Probabilistic prediction of taxi origin–destination demand via multi-perspective graph neural networks and quantile regression

Intrinsic capacity (IC), proposed by the World Health Organization, reflects the composite of an individual's physical and mental capacities and is considered a key indicator of healthy aging. Although air pollution has been associated with multiple adverse health outcomes, evidence on its relationship with IC-a multidimensional indicator of overall functional ability-remains limited. This study aimed to examine the longitudinal associations between long-term exposure to particulate matter of different sizes (PM1, PM2.5 and PM10) and declines in intrinsic capacity among middle-aged and older adults in mainland China. Data were obtained from the nationally representative longitudinal cohort of the China Health and Retirement Longitudinal Study (CHARLS), including adults aged ≥ 45 years. Annual average concentrations of PM1, PM2.5 and PM10 were estimated using satellite-based models based on participants' residential locations. A composite IC score was constructed according to the WHO intrinsic capacity framework. Multivariable logistic regression models were used to evaluate associations between particulate matter exposure and IC decline. Restricted cubic spline models were applied to assess nonlinear dose-response relationships. Mixed-exposure effects were further evaluated using weighted quantile sum regression, Bayesian kernel machine regression and quantile g-computation. Long-term exposure to PM1, PM2.5 and PM10 was significantly associated with a higher risk of IC decline and lower composite IC scores. Nonlinear dose-response relationships were observed between particulate matter exposure and IC decline. In mixed-exposure analyses, PM10 and PM2.5 showed the strongest contributions to the joint effects, with PM10 having the highest weight and posterior inclusion probability. Each one-quantile increase in the overall pollutant mixture was significantly associated with a decline in composite IC scores.

The American Heart Association (AHA) recently proposed the concept of Cardiovascular-Kidney-Metabolic (CKM) syndrome, highlighting the strong pathophysiological links among metabolic disorders, chronic kidney disease, and cardiovascular disease (CVD). Insulin resistance (IR) is regarded as a central mechanism underlying CKM syndrome. However, studies comparing the predictive value of different IR surrogate markers for incident CVD are still limited. This study aimed to evaluate the associations between multiple IR surrogate markers and incident cardiovascular disease and to further assess their predictive performance using machine learning approaches. Using data from the China Health and Retirement Longitudinal Study (CHARLS), this prospective cohort study included 5,528 participants. Twelve IR surrogate indices were assessed, including TyG-related indices, TG/HDL-C, METS-IR, CTI, CHG, and eGDR. Incident CVD was defined as self-reported physician-diagnosed heart disease or stroke during follow-up. Cox proportional hazards models were used to estimate associations between standardized IR indices and incident CVD. Restricted cubic splines, weighted quantile sum regression, and quantile g-computation were used to examine dose-response patterns and the relative contribution of correlated IR indices. Predictive performance was evaluated using ROC analysis, calibration, Brier score, decision curve analysis, NRI, IDI, and machine-learning models. During a median follow-up of 7.0 years, 741 participants developed incident CVD. In fully adjusted Cox models, several IR surrogate indices were associated with incident CVD. TyG-related composite indices incorporating adiposity-related information, particularly TyG-WC, TyG-CVAI, TyG-WHtR, and TyG-BMI, showed stronger positive associations with CVD risk, whereas eGDR showed an inverse association. Restricted cubic spline analyses showed significant overall associations for most indices, with nonlinear patterns observed for METS-IR, CTI, and eGDR. Mixture-based analyses suggested relatively larger contributions of CTI, TyG-BMI, and TyG-WC. Among individual indices, eGDR showed the highest discrimination for incident CVD, followed by TyG-CVAI and TyG-WC. Adding selected IR indices, particularly eGDR, to the covariate-based model modestly improved discrimination and reclassification. Among adults with CKM syndrome stages 0-3, several IR surrogate indices were prospectively associated with incident CVD, with stronger and more consistent associations observed for TyG-based indices incorporating adiposity-related measures and for eGDR. These results suggest that the combined assessment of metabolic dysfunction, adiposity, and insulin sensitivity may provide useful information for identifying individuals at higher cardiovascular risk in early-stage CKM syndrome.

