Abstract
Obesity is a major public health challenge and childhood obesity represents a growing threat to health. The aim of this epidemiological PhD project was to describe the current prevalence of childhood overweight and obesity and sociodemographic predictors.
Anthropometry was objectively measured in a nationally representative sample of 3166 eight year-olds in the Norwegian Child Growth Study in 2010, with a cross-sectional design. Measurements were conducted by school nurses at participating schools according to standardised procedures. Individual anthropometric data were linked to national register information. Generalized linear models (log-binominal regression) with a logarithmic link function to calculate relative risk or prevalence ratio were used. A methodological study investigated the impact of instrument error of stadiometers and scales, as this could lead to increased variance of the BMI distribution. Simulations were used and we found that instrument error might contribute to overestimation of overweight/obesity in population-based surveys.
The findings showed that general- and abdominal obesity were 1.5 – 2 times as prevalent among children from smaller municipalities as larger municipalities. Similarly, abdominal obesity was almost twice as prevalent among children of mothers with low education compared with highly educated mothers. Furthermore, general- and abdominal obesity were 50% and 75% more prevalent among children of divorced compared with married parents, respectively. These differences could not be explained by maternal education. Particularly high rates of abdominal obesity were found among boys of divorced parents compared with married parents but we did not find similar differences among girls.
Findings from epidemiological research are invaluable and enable health authorities to follow trends and identify high-risk groups; a means to prevent childhood overweight and obesity.