Obesogenic family types identified through latent profile analysis

Brian C. Martinson, Gabriela Vazquezbenitez, Carrie D. Patnode, Mary O. Hearst, Nancy E. Sherwood, Emily D. Parker, John Sirard, Keryn E. Pasch, Leslie Lytle

Research output: Contribution to journalArticlepeer-review

8 Scopus citations

Abstract

Background: Obesity may cluster in families due to shared physical and social environments. Purpose: This study aims to identify family typologies of obesity risk based on family environments. Methods: Using 2007-2008 data from 706 parent/youth dyads in Minnesota, we applied latent profile analysis and general linear models to evaluate associations between family typologies and body mass index (BMI) of youth and parents. Results: Three typologies described most families with 18.8% "Unenriched/Obesogenic," 16.9% "Risky Consumer," and 64.3% "Healthy Consumer/Salutogenic." After adjustment for demographic and socioeconomic factors, parent BMI and youth BMI Z-scores were higher in unenriched/obesogenic families (BMI difference=2.7, p<0.01 and BMI Z-score difference=0.51, p<0.01, respectively) relative to the healthy consumer/salutogenic typology. In contrast, parent BMI and youth BMI Z-scores were similar in the risky consumer families relative to those in healthy consumer/salutogenic type. Conclusions: We can identify family types differing in obesity risks with implications for public health interventions.

Original languageEnglish (US)
Pages (from-to)210-220
Number of pages11
JournalAnnals of Behavioral Medicine
Volume42
Issue number2
DOIs
StatePublished - Oct 2011

Bibliographical note

Funding Information:
Acknowledgments For study support, the authors would like to thank Transdisciplinary Research in Energetics and Cancer (TREC) Initiative (grant #1U54CA116849-01) and the National Heart, Lung and Blood Institute (grant #R01HL085978).

Keywords

  • Family types
  • Latent profile analysis
  • Obesogenic environment
  • Youth

Fingerprint

Dive into the research topics of 'Obesogenic family types identified through latent profile analysis'. Together they form a unique fingerprint.

Cite this