OTIS - Optimized Complementary Feeding Study

Introduction Evidence is accumulating that early feeding affects health and development later in life. The positive effects of breast feeding are well known, but less is known on how different types and timing of complementary foods, i.e. the foods used during the transition from breastfeeding or formula-feeding to family foods affect short and long-term health. During the time of complementary feeding, the demands for energy and nutrients remain high, especially during the second half of the first year of life. We and others have shown a link between protein intake during this time period and later risk for overweight. Earlier studies have implicated very high protein intakes during infancy but the latest dietary recommendations for infants and children suggest lower intakes, which may have contributed to the decreasing prevalence of obesity seen in some countries. Early introduction of carbohydrate-rich complementary foods has also been associated to alterations leading to later obesity and insulin resistance. Finally, we and others have shown that higher intakes of polyunsaturated fatty acids and lower intakes of saturated fatty acids in the diet of infants improve blood lipids and that these influences track into childhood.

The composition of the complementary food is important also from the perspective of food flavour, texture and presentation. Experiences of flavour during foetal and breast feeding period prime the newborn infant to accept the same types of foods as the mother usually takes, including foods with flavours outside the normal preferences of the infant, i.e. sweet, salty and fatty. Also the acceptance of different textures of foods changes during the infancy year. As the oral motor skills of the infant develop, the child is able to handle increasingly lumpy and solid foods. A normally developing infant will have periods of greater acceptance of novel foods both in terms of flavour and texture, usually coinciding with the time when complementary foods are introduced. However, this window of opportunity to introduce new foods begins to close around 10-12 mo. of age, when the child grows increasingly suspicious of unfamiliar tastes, i.e. food neophobia, but also, if solids haven't been introduced at this time, to the texture of common food items. Studies have shown that the types and numbers of different foods that a child accepts can be influenced by the manner and frequency of how they are introduced during this time. Numerous studies have shown that fruits and vegetables are an important part of healthy eating habits in all ages, but that the actual intake of these food items among young children is low. Instead, population-based studies from Sweden have shown that the dietary intake of children is not satisfactory from a public health perspective, with intakes of refined sugars, fatty foods and salt being too high and intakes of fruits and vegetables being too low.

In 2005 the Nordic Council of Ministers launched the New Nordic Food programme. This initiative highlights the health, gastronomic and sustainability aspects of foods produced within the Nordic region. Compared to the regular Swedish diet, the Nordic diet (ND) stresses higher intakes of regionally produced fruits, berries, vegetables, tubers, and legumes, higher intakes of whole-wheat, vegetable fats and oils, fish and egg, and lower intake of sweets, desserts and dairy, meat and poultry products, but at the same time being in line with current dietary recommendations. In adults, studies on ND have shown beneficial effects on weight and metabolic and cardiovascular disease markers of the same magnitude as e.g. the Mediterranean diet.

We aim to improve body composition, metabolic markers and the faecal microbiota composition of young children by using ND as a basis for complementary foods, with its emphasis on regionally produced fruits, berries, vegetables, tubers and legumes, also increasing the intake of fish and vegetable fats and oils, particularly using rapeseed oil but decreasing total protein intake. Effects on faecal microbiota composition have been a proposed pathway linking early dietary intake with later metabolic outcomes.

Relevance The early diet has longstanding effects on human health. Apart from breastfeeding little is known on which infant diets would confer long-term health benefits. ND has several potential advantages in this respect, and studies in adults have shown favourable effects on metabolic and cardiovascular disease markers and mortality similar to those gained through the Mediterranean diet. The present study, using ND as the basis for complementary foods but also reducing the total protein intake, improving fatty acid composition and expanding the introduction of fruits and vegetables into the diet of small children is the first study of its kind and will add vital information in the search to prevent obesity, dyslipidaemia, insulin resistance, type 2 diabetes, hypertension and cardiovascular disease and will have implications for future national and international dietary recommendations to young children. Through early intervention, i.e. already when complementary foods are introduced into the diet of the infant, we believe that we can initiate healthy food preferences from the earliest date thus maximizing the long-term public health benefits of healthy eating. By adopting the concept of ND we will also add gastronomical aspects to the development of optimal complementary foods as well as support regional food production and contribute to a sustainable environment.

Objectives The overall aim of the study is to compare the effects of a ND portfolio intervention on body composition, metabolic markers, blood pressure and faecal microbiota composition to conventional complementary feeding from 6-18 months of age.

