Approval for the study was obtained from Hunter New England (HNE) Human Research Ethics Committee and the University of Newcastle Human Ethics Research Committee in 2009.
HNEHREC Reference No: 08/12/17/4.02 NSW HREC Reference No: HREC/08/HNE/403.
HNE SSA Reference No: SSA/08/HNE/UoN; H-2009-0106.
The trial was registered with the Australian Clinical Trials Registration in 2009, trial number ACTRN12609000356268.
Written informed consent was obtained from all participants prior to their enrolment in the FHFK study.
Design - randomized control trial
A prospective randomized control trial (RCT) to measure the impact of providing parents of pre-school aged children with self-directed nutrition and parenting resources on selected dietary and child feeding factors variables, following intention-to-treat principles, with secondary per-protocol analysis.
The criteria for inclusion in the RCT are shown in Table .
Inclusion and exclusion criteria for recruitment into the feeding healthy food to kids randomized controlled trial
Participants will be recruited primarily from Children’s Services (Long Day Care Centres, Pre-schools, Family Day Care, in-home care, Playgroups) and by Early Childhood health professionals who have direct contact with parents of children in target age group, and work in study locations. The primary recruitment strategy will be reinforced by strategic dissemination of flyers and newspaper advertisements in order to maximise the effect of successive approximation, whereby potential participants are exposed to the study more than once, to increase likelihood of participation.
Returned baseline survey envelopes will be numbered sequentially from one to 150 before being opened. A computerised random number generator will be used to generate 75 random numbers between 1 and 149, which constitute the ‘intervention’ numbers, with the remaining numbers being the ‘control’ numbers. Allocation of participants to the control or intervention group will be conducted by one member of the research team (KD), who is blinded to participant identity. Participants will be blinded to group allocation throughout the trial. Randomisation numbers will only be used to assign participants to their group, before switching to the use of unique research codes based on location, research phase and consent form numbers. The flow of participants through the trial is outlined in Figure .
Flow of participants through the feeding healthy food to kids randomized controlled trial.
Vegetable and energy-dense, nutrient-poor food (and beverage) consumption are the primary outcomes of FHFK. Vegetable consumption was used to determine sample size because evidence suggests this to be the most challenging dietary factor to change [9
]. Based on 80% power to detect a significant difference between the intervention and control groups of 37.5 grams (0.5 serves) of vegetables (P
0.05, two-sided), a sample size of 100 participants is needed at 12
months. Allowing for a drop-out rate of 30%, up to 150 participants will be recruited.
Theoretical framework for FHFK study design
Changing the dietary intake of children requires change to the feeding practices of parents. The Theory of Planned Behaviour (TPB), originally developed to predict and explain human social behaviour, is being used in FHFK to serve as a framework for a behaviour change interventions. In the context of FHFK, the key components of the TPB are proposed to predict the child feeding practices or behaviours of parents. Table summarises how TPB has been applied to change parental behaviour in order to increase child intake of vegetables within the FHFK trial.
Application of the theory of planned behaviour to feeding healthy food to kids randomized controlled trial
The resource selection process for use in the FHFK study involved three sequential steps.
1) An initial review of literature to identify key features of successful nutrition interventions (see Table ).
Enablers, barriers and resources for parent engagement in health behaviour change
2) Mapping of available Australian nutrition and parenting resources against key success factors and capacity for application of resources as a low intensity population level intervention.
3) Mapping of outcome measures for FHFK RCT against resource modules/sections; vegetable consumption: reducing energy-dense, nutrient-poor foods: parenting skills: and child feeding practices (see Table ).
While none of the reviewed resources contained all of the nutrition and parenting elements measured in FHFK, two resources were selected for use together. The evaluation of resources by three research team members indicated that the combination of these two resources provided an optimal balance of parenting and nutrition information in a desirable format for an information technology intervention, with capacity to be applied across a population.
The Tummy Rumbles [45
] interactive nutrition education CD is a self-directed resource that was adapted from an early childhood nutrition education program for Childcare staff and parents. The resource is divided into modules that include: the five food groups, dietary fats, fussy eaters, healthy lunchbox ideas, food budgeting and reading food labels. It has been evaluated by users as a useful and effective resource for early childhood nutrition education [46
Raising children [22
] is ‘a guide to parenting from birth to 5’, the content of which is based on the principles of the Raising Children [22
Australia’s definitive parenting resource. It contains different sections for parents of newborns, baby and child. Participants in this study were requested to view the DVD’s child section, specifically the segments on eating strategies, junk food, encouraging behaviour, minimising choking risk, play and learning.
