A randomized clinical trial comparing complementary feeding methods in mother-infant pairs: (A) strict PLW, which is the control arm; (B) strict BLISS; and (C) mixed method: a combination of PLW and BLISS created especially for this study. Further information about the randomized trial can be found in the published study protocol 16,17.
Participants were recruited online between 2019 and 2020 on university social networks, a university hospital, personal pages of health professionals, and aimed groups at mothers. The invitation to participate in the research that was published on the Internet (supplementary material) contained the telephone number and e-mail of the researchers for the mothers interested to contact. When contacting the researchers, the mothers were asked questions about the inclusion criteria and received guidance on the research procedures.
The inclusion criteria were mothers living in Porto Alegre, Rio Grande do Sul, Brazil, or nearby cities, with healthy singleton infants born at term, with birth weight ≥ 2,500 g, internet access, and who had not yet started the CF process. Children who did not present data on food consumption at nine and 12-17 months were not included in this study.
After signing the consent form, participants were sequentially numbered and had their names entered into a randomization list of three blocks and equal numbers, previously computer-generated (http://www.randomization.com) by blinded research staff. All families were advised to not start the complementary feeding until the sixth month. Only at the time of the first intervention, the mothers were informed of their group allocation. Assessment data were collected by a different researcher from the one responsible for the intervention, blinded to the allocation group, in a different room.
All participants received one of the three interventions during the study. First, when infants were 5.5 months old, there was a dietary workshop at a private nutrition office equipped with a test kitchen, in which nutritionists taught parents how to begin CF as appropriate for the group to which they were randomized. Participants and nutritionists cooked example meals together in real-time in the test kitchen.
At the same visit, speech therapists advised on choking prevention, showing videos with examples of choking and gagging. All advice was standardized; each group had its own set of printed material on topics to be addressed by the professional, and each participant received a copy to take home. All groups received information about continued breastfeeding and healthy feeding. The major differences between the intervention groups were the consistency of foods and the level of adult participation during each meal.
The intervention groups by methods included (more details in the supplementary material):
(A) Parent-Led Weaning group: mothers allocated to the PLW group, besides the standard information described above, were encouraged to spoon-feed pureed foods to their infants, following the traditional feeding method 2.
(B) BLISS group: in addition to the standard information, participants from the BLISS group were taught to prepare meals shaped into sticks, allowing the child to self-feed without adult interference 6.
(C) Mixed group: this group was instructed to combine the two methods explored above according to the child’s wishes.
After the first intervention, every mother received a dietitian’s phone number to call if they had questions related to the study of the CF process.
In the second interview, in a home visit at nine months of age, it was applied the 24-h dietary recall and any extra individualized advice or support for the CF process was provided, always appropriate for the approach to which each mother-child pair was allocated. During the COVID-19 pandemic, the 24-h dietary recall was applied by phone connection, and CF support was done online. The interview and the 24-h dietary recall at 12-17 months were applied at the Hospital de Clínicas de Porto Alegre (HCPA) in Centro de Pesquisa Clínica (CPC). During the COVID-19 pandemic, the 24-h dietary recall was also applied by phone connection and CF support was done online.
The 24-h dietary recalls were administered by trained researchers (referring to the preceding day) and designed to estimate the quantity, type, brands, time of the day, and cooking methods (including a description of any recipes used, with amounts of raw ingredients) of all foods and drinks consumed by the infant. If the mother considered the previous day to have been atypical, she was instructed to replace it with a day of the week more like the family's eating routine.
All diet records were entered into the dietary analysis in the nutrition software Dietbox® (using the Brazilian Food Composition Table, and Food Composition Table: support for nutritional decision 18 to analyze energy (Kcal/day), protein (g/day), carbohydrate (g/day), total lipid (g/day), total fiber (g/day), calcium (g/day), sodium (mg/day), vitamin C (mcg/day), vitamin B9 (mcg/day), iron (mg/day), zinc (mg/day), total sugar (g/day), trans fat (g/day), cholesterol (g/day), saturated fat (g/day), polyunsaturated fat (g/day), and monounsaturated fat (g/day).
