INTRODUCTION

Breastmilk supplies newborn infants with the specific compounds they need to grow and develop, not only from a nutritional point of view, as breastmilk also provides hormones, anti-inflammatory and anti-infectious agents, cells, growth factors and particles such as DNA and RNA. The close physical contact between mother and child that breastfeeding implies is also essential for the physical and emotional development of the human newborn. Breastfeeding protects the mother-child relationship in the short, medium and long term and strengthens their emotional bond. Society as a whole benefits when mothers breastfeed, unnecessary costs are saved and it protects the environment^1,2.

Global goals have been established to achieve at least 50% of exclusive breastfeeding during the first 6 months of baby’s life³, but in Spain, only 39.9% of infants benefit from it⁴.

The recommendations need to be accompanied by effective support measures⁵. This practical support can be very varied between centres and among different health workers. The main reasons for breastfeeding cessation are related to common problems⁶, such as perceived insufficient milk, difficulties with breast attachment, mastitis or poor weight gain. A lack of training among professionals in the management of these issues, along with a low awareness of the disadvantages of formula feeding can hinder many mothers’ breastfeeding goals.

The BFHI strategy (Baby-friendly Hospitals Initiative) promotes the best practices for care during childbirth and breastfeeding. Its implementation improves breastfeeding and child health rates⁷ and it is recommended as a minimum standard practice in maternity hospitals^6,8. BFHI designated hospitals must meet certain conditions stipulated in their “Global Criteria” (summarized in the Ten Steps to Successful Breastfeeding Guide). Among the criteria is to demonstrate an exclusive breastfeeding rate at hospital discharge above 75%, a requirement that rose to 80% in 2018⁸, and to adhere to the International Code of Marketing of Breast-milk Substitutes. In Spain, facilities obtain this designation in four progressive phases: 1D, 2D, 3D and 4D⁹. Phase 1D only documents the commitment of the hospital administration and the breastfeeding committee to bring about change towards good practice. When a hospital reaches phase 4D it is conferred the Baby-Friendly (BF) designation.

Training health workers in breastfeeding improves breastfeeding rates^10,11 and the pediatricians opinion is especially important in breastfeeding support^10,12. However, scientific literature reports that in many occasions pediatricians and residents in pediatrics are lacking in sufficient training^13,14,15 and they are often influenced by their personal experience¹⁶. The Residency in Pediatrics training program in Spain covers training in breastfeeding although not in a standardized way¹⁷. Participation in breastfeeding courses is voluntary and dependent on the interest of each resident and hospital.

Since the publication of the most comprehensive report about training in breastfeeding among Spanish residents in pediatrics in the year 2003¹⁸, promotion and protection of breastfeeding has improved substantially in Spain¹⁹: the number of BF designated or working towards BF designation facilities has increased and the Spanish Pediatric Association has developed an active Breastfeeding Committee.

The aim of this study was to assess the current state of knowledge regarding breastfeeding among Spanish residents in pediatrics.

SUBJECTS AND METHODS

A cross-sectional study was conducted by means of a questionnaire-based survey among Spanish residents in pediatrics. The Residency in Pediatrics in Spain is a 4-year program (residents are known as R1, R2, R3 or R4 depending on the year they are in) and the curricular year begins in June. The survey took place in April-May 2016. In Spain, in the year 2016 there were 1,612 places for residents in pediatrics in 105 hospitals. Of these residents, 70.5% trained in non-BF designated hospitals, 20.3% in hospitals working towards BF designation (Phases 1D-3D) and 5.7% in BF designated hospitals. For a type I error of 5%, a standard deviation of 15.9 points²⁰ and a standard error of measurement of 5 points we calculated a sample size of 142 surveys. We defined three strata according to BF designation, with a proportionate number of residents each. We performed a stratified and cluster (hospitals) randomized sampling. We contacted the Head Resident of each hospital personally and asked them to distribute the surveys among their residents. To improve each Head Resident’s response rates we sent a pre-invitation, a link to the survey and three reminders spread out over 7 and 14 days to their residents. The authors contacted the Head Residents on several occasions to improve response rates²¹. In order to compensate the losses, 315 surveys were sent. We requested the participation of all the residents of each hospital, but, in facilities that had more than 20 residents, we selected a maximum of five surveys per year by randomized sampling to avoid the excessive weight of the large hospitals.