The resources selected for the control group were provided to participants with the intention of enhancing retention of control group participants, providing placebo contact with researchers, preventing resentful demoralisation due to not receiving active intervention [47
] and blinding the participants to study group allocation. “Here’s 3 steps you should know
…” is a generic three-fold brochure that is widely available across the study locations, and promotes three key health messages; drink water, get active and eat fruit and vegetables [48
]. Active alphabet
] is a set of two books, one targeted at parents and one version specifically for use by children. They are colourful, informative and provide an A to Z guide to increasing physical activity with young children through development of fundamental movement skills.
Participants who complete baseline surveys and are randomised will be sent their respective intervention or control group resources by mail within one week of receipt of baseline surveys. They will be followed up by their preferred communication method (telephone call, text message or email) after another week to ensure that the resources had been received. Resources are accompanied instructions explaining how to access the information on the interactive resources. Minimal guidance regarding the recommended frequency of utilisation will be given, to mimic real-life dissemination of resources, and to simulate how parents might access resources within a low intensity intervention.
Participants will receive a reminder by telephone, text or email about using the resources when the three month surveys are administered. No further prompting of participants to use the resources will be provided prior to the final data collection at 12
months. At the trial conclusion participants will be sent an order form for a range of free resources for participating in the study (see Table ). These include the Tummy Rumbles CD and Raising Children DVD (for control group participants), plus a recipe book and supermarket pocket guide. The order form also contains a section to be completed by participants if they are willing to be involved in further studies related to the Feeding Healthy Food to Kids RCT.
Time line and participant requirements for FHFK study
Data collection procedures
Parents who provide written consent to participate will be mailed baseline surveys; the Australian Toddler Eating Survey (ATES) Food Frequency Questionnaire (FFQ), Child Feeding Questionnaire, the parenting questions from the Longitudinal Study of Australian Children (LSAC) questionnaire, demographic data questionnaire, instructions for the completion and return of surveys, and a stamped return envelope. Participants will be informed in the instructions for completion that the surveys will take a total of 30 to 45 minutes to complete. A period of six weeks will be allocated for return of surveys. If the surveys are not returned after four weeks, participants will receive a reminder via their preferred contact method (phone call, text message or email) to encourage them to return the surveys. Returned survey envelopes will be numbered for blinded randomisation prior to opening envelopes. The researchers will then check incoming surveys for completeness, and contacted participants for clarification or completion of missing questions.
Three and twelve month data collection involves repeated administration of all questionnaires administered at baseline, except for the demographic data questionnaire being excluded and a resource utilisation questionnaire being included (see Table ).
Quality control procedures will be employed to optimise the quality of the study and maximise validity and reliability of the program delivery and outcome assessments. These include:
Assessor blinding: Assessors of the main outcome measures are blinded to participant group allocation. Food frequency questionnaires will be checked for completeness and sent directly to the University of Newcastle by data collection personnel. Child feeding and parenting questionnaires will be analysed by an individual other than the assessor.
Written documentation: All written documentation, including assessment protocols and letters sent to participants, are standardised and subject to institutional ethics committee approval.
Training: Data collection personnel will be trained prior to data collection. Standardised procedures for data cleaning will be documented and provided to data collection personnel. Where possible, the same assessors will be used for all assessments.
CONSORT guidelines: The study protocol followed the CONSORT guidelines for Randomised Controlled Trials.
Measurement instruments and outcome measures
Data used to establish the demographic profile of the study population will be collected using a demographic data questionnaire at baseline.
Demographic outcome measures
Categorical responses will be reported for the following outcome measures:
Parent: age group (five year increments), education level and Aboriginal Torres Strait Islander status
Study child: age group (six month increments), type of child care, Aboriginal Torres Strait Islander and health status
Child dietary intake
Dietary intake will be assessed using the Australian Toddler Eating Survey (ATES), a 120-item semi-quantitative food-frequency questionnaire (FFQ). The ATES is currently being tested for reliability and relative validity [50
] and demonstrates acceptable accuracy for ranking nutrient intakes in Australian toddlers 2 to 4
years. Additional validation studies are currently underway.
The ATES tool is used by parents to record their child’s frequency of consumption of a comprehensive, defined list of foods over the previous six month period. Each question contains between four and eight categorical responses. Parents will complete the ATES at baseline, 3
months and twelve months. The ATES surveys are computer analysed and each response is converted from the frequency of serves into an estimated gram weight, by auto calculation of serves, using the median intake per serve from the Australian National Nutrition Survey.
Nutrient intakes from the FFQ will be computed from the Australian AusNut 1999 database (All Foods) Revision 17 and AusFoods (Brands) Revision 5 (Australian Government Publishing Service, Canberra) to generate individual mean daily macro-and micro-nutrient intakes, and percentage Energy (%E) contributions of foods or foods groups.
