|Project Title: A population-level evaluation of a family-based community intervention for childhood overweight and obesity
||Professor Catherine Law
Professor of Public Health & Epidemiology
||UCL Institute of Child Health
||1 December 2010
||Research in progress
|Plain English summary:
Childhood overweight (which includes obesity) is an important health and societal problem. Children who have excess fat have an increased risk of poor health and lower achievement at school and socially, compared to children who are not overweight, and they are more likely to be overweight as adults. Rates of overweight are relatively high amongst children living in poorer socio-economic circumstances and those from some minority ethnic groups. Research has shown that some community-based programmes can help children and their families adopt healthier lifestyles and lose excess fat. However, these programmes have not been tested in the real world where children and their families live, so we do not know (outside of a research setting) for whom, where, how, and why they work.
We will study the MEND programme (Mind, Exercise, Nutrition, Do It!), a family-based community intervention for promoting healthy lifestyles in overweight children, currently running in more than 300 schools and leisure centres across England. We will use information already collected by the MEND programme on several thousand children to see who comes to MEND and whether they become less overweight. We will relate this to information about the local area and population, to find out whether people living in certain areas or from particular groups are more likely to use MEND than others.
We will also use the MEND programme information to investigate whether MEND works better/worse among certain groups (for example, children living with one parent or children from particular ethnic groups). With family groups, we will discuss whether taking part in MEND is easy or difficult for families, and why. We will ask about the types of things that families feel they have to change for example, the way they spend their leisure time or their money. We will also ask the views of families who decide not to go to MEND, or who start a MEND programme but don't complete it. We will ask those who plan public services why they decided to pay/not pay for a MEND programme in their area.
At the end of the project we will have some information on who comes to MEND, who finds that MEND helps them and in what circumstances. We should also have discovered something about why this happens. We will be able to use this information to develop weight management programmes for children, particularly for groups of children who don't seem to benefit from them at present. We may also learn more about how we can take information from research and use it to help improve children's lives, particularly children living in disadvantage.
Researchers in our team have skills in research and analysis on programmes, people, places and costs, and on what we can learn from children and families. We have done a lot of research on children's lives (including their physical growth and health), with a focus on children living in poorer socio-economic circumstances. The application has drawn on the views of children and young people and those who plan public services. The funds requested are for analysis of information already collected by MEND, and to collect and analyse information on what families and public services think about MEND.
Setting: Community-based health promotion intervention operating in over 300 schools and leisure centres throughout England from January 2006 - October 2010.
Target population: Overweight (body mass index (BMI) greater than 91st centile) or obese (greater than 98th centile) children aged 7-13 years.
Intervention: MEND, a multi-component, community-based intervention aiming to empower families of overweight/obese children to adopt and sustain healthy lifestyles.
Measurement of outcomes and duration of follow-up in secondary analysis: Primary outcomes: change in BMI after participation in MEND, adjusted for age. Secondary outcomes: post-intervention changes in body esteem, self-esteem and 'Strengths and Difficulties' questionnaire (SDQ) - all validated. Duration of follow-up: 10 weeks using service data, benchmarked against 10 weeks, and 6 and 12 months in original and concurrent RCTs (funded separately).
Secondary data analysis sample size: we will use all the data that are available, as there is no resource implication or rationale for selecting a smaller sample. If differences in BMI change between sub-groups (e.g. children of lone versus couple parents) were 0.3 kg/m2, then 260 children would be required in each group, using a conservative estimate (0.025) of the intra-class coefficient. Power will be less for multilevel modelling and for testing interactions.
Qualitative component sample size: for the family-based focus groups, 30 families (approximately 100 individuals), covering a range of demographic and socio-economic circumstances, and 30 individual interviews.
Also for interviews, 30 Primary Care Trust/Local Authority commissioners.
- RQ 1 Who participates in MEND? Description of participant characteristics and comparison with that expected, estimated from routinely collected population data (e.g. to examine whether MEND reaches disadvantaged children proportionately).
- RQ 2 Does outcome after participation in MEND vary by sub-group of participants? We will first describe the outcome (primary outcome is BMI change) seen in service-level data, using random effects modelling after adjusting for clustering within programmes. We will benchmark this against results seen in the original trial. We will then use multi-level models to estimate outcome by characteristics of individuals (e.g. children of lone vs couple parents) and area of residence of the child. In the models, the first level is the individual and the second level is the area of residence, with MEND centre/programme as a cross-classification. We will test for pre-specified interactions. Models will be adjusted for BMI at baseline to adjust for regression to the mean
- RQ 3 What is the cost of providing MEND? This will be estimated using financial data provided by MEND
- RQ 4 What is the salience and acceptability of MEND?
- RQ 5 What types of costs are borne by families through participation in MEND and sustaining healthy lifestyles thereafter? Data from the qualitative component will be transcribed, coded and analysed using the constant comparative method.
Months 1-4: confirmation of ethics permission, collation and anonymisation of data at MEND, data transfer to Institute of Child Health, prepare qualitative component. Months 5-7: obtain routine data on areas, assessment of data quality, link MEND data with area-level descriptors, construct optimal data sets, obtain customised prevalence rates, qualitative sampling and data collection, first meeting of Study Steering Committee (SSC). Months 7-20: secondary data analysis, including multi-level modelling, qualitative data collection, transcription and analysis. Months 18-20: meeting of SSC, discussion with young people's group. Months 20-24: preparation of papers, synthesis of results, final report, meeting of SSC (to agree final report), dissemination.