Kernel quantile regression for semiparametric partially linear time-varying-coefficient model based on a history process of longitudinal data

Regional economy, physical environment and self-rated health among older adults in China: Evidence from XGBoost and SHAP.

Determinants of hyperfocus in the context of escapism and gaming motivations, flow, and ADHD symptoms among adult video gamers.

How fintech and institutional governance clean up transport emissions?

Multimodal computed tomography angiography and perfusion (CTA/CTP) is increasingly used in the emergency department (ED) evaluation of acute ischemic stroke (AIS) but may delay thrombolysis. We examined whether the sequencing of multimodal CTA/CTP relative to thrombolytic administration affects door-to-needle (DTN) time. We conducted a retrospective cohort study of adult patients with AIS treated with intravenous alteplase at a tertiary, academically affiliated ED that is part of a certified Comprehensive Stroke Center between January 1, 2012, and August 1, 2023. All patients underwent an initial noncontrast CT (NCCT), followed by either a contrast-first workflow (i.e. CTA/CTP) or were given alteplase before proceeding to contrast administration. The primary outcome was DTN time, analyzed using univariate methods and multivariable quantile regression adjusting for stroke severity, pre-alteplase antihypertensive use, thrombectomy (as a correlate for large-vessel occlusion), primary language, and bedside ED pharmacist presence. Among 1,382 AIS encounters, 167 patients met inclusion criteria; 148 underwent CTA/CTP before alteplase and 19 received alteplase before contrast administration. Median DTN time was significantly shorter in the alteplase-first group (20 minutes [interquartile range (IQR), 15-26]) compared with the CTA/CTP-first group (44 minutes [IQR, 32-56]; P<0.001). After adjustment, alteplase-first administration remained independently associated with faster DTN time (-25 minutes; 95% CI, -34 to -16). Higher stroke severity and ED pharmacist presence were also associated with shorter DTN, whereas pre-alteplase antihypertensives, primary language, and thrombectomy were not. A contrast-first strategy incorporating multimodal CTA/CTP before alteplase administration was associated with approximately twofold longer DTN times. These findings suggest that obtaining CTA/CTP before thrombolytic administration may substantially prolong DTN times. Larger studies are needed to confirm the magnitude of this association and to identify patient populations most likely to benefit from alternative imaging workflows.

To explore differences in human milk hormones with metabolic conditions (obesity and/or diabetes) during pregnancy and the relationship with infant adiposity. In a prospective cohort of term and preterm infants, hormones were measured in early human milk, and growth was assessed at discharge. Differences in hormones by metabolic conditions and the relationship with growth were modeled using quantile and linear regression. Among 38 infants (mean gestational age 35.2 weeks), 50% were in the metabolic-group. Milk leptin was higher in the metabolic-group, with median difference 427 pg/mL (95% CI: 3.2, 850.9). After adjustment for exposure to maternal milk, each 10-fold change in leptin related to 2.1 cm (95% CI: 0.3, 4.2) increase in abdominal circumference and each 10-fold change in insulin related to 1.1 cm (95% CI: 0.03, 2.1) increase in mid-arm circumference, while other effects were attenuated. In this explorative study, leptin was higher in human milk from women with metabolic conditions, and human milk hormones predicted growth at discharge.

