Modern Methods for Modeling Change in Obesity Research in Nursing

Susan M. Sereika, Yaguang Zheng, Lu Hu, Lora E. Burke

Research output: Contribution to journalArticlepeer-review


Persons receiving treatment for weight loss often demonstrate heterogeneity in lifestyle behaviors and health outcomes over time. Traditional repeated measures approaches focus on the estimation and testing of an average temporal pattern, ignoring the interindividual variability about the trajectory. An alternate person-centered approach, group-based trajectory modeling, can be used to identify distinct latent classes of individuals following similar trajectories of behavior or outcome change as a function of age or time and can be expanded to include time-invariant and time-dependent covariates and outcomes. Another latent class method, growth mixture modeling, builds on group-based trajectory modeling to investigate heterogeneity within the distinct trajectory classes. In this applied methodologic study, group-based trajectory modeling for analyzing changes in behaviors or outcomes is described and contrasted with growth mixture modeling. An illustration of group-based trajectory modeling is provided using calorie intake data from a single-group, single-center prospective study for weight loss in adults who are either overweight or obese.

Original languageEnglish (US)
Pages (from-to)1028-1044
Number of pages17
JournalWestern Journal of Nursing Research
Issue number8
StatePublished - Aug 1 2017


  • calorie intake
  • group-based trajectory modeling
  • growth mixture modeling
  • longitudinal data
  • obesity

ASJC Scopus subject areas

  • General Nursing


Dive into the research topics of 'Modern Methods for Modeling Change in Obesity Research in Nursing'. Together they form a unique fingerprint.

Cite this