Progressive universalism is a core principal in the Healthy Child Programme and obesity prevention one of its priority areas. Guidance has yet to be developed on how health professionals should identify those babies and toddlers at increased risk for obesity. This is important in order to ensure that valuable resources are targeted for those at highest risk. Factors such as family history, events in pregnancy and family lifestyle may be as important as weight gain.

We propose to come up with a simple tool to help professionals in primary care identify babies at higher risk by analysing data from the ALSPAC cohort. If this initial work looked promising it would need to be validated against other data sets and piloted before it could be introduced into the Healthy Child Programme.

This work has another potential benefit. It would allow a more complex tool to be developed for use by researchers working in the area of antenatal and infant obesity prevention. This would allow them to target babies at highest risk for obesity for epidemiological purposes and recruitment to trials. To date maternal obesity alone has been used as a too simplistic proxy for infant risk.

The project would be coupled with another piece of work for the Department of Health. Analysis of the ALSPAC dataset would also be used to contribute to discussions about the optimum age for measuring children if screening for obesity was introduced during the preschool years.

Both aspects of this work have relevance beyond the UK and extend to antenatal as well as preschool guidance. The Centers for Disease Control and Prevention in the USA has committed support and the funds needed for the first phase of the project. If the initial analysis looks promising application for further funding would proceed in partnership with them.

AIMS:

1. To develop a tool that would be of practical use to clinicians, particularly health visitors (and potentially midwives), to help them identify babies/toddlers at high risk of obesity and ensure that they are targeted for appropriate guidance.

2. To develop a more complex tool to help researchers target populations within cohort studies and for inclusion in trials

3. To advise the Department of Health on the optimal time to weigh and measure preschool children for screening or monitoring purposes

METHODS

A. Development of a simple tool to support anticipatory guidance

PHASE ONE

Literature review

We propose to start by reviewing the literature to explore perinatal and infant factors that are known to be associated with obesity later in life. These include:

* Birth weight (LGA or SGA)

* Rapid weight gain in first year

* One or two parents obese

* Maternal obesity pre- and during pregnancy

* Maternal weight gain during pregnancy

* Maternal age at birth

* Gestational diabetes

* Smoking in pregnancy

* Family history of type2 diabetes and cardiovascular disease

* Race and ethnicity

* SES

* Bottle feeding

Data analysis

The ALSPAC data set will be examined to ascertain the risks for obesity associated with these factors. Obesity at the age of 13 years will initially be used (both on clinical measures and by DEXA scanning). This will be re-assessed once the results at the age of 15 years become available. Some of these variables are fixed for a child (eg ethnicity) and others will vary by age (eg weight at the time of examination). We are likely to use multi-level models to make use of all the data.

We will then assess the extent of the risks for each variable, and for the variables in combination. We will judge whether the final model predicts risk adequately to make it worthwhile to proceed to developing a tool for clinical use.

Development of the clinical tool

The tool will be developed presenting the data in a simple way so that it can be readily used in the clinical situation. This should be of value for identifying babies at risk and for explaining risks to parents directly.

PHASE TWO (Funding to be sought through the CDC if Phase 1 is promising

Piloting and validating the clinical tool

A small number of primary care professionals will be recruited to pilot the tool - health visitors and midwives in the UK, and primary care paediatricians in the USA. We will seek their views about its value and ease of use. If the tool seems promising a properly evaluated trial will be required before more widespread introduction is considered.

Validation is also needed at this stage. We propose to approach the Gateshead Millenium cohort in the UK and the VIVA cohort in Boston to ask if they would like to collaborate to test the accuracy of the tool in these different populations. Other cohorts that might be considered are the Raine cohort in Perth and the Southampton Women's study in the UK.

The research tool

A more complex measure for use by academics and epidemiologists will also be developed. This will be published in the academic literature and offered as a way to identify babies at risk more effectively. This too would require validation against other data sets.

B. Exploration of the predictive value of weight and/or BMI at different time points through the preschool years

The Department of Health is considering whether there might be value in reintroducing growth monitoring during the preschool years. It would be helpful to know if there is an optimal time point for collecting these data for the purposes of providing guidance or more intense input.

The ALSPAC data set will be analysed for the predictive value of weight and BMI at the various time points when weight and height are measured. This will help inform discussion about the value of screening and the best time to measure children taking into account other practical issues such as contact for other reasons.

COLLABORATION BETWEEN THE CDC AND UK DEPARTMENT FO HEALTH

Professor Mary Rudolf, paediatric consultant to the DH and currently on sabbatical at the CDC will work with Dr David Freedman, a senior epidemiologist at the CDC, on the preliminary stages of the project. They may be assisted by an intern at the CDC who will conduct the literature review. A more detailed funding proposal to pilot the tools will be developed if the data analysis shows evidence that a clinical or research tool might be promising.