Clinicians rely on reference intervals (RIs) to interpret laboratory test results. In pediatric populations, estimating RIs typically requires partitioning data by age, sex, and other relevant factors, which can lead to limited sample size and imprecise estimates; these limitations are addressed by using curve estimation, modeling hemoglobin level as a continuous function of age. To establish hemoglobin reference curves (RCs) for children and to complement recently published World Health Organization (WHO) thresholds by estimating hemoglobin optimal curves (OCs) that may inform more appropriate reporting standards. This cross-sectional study included healthy Canadian children aged 2 weeks through 10 years attending scheduled primary care visits from June 3, 2008, to February 26, 2020, in Toronto, Ontario, Canada. Data were analyzed from October 16, 2024, to February 1, 2026. Blood samples were collected and analyzed for hemoglobin, ferritin, and C-reactive protein levels. Parents completed a questionnaire to collect variables used as optimality criteria. Sex-specific RCs and OCs were estimated using nonparametric quantile regression with restricted cubic splines. RCs were based on the full sample, whereas OCs excluded children with indicators of suboptimal iron status. A web-based platform was developed to visualize these curves and calculate sex-specific reference and optimal limits by age. Findings were examined in relation to WHO hemoglobin thresholds. Blood samples from 4597 children (2451 males [53%]; median age, 38 months [IQR, 18-63 months]) were used to estimate hemoglobin RCs, and samples from a subgroup of 3426 children (1798 males [52%]; median age, 45 months [IQR, 24-68 months]) were used to estimate OCs. For females, lower OC hemoglobin limits were slightly below the lower RC limits up until age 2 years and became higher after age 6 years (eg, at 6 months, the OC lower limit was 9.91 g/dL [90% CI, 9.70-10.13 g/dL] vs 10.00 g/dL [90% CI, 9.78-10.23 g/dL] for RC). For males, lower OC limits were higher than lower RC limits until age 20 months (eg, at 6 months, the OC lower limit was 9.74 g/dL [90% CI, 9.46-10.02 g/dL] vs 9.28 g/dL [90% CI, 8.94-9.63 g/dL] for RC) and were similar afterwards. Differences in the upper limits were minimal for both sexes. WHO hemoglobin thresholds were consistently higher than lower limits of OCs across all ages but exceeded the 5th percentile curve only among children aged 5 through 10 years for both sexes (eg, for males aged 1 year, the OC lower limit was 10.06 g/dL [90% CI, 9.92-10.21 g/dL] vs the 10.5 g/dL WHO threshold). This cross-sectional study estimated sex-specific pediatric RCs for hemoglobin, modeled as a continuous function of age, to eliminate the need for age partitioning and overcome the associated sample size limitations. OCs, developed using health-based criteria, offered additional clinical context beyond traditional RIs. The findings highlight potential misalignments with existing WHO thresholds, particularly at younger ages.

First, to evaluate whether the ratio of soluble fms-like tyrosine kinase-1 to placental growth factor (sFlt-1/PlGF) is associated with hemodynamic changes in ophthalmic artery (OA) Doppler in near-term pregnancy. Second, to assess the performance of OA Doppler to rule in and rule out angiogenic factor imbalance. This was a cross-sectional cohort study nested within the PE37 randomized controlled trial, involving nulliparous women recruited between January 2023 and January 2025 who underwent sFlt-1/PlGF ratio measurement between 35 + 0 and 36 + 6 weeks' gestation. We included a subsample of women who underwent OA Doppler evaluation, including measurement of OA peak systolic velocity (PSV) ratio and OA pulsatility index (PI), as well as assessment of mean arterial pressure (MAP) and mean uterine artery (UtA) PI. The operator was blinded to sFlt-1/PlGF ratio values. Trends in median values of OA and maternal-fetal Doppler parameters across sFlt-1/PlGF tertiles were analyzed using the Jonckheere-Terpstra test and quantile regression, adjusting for maternal body mass index, age and smoking status. The predictive performance of the OA-PSV ratio for sFlt-1/PlGF ratio ≥ 38 was evaluated using receiver-operating-characteristics-curve analysis. We included 203 women, of whom 62, 71 and 70 were in the lowest, middle and highest tertiles of the sFlt-1/PlGF ratio, respectively. With increasing sFlt-1/PlGF ratio tertile, there was a significant increase in the OA-PSV ratio (median, 0.45 (interquartile range (IQR), 0.39-0.53) vs 0.48 (IQR, 0.41-0.58) vs 0.59 (IQR, 0.50-0.66); adjusted P < 0.001), a significant decrease in OA-PI (median, 2.20 (IQR, 1.92-2.61) vs 2.13 (IQR, 1.86-2.37) vs 1.86 (IQR, 1.60-2.26); adjusted P = 0.031) and a significant increase in MAP (median, 87.0 (IQR, 82.7-92.0) mmHg vs 88.7 (IQR, 83.7-93.7) mmHgvs 94.7 (IQR, 89.3-100.0) mmHg; adjusted P < 0.001). In contrast, no significant trend was observed in mean UtA-PI across sFlt-1/PlGF ratio tertiles. Among those individuals with an OA-PSV ratio < 0.61, 90.5% truly had a sFlt-1/PlGF ratio < 38, at a 15% false-positive rate. This study provides new evidence of a significant association between the sFlt-1/PlGF ratio and OA Doppler parameters in near-term pregnancies, suggesting that OA Doppler indices, particularly the PSV ratio, reflect angiogenic imbalance. Given its non-invasive nature, accessibility and low cost, OA Doppler emerges as a promising surrogate tool for ruling out angiogenic imbalance. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

