- Browse by Author
Browsing by Author "Pi-Sunyer, Xavier"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Predictive utilities of lipid traits, lipoprotein subfractions and other risk factors for incident diabetes: a machine learning approach in the Diabetes Prevention Program(BMJ, 2021-03) Varga, Tibor V.; Liu, Jinxi; Goldberg, Ronald B.; Chen, Guannan; Dagogo-Jack, Samuel; Lorenzo, Carlos; Mather, Kieren J.; Pi-Sunyer, Xavier; Brunak, Søren; Temprosa, Marinella; Medicine, School of MedicineIntroduction: Although various lipid and non-lipid analytes measured by nuclear magnetic resonance (NMR) spectroscopy have been associated with type 2 diabetes, a structured comparison of the ability of NMR-derived biomarkers and standard lipids to predict individual diabetes risk has not been undertaken in larger studies nor among individuals at high risk of diabetes. Research design and methods: Cumulative discriminative utilities of various groups of biomarkers including NMR lipoproteins, related non-lipid biomarkers, standard lipids, and demographic and glycemic traits were compared for short-term (3.2 years) and long-term (15 years) diabetes development in the Diabetes Prevention Program, a multiethnic, placebo-controlled, randomized controlled trial of individuals with pre-diabetes in the USA (N=2590). Logistic regression, Cox proportional hazards model and six different hyperparameter-tuned machine learning algorithms were compared. The Matthews Correlation Coefficient (MCC) was used as the primary measure of discriminative utility. Results: Models with baseline NMR analytes and their changes did not improve the discriminative utility of simpler models including standard lipids or demographic and glycemic traits. Across all algorithms, models with baseline 2-hour glucose performed the best (max MCC=0.36). Sophisticated machine learning algorithms performed similarly to logistic regression in this study. Conclusions: NMR lipoproteins and related non-lipid biomarkers were associated but did not augment discrimination of diabetes risk beyond traditional diabetes risk factors except for 2-hour glucose. Machine learning algorithms provided no meaningful improvement for discrimination compared with logistic regression, which suggests a lack of influential latent interactions among the analytes assessed in this study.Item The moderating role of the built environment in prenatal lifestyle interventions(Springer Nature, 2021) Phelan, Suzanne; Marquez, Fred; Redman, Leanne M.; Arteaga, Sonia; Clifton, Rebecca; Grice, Brian A.; Haire-Joshu, Debra; Martin, Corby K.; Myers, Candice A.; Pomeroy, Jeremy; Vincent, Eileen; Van Horn, Linda; Peaceman, Alan; Ashby-Thompson, Maxine; Gallagher, Dympna; Pi-Sunyer, Xavier; Boekhoudt, Trisha; Drews, Kimberly; Brown, Greg; LIFE-Moms consortium; Emergency Medicine, School of MedicineThis study examined whether the neighborhood built environment moderated gestational weight gain (GWG) in LIFE-Moms clinical trials. Participants were 790 pregnant women (13.9 weeks’ gestation) with overweight or obesity randomized within four clinical centers to standard care or lifestyle intervention to reduce GWG. Geographic information system (GIS) was used to map the neighborhood built environment. The intervention relative to standard care significantly reduced GWG (coefficient = 0.05; p = 0.005) and this effect remained significant (p < 0.03) after adjusting for built environment variables. An interaction was observed for presence of fast food restaurants (coefficient=−0.007; p = 0.003). Post hoc tests based on a median split showed that the intervention relative to standard care reduced GWG in participants living in neighborhoods with lower fast food density 0.08 [95% CI, 0.03,0.12] kg/week (p = 0.001) but not in those living in areas with higher fast food density (0.02 [−0.04, 0.08] kg/week; p = 0.55). Interaction effects suggested less intervention efficacy among women living in neighborhoods with more grocery/convenience stores (coefficient = −0.005; p = 0.0001), more walkability (coefficient −0.012; p = 0.007) and less crime (coefficient = 0.001; p = 0.007), but post-hoc tests were not significant. No intervention x environment interaction effects were observed for total number of eating establishments or tree canopy. Lifestyle interventions during pregnancy were effective across diverse physical environments. Living in environments with easy access to fast food restaurants may limit efficacy of prenatal lifestyle interventions, but future research is needed to replicate these findings.