More specifically, we will investigate if:

1. A complementary diet lower in protein (total reduction 30%) will affect growth, body composition, i.e. reduce total body fat mass and improve metabolic markers, i.e. decrease plasma (P) levels of insulin, glucose and insulinlike growth factor 1 (IGF1) without detrimental effects on linear growth.
2. A complementary diet based on Nordic foods will affect growth, body composition, i.e. reduce body fat mass and improve metabolic markers and the faecal microbiota composition, i.e. a microbiota composition associated with less obesity and inflammation without detrimental effects on linear growth.
3. Complementary foods with a modified fat intake, i.e. a higher intake of LCPUFA through an increased intake of fish and substituting fats and oils found in regular complementary foods with milk fat and rapeseed oil will improve body composition, metabolic markers, cognitive development and the faecal microbiota composition without detrimental effects on linear growth.
4. A systematic introduction of plant foods from the Nordic diet during weaning will increase the acceptance of new fruits and vegetables into the diet of the young child.

Methods Participants and recruitment Healthy, full-term singletons will be recruited at 4 months of age. When parents consider it appropriate to introduce complementary foods into the diet of their child, but no later than at 6 mo. of age, the child will be randomly allocated to one of two study groups, i.e. the Nordic diet group (intervention, ND) or the Regular diet group (control). The group allocation will be masked for the participants and researchers.

Study products All participants will be provided with study products free of charge. In the ND group these will be specially prepared, protein-reduced, age-adjusted milk cereal drinks (MCDs), baby cereals and baby milk, and commercially available baby foods in glass jars (BIG). The control group will be offered commercially available, age-adjusted MCDs, baby cereals, baby milk and BIG. All products will be manufactured by Semper AB, Sweden.

Introduction of complementary feeding The ND group will be given recipes of fruit and vegetable purees and instructions how to prepare these recipes. The main ingredients will be fruits, berries and vegetables which grow and are available in the Nordic region according to a predefined list. These taste meals will be given to the participating infant when the parents judge it prudent to start offering the infant other foods beside breast milk or formula and the infant is deemed ready by the parent.

The control group will be given the standard, oral and written advice that is given from well-baby clinics to parents when beginning with small taste meals.

Parental support Parents with children in the ND group will be given support to continue following the dietary intervention through social media (Face book). In this forum parents can discuss with the study dietician and among themselves, sharing experiences. Parents will also have instructional videos on the preparation of certain study foods and recipes. Participants will share experiences and other information at their own discretion. No data will be collected through this group, but the group may function as a way to inform the families on the group on upcoming study events and remind participants to send in e.g. questionnaires. Both the intervention and the control group will get monthly telephone calls from the research nurse or the study dietician.

Data collection Main data collection points will be at baseline, i.e. when the child is between 4-6 months of age, and at 9, 12 and 18 months of age. At these time points the participants will come to the Paediatric Research Unit for measurements and sampling.

Procedures The baseline visit and the 9mo. visits will take place at the Paediatric Research Unit at the Department of Clinical Sciences, Paediatrics, Umeå University Hospital (Figure). At ages 7, 8, 10, 11 and 13-17 mo. the research nurse or dietician will have a telephone follow-up with the participants checking adherence, possible study related problems, supply and consumption of study products and the symptoms registrations. At 12 and 18 mo. the participants will again come to the Paediatric Research Unit for samples. Also, the research nurse will do home visits for anthropometric measurements, TBW and the two videotaped tests, i.e. food acceptance test and the free play test.

Informed consent and withdrawal A written Informed Consent from the subject's parents/caregiver is an inclusion criterion without which the subject cannot be included in the study. The participants may withdraw from the study at any time by withdrawing consent, through noncompliance or if the participant experiences adverse events requiring the participant to withdraw on advice of the research physician or by his/her own accord. If possible the reasons for withdrawal will be recorded. Participants that are withdrawn will not be replaced.

1. Statistical analyses and ethical clearance Comparisons of differences in main outcomes between the ND and control groups will be the focus of the statistical analyses. The primary outcomes will be analysed according to the intention-to-treat principle when comparing the two study groups, but per protocol analyses will also be done. Both parametric and nonparametric tests will be used depending on the distribution of data. Possible effect modifiers and confounders will be included in the analyses as needed to explain group differences. Data on dietary intake will be converted to energy and nutrients through nutrient calculation software (Dietist NET, Kost och Näringsdata AB, Bromma, Sweden). The study has ethical clearance from the Regional Ethics committee, Umeå University (Dnr 201436331M).