ABSTRACT Global trade has shifted from the traditional exchange of finished goods to the widespread integration of global value chains (GVCs). The expansion of GVCs has important implications for both developed and developing economies. We construct a framework to explain how GVC participation affects GDP per capita and how these effects differ across countries at different income levels by analysing types of GVC activities and the mediating role of total factor productivity (TFP). We use advanced econometric methods, including fixed-effects models, two-stage least squares estimators, system generalized method of moments estimators, and fixed-effects quantile regression, to provide detailed insights into the differential effects of GVC participation. Our results show that the impact of GVC participation on economic growth varies by income level. For low-income countries, backward GVC participation significantly affects GDP per capita, while forward GVC participation has significant effects only for upper-middle- and high-income countries. We also find that TFP mediates the impact of GVC participation on GDP per capita.

Many drugs influence blood potassium levels and are often prescribed to older adults. Aim of this study was to estimate the extend of potassium level changes of drugs in the case of older, multi-medicated patients respecting renal function. Data of patients from an outpatient clinic (polypharmacy consultations hours) at the University Hospital RWTH Aachen and from a prospective, multicenter study on adverse drug reactions in emergency departments (ADRED) was used. Data of patients aged ≥ 70 years taking ≥ 3 drugs with available blood potassium levels were included. Quantile regression models were used to analyze drug effects in the context of multi-medication. In total, N = 1097 patient cases were included with hypokalemia in 28.4% (n = 311) and hyperkalemia in 20.3% (n = 223). Male sex, use of low-ceiling diuretics, loop diuretics, mineralocorticoid receptor antagonists (MRAs), angiotensin-converting-enzyme inhibitors/ angiotensin II receptor blockers, number of other drugs, and CKD stage were associated with serum potassium levels and included in quantile regression models. In CKD stages 1 and 2, the use of MRAs (+ 0.49, p = 0.002), together with low-ceiling diuretics (-0.54, p < 0.001) had highest impact on serum potassium levels. In CKD stage 3, use of low-ceiling diuretics (-0.52, p < 0.001) and in CKD stages 4 and 5, the use of MRAs (+ 0.86, p < 0.001) had highest impact on serum potassium levels. Hypo- as well as hyperkalemia occur frequently in geriatric, multi-medicated patients. While renal function is an important predictor of blood potassium levels, drug effects can differ per drug class in the context of multi-medication. The ADRED-study is registered at the German Clinical Trials Register under: DRKS00008979. The cohort study of the polypharmacy consultation hours at the University Hospital RWTH Aachen is registered at the ClinicalTrials.gov under: NCT05247814.

