Child dietary and eating behavior outcomes up to 3.5 years after an early feeding intervention: The NOURISH RCT.

25th Aug 2021

Magarey A1, Mauch C1, Mallan K2,3, Perry R1, Elovaris R1, Meedeniya J2, Byrne R2, Daniels L1,2.

  • 1Nutrition and Dietetics, School of Health Sciences, Faculty of Medicine, Nursing and Health Sciences, Flinders University, Bedford Park, South Australia, Australia.
  • 2School of Exercise and Nutritional Sciences, Institute of Health and Biomedical Innovation, Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia.
  • 3School of Psychology, Faculty of Health Sciences, Australian Catholic University, Brisbane, Queensland, Australia.



To evaluate dietary intake impact outcomes up to 3.5 years after the NOURISH early feeding intervention (concealed allocation, assessor masked randomized controlled trial).


In this study, 698 first-time mothers with healthy term infants were allocated to receive anticipatory guidance on protective feeding practices or usual care. Outcomes were assessed at 2, 3.7, and 5 years (3.5 years post-intervention). Dietary intake was assessed by 24-h recall and Child Dietary Questionnaire. Mothers completed a food preference questionnaire and Children’s Eating Behavior Questionnaire. Linear mixed models assessed group, time, and time × group effects.


There were no group or time × group effects for fruit, vegetable, discretionary food, and nonmilk sweetened beverage intake. Intervention children showed a higher preference for fruit (74.6% vs. 69.0% liked, P < 0.001), higher Child Dietary Questionnaire score for fruit and vegetables (15.3 vs. 14.5, target ≥18, P = 0.03), lower food responsiveness (2.3 vs. 2.4, of maximum 5, P = 0.04), and higher satiety responsiveness (3.1 vs. 3.0, of maximum 5, P = 0.04).


Compared with usual care, an early feeding intervention providing anticipatory guidance regarding positive feeding practices led to small improvements in child dietary score, food preferences, and eating behaviors up to 5 years of age, but not in dietary intake measured by 24-h recall.

PMID: 27193736

Supplementary information

The NOURISH trial was developed as a response to the increasing prevalence of overweight in children under the age of eight years and the importance of prevention in managing this public health problem. Its design was based on our understanding of the importance of food and beverage intake in the aetiology of overweight and the role of parent feeding practices in determining children’s eating patterns. The eating environment of infants and young children is predominantly controlled by parents, particularly mothers. As ‘gatekeepers’ they determine infant exposure to food i.e. how frequently food is offered, the type of food offered and the amount; and respond to infant feeding behaviour such as food refusal. These feeding practices influence children’s eating patterns which are well established by the age of five years and tend to continue into adulthood. The degree to which parents control early feeding through restriction, monitoring and pressure has been associated with child food preferences and intake, and weight status.

Despite strong evidence for the role of parenting in determining food intake in young children there were few studies examining the impact of a parenting intervention on the eating behaviour of infants. The NOURISH trial was designed to fill this evidence gap (1) by evaluating an intervention that provided anticipatory guidance to first time mothers on early feeding. The intervention consisted of two modules (each comprised of six interactive group sessions), with content – the what (nutrition) and the how of feeding (parenting) – matched to developmental age at commencement. Mothers attended Module 1 when infants were approximately four months of age, just prior to the introduction of complementary foods, and module 2 at 14 months.

Five principles guided program content (i) the way we feed young children affects the foods they will like and their health, encapsulated in the slogan “Learning to like, liking to eat”, (ii) listen to and trust your child represented by “Parent provides, children decides” i.e. division of responsibility, (iii) habits are formed early and track to adulthood, (iv) set good examples for your child, and (v) your relationship with your child is important.

Module 1 focussed on establishing solid feeding and positive exposure to healthy foods, including type (variety, texture) and quantity (serve size). There was an emphasis on neutral repeated exposure to vegetables; limited exposure to high fat/sugar foods; recognition and trust in appetite, healthy infant growth and nutritional requirements. Module 2 provided strategies to manage the toddler’s growing need for autonomy and independence, namely neutral response to food refusal, dawdling or ‘fussiness’; avoiding use of coercion, rewards (food and non-food) or emotional feeding; supporting self-feeding, evolution of structured eating pattern; and role modelling healthy food choices. Both modules covered parenting warmth, consistency and self-efficacy, with a session specifically on the principle of attachment, such that parents recognise and respond appropriately to child feeding cues.

We hypothesized that compared with self-directed usual care, anticipatory guidance would result in an increased prevalence of protective feeding practices, healthy child eating patterns and behaviours, and a reduction in anthropometric indicators of obesity risk. A comprehensive range of outcome measures for child and mother were assessed and included maternal feeding styles and practices, parenting skills, child food intake, food preferences, feeding behaviour and weight and growth (1). Initially assessment took place at baseline, before randomisation (infant age 4 months) and 6 months after each module (mean infant age 14 months and 2 years respectively).

