Article Text

Download PDFPDF
Emotional, behavioural and social correlates of missing values for BMI
  1. H Fonseca1,
  2. M Gaspar de Matos2,
  3. A Guerra3,
  4. J Gomes-Pedro1
  1. 1
    Department of Pediatrics, Hospital de Santa Maria, Lisbon, Portugal
  2. 2
    Faculty of Human Kinetics, Technical University of Lisbon, Lisbon, Portugal
  3. 3
    Department of Pediatrics, Hospital de S. João, Oporto, Portugal
  1. Dr Helena Fonseca, Hospital de Santa Maria, Department of Pediatrics, Av. Prof Egas Moniz 1, 1649-028 Lisboa, Portugal; helena.fonseca{at}


Objective: To examine the emotional, behavioural and social correlates of missing values for body mass index (BMI) in a nationally representative sample of Portuguese youth.

Design and methods: 6131 Portuguese public school students aged 11–16 years in 6th, 8th and 10th grades who participated in the 2002 HBSC/WHO survey of adolescent health were included. Those not reporting their weight and/or height were compared with their peers. Bivariate analyses of psychosocial and behavioural variables were conducted to compare the two groups. Multivariate logistic regression was used to determine if variables significantly associated with missing values for BMI at a bivariate level predicted missing BMI when controlling for all other variables in the model.

Results: Of 6131 adolescents answering the questionnaire, 661 (10.8%) did not report their weight and/or height. Gender was not associated with missing BMI. Missing values were significantly predicted by younger age (adjusted odds ratio (aOR) = 2.56, 95% CI 1.99 to 3.29, p<0.001), sedentary lifestyle (aOR = 1.53, 95% CI 1.16 to 2.01, p<0.01), poor body satisfaction (aOR = 1.34, 95% CI 1.19 to 1.51, p<0.001), absence of father (aOR = 1.62, 95% CI 1.14 to 2.30, p<0.01), lack of friends of the opposite sex (aOR = 1.65, 95% CI 1.03 to 2.66, p<0.05) and poor perception of academic achievement (aOR = 1.23, 95% CI 1.06 to 1.42, p<0.01).

Discussion: Our findings suggest that those with missing values for BMI tend to have poorer body image, poorer health behaviours and poorer social networks. These results have implications for potential bias in the results of studies that do not account for missing BMI.

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.


  • Funding:None.

  • Competing interests: None.

  • Ethics approval: This study was approved both by a university ethics committee and by the school teachers’ ethics board.

  • Patient consent: Obtained.