Objectives To establish gestational age‐specific reference ranges for the intraplacental vascular index measured using microvascularity‐flow ultrasound software (VI MV ) and to evaluate its variation by placental location, parity, and pregnancy outcome. Methods This prospective study enrolled 354 singleton pregnancies, of which 243 (68.64%) uncomplicated cases were used to construct reference ranges. A residual bootstrap method with 500 iterations was applied to square‐root–transformed VI MV values, followed by back‐transformation to derive the 5th, 50th, and 95th percentiles with 95% confidence intervals. Group differences were assessed using quantile regression, and associations with Doppler parameters were examined. Results VI MV increased significantly with advancing gestational age (linear coefficient 1.88, p = .007), following a curvilinear pattern with mild late‐gestation downturn (quadratic coefficient − 0.036, p = .011). Posterior placentas showed slightly higher early values, although this difference was not significant; from 31 weeks onward, anterior placentas demonstrated significantly higher VI MV (mean difference + 15.4, p = .010 at 31–35 weeks; +11.6, p = .003 at 36–40 weeks). Multiparous women showed a modestly higher fitted trajectory and a similar non‐significant trend toward higher birthweight, but parity was not independently associated with VI MV . Pregnancies complicated by maternal disease, hypertensive disorders, or small‐for‐gestational‐age neonates exhibited lower trajectories than uncomplicated pregnancies, although these differences were not statistically significant. VI MV was inversely associated with uterine and umbilical artery pulsatility indices and positively associated with umbilical vein time‐averaged mean velocity. Conclusions Gestational age–specific VI MV reference ranges were established. VI MV may provide an adjunctive quantitative marker for assessing placental vascular adaptation and identifying pregnancies with altered placental perfusion.

To determine gestational age-specific reference ranges and develop a Z-score equation for the fetal ductal pulsatility index (PI), between 24 and 33+6weeks of gestation. A prospective cross-sectional study conducted from 2023 to 2024, 226 singleton fetuses between 24 and 33+6weeks were enrolled. The ductus arteriosus (DA) was assessed in the longitudinal plane of the fetus using color Doppler to confirm its course and flow direction, and pulsed Doppler sampling was placed at the proximal DA with an insonation angle <15°. Three consecutive measurements of PI were recorded and averaged. Z-scores were derived from predicted means and standard deviations and reference percentiles (5th, 50th, and 95th) were computed for gestational age groups using quantile regression. Intraobserver reproducibility was evaluated in 30 randomly selected cases with an interval≥2weeks using the intraclass correlation coefficient (ICC). The mean gestational age was 28.2±3.1weeks. The ductal PI showed a mean of 2.4±0.26 with normal distribution (p=0.2525). The Z-score equation for PI was successfully derived Z-score = (Measured PI- 2.4)/0.26. Reference ranges for PI were: 24-27weeks (5th=2.0; 50th=2.4; 95th=2.7), 28-31weeks (5th=2.1; 50th=2.5; 95th=2.9), and 32-34weeks (5th=2.2; 50th=2.4; 95th=2.5). Intraobserver reproducibility for PI was excellent (ICC >0.90). We established gestational age-specific reference ranges and a Z-score equation for the ductal PI in normal fetuses between 24 and 33+6weeks. These findings provide a standardized tool for fetal cardiologists to identify subtle ductal flow abnormalities and support perinatal risk stratification.

