Determining rates of overweight and obese status in children using electronic medical records: cross-sectional study

Background — Routinely collected data from electronic medical records may be used to determine child overweight and obesity prevalence. The objectives were to determine the prevalence of child overweight and obesity, by age, sex, and visit type, using data from the Electronic Medical Record Administrative data Linked Database (EMRALD) database in Ontario.

Methods — Heights and weights were abstracted for children 0-19 years of age in EMRALD with at least one well-child visit from January 2010 to December 2011. Using the most recent visit, the proportion and 95% CIs of subjects defined as overweight, and obese, were compared by age group, sex, and visit type using the WHO growth reference standards.

Results — There were 28,083 well-child visits over this period. Of those children who attended well-child visits, 84.7% of visits had both a height and weight documented. Obesity rates were significantly higher in 1-4 year olds compared to children under 1 (6.1% vs. 2.3%), and in 10-14 compared to 5-9 year olds (12% vs. 9%). Both 1-4 (7.2% vs. 4.9%) and 10-14 year old boys (14.5% vs. 9.6%) had higher obesity rates compared to girls. Overweight and obesity rates were lower using data from well-child visits, compared to other visits.

Interpretation — EMR may be a useful to conduct population-based surveillance of child overweight and obesity. Methodological standards, however, should be developed.