacy is the most frequent reason for BF abandonment, which turns self-efficacy (SE) into an essential factor for EBF initiation [8,14].

A large number of the studies on SE are based on Albert Bandura's concept [8,9,10,11,12,13,14,15]. Dennis adapted this concept to the reproductive field, defining SE regarding EBF as the mother's self-confidence in her ability to breastfeed her newborn [19]. Since then, maternal SE has received considerable attention as a predictor of health-related behaviors [20,21,22,23] in addition to being a factor that exerts an influence on maternal satisfaction with EBF [24].

In relation to the interventions performed, several programs have been developed in an attempt to increase maternal breastfeeding self-efficacy (BSE), mainly targeted at mothers at a high risk of EBF abandonment in the first weeks postpartum [11,25,26].

The aim of the review is to assess how mothers' perceived level of self-efficacy during breastfeeding affects their ability to breastfeed and the rates of exclusive breastfeeding up to six months postpartum.

2. Materials and Methods

A systematic literature review has been conducted according to the guidelines established by the 2020 Preferred Reporting Items for Systematic Reviews (PRISMA) (Supplementary Table S1) [27], using qualitative literature synthesis. Table 1 shows the PICO criteria (participants, intervention, comparison, and outcomes) used for the inclusion of studies.

2.1. Search Strategy

A literature search was conducted from January 2000 to December 2023. The systematic literature search was conducted in the following electronic databases: Cochrane, Web of Science (WoS), Scopus, PubMed, Science Direct, and Cumulative Index to Nursing and Allied Health Literature (CINAHL).

The designed search strategy was conducted by combining the Medical Subject Headings (MeSH) and Descriptors in Health Sciences thesaurus with free terms, as they are found as synonyms of the descriptors in the literature, through the use of the AND and OR Boolean operators. The MeSH term employed was “self-confidence”; and those from DeCS were “self-efficacy”, “breastfeeding”, “lactation”, and “exclusive breastfeeding”. The free terms chosen were the following: “self-reliance” and “breastfeeding rate” (Table 2).

2.2. Inclusion/Exclusion Criteria

The inclusion criteria were as follows: (1) clinical trials or observational studies published in English and/or Spanish, (2) the subjects studied must be postpartum women in whom the variable SE in their ability to breastfeed is present or, in its absence, terms that are used as synonyms, (3) SE is measured using a validated tool, (4) the influence of this SE on the rate of EBF is studied, and (5) the studies meet the PICO criteria described.

The studies excluded were: studies not published in English or Spanish, studies carried out in puerperal women with any disease during pregnancy or puerperium (diabetes, pre-eclampsia, etc.), and studies carried out in preterm newborns, studies that do not study the variable SE or do not study EBF rates, and studies that do not meet all the PICO criteria.

2.3. Data Extraction

The article selection process began with the paired search conducted by two of the authors. The eligible studies retrieved from the six bibliographic databases (n=250 records) were imported into the Mendeley^® bibliographic reference manager and duplicates were removed (n=44). The search in the grey literature did not yield relevant results. The pre-selected records (n=206) were examined in two stages. As a first step, the titles and abstracts were evaluated considering eligibility regarding the inclusion criteria defined according to the PICO framework, eliminating n=171 records. Secondly, the remaining full-text articles that were selected (n=35) were thoroughly read to evaluate their inclusion in the review. Once the eligibility process was complete, another two authors assessed the methodological quality and the biases of the potentially useful studies; this allowed improving the screening of the results to obtain more complete and relevant information, thus enhancing the quality of the study. The agreement degree between both researchers in terms of evaluating eligibility of the studies was assessed using Kappa's statistical test, with high agreement as a result (Kappa statistics = 0.85). The articles excluded were those that did not meet all the inclusion criteria, did not respond to the review objective, or were focused on studying any variable other than SE regarding ability to breastfeed (n =17). Finally, the set of articles included in the current systematic review amounted to a total of 18 records. The PRISMA flow diagram is presented in Figure 1.