Questions relating to specific food items from the FFQ will be aggregated categorically based on food groups corresponding to the Australian Guide to Healthy Eating: breads and cereals, fruits, vegetables, dairy foods, meat or alternatives and “extras” (energy-dense, nutrient-poor foods). The energy-dense, nutrient-poor foods category will be broken down into sweetened drinks, sweet packaged snacks, savoury packaged snacks, baked goods, confectionery, take-out foods and fatty meats. Sub categories are consistent with previous studies [51
Serves of fruits and vegetables will be calculated by summing the weight of food items in the FFQ coded as fruits or vegetables and dividing by the serve size dictated in the Australian Guide to Healthy Eating (fruits; 150
g and vegetables; 75
g). All other foods in the ATES are quantified using multiples of standard children portions from the 1995 Australian National Nutrition Survey of Children and Adolescents or Foodworks computer nutrition analysis tool.
Dietary intake outcome measures
Primary outcome measures are: Vegetables consumption (serves/day); Fruit consumption (serves/day and %Energy (E)); Total Energy-dense, nutrient-poor foods food intake (serves/day and %E).
Secondary outcome measures are: Total Energy (kJ and kCal); Dairy intake (serves/day and %E), Bread, grains and cereals (serves/day and %E), Meat and alternatives (serves/day and %E), Micronutrients: Iron and calcium (mg/day).
Child feeding measurement instrument
Parents will complete the self-report, 31–item Child Feeding Questionnaire (CFQ) [17
] at baseline, three and 12
months to identify whether the provision of resources influenced child feeding practices. The CFQ is designed for use by parents of children aged two to 12
years. Items are measured using a five-point Likert-type scale, with each point on the scale represented by a word anchor. Parental beliefs and attitudes regarding child feeding practices are measured in seven domains; perceived responsibility (mean of three items), parent perceived weight (mean of four items), perceived child weight (mean of three items), parents' concerns about child weight (mean of three items), monitoring (mean of three items), restriction (mean of eight items), pressure to eat (mean of four items).
Scores for each question in each of seven domains are aggregated and divided by the number of questions in each respective domain. Mean scores for each participant can be compared with nationally representative data, and within the study at each time point and between groups.
Child feeding outcome measures
Parent perceived weight
Perceived child weight
Parents' concerns about child weight
Pressure to eat
The Longitudinal Study of Australian Children (LSAC) [53
] parenting questions will be used to evaluate parenting across a variety of domains. Six domains from the parenting section of the LSAC dataset will be analysed in the FHFK RCT. Warmth, Inductive Reasoning, Parental Efficacy
are assessed using a 1-5 point Likert scale with responses ranging from ‘never/almost never’ (=1) to ‘always/almost always’ (=5). Warmth
is a 6-question domain that identifies the level of affection that a parent displays towards their child. Inductive Reasoning
is a 3-question domain that analyses a parent’s level of communication related to discipline. Parental Efficacy
is a 4-question domain that identifies child behaviours related to parental discipline. Overprotection
is a 3-question domain that identifies how much a parent attempts to shelter their child.
Self-efficacy is measured by a single question that asks parents how they feel, using a 5-point Likert scale ranging from ‘a very good parent’ (=1) to ‘not very good at being a parent” (=5). Parental Hostility is 5-question domain that assesses how often parents display anger related behaviour towards their child. This domain is measured on a 10-point Likert scale ranging from ‘not at all’ (=1) to ‘all the time’ (=10).
Scores for each question in each of the four selected domains are aggregated and divided by the number of questions in each respective domain. Mean scores for each participant can be compared with nationally representative data, and within the study at each time point and between groups.
Parenting outcome measures
Parental Efficacy and
Demographic data, child feeding and parenting data and resource evaluation data will be manually entered into STATA by one of the researchers (KD). The ACAES surveys will be cleaned and scanned in a customised nutrition analysis program and imported into STATA format for analysis. Normality checks on all data will be conducted prior to further analysis. No imputation of missing values will be carried out for subjects; the study and subjects will be analysed in their allocated, randomisation group.
Analysis will be on an intention-to-treat basis. Differences in vegetable consumption, fruit consumption, total energy-dense, nutrient-poor foods food intake will be analysed using the random effects model, with the effects being time (baseline, three and 12
months), group (intervention or control) and group-by-time interaction. Statistical significance will be set at P
0.05. Statistical analysis will be completed using STATA statistical software (Version 10, College Station, Texas USA). The primary intention-to-treat analysis involved all participants who are randomly assigned and complete baseline and 3
month and/or twelve month surveys, regardless of whether they reported using the resources.
Secondary per-protocol analysis will be conducted to determine the relative impact of resource utilisation. Participants will be considered to have adhered to the study protocol if they report using Tummy Rumbles for at least one hour and Raising Children for at least one hour over the course of the 12-month trial. The control group will have met the protocol if they report using Active Alphabet at least once within the 12-month trial. The same analysis protocol will be followed after removal of data for participants who did not follow the study protocol.