The triglyceride-glucose (TyG) and its obesity‑derived indices are important composite parameters of insulin resistance coupled with obesity, and have been associated with cardiometabolic diseases. However, their associations with cardiovascular‑liver‑metabolic multimorbidity (CLMM) remain unclear. This study aims to systematically compare TyG and its obesity-derived indices to identify the optimal surrogate marker for CLMM, and to explore the potential impact of CLMM status on mortality. Based on a multi-stage sampling design of a national cohort, weighted logistic and Cox regression models were applied to evaluate the associations of each index with CLMM and long-term mortality. Discriminatory performance was compared using the area under the receiver operating characteristic curve (AUROC), net reclassification improvement, and integrated discrimination improvement. Additionally, weighted quantile sum regression, and mediation analysis were employed to explore the underlying disease mechanisms. The study included 20,235 adults. Weighted logistic regression analyses for the individual indices revealed that the TyG and its obesity‑derived indices were all positively associated with CLMM, with TyG‑WHtR, TyG‑WWI, and TyG‑CVAI showing the strongest associations. Further longitudinal weighted interaction analyses revealed that CLMM status significantly amplified the mortality risk associated with the TyG and its obesity-derived indices. For CLMM identification, all indices showed favorable diagnostic performance (AUROC > 0.76), with TyG‑WWI and TyG‑CVAI achieving the best performance (AUROC > 0.83) and exhibiting the largest net reclassification improvement and integrated discrimination improvement. Joint XGBoost-SHAP analyses further indicated that TyG-CVAI and TyG-WWI were the major contributing variables for discriminating CLMM status. Exploratory mechanistic analyses suggested glucose metabolism as a core driver underlying CLMM development, with inflammatory factors partially mediating these associations. The TyG and its obesity-derived indices, particularly TyG-WWI and TyG-CVAI, exhibited significant incremental value for identifying CLMM, with CLMM status also conferring elevated subsequent mortality risk. Incorporating these indices into primary prevention monitoring systems for chronic diseases could facilitate the identification of individuals at high risk for CLMM and mortality.

Hypertension remains a major public health challenge in developing countries, with its prevalence continuing to rise and risk factors varying across populations. Traditional logistic regression often fails to capture distributional heterogeneity in health outcomes, thereby motivating the use of more flexible approaches. This study applied Bayesian Binary Quantile Regression with adaptive Lasso (BALBqr) to hypertension data from 653 patients at Arosuka Solok Hospital, Indonesia. The method was further integrated with bootstrap resampling as a consistency check for posterior estimates. The Bayesian quantile framework enabled the evaluation of covariate effects across different quantiles of the conditional distribution, while adaptive Lasso improved interpretability by addressing multicollinearity and ensuring parsimonious variable selection. Bootstrap validation complemented posterior inference by assessing the stability of parameter estimates within bootstrap confidence intervals, with the optimal replication identified at 250 iterations based on the smallest bias relative to the posterior mean. Simulation results demonstrated that the proposed approach consistently outperformed conventional methods in terms of bias, RMSE, and MAE across various error distributions. Based on this study, age, body weight, cholesterol, triglycerides, and blood sugar levels were found to be statistically significant predictors of hypertension status at the 0.95 quantile.

The lives of people with eating disorders (EDs) habitually revolve around food and diet, and they exhibit disturbances in attitudes and feelings about their bodies and appearance. Both aspects are also described in people at risk of orthorexia nervosa (ON). Social support plays an important role in the treatment of EDs, but less is known about the role of social support in ON. Therefore, the study aimed to estimate the levels of ON risk, perceived social support, body esteem (BE), and diet quality among students by biological sex, body mass index (BMI) group, and current education stage and program; and evaluate the associations between the levels of ON risk, perceived social support, BE, and diet quality. Data were collected from 2022 to 2025 through a cross-sectional quantitative survey among 261 Polish high school students (aged 16-19years) and university students (aged 18-23years). Students completed the Polish version of the Düsseldorf Orthorexia Scale (PL-DOS), the Multidimensional Scale of Perceived Social Support (MSPSS), the Body Esteem Scale for Adolescents and Adults (BESAA), and the Dietary Habits and Nutrition Beliefs Questionnaire (KomPAN). Pearson Chi-squared test, the Mann-Whitney U test, the Kruskal-Wallis rank ANOVA, and quantile regression analysis were used for further analyses. Females exhibited a significantly higher ON risk than males. University students from nutrition-related majors demonstrated a significantly higher ON risk than those from non-nutrition-related majors. No significant association was found between ON risk and BMI. An increased risk of ON was significantly associated with lower social support from friends, lower BE weight, and a lower intensity of harmful dietary characteristics for health. Meanwhile, higher BE attribution and a higher intensity of beneficial dietary characteristics for health were significantly associated with increased ON risk. Overall, the findings suggest that ON risk is associated with selected psychosocial characteristics, such as perceived social support and body weight-related self-esteem. These findings highlight the need for greater integration of psychological aspects of eating and body image into school and university education programs, as well as interventions aimed at enhancing support from friends to reduce the risk of ON and EDs. Level III. Evidence obtained from a cross-sectional quantitative study.

