What does this research mean for the field?

Fractional Polynomials and Reed2 models, utilizing an independent Laplace likelihood, outperform Breakpoint models in capturing the complex non-linear trajectories of childhood BMI, with the majority of variance explained by individual random effects. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To evaluate robust Bayesian mixed-effects models for characterizing longitudinal childhood body mass index trajectories from birth to age 13 across multiple cohorts.

September 15, 2026The American StatisticianOpen Access

Evaluation of robust Bayesian mixed-effects models of longitudinal childhood BMI

View Full Paper

Ask AI

Bookmark

Authors

JHJordan HedgesUniversity of Surrey SCSarah R. CrozierUniversity Hospital Southampton NHS Foundation Trust ACAdam CollinsUniversity of Surrey

Discussion

Member takes

Implication

Longitudinal cohort analysis reveals robust fit for fractional polynomial and Laplace models of childhood BMI, highlighting the value of personalized tracking for early obesity intervention.

Key Points

To evaluate robust Bayesian mixed-effects models for characterizing longitudinal childhood body mass index trajectories from birth to age 13 across multiple cohorts.
Evaluated Fractional Polynomials, Reed2, and Breakpoint models using longitudinal data from three cohorts spanning birth to 13 years of age.
Assessed model performance based on bias, heteroskedasticity, and Bayesian goodness-of-fit across five likelihood functions, including Laplace and normal distributions.
Decomposed body mass index variance into fixed effects, random effects, and residual error using a Bayesian marginal posterior sampling approach.
Fractional Polynomials and Reed2 models outperformed Breakpoint models in modeling non-linear growth trajectories, with residual errors accounting for less than 10% of total variance.
The independent Laplace likelihood provided the most robust fit among all tested distributions, outperforming a normal likelihood with stationary correlation.
Random effects explained the vast majority of variance in body mass index, demonstrating substantial individual heterogeneity across child growth trajectories.

Cite This Study

Hedges et al. (2026) studied this question.

synapsesocial.com/papers/6aa913ba9013453be30a1d0e https://doi.org/10.1080/00031305.2026.2730371

View Full Paper

Ask AI

Bookmark

September 15, 2026The American StatisticianOpen Access

Evaluation of robust Bayesian mixed-effects models of longitudinal childhood BMI

View Full Paper

Ask AI

Bookmark

Authors

JHJordan HedgesUniversity of Surrey SCSarah R. CrozierUniversity Hospital Southampton NHS Foundation Trust ACAdam CollinsUniversity of Surrey

Discussion

Member takes

Implication

Longitudinal cohort analysis reveals robust fit for fractional polynomial and Laplace models of childhood BMI, highlighting the value of personalized tracking for early obesity intervention.

Key Points

To evaluate robust Bayesian mixed-effects models for characterizing longitudinal childhood body mass index trajectories from birth to age 13 across multiple cohorts.
Evaluated Fractional Polynomials, Reed2, and Breakpoint models using longitudinal data from three cohorts spanning birth to 13 years of age.
Assessed model performance based on bias, heteroskedasticity, and Bayesian goodness-of-fit across five likelihood functions, including Laplace and normal distributions.
Decomposed body mass index variance into fixed effects, random effects, and residual error using a Bayesian marginal posterior sampling approach.
Fractional Polynomials and Reed2 models outperformed Breakpoint models in modeling non-linear growth trajectories, with residual errors accounting for less than 10% of total variance.
The independent Laplace likelihood provided the most robust fit among all tested distributions, outperforming a normal likelihood with stationary correlation.
Random effects explained the vast majority of variance in body mass index, demonstrating substantial individual heterogeneity across child growth trajectories.

Cite This Study

Hedges et al. (2026) studied this question.

synapsesocial.com/papers/6aa913ba9013453be30a1d0e https://doi.org/10.1080/00031305.2026.2730371

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark