Early introduction of solid foods and tiny tastings shortens the duration of breastfeeding

Background: Since 2011, the National Food Agency in Sweden has informed parents that they can introduce tiny tastings (1 mL of solid food, i.e., other sources of nutrition than breastmilk/formula) to infants from four months of age. It is unknown how national recommendations, which differ from the Word Health Organization’s recommendation, affect breastfeeding. We hypothesized that introduction of tiny tastings of solid foods would shorten the duration of partial breastfeeding. Methods: This retrospective study utilizes data from the longitudinal ‘Swedish Pregnancy Planning Study’ in which mothers were recruited at antenatal clinics and answered three questionnaires up to one year after birth (n=1,251). Linear regression models were used to analyze the association between the introduction of solid foods and the duration of breastfeeding. Results: The main ndings of this study were that half of all infants (48%) were fed with tiny tastings already in the fourth month, and correlation analysis showed that the earlier the infants started with tiny tastings, the earlier they ate larger amounts of solid food. In a multivariate linear regression analysis, ve factors were identied as having a negative effect on the duration of breastfeeding: low infant age at the introduction of tiny tastings, low maternal age, low level of maternal education, high maternal BMI, and twin birth. Conclusions: Early introduction of tiny tastings of solid foods shortened the duration of breastfeeding.


Introduction
The Word Health Organization (WHO) recommends exclusive breastfeeding for six months and partial breastfeeding for at least two years or longer (1). Exclusive breastfeeding provides more health bene ts than partial breastfeeding and is su cient as the sole nutrition for the rst six months of an infant's life (2). In both high-and low-income countries, breastfeeding has positive effects on both the child and maternal health. Exclusive breastfeeding decreases the risk of respiratory infections during infancy, type 1 diabetes, and childhood leukemia. Moreover, long-term breastfeeding reduces the odds of being overweight or obesity in children. Maternal health bene ts are related to a reduced incidence of breast cancer, ovarian cancer, and diseases related to the metabolic syndrome (3)(4)(5).
It is well known that mixed milk feeding increases the risk for breastfeeding cessation (6). However, research has shown con icting results if also early introduction of solid foods is associated with a shorter duration of breastfeeding. In the Swedish study above, there was no association, while one from UK both showed association and a dose-response effect (7). In this context, it is notable that these two countries differ a lot in breastfeeding habits. Early introduction of solid foods, i.e., other sources of nutrition than breastmilk/formula, is related to maternal sociodemographic factors, such as lower educational level, smoking, and lower age (8).
Even though breastfeeding is associated with health bene ts for both the mother and the child, breastfeeding recommendations in several high-income countries are in con ict with the WHO's recommendation.(1) It is unknown how national recommendations affect breastfeeding. Sweden is considered a pro-breastfeeding country and the breastfeeding incidence peaked during the 1990s when more than 40% of infants were exclusively breastfed for six months (9). This was partly a result of a planned strategy to increase breastfeeding rates, the Baby Friendly Hospital Initiative (10). However, the prevalence of exclusive breastfeeding for six months has decreased over the last few decades to 11% in 2019 (9). One recommendation, in particular, in the national guidelines has possibly had a signi cant impact on breastfeeding. Since 2011, the National Food Agency has informed parents that they can introduce tiny tastings (1 mL of solid food) from four months of age. Due to a non-evidence-based belief that exclusive breastfeeding might delay and complicate the introduction of solid foods, the Child Health Care had not recommended exclusive breastfeeding for six months. Based on this misunderstanding, the National Food Agency made the change in the national guidelines (11). This change has been utilized by the Swedish baby food industry, offering a range of products intended from four months of age. Furthermore, the Child Health Care continue to encourage parents to start providing food at four months and in contrast, women express that they want to breastfeed exclusively for six months (12). In the present study, we hypothesized that early introduction of solid foods, including tiny tastings, was associated with shorter duration of breastfeeding. This retrospective study was a part of the longitudinal 'Swedish Pregnancy Planning Study' (13) and approved by the Swedish Ethical Review Authority, d.nr. 2010/085, with supplemental applications during the years. The aim was to investigate at what age solid foods, including tiny tastings, were introduced to Swedish infants and to investigate the effect of introduction of solid foods on the duration of breastfeeding.
Antenatal clinics (n = 215) in ten (n=10/21) regions were invited to participate, and 153 (71%) agreed to participate. Women were asked to complete questionnaires during registration at antenatal clinics in early pregnancy (Q1), in the third trimester (Q2), and one-year post-partum (Q3). The recruitment was conducted in 2012 to 2015. A total of 5,494 women were initially approached, out of which 4,969 women accepted participation. In the end, 3,389 completed and returned Q1. The rst follow-up questionnaire (Q2) was sent to a total of 3,215 women, and 2,583 completed and returned the questionnaire. The second follow-up (Q3) was sent to 2,018 women, and 1,263 returned the questionnaire. A more detailed description of the procedure is previously published (14).
Four of these questionnaires lacked an ID and were not possible to match to previous questionnaires. For the current study, we also excluded women who had not provided data for all three questionnaires (n=4), who's child was no longer alive (n=3), and who provided data for an older sibling (n=1). The nal study sample comprises 1,251 women.
The self-reported questionnaires included sociodemographic questions about the mother (age, sex, previous children, country of birth, level of education, and household income), the pregnancy (level of pregnancy planning, single/multiple pregnancy), mode of delivery (how it started, ended, and if there was hemorrhage >1000ml), and the infant (birth weight, gestational age, sex, neonatal care, congenital states, and twins), see Table 1. Furthermore, there were a detailed question about nutrition for the infant's rst year (0-12 months): duration of breastfeeding, and introduction of solid foods including tiny tastings, see supplemental le.