2.4. Data Analysis

The tool described by López de Argumedo et al. [28] for systematic reviews was used to assess the quality level and evaluate the risk of bias of the studies (Supplementary Table S2). This tool appraises six areas to assess the quality of the evidence contributed by each study included. Its purpose is to provide a structured and standardized way to identify limitations in the studies included in a review, in order to improve the interpretation of the findings and assess the strength of the scientific evidence. A narrative synthesis of the data was undertaken. From each study we extracted general data, the assessment made by the authors, and key findings in reference to the

3. Results

3.1. Characteristics of the Included Studies

Of the total of 18 studies included, 13 (72.2%) were conducted in Asia, four (22.2%) in North America and one in Europe (5.6%); 72.22% were experimental studies and 27.8% area observational studies.

All the studies were conducted with women who wanted to breastfeed, with sample sizes varying from 30 to 781, accounting for a total of 5,771 participants in the 18 studies included. Eleven studies (61.1%) included only primiparous women while in the other seven (38.9%), the participants were primiparous and multiparous women.

The author, date of publication, study aim, sample, and design of the studies included in the review are shown in Table 3.

All the detailed and relevant information for data analysis, synthesis and interpretation was collected, encompassing the following: bibliographic data, country, study design, tool used to assess SE, evaluation of the EBF rates, main results, and quality level of each study (Table 4) (Table 5).

3.2. Measuring Instruments and Interval

To assess the self-efficacy level, all 13 studies reviewed in this study used the BSES [32,40,41,46] or BSES-SF [29,30,31,34,35,37,39,42,45]. The BSES was created by Dennis and Faux in 1999 to assess confidence in breastfeeding [47]. This self-administered tool consists of 33 items, all preceded by the phrase "I can always", and is rated on a five-point Likert scale, ranging from 1 (not at all confident) to 5 (very confident). Higher scores indicate greater levels of breastfeeding self-efficacy. In 2003, Dennis reduced the BSES from 33 to 14 items and renamed it the BSES-SF [48]. There is substantial evidence supporting the reliability and validity of this version as a global measure of breastfeeding self-efficacy. The reliability and validity of this instrument have been satisfactory in different countries and populations [49,50,51,52,53,54].

The SE levels and the EBF rates were evaluated at different points during the postpartum period: early postpartum, 1 month postpartum (4 weeks), 2 months postpartum (8 weeks), 3 months postpartum (12 weeks), and 6 months postpartum [29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46].

Table 6 shows the author, date of publication, variables measured, instruments, reliability, and validity of the instruments of the studies included in the review.

3.3. Quality Assessment

The results of the analysis carried out according to the tool described by López de Argumedo et al. [28], in relation to the evaluation of methodological quality, are presented in the Supplementary Table S3.

3.4. Self-Efficacy Levels Perceived by the Mothers About Their Ability to Breastfeed

The difference between the two scales is that the BSES consists of 33 items with a maximum score of 165 points, and the BSES-SF consists of 14 items with a maximum score of 70 points [47,48].

In the studies that used BSES, the scores were 105.28 points at 6 months postpartum (IG) in the study by Ansari et al. [32], 121.44 points at 6 months postpartum (IG) in Shariat et al. [40], and 141.44 points at 2 months postpartum (IG) in Yesil et al. [46].

In relation to the other studies that used BSES-SF, the scores in early postpartum were as follows: 34.8 points (IG) in the study by Awano & Shimada [30], 51.6 points in Otsuka et al. [34], 55.89 points (IG) in Chan et al. [37], 46.2 points (IG) in Tseng et al. [42], 63.66 points (IG) in Vakilian et al.[41], and 43.05 points (IG) Wong & Chien [45].

At 1 month postpartum, the scores were 59 points in McQueen et al. [31], 53.38 points in Noel-Weiss et al. [29], 53.5 points in Otsuka et al. [34], 58.8 points in Wu et al. [35], and 48.1 points (IG) in the study by Tseng et al. [42].

At 2 months postpartum, the scores were 62.46 points in the study by Araban et al. [39] and 59.85 points in that by Wu et al. [35]. At 3 months postpartum, the scores were 49 points (IG) in the study by Tseng et al. [42]. At 6 months postpartum, the scores were 49.9 points (IG) in the study by Awano & Shimada [30] and 46.7 points (IG) in Tseng et al. [42].

