Overall study design
The design of this proposed study is a randomised controlled trial (RCT) (see Figure ). Two phases of the study are planned: an intervention phase for the first two years and a follow-up phase for the following three years. The current funding we have received covers the first two years of the study. Additional funding will be sought for further follow-up of participants, taking the total study period to 5 years. Ethical approval to conduct this trial has been granted by the Research Ethics Review Committee of the Sydney South West Area Health Service – Eastern Zone (ID number X04-0189).
Participants and recruitment
All pregnant women who attend antenatal clinics of Liverpool and Campbelltown Hospitals will be approached by research nurses with a letter of invitation and information about the study. These two hospitals have been chosen since they attract a large number of patients from the areas of South West Sydney with the relatively high levels of disadvantage based on Social Economic Indexes For Areas (SEIFA) rankings [30
] and a broad ethnic mix. Once eligibility is established and consent obtained, women will be asked to fill in a registration form with their contact information to allow the nurses to make further arrangement for the baseline data collection and random allocation to study condition. Mothers will only be entered into the study after written consent is obtained and eligibility criteria are met.
Women will be eligible to participate if they are aged 16 years and over, are expecting their first child, are between weeks 24 and 34 of pregnancy, are able to communicate in English and live in the local areas.
At the initial stage, women will be excluded from the study if they have a severe medical condition based on advice given by their doctors. Women who cannot give informed consent and those with known major fetal anomalies will also be excluded.
The further exclusion criteria after giving birth extend to:
• presence of a physical or intellectual disability in a child which is likely to influence physical activity or food habits;
• presence of a chronic health problem in a child likely to influence weight or height;
• inability of parents or carers to communicate in spoken English.
A consecutive sample of 782 first time mothers (391 in each arm of the study) will be required for the study. The sample size calculation is based on detecting changes in the main outcomes: weight status and key behaviour factors. We selected the larger result from the following two calculations:
1. To detect a difference in mean body mass index (BMI) z-score of 0.25 units between the groups at age 3 years a sample size of 252 for per group is needed. This was based on an estimated mean body mass index (BMI) of 16.8 and SD of 1.5 from an earlier pilot study of 174 children aged 3 years in Central Sydney [11
2. One of the main predictors for childhood obesity is television-viewing time [31
]. From previous studies we conservatively predict a 10% reduction in TV viewing after our intervention. A total sample size of 626 families (313 per arm) is required to have 80% power to detect a 10% difference between groups in the proportion of children watching more than 2 hours of TV/video per day at 2 years of age, at the two-sided 5% significance level. To allow for an estimated 20% drop out, we will recruit a total of 782 first-time mothers.
Registration and randomisation
Information collected from the registration form will be entered into a password-protected Microsoft Access database. A letter will be sent by the research nurse to inform the hospital obstetrician that the mother is enrolled in this research trial. An appointment for the first home visit will be made with all participating mothers by a letter, followed by a telephone call. The first home visit will take place between weeks 30 to 36 of pregnancy.
Random allocation to either the intervention or control group is concealed by sequentially numbered, sealed opaque envelopes containing the group allocation, which will be determined by a computer generated random number. Randomisation will be stratified by hospital, with a block size of 50. A research assistant who has no direct contact with participating mothers will be responsible for generating the random numbers and preparing the envelopes.
During the first home visit, immediately after the baseline data collection, the nurse will open the sealed envelope and inform the mother of the outcome of randomisation.
The intervention comprises eight home visits from a specially trained community nurse delivering the staged intervention, which includes one home visit at the gestation age of 30–36 weeks and seven visits at 1, 3, 5, 9, 12, 15 and 24 months after birth, together with pro-active telephone support. The timing of the visits corresponds to milestones in early childhood development, in particular with regard to healthy feeding practice, nutrition and physical activity as well as parent-child interactions (see Table ).
Stage and focus of the intervention
At each visit, the nurse will spend approximately one hour with the mother and infant, monitoring the parent-child feeding interaction and practice, and behaviours promoting physical activity/inactivity in the child. The nurse will identify the parental needs in relation to nutrition, feeding and age-appropriate physical activity/inactivity using a checklist. All information and recommendations provided to parents will be in keeping with National Health and Medical Research Council Dietary Guidelines [35
], the Australian Guide to Healthy Eating [36
] and the National Physical Activity Guidelines [37
]. One-to-one consultation focusing on feeding behaviour and recommended problem-solving activities will be conducted. Pro-active telephone support will be provided between home visits to support behaviour maintenance and change. An individualised information kit will be provided to the parent.
The control group families will receive the usual childhood nursing service from the Area Health Service, comprising one home visit by a community nurse within a month of birth plus possible baby clinic visits. However, for study participants in the control group there will be additional visits at baseline, 12 and 24 months, for the purpose of data collection only. To maximise the retention rate, we will also offer home safety promotion materials to the control group.
The proposed main outcome measures are set out in Table . Short questions about selected breastfeeding practices, food habits and physical activity levels will be used, which have been widely employed in population health surveys including the NSW Child Health Survey 2001 [38
] and the Childhood asthma Prevention Study [39
], and are supported by validity testing [40
Outcome measures for the study
In addition, two validation studies of measures of physical activity and nutrition for children aged 2–4 years are currently underway. These studies are being led by the NSW Centre for Public Health Nutrition, and the Discipline of Paediatrics & Child Health at the University of Sydney. The measurement tools will allow us to make detailed assessments of dietary intake and levels of physical activity and sedentary behaviour in order to detect the impact of the intervention.
Standard demographic and socio-economic information, and secondary measures will also be collected using questions from the NSW Child Health Survey 2001 [38
Measurement points for both control and intervention groups for Phase One of the study will occur in participants' homes at baseline, 12 and 24 months. Phase Two measurements will take place at 3, 4 and 5 years.
Nurses trained in anthropometric techniques will collect measurements of height, weight and waist size in children at 2, 3, 4 and 5 years using stadiometers, scales and tape measures.
Phase two study
Additional funding will be sought for a further follow-up of the cohort to assess the outcomes indicated above at 3, 4 and 5 years of age. This is an important dimension of the project, which will strengthen the overall study in the following ways. First, longer-term follow up will allow an assessment of whether any behavioural changes resulting from the intervention are sustained over time. Second, nutritional and physical activity behaviours in the first two years of life may not have a significant bearing on weight status during this immediate period. Finally, longer-term follow-up of the control group will permit prospective, observational analysis of the prevalence and determinants of overweight and obesity within a large, population-based sample during the early years.
BMI will be calculated as weight (kg)/height (metres) squared and children at two to five years of age will be classified as not overweight or obese, overweight or obese, based on the International Obesity Taskforce recommended age-standardised BMI cut points [43
The outcomes will be compared between the intervention and control groups. For continuous variables, such as knowledge and attitudes scales and BMI, means will be compared using t-tests, or non-parametric equivalents for non-normally distributed variables. For categorical variables, chi-squared tests will be used. All analyses will be by intention to treat.