Statistical analyses
The primary aim and sample characteristics were explored with descriptive statistics. Linear regression was used to analyze the effect of introduction of solid foods (independent variable) on duration of breastfeeding (dependent variable). Independent covariables were chosen based on previous knowledge on breastfeeding and are presented in Table 1. Independent variables were analyzed at univariate level, and all signi cant variables were then included in the analysis at multivariate level. Cox and Snell pseudo-R2 and Nagelkerke pseudo-R2 are presented as measures of the proportion of variation of outcomes explained by the model. For all statistical analysis, a two-sided p-value <0.05 was considered signi cant. Data were entered and analyzed using IBM SPSS Statistics version 26 (IBM Corp. Armonk, NY, U.S.A.).

Results
The age at which solid foods were introduced Background characteristics of the mothers and their pregnancies, deliveries, and infants are presented in Table 2. The median age for introducing solid foods was during the fourth month. Almost all participants (94%) introduced their infant to solid foods during their fourth to seventh month, most commonly during their fourth month (48%). Tiny tastings (1ml) were the most common kind of food intake during the third to fth month, tastings (5-10ml) the most common during the sixth month, and thereafter food in larger servings (15ml or more), see Figure 1. During their seventh month, more infants were fed with solid foods than with breastmilk. The regression analysis showed that the age of introduction of tiny tastings (1ml) was associated with age of introduction of food (15ml or more), i.e., the earlier infants started with tiny tastings, the earlier they ate larger amounts of solid foods (β 0.813, p<0.001).
The effect of introduction of solid foods on the duration of breastfeeding Page 4/11 Variables associated with the duration of breastfeeding on univariate level were included in the multiple regression model and are presented in Table 3. In the multivariate linear regression analysis, higher maternal age, higher maternal education, lower maternal body mass index, singleton pregnancy, and higher age at introduction of tiny tastings were associated with longer duration of breastfeeding (Table 3).

Discussion
The main ndings of this study were that half of all infants were fed with tiny tastings already in the fourth month and that the earlier the infant started with tiny tastings, the earlier they ate larger amounts of solid food. In the multivariate linear regression analysis, ve factors were identi ed as having a negative effect on the duration of breastfeeding: the infant's age at the introduction of tiny tastings, low maternal age, low level of maternal education, high maternal BMI, and twin birth.
The Swedish recommendation of tiny tasting from four months of age is contradicting to the WHO recommendation (1) and the impact on breastfeeding has not been studied before. This study shows negative effects on breastfeeding duration. To recommend exclusive breastfeeding for six months could help scaling up breastfeeding and generate bene ts besides those of the breastmilk itself, since breastmilk intake among children is associated with lower odds of consuming non-recommended foods, such as cookies, crackers, and sweetened drinks (15,16). Instead, Nutrition Committees continue to emphasize introduction of solids from four months (17). One common concern about exclusive breastfeeding for six months is the risk for iron-de ciency anemia. However, the risk can be successfully mitigated by delayed umbilical-cord clamping (18).
Con icting advices and non-evidence-based recommendations have a negative effect on breastfeeding (12). In many ways, it can be perplexing for women to breastfeed in a society that is not infused by a favorable attitude toward breastfeeding. Consequently, breastmilk substitutes have become a "multi-billion dollar industry" that has the opportunity to devote considerable nancial resources in uencing women not to breastfeed (19).

