A trajectory analysis of daily step counts during a physician-delivered intervention

Great diversity in the utilization and reporting of latent growth modeling approaches in type 2 diabetes: A literature review

“… 3-To examine antecedent and subsequent factors associated with each trajectory. Psychological science Incident cases, no diabetes in 1996, but diabetes in 1999 (self-reported) (n = 487) Predictive model: Predictors of trajectories: laboratory and clinical measures, socio-demographic measures, mobility score, self-rated health, comorbidities, lifestyle behaviors Explicative model: trajectory groups Predictive model: Depression symptoms trajectories Tool : Center of Epidemiological Studies – Depression scale Explicative model: future disability Tools: Activities of daily life score; Instrumental activities of daily life Nagi's mobility item Comparison between groups: X 2 and ANOVA Predictive model: Multinomial logistic regression Explicative model: Multiple regression Time-point studied: 1 year Trajectory identification: 8 years Outcome assessment: 2007 Latent class growth modeling [ 32 , 35 , 39 ] Cooke 2020 [ 43 ] Cohort: SMARTER randomized controlled trial (2012–2016) Type of data source: Primary data (clinical trial) 1-To examine indicators of trajectory membership of both steps/day and changes from baseline steps/day over the 1-year intervention. Clinical research Overweight/obese adults with T2D and/or hypertension (n = 118) Predictors of trajectory group membership Tools: Baseline sociodemographic, T2D, time of intervention, anthropometric, clinical data Trajectories of mean septs/day Tool: Average steps/day on a 30-day period Predictive model: Cumulative logistic regression Time-point studied: 30 days Follow-up period : 1 year Group-based trajectory modeling [ 3 , 34 ] Davis 2016 [ 44 ] Cohorts: Fremantle Diabetes Study Phase 1 (1993–1996) and Western Australia data linkage system (1998–2001) Type of data source: Primary data (prospective cohort) 1-To determine whether there was a mortality benefit of tight glycemic control beyond the period in which it was implemented in recently diagnosed patients; a neutral or increased risk of death in those with long-duration diabetes.…”

“… Yes 7 18 Six studies considered within-class heterogeneity using either LGMM or GGCMM. Two studies with unclear reporting added within-class confidence intervals on time intervals, although reporting using LCGA [ 43 , 67 ] and 1 study mentioned considering within-class heterogeneity in text, although reporting using LCGA. No 28 74 Unclear 3 8 6b.…”