3.5. Self-Efficacy and Exclusive Breastfeeding Rates

The study with the largest sample included in the current review (781 women) is a clinical trial conducted by Otsuka et al. [34] in Japan. This study evaluated the EBF rates in early postpartum, at 4 weeks, and at 12 weeks. In early postpartum, the EBF rate was 88% (mean of both groups). However, at 4 weeks postpartum, the EBF rates were higher in the women whose maternal self-efficacy levels increased (IG) than in the other groups; 73.4% maintained EBF in the group with the highest score (53.5 points in BSES-SF). Despite this, the EBF rates presented a marked reduction at 12 weeks postpartum, falling to 47%, and the study argued that the intervention was therefore not effective in increasing or maintaining the SE levels at 12 weeks postpartum.

In this same line, Shariat et al. [40] evaluated the EBF rates in early postpartum and at 6 months postpartum, showing how they drop from 87.5% to 40.9% between early postpartum and 6 months postpartum in the IG, and from 75.4% to 23.5% in the CG. The difference in the percentage of the BF rates lies in the SE levels: in the IG, the BSES scores increased after one month of intervention, up to 20 points more than in the CG. Despite that, the mean BSES scores at 6 months were similar (121.44 vs. 122.52). The variable that marked the difference was “anxiety” (which is not studied in this review): higher anxiety levels were related to lower SE levels, and the mothers in the IG presented lower anxiety levels. As in the aforementioned study [34], the EBF rates at 6 months were markedly reduced when compared to early postpartum or to the first postpartum weeks, where the maternal SE levels were one of the factors that exerted a notable influence.

Unlike the previous studies, an intervention performed by Ansari et al. [32] to improve SE shows that the increase in BSE sustains the high EBF rates over time, with 73.3% at 6 months postpartum.

In the study by Awano & Shimada [30], which starts from a sample consisting exclusively of primiparous women (which can exert an influence on the results because of the lack of previous experience when compared to other studies that include multiparous women), the maternal SE levels exerted a major impact on the EBF rates at 4 weeks postpartum. In this study, the EBF rates in early postpartum were 90% in the IG and 89% in the CG; however, they remained at 90% in the IG at 1 month postpartum and dropped to 65% in the CG, with the SE levels also increasing in the IG throughout this period. The SE levels at 1 month postpartum were higher in the IG than in the CG (49.9 vs. 46.5 points in the BSES-SF scale, respectively).

However, in the study by Wu et al. [35], which was also conducted with primiparous women, the EBF rates were only 60% at 8 weeks postpartum in both groups, although the BSES-SF scores were similar to those found in Awano & Shimada [30]. The main cause was the mothers' perception of supplying insufficient milk to their newborns. Nevertheless, higher SE levels were associated with EBF maintenance.

Other factors, such as maternal educational level or the mothers' previous BF experience, are not addressed in the current review; however, it is necessary to take them into account in future studies, as they also exerted an influence on the EBF rates.

Two of the most recent studies included—Tseng et al. [42] and Vakilian et al. [41]—conducted randomized controlled trials with primiparous mothers. Vakilian et al. [41] evaluated only the EBF rates at 1 month postpartum, obtaining 89.2% in the IG (63.66 points in BSES-SF) and only 55.5% in the CG (57.04 points in BSES-SF). The positive impact of the interventions during the first weeks on the maternal SE levels is consistent with the results found by Otsuka et al. [34] and by Shariat et al. [40]. In these studies, the same result is not obtained in the subsequent weeks: although the mothers' SE levels are maintained, the rates are markedly reduced.

This is also the case in the study by Tseng et al. [42] After the intervention, the study determined the EBF rates at four time points: early postpartum and at 1, 3, and 6 months postpartum. The EBF rates were higher in the IG when compared to the CG, although they varied in both groups at the different time points studied. The differences in the EBF rates are related to the BSES-SF scores, with a mean of 48 points in the IG versus 40 points in the CG. Another study, Chan et al. [37], where the EBF rates were assessed in early postpartum and at 1, 2, and 6 months postpartum, shows similar results to those of the previous study. The EBF rates presented a marked reduction at 6 months postpartum: 40%, 37.2%, 31.4%, and 11.4% respectively in the IG with a mean score of 50 points in BSES-SF, and 22.2%, 13.9%, 5.5%, and 5.6% in the CG with a mean score of 40 in BSES-SF.