Socioeconomic factors with an impact on breastfeeding
The multivariate linear regression analysis showed that low maternal age and low education had a negative impact on breastfeeding. In previous studies, several socioeconomic factors have shown an association with a shorter period of breastfeeding. For example, mothers with less privileged economic situation and less education have a shorter duration of breastfeeding (20,21) and mothers with lower age breastfeed for fewer months (21,22). This contributes to unequal starting points for children, already from birth. Thus, it should be in focus for targeted interventions from the Child Health Care in order to promote equal health. The United Nations' Sustainable Development Goals obligate governments to promote healthy lives and welfare for all. This makes breastfeeding a central part of the 2030 Agenda, since it contributes to the achievement of an equal, healthy, fair, a uent, and sustainable future for both people and the planet (23).

Obesity and breastfeeding
This study showed that high BMI in the mother was a signi cant factor for shorter duration of breastfeeding. Maternal obesity is linked to many risks, and one of them is lower initiation rate of breastfeeding and also a greater risk of early breastfeeding cessation (24). It has been suggested that the causes can be a mix of physiological, behavioral, sociocultural, psychological, and medical reasons. For example, obese women can have higher progesterone levels which may impair lacto genesis. Furthermore, large breasts may lead to problems for the infant to latch on, and the obese mother may lack con dence in the breastfeeding situation because of low body image (25).

Less breastfeeding among twins
Another factor identi ed as showing a negative effect on the duration of breastfeeding in the multivariate linear regression analysis was twin births. Women who have given birth to twins face special challenges, and breastfeeding rates are lower among these infants (26). The reasons for weaning twins, according to the mothers, are insu cient milk supply and infants' problematic breastfeeding behavior (27). This indicates that mothers of twins need targeted breastfeeding support that takes into account these mothers' unique situation.

Strengths and limitations
This study was the rst to investigate the impact of tiny tastings on breastfeeding. Data were provided from a large number of mothers (n=1,260), and the sample represents different geographical areas including both high and low socioeconomic statuses. The question measuring breastfeeding duration (exclusive and partial) is very detailed, consequently, it may be more reliable than the Swedish national data (9),eventhough the retrospective data this is a limitation, because memories might be less accurate. Conversely, the study design cannot provide causes, rather, it shows associations. The response rate for the follow-up (Q3, n=1,251), compared with baseline data (a total of 3,389 women completed and returned Q1), was lower. In addition, there might be selection bias since the study design excluded non-Swedish speaking parents.

Conclusion
This is the rst study to investigate the impact of tiny tastings on duration of breastfeeding. The results revealed that half of all Swedish infants taste their rst solid foods during their fourth month, and that the earlier the tiny tastings were introduced, the shorter the duration of breastfeeding. Most conditions that affect breastfeeding are di cult to in uence, for example, the mother's educational level, BMI, age, and if she has given birth to twins. In contrast, national guidelines can always be updated. Swedish recommendations from the national authorities must adhere to international consensus and be designed to support breastfeeding.

WHO -World Health Organization
BMI -Body Mass Index Q1, Q2, Q3 -Questionnaire 1, questionnaire 2, questionnaire 3 Declarations Ethics approval and consent to participate The study was approved by the Swedish Ethical Review Authority, D.nr. 2010/085 and the mothers were ensured anonymity and the right to withdraw from their participation at any time without giving any reason.

Funding/Support:
No funding was secured for this study. The SWEPP-study was funded by the Family Planning Fund of Uppsala, the Uppsala County Council, the Faculty of Medicine, Uppsala University and the Uppsala-Örebro Regional Research Council, Sweden.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Competing interests
The authors declare that they have no competing interests.  Figure 1 Infants' nutrition during their rst 12 months

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.