Data was obtained by completion of an online form from Google platform with control of duplicates, no blank questions allowed and direct import of the data to a LibreOffice spreadsheet. We gathered general information about the participants (gender, age, year of residency, previous training in breastfeeding and personal experience in breastfeeding). To measure the knowledge about breastfeeding we used the ECoLa questionnaire (Encuesta de Conocimientos en Lactancia, “Questionnaire about breastfeeding knowledge”)²⁰. The complete data dictionary as well as the study database and complementary resources are available at aelama.org²². It also includes supplementary information on sampling and variables available as additional resources.

We considered the residents’ scores according to their hospitals BF designation and stage of training: first and second year residents (younger residents) or third and fourth year residents (older residents). The comparisons between groups were performed using confidence intervals at 95% (CI 95%) and the hypotheses contrasts were performed using Wilcoxon and Kruskal-Wallis tests, considering P value with Bonferroni correction p<0,025. We used R for the statistical analysis²³.

The study was approved by the Research Committee of the “Hospital Universitario 12 de Octubre” in Madrid. The data was collected anonymously.

RESULTS

We selected 21 hospitals: 2 BF designated hospitals, 4 hospitals working on BF designation and 15 non-BF designated hospitals. Inside 315 surveys that were sent, we received 189 (response rate of 60%). We randomly eliminated 12 answers from two hospitals with more than 20 residents, leaving 177 surveys for analysis.

The socio-demographic characteristics are shown in table 1. We found no socio-demographic differences among the residents according to the type of hospital, however, the residents from non-BF designated hospitals qualified their training in breastfeeding more poorly and their attendance to courses in breastfeeding was lower. Independent of the type of hospital, most of the residents considered the importance of receiving training in breastfeeding.

Table 1. Sample characteristics 

Total analyzed size is 177, as a maximum of 5 residents per training year were allowed for each hospital; BF: Baby Friendly hospital according to Baby Friendly Hospital Initiative; sd: standard deviation;

^(*)Of whom who have had personal breastfeeding experience or atten- dance to courses, respectively;

^(†)About breastfeeding;

^(‡)Full text of the question was: ‘Do you think you should be an expert in diagnosing and providing solutions to specific breastfeeding problems?

In table 2 the overall percentage of correct answers and percentages according to year of residency program, type of hospital, attendance to courses and personal experience in breastfeeding are shown. Training at a BF designated hospital and attending specific courses are the factors associated with better scores. The percentage of correct answers of residents with personal experience in breastfeeding was 82.7% and that of residents without was 73.1% (p=0.03). The results per hospital are shown in figure 1.

Table 2. Main results. Score as percentage of correct answers: globally and by sample subgroups. 

CI: Confidence Interval; IQR: Interquartile Range; BF: Baby Friendly hospital according to Baby Friendly Hospital Initiative;

^(*)Value at a distance from percentil 25 greater than 1.5*IQR, so it can be considered an extreme value (outlier).

Scores of the different hospitals ordered within their strata. Box and whisker plot. For each hospital, a box represents the interquartile range (IQR), the central line represents the median, the upper and lower whiskers reach respectively the maximum or minimum value at a distance of 1.5*IQR from the box. Extreme values are represented individually. Box width is proportional to the square of the sample size.

Figure 1. Successes percentage by hospitals according to BF accreditation. 

The percentage of correct answers to each question of the survey (both overall and stratified) are shown in table 3. It is worth noting that 51% of residents did not know of the International Code of Marketing of Breast-milk Substitutes (question 20); 50% did not know how to recognize basic warning signs in the breastfed newborn in the first days of life (question 9); 31% was not able to mention quote 2 correct criteria of breastfeeding assessment (question 18); 34% would limit the number of feeds in infants older than 6 months (question 12), and 36% do not acknowledge the nutritional value of breastmilk in young children older than 12 months (question 10). 28% percent of Spanish residents in Pediatrics would recommend limiting the number and duration of feeds (question 2).

Table 3. Percentage of correct answers on each test item. 

^(*)Full text can be found in the http://www.aelama.org/investigacion/conocimientos-en-profesionales/encuesta-nacional2016

BF: Baby Friendly hospital according to Baby Friendly Hospital Initiative;

^(†)Answer to this ítem is not considered as correct, but as desirable.

The residents that had taken part in breastfeeding courses obtained better results with a median of correct answers 22.3 percentage points higher than residents who had not attended courses (table 2). The relationship between self-confidence in their own competence and the level of knowledge is shown in figure 2.

Figure 2. Confidence in the own capabilities according to successes percentage. 

DISCUSSION

Residents in Pediatrics in Spain as a whole show acceptable knowledge, and higher compared to the results observed 13 years ago¹⁸ and to Canadian residents¹⁵.

The improvement of knowledge has been described by other authors: an American revision describes an improvement of knowledge among pediatricians¹⁷. It is important to note that the questionnaire used evaluates knowledge and basic skills²⁰ and that as the survey was performed towards the end of the curricular year the R4 residents would obtain their specialist qualification a few days after the collection of data.