In the study conducted by Wu et al. [44], only 25% of the participants reported BF during their postpartum hospitalization, a percentage much lower than that reported in previous studies, although the study also established a significant relationship between SE and the EBF rates. In the oldest study, Noel-Weiss et al. [29], EBF rates at 8 weeks were maintained at 64%, which correlated with the increase in SE levels in both groups.

Finally, the two most recent studies were by Wong & Chien [45] and Yesil et al. [46] (2023). In the study by Yesil et al. [46], there were significant differences in EBF rates, with 72.5% in IG providing EBF at birth compared to only 30% in CG.

In a study during the COVID-19 pandemic in China [45], where SE scores were low (43.05 points on the BSES-SF), only 53-4% of mothers gave EBF at 2 months postpartum.

4. Discussion

This systematic review assessed how mothers' perceived level of self-efficacy during breastfeeding affects their ability to breastfeed and their rates of EBF up to six months postpartum. It is necessary to take into account the demographic and cultural characteristics of mothers that may influence breastfeeding, whether they are primiparous or multiparous, and whether or not they have previous breastfeeding experience. It should also be remembered that exceptional situations such as the COVID-19 pandemic may create additional difficulties in promoting BF.

The instruments used to perform the SE measurements in the articles included in the current review [29,30,31,32,34,35,37,39,40,41,42,43,45] are among the most frequently employed in the international scientific community, showing homogeneity in the conceptualization of development, content, construct and predictive validity [55]. It is indeed a reality that certain disparity is observed in the measuring intervals of the studies reported, which can hinder interpretation of how SE evolved during the postpartum period.

The SE levels found in the articles reviewed [29,30,31,32,34,35,37,39,40,41,42,43,45,46] show scores that are in line with what is described in the literature in all their measurements, if we compare them to the study conducted by Degrange et al. [56] where the threshold score for BSES was defined at 116/165, which would be equivalent to 49/70 in BSES-SF.

Several publications [30,32,34,35,37,39,40,41,42,43] have studied the impact of maternal SE on the ability to maintain EBF by comparing two groups. The results show that pregnant women with less SE in their ability to breastfeed presented significantly more chances of interrupting it [34,40,41,42,43].

Ansari et al. [32] reported that there was a significant relationship between SE and the duration of EBF.SE is a modifiable variable that can be improved through the implementation of appropriate programs, but factors such as gestational age or maternal education level must be taken into account. Another intervention conducted in Iran [39] agrees with Ansari et al. [32] that interventions that help increase SE levels improve EBF rates.

In the intervention conducted by Wu et al. [35] in China, EBF rates were below 60% at 4 and 8 weeks postpartum in both groups, despite having SE scores similar to those in other studies such as Awano & Shimada [30]. The main reasons were the maternal perception of insufficient milk supply, lack of family support, and the limited knowledge of professionals regarding breastfeeding. Regardless of the intervention's effectiveness, these cultural considerations in China have influenced SE levels and, consequently, EBF outcomes. This may be a differentiating factor compared to results from other studies such as Awano & Shimada [30], conducted in a different culture. These cultural aspects of BF also influenced the results of interventions in primiparous women in the studies by McQueen et al. [31]and Noel-Weiss et al. [29]. Nearly half of the mothers intended to breastfeed beforehand, with BF rates remaining around 60-70% at 8 weeks postpartum in both studies. This was correlated with increasing SE levels as the weeks passed.

Otsuka et al. [34] found a positive correlation between maternal SE and EBF rates in hospitals adhering to United Nations and WHO breastfeeding guidelines compared to those that did not. Hospitals supporting these guidelines showed higher SE scores and EBF rates. Increased maternal SE levels generally led to higher EBF rates and duration, though prior BF experience also played a role. The study included both first-time and experienced mothers but did not specifically analyze how prior experience affected the outcomes.