Previous papers report maternal and child outcomes from these assessments (2-4). At two years, there were a number of positive outcomes with respect to child eating behaviour, child food preferences and parenting. Compared with children of mothers in the control group (usual care) children of mothers who received the intervention displayed more desirable eating behaviours. These included emotional eating e.g. less likely to eat when anxious; fussiness e.g. less likely to refuse new foods at first; food responsiveness e.g. less likely to always be asking for food; and satiety responsiveness e.g. more likely to get full easily. Further, intervention children were reported to have more favourable food patterns e.g. they liked more fruits, but less non-core beverages (i.e. high in sugar/energy but provide few nutrients), there were fewer vegetables and more non-core beverages that they had never tried compared with control children. While a high proportion of women in both groups reported using favourable parenting practices in both groups, intervention mothers reported that they were more likely to use practices that allowed their child to regulate his/her own intake. There were no significant differences between groups with respect to actual food intake based on a 24-hour recall although a number of outcomes were consistently in the desirable direction. While overall these results indicate that providing information to parents on feeding practices that support healthy eating behaviours early in their child’s life is associated with modest improvements in child eating behaviours we postulated that it may not be until the children are older that the full effect is discernible.

Thus we undertook further assessments when children were 3.7 and 5 years i.e. 24 and 42 months, after completion of the intervention. The present study combines outcomes at these later points with those at 6 months post the complete intervention (children aged 2 years).

Of the 698 mother-infant dyads enrolled in the study at approximately 4 months of age, there were 541 (281 controls) at 2 years, 504 (254 controls) at 3.7 years, and 424 (211 controls) at 5 years representing 61% retention at the final point. Parents completed a 24-hour recall, the Child Dietary Questionnaire (CDQ) which assesses intake patterns at group level (5), child food preferences using a questionnaire adapted for Australian foods (6), and the child eating behaviour questionnaire(7).

Based on the 24-hour recall data we did not find any differences between the groups, or time x group differences in consumption of the amount of fruit and the amount of vegetables per kg body weight. However intervention children had a significantly higher fruit and vegetable score (a component of the CDQ) than control children. This score (where a higher value is more favourable) incorporates a range of elements of consumption and is potentially a better reflection of intake than one based on what was eaten yesterday. Intervention children were reported to ‘like’ a greater proportion of the listed fruits than control children although there were no such differences for ‘liked’ vegetables.

With respect to discretionary foods (high in fat/sugar/salt/energy and low in nutrients), we did not find any group or time x group differences in intake expressed as the contribution made to overall energy intake in the 24 hour recall, or the discretionary food score from the CDQ. Of note is the significant time difference in (i) the proportion of ’liked’ discretionary foods with this proportion increasing with increasing age and (ii) the contribution of discretionary foods to total energy intake which also increased with age.

There were no time x group differences for child eating behaviours but food responsiveness and satiety responsiveness were reported as more favourable in intervention children compared with control children. Interestingly, all eating behaviour scores changed significantly over time to become less favourable.

Although this intervention effect did not increase over time from what we observed at 2 years (6 months from the end of the intervention) as we hypothesized, it was maintained. A number of factors may have contributed to this. With module 2 delivered when children were approximately 14 months of age, families may need ongoing support into the pre-school years to maintain healthy habits, particularly as children are exposed to a wider range of influences outside the home. Also intervention dose of module 2 was relatively low, with only 45% of mothers in the intervention group attending ≥ 2 sessions. It may be that mothers had returned to paid employment or were expecting their second child, and these additional demands limited their attendance at face-to-face sessions. It is worth considering how new technologies could deliver information to parents at this busy time in their lives.

Overall, our results show promising effect, with anticipatory guidance regarding early feeding resulting in modest improvement of eating behaviours and dietary intake of young children.


  1. Lynne A Daniels, Anthea Magarey, Diana Battistutta, Jan M Nicholson, Ann Farrell, Geoffrey Davidson and Geoffrey Cleghorn. The NOURISH randomised control trial:Positive feeding practices and food preferences in early childhood – a primary prevention program for childhood obesity. BMC Public health 2009;9:387
  2. Daniels L, Mallan K, Battistutta D, Nicholson J, Perry R, Magarey A. Evaluation of an intervention to promote protective infant feeding practices to prevent childhood obesity: outcomes of the NOURISH RCT at 14 months of age and 6 months post the first of two intervention modules. Int J Obesity (2012) . doi:10.1038/ijo.2012.96
  3. Daniels LA, Mallan KM, Nicholson JM, Battistutta D, Magarey A. Outcomes of an early feeding practices intervention to prevent childhood obesity: the NOURISH RCT. Pediatrics accepted 27 March 2013 originally published online June 10, 2013 DOI 10.1542/peds.2012-2882
  4. Lynne Allison Daniels, Kimberley Margaret Mallan, Diana Battistuttaa, Jan Maree Nicholsond, Josephine Emma Meedeniya, Anthea Magarey. Outcomes of a child obesity prevention RCT (Part II): Child eating behaviours and weight status. Obesity (2014) 22, E104-E111. doi:10.1002/oby.20693
  5. Anthea Magarey, Rebecca Golley, Nicola Spurrier, Emma Goodwin, Fangyi Ong Reliability and validity of the Children’s Dietary Questionnaire; a new tool to measure children’s dietary patterns. Int J Paed Obesity 2009;4:257
  6. Wardle J, Guthrie CA, Sanderson S, Birch L, Plomin R. Food and activity preferences in children of lean and obese parents. Int J Obes 2001;25:971-7
  7. Wardle J, Guthrie CA, Sanderson S, Rapoport L. Development of the children’s eating behaviour questionnaire. J Child Psychol Psychiatry 2001;42;963-70