Weight Development in Children With Obesity After Declining Treatment

Childhood obesity has been associated with increased risk of poorer mental health, later non-communicable lifestyle diseases (e.g. prediabetes, sleep apnea), and of continued excess body weight into adulthood. Multi-component family-centered lifestyle intervention seems to be the cornerstone in the treatment. However, only limited knowledge exists regarding weight development in children with obesity who decline to participate in such a lifestyle intervention. The aim of this study is therefore to investigate the long-term development in BMI z-score in children with obesity from 5 to 7 years of age who declined to participate in a lifestyle intervention. The development in BMI z-score for these children will be compared to children accepting the treatment and children who were never invited. A secondary aim is to investigate the potentially modifying effects of socioeconomic status (SES).

The participants:

This cohort study will include children from the city of Aarhus with a baseline visit between January 1st 2010 and June 30th 2020 children. The inclusion criteria are 1) 5-7 years of age and 2) obesity at baseline. Obesity will be defined by the International Obesity Task Force (IOTF) as BMI ≥ 30kg/m2 for age and gender. BMI z-score will be calculated be external reference population (WHO).

The expected number of children included:

• App. 150-170 children with obesity who declined to participate in the Aarhus intervention (Declining group).
• App. 400 children with obesity accepted participation and received treatment in a multicomponent family-centered lifestyle intervention (intervention group).
• App. 700-800 children with obesity living in Aarhus, who were not invited to participate in the lifestyle intervention (Non-intervention group)

Data sources:

• Participants will be identified using data from the municipality register.
• Anthropometrical data will be extracted from TM-Sund (Aarhus). TM-Sund is a database used to store data on anthropometrics obtained at health checks by specialized school nurses, Denmark.
• Data on SES, immigration and psychiatric illness will be obtained from Danish National registries.

Statistics:

A multiple imputation (MI) with chained equations will be applied to replace missing data regarding socioeconomic status and immigration, if missing data causes the models to lose substantial amount of observations, m=100. An adjusted mixed effects model with splines will be used to describe the development in BMI z-score for children declining participation and compare those to the reference and to the intervention groups. Knots will be placed at baseline and after ½, 1, 3 and 10.5 year. The model will be adjusted for age, BMI z-score, gender, family type, highest completed household education, equalized household income, immigration status, psychiatric diagnosis and psychiatric diagnosis (parents), zip code / distance to treatment center at baseline.

Ethics & permissions:

The local health ethics committee has approved the project, including data transfer (rec.no 1-45-70-27-20).

The project is internally reported to the University of Aarhus (rec no. 2916) The project has achieved approval from The Danish Data Protection Agency and the Principal Investigator has been granted accessed to register data from Statistics Denmark.

A full plan for analytics has been completed.

Inclusion criteria

• 5-7 years of age
• obesity at baseline as defined by the International Obesity Task Force (IOTF) as BMI ≥ 30kg/m2 for age and gender.

Exclusion criteria

• More than 6 months between a baseline visit and the next observation with a BMI z score
• First declining and subsequently accepting the intervention.