It is crucial to consider cultural variations when analyzing the impact of maternal self-efficacy on breastfeeding, as different cultural contexts can significantly influence beliefs, practices, and the support mothers receive regarding breastfeeding. For example, studies conducted in Asia have shown that educational interventions based on self-efficacy theory have a particularly strong impact on maternal self-efficacy and exclusive breastfeeding rates compared to other cultural contexts, due to the central role of community norms and family support in these societies [57]. These cultural differences not only impact how interventions are implemented but also how mothers interpret and act upon the information received. Additionally, it is important to highlight the distinction between clinical and statistical significance in these studies. While the results show a statistically significant increase in maternal self-efficacy and breastfeeding duration, it is also necessary to interpret the clinical relevance of these changes—specifically, how these results translate into tangible long-term health benefits for both mothers and infants. Including a broader discussion of clinical significance could better contextualize the findings across different cultures and populations.

The most recent studies included primiparous mothers [42,43] showed that EBF rates positively correlated with SE scores. In the most recent study [46], which included both primiparous and multiparous mothers, previous BF experience influenced SE levels and EBF rates. The intervention was effective in boosting SE levels, maintaining EBF rates up to 12 weeks postpartum. An interesting study because it was developed in the middle of the Covid-19 pandemic in China [45], the SE scores were low and only half of the mothers gave EBF at 2 months postpartum. The pandemic had negative influences on SE levels and EBF rates because direct contact with mothers could not be had.

Unlike other factors, SE is potentially modifiable with interventions conducted by health professionals [58,59] and its role as one of the factors positively associated with EBF initiation and duration is acknowledged [21,22] even in premature newborns.[25]. In addition, SE has shown an additional positive impact on the EBF rates from early postpartum to 6 months postpartum [32,36,40,41,56,58] and has also been identified as a significantly relevant factor for BF in future pregnancies [560].

On the other hand, maternal education is a key factor in SE and EBF rates, as several studies suggest. Women with higher educational levels tend to exhibit greater perceived self-efficacy, which translates into a higher ability to initiate and maintain exclusive breastfeeding [57]. Additionally, education level may influence how breastfeeding information is interpreted, the capacity to seek support, and decision-making during the postpartum period [60]. In this sense, formal education provides mothers with cognitive and critical tools that allow them to better handle breastfeeding challenges, facilitating the adoption of healthy practice [57]. However, education is not limited solely to academic training; specific educational interventions on breastfeeding, led by healthcare professionals, play a crucial role in increasing maternal SE, especially in women with fewer formal educational resource [62]. This highlights the need to design and implement inclusive and accessible educational programs for all mothers, addressing their particular needs and promoting successful breastfeeding, regardless of their educational background [57]. Although the studies reviewed do not deeply analyze this aspect, the data suggest that educational interventions may be essential in improving breastfeeding outcomes in more vulnerable populations [57].

Despite the impact of the interventions on the levels of SE and the rates of EBF at 6 months postpartum, the figures achieved are still insufficient with respect to the WHO [4] criteria, so it is necessary to continue investigating how to increase these figures.

Furthermore, it is the generalization of the included studies (external validity), the results obtained cannot be extrapolated to the general population because the studies have not covered a sufficiently heterogeneous population (most of the samples come from hospitals in a single city).

5. Conclusions

The SE level in relation to mothers' ability to breastfeed affects the EBF rates up to 6 months postpartum. Consequently, higher SE levels are positively related with EBF initiation and maintenance. The EBF rates are maintained in early postpartum and at 1 month after birth when the breastfeeding SE levels are high. However, despite the interventions' positive impact on the EBF rates at 6 months postpartum, and even maintaining high SE numbers, these rates are markedly reduced during this period and are insufficient in relation to the objectives proposed by the WHO.

Perceived SE on the ability to breastfeed is a modifiable factor, so it is pertinent to identify early mothers who lack this self-confidence in their breastfeeding ability and to implement effective interventions to increase the maternal SE levels.

Given the importance of maternal SE in EBF, it is recommended that future research focuses on the development and evaluation of specific interventions aimed at increasing SE levels in BF mothers. These interventions could include training programs that provide information on BF techniques, as well as support groups where mothers can share experiences and receive positive feedback. Additionally, it would be valuable to implement longitudinal studies that assess the impact of these interventions on SE levels and EBF rates, thereby allowing the identification of effective strategies and their adaptation according to the individual needs of mothers. In this way, we can contribute to improving BF outcomes and supporting mothers on their journey toward a successful BF experience.