Article Text


Chronic disease
P2-286 Systematic review with meta-analysis of population-based vs school-based studies of obesity prevalence among adolescents from Brazil
  1. B Soldateli1,
  2. S Fuchs1,2,
  3. F Schmidt1
  1. 1Postgraduate Studies Program in Epidemiology, Universidade Federal do Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil
  2. 2Hospital de Clínicas de Porto Alegre, Porto Alegre, Rio Grande do Sul, Brazil


Introduction In Brazil, obesity prevalence in adolescents has been investigated in different regions and periods of time. Studies of school-based samples have low cost and are easily conducted, being widely used. Otherwise, population-based studies require greatest investments, more time and researches, besides high complexity to minimise bias.

Objective To undertake meta-analysis of observational studies conducted in Brazilian adolescents to calculate pooled estimates for obesity prevalence for school-based vs population-based samplings.

Methods Data sources included Medline, Embase, Bireme, and CAPES-Thesis database searching for articles and grey literature about obesity prevalence in adolescents aged 10–19 years, from studies conducted in Brazil from 1980 to 2010. Studies were systematically reviewed, with no language restriction, and independently extracted by two investigators using a standardised protocol. Obesity was defined by body mass index (BMI) ≥95 Percentile, for sex and age. Data analysis was performed by Comprehensive Meta-Analysis 2.0®, using random effect model.

Results Among 1939 articles reviewed, 138 met the inclusion criteria, 56 were excluded, and 25 were not retrieved after three attempts to contact authors. Among 57 studies included in meta-analysis 14 were population-based and 43 school-based samples. No significant heterogeneity was detected (Q=8.25, p=0.001). Obesity prevalence in population-based studies was 5.5 (95% CI 4.2 to 7.1) and in school-based studies it was 5.1 (95% CI 4.3 to 6.0). School-based studies showed higher inconsistency measure of results when compared with population-based samples studies (I2=7.74 vs 0).

Conclusion School-based studies as well population-based studies provide similar estimates of obesity prevalence, since they are rigorously conducted regarding methodological aspects.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.