The family's sociodemographic data were collected through an online questionnaire after signing the consent form at the beginning of the research. This questionnaire contained questions about maternal age (years), maternal education (years), total family income (Brazilian real), maternal race/ethnicity (White or non-White - Brown, Black, Yellow, or Indigenous), maternal marital status (with a partner or without a partner), sex infant (female or male), and the number of children (multiparous or primiparous).
In all moments, mothers and children were expected to answer questions about exclusive breastfeeding (EBF), and any breastfeeding (BF). EBF practice was defined as when the child received no liquid or solid other than human milk - not even water - except for oral rehydration solution, or drops/syrups of vitamins, minerals, or medications. Any BF practice was defined as receiving any amount of human milk by bottle, cup, or breast, independent of any other food offering 19.
The sample size for the main outcome (body mass index - BMI - for age) was calculated using the WinPepi® software version 11.65, based on previously published studies on the subject. Considering a unit standard deviation of 1, a statistical power of 80%, and a significance level of 5% to detect a difference in BMI of 0.8 kg/m², the sample analyzed for a difference of half a standard deviation consisted of 48 mother-infant pairs for each of the three intervention groups, totaling 144 mother-infant pairs. For the result of estimating nutrient intake and intake inadequacy, the sample power was calculated to estimate the impact of the results on the total number of infants followed at 12-17 months. The power to test whether there is a difference between the percentages of insufficient iron intake in the BLISS and PLW groups was 94.9%. This value, applying continuity correction, was obtained considering a significance level of 5%, a sample size equal to 35 subjects for the BLISS group and 43 subjects for the PLW group, the prevalence of inadequate intake was considered 26% for the BLISS group and 67.7% in the PLW group 9,20. This calculation was performed using the PSS Health® tool online version 21. Insufficient iron intake was used as a reference for the most common nutritional difficulty in childhood 22.
The database was created using the Statistical Package for the Social Sciences (SPSS)® program version 21.0, with double entry and subsequent validation. The data were analyzed according to intention-to-treat. The normality of quantitative variables was verified by the Kolmogorov-Smirnov test. Quantitative variables were described as mean and standard deviation (±SD) or median and interquartile range [P25 - P75] and categorical variables using absolute (n) and relative (%) frequencies. Quantile regression was used to estimate group differences in the intake of energy, protein, carbohydrate, total lipid, fiber, calcium, sodium, vitamin C, vitamin B9, iron, zinc, total sugar, trans fat, cholesterol, saturated fat, polyunsaturated fat, and monounsaturated fat. “Usual” intake data were used to assess the prevalence and relative risk of inadequate intakes, which were calculated based on the number of participants at nine and 12-17 months with an intake below the estimated average requirement (EAR) or recommended dietary allowances (RDA), as well as those above the tolerable upper intake level (UL) at 9 months and 12-17 months. Breast milk intake could not be directly measured, so it was assumed to be 646 g/day at nine months and 448 g/day at 12 months, based on breast milk volumes 23.
The EAR at nine months is 6.9 mg/day of iron, 2.5 mg/day of zinc, and 1 g/kg/day of protein. The RDA at nine months is 270 mg/day of calcium, 50 mg/day of vitamin C, 80 µm/day of vitamin B9, 95 g/day of carbohydrate, 30 g/day of total lipids, 19 g/day of fiber, 500 mg/day of calcium, 1 g/day of sodium, and 370 mg/day of potassium. The EAR at 12-17 months is 0.87 g/kg/day of protein, 100 g/day of carbohydrate, 13 mg/day of vitamin C, 120 µm/day of vitamin B9, 7 mg/day of iron, and 2.5 mg/day of zinc 24.
Ethical statement
This study was approved by the Research Ethics Committee of Hospital de Clínicas de Porto Alegre (under number 2019-0230, CAAE: 1537018500005327) and all methods were carried out by relevant guidelines and regulations, followed Resolution number 466, of December 12, 2012, of the National Health Council of Brazil. Also, written informed consent to participate in this study was provided by the participants’ legal guardian(s).
The trial was registered in the Brazilian Registry of Clinical Trials (ReBEC) with identifier [RBR-229scm U1111-1226-9516], [https://ensaiosclinicos.gov.br/rg/RBR-229scm]. The full data of the first registration was on 24/09/2019.