Abstract MP95: Title: Self-Reported versus Objective Sedentary Behavior Assessment in Young Adults
Background: Sedentary behavior (SED) has emerged as an independent risk factor for cardiovascular events and mortality in longitudinal studies, but has almost exclusively been measured by self-reported total SED or television (TV) time.
Objective: Our goal was to compare self-reported SED vs. objective SED by the BodyMedia® armband in young adults
Methods: Among overweight and obese young adults (age: 31 ± 4 years, BMI: 32 ± 4 kg/m2) enrolled in a clinical weight loss trial, baseline SED was assessed by 3 methods: 1) self-reported time spent in 8 common SEDs (TV, computer, transportation, etc.) from a questionnaire adapted from the Coronary Artery Risk Development in Young Adults Study (SEDQ), 2) a single question about total SED in a ‘typical day’ from the Global Physical Activity Questionnaire (SEDGPAQ), and 3) objective SED was calculated as the average awake time spent at <1.5 METS over ≥4 days with ≥10 hours of wear time (SEDOBJ). Also, because it is often used as a surrogate for SED, TV time (SEDTV) was isolated from the SEDQ. SED measures were compared using nonparametric sign-tests, Spearman’s ρ, and Bland-Altman plots.
Results: Valid data were available in 457 of 470 subjects. On average, compared to SEDOBJ, SEDQ tended to overestimate while SEDGPAQ tended to underestimate (Table). SEDQ and SEDGPAQ were weakly correlated with SEDOBJ; the correlation with SEDGPAQ was stronger (Table). In the Bland-Altman plots, SEDQ and SEDGPAQ had highly patterned errors where SEDQ and SEDGPAQ underestimated at low values and overestimated at high values. Limits of agreement (LOA) were wide for SEDQ[-7.4, 10.5 hr/day] and SEDGPAQ [-8.4, 6.0 hr/day]. Average SEDTV differed substantially from SEDOBJ, as expected, but was also not correlated.
Conclusions: SEDQ,SEDGPAQ, and especially SEDTV were poorly related to SEDOBJ in overweight and obese young adults.
The poor validity of these subjective assessment methods may have implications for the interpretation of observational research relating self-reported SED to adverse health outcomes.
Author Disclosures: B. Barone Gibbs: None. W.C. King: None. K.K. Davis: None. A.D. Rickman: None. R.J. Rogers: None. J.M. Jakicic: None.
This research has received full or partial funding support from the American Heart Association, Great Rivers Affiliate (Delaware, Kentucky, Ohio, Pennsylvania & West Virginia).
- © 2014 by American Heart Association, Inc.