Cardiovascular disease has been consistently associated with cognitive impairment; however, the relationship between Life’s Essential 8 (LE8), a novel composite metric of cardiovascular health (CVH), and cognitive function in older adults remains inadequately characterized. This study aimed to investigate the cross-sectional association between LE8 scores and cognitive function among United States older adults and identify the relative contributions of individual LE8 components. We conducted a cross-sectional analysis using data from the National Health and Nutrition Examination Survey 2011 to 2014 cycles. A total of 2108 participants aged ≥ 60 years with complete cognitive assessment and LE8 data were included. Cognitive function was evaluated using 4 validated neuropsychological tests: Consortium to Establish a Registry for Alzheimer Disease Word Learning and Delayed Recall, animal fluency test, and Digit Symbol Substitution Test. Composite and domain-specific z-scores were computed. Weighted multiple linear regression models were employed to examine LE8-cognition associations, adjusting for demographic and clinical covariates. Weighted quantile sum regression was applied to determine the relative contribution weights of individual LE8 components. In fully adjusted models, each 10-point increment in LE8 score was associated with a 0.09-unit increase in composite cognitive z-score (95% confidence intervals: 0.04–0.14, P < .001). Compared with low CVH (LE8 < 50), high CVH (LE8 ≥ 80) demonstrated significantly higher cognitive z-scores (β = 0.41, 95% confidence intervals: 0.25–0.57, P < .001). Weighted quantile sum regression identified moderate physical activity as the predominant contributor (weight = 0.56) to cognitive performance, followed by blood glucose (0.19) and blood pressure (0.14). Higher LE8 scores are significantly associated with better cognitive function in older adults, with moderate physical activity emerging as the most influential modifiable factor. These findings underscore the potential of integrated CVH optimization as a strategy for cognitive preservation in aging populations.

Obesity is a highly heritable trait, but rising obesity rates suggest environmental change is also of profound importance. We conducted a cross-cohort analysis to examine how associations between genetic risk for high BMI and observed BMI differed in four British birth cohorts born before and amidst the obesity epidemic (1946, 1958, 1970 and ~2001; N = 19,379). BMI (kg/m2) was measured at multiple time points between ages 3 and 69 years. We used polygenic indices (PGI) derived from GWAS of adulthood and childhood BMI, respectively, with mixed effects models used to estimate associations with mean BMI and quantile regression used to assess associations across the distribution of BMI. We further used linear regression to estimate PGI-heritability (PGI-h2; incremental variance explained by the PGI) and Genomic Relatedness Restricted Maximum Likelihood (GREML) to calculate SNP-heritability (SNP-h2) by cohort and age. Adulthood BMI PGI was associated with BMI in all cohorts and ages but was more strongly associated with BMI in more recently born generations, e.g., at age 16y, a 1 SD increase in the adulthood PGI was associated with 0.46 kg/m2 (0.37, 0.55) higher BMI in the 1946c and 0.90 kg/m2 (0.83, 0.97) higher BMI in the 2001c. Cross-cohort differences widened with age and were larger at the upper end of the BMI distribution, indicating disproportionate increases in obesity in more recent generations for those with higher PGIs. Differences were also observed when using the childhood PGI. There were no clear, consistent differences in PGI-h2 or SNP-h2, possibly due to limited statistical power, except that PGI-h2 was highest in the most recently born cohort (2001c) when using the most predictive PGI for adulthood BMI. Findings highlight how the environment can modify genetic associations; genetic associations with BMI differed by birth cohort, age, and outcome centile.