We have not found better scores among the older residents. We have not found differences between global scores or in each particular question according to year of residency training. It has been reported that specific training in breastfeeding improves the knowledge of health professionals^24,25. In our sample, 64.4% of the total number of residents had attended specific courses in breastfeeding, a percentage much higher than among the Canadian sample (48%)¹⁵. This progress in training is likely influenced by the efforts directed specifically at residents in Pediatrics carried out by the BFHI and the Breastfeeding Committee of the Spanish Pediatrics Association. These efforts include national courses at a very low cost, such as the course in Teruel run by BFHI-Spain, which has gathered annually around 90 first and second year residents for the last 10 years. Other training examples are different online educational options²⁶. In any case, it is noteworthy that a significant 35.6% of residents refer not having attended any courses in breastfeeding and that the attendance to such courses was a lot higher among residents from BF designated hospitals or hospitals working on achieving the designation compared to non-BF designated facilities. This is probably because, in our country, despite the evidence regarding its importance, training in breastfeeding is not orgname, so undertaking a medical residency in pediatrics does not necessarily imply the acquisition of the minimum knowledge and skills required²⁰, but rather depends on the will of the physician or each hospital¹⁹.

In the BF designated hospitals, residents’ knowledge was higher compared to the other hospitals, with a median of correct answers 11.5 percentage points higher compared to non-BF designated hospitals. In the 1D BF hospitals and especially in the non-designated hospitals the dispersion of the scores was large and scores in all ranges were found: hospitals with high scores and hospitals in which the median of correct answers was close to 50% and where none of the residents attain 70% of correct answers. It is also noticeable that among the group of non-BF designated hospitals, the lowest score, at 19% of correct answers, was not considered an anomalous value. This contrasts with the little dispersion among the scores from BF designated hospitals in which the scores were found in a high and limited range. In BF designated hospitals, practically all residents surpass 70% of correct answers.

Other authors have reported differences in health professionals knowledge depending on their personal experience^15,27. In our study, only 14 residents had personal experience in breastfeeding and the difference (p=0.035) does not allow conclusions to be drawn on this point.

Regarding the survey’s questions, it is worth noting that most residents were familiar with the international recommendations on duration of breastfeeding and knew where to search for information about medications and breastfeeding. We also observed significant improvements in the knowledge of Spanish residents compared to similar questions surveyed years ago by Temboury¹⁸, although, when it comes to knowing the importance of assessing a feed or knowing how to recognize a growth spurt Spanish residents obtain lower scores than their Canadian peers¹⁴.

Even though the current recommendation is feeding on demand, other authors report worse results in their countries (32.3% recommend limitation in the US¹⁶ or 88.2% in Canada); in Spain, the result is worse compared with a similar question asked in 2003¹⁸, when only 18.8% defended limiting feeds. 36% percent of residents are not aware of the greater morbidity and mortality associated to formula-based feeding (26% in the US²⁸, 29.9% in Canada¹⁴). Only 60% of residents in pediatrics recognize a breastfeeding crisis and would give adequate advice (question 17); and faced with a child who is not gaining the expected weight without other signs of sickness (question 11), 20% would prescribe formula as the first course of action (64% would do so in Canada¹⁴). To summarize, residents have a better domain of the theoretical aspects than the practical ones. On another note, it is worrying that an important number of residents with insufficient knowledge feel confident about their competences.

We have assessed the knowledge about breastfeeding in a representative sample of the medical residents in Pediatrics in Spain, using a validated tool²⁰. The global response rate can be considered good²⁹(details on the response rates per hospital can be found in the appendix) and superior to similar studies^14,15,16.

One of the limitations of the study is that attitudes and communicative abilities were not assessed, reported by other authors as relevant aspects¹⁷. It may be interesting to include them in future assessments. Another limitation is that residents from hospitals in phases 2D or 3D were not surveyed. It should be taken into account in new studies, maybe having more strata.

Residents in Pediatrics in Spain are better trained today than a few years ago and probably better than their American or Canadian peers although there are important areas for improvement, particularly regarding practical issues and the International Code of Marketing of Breast-milk Substitutes and the risk of feeding with substitutes. It is worrying that even those with the worst training may see themselves as prepared. All this considered, along with the great disparity in training in different hospitals, shows the need for establishing a regulated breastfeeding training program, with minimum standards within the residency in pediatrics training program.

BF designation appears to have a positive influence in the training of future pediatricians, therefore reinforcing the advancement and support towards this initiative.