Non-communicable diseases

Background: Low physical activity (PA) and high adiposity are associated with cancer risk. It is unclear whether different amounts and intensities of PA can mitigate this association. Purpose: We aimed to examine the joint dose-response associations between body mass index (BMI) and accelerometer-measured PA with incident cancer. Methods: UK Biobank participants were classified as normal weight (18.5-24.9 kg/m2), overweight (25-29.9 kg/m2) or obese (>30 kg/m2) BMI. PA data included weekly light-intensity (LPA), moderate-to-vigorous-intensity (MVPA), vigorous-intensity (VPA), and total volume. Cancer diagnoses and deaths were obtained from registries. We excluded participants with events in the first two years of follow-up, prevalent diseases, or physical limitations. Hazard ratios were calculated from proportional sub-hazards models with multivariate adjustment and age as timescale to examine the joint dose-response associations between BMI categories and PA levels with incident fatal and non-fatal cancer, accounting for other deaths as competing events. Results: Analysis included 70,747 participants (mean age 61.6 ± 7.9, 56.4% women) over a median 6.1-year follow-up, with 2,625 incident cancer cases. Compared with the referent (normal-weight and 5th centile of total PA/LPA and zero MVPA/VPA), participants with overweight and obesity showed a 25% to 45% higher cancer hazard at zero MVPA, VPA and low total PA levels. Some attenuation of these associations was found from MPVA levels within international guidelines, from VPA below the guidelines, and with higher total PA. Normal weight participants did not show a higher cancer hazard regardless of PA levels. No clear associations were found for LPA. Conclusions: High levels of MVPA, VPA and total PA may attenuate but not clearly offset the cancer hazard associated with overweight and obesity. Maintaining a healthy weight seems comparatively more important. Practical implications: Public health efforts should encourage both weight control and promoting higher PA to mitigate cancer risk effectively. Funding: None

Background:
Changes in walkability may be linked to cardiovascular disease incidence, but longitudinal studies are lacking.

Purpose:
To investigate the relationship between changes in residential neighbourhood walkability and CVD incidence in adults in the Netherlands.

Methods:
We conducted a large-scale population-based cohort study. Using data from Statistics Netherland we included all Dutch residents aged ≥40y at baseline (2009), without a history of CVD, and who did not move house after baseline (n=3,019,069). A nationwide, objectively measured walkability index was calculated for Euclidean buffers of 500m around residential addresses for the years 1996, 2000, 2003, 2006 and 2008. To identify changes in neighbourhood walkability, latent class trajectory modelling was applied. Incident CVD between 2009-2019 was assessed using the Dutch Hospital Discharge Register and the National Cause of Death Register. Cox proportional hazards modelling was used to analyse associations between walkability trajectories and subsequent CVD incidence, adjusted for individual- and area-level sociodemographic characteristics.

Results:
Four distinct neighbourhood walkability trajectories were observed: a stable but relatively low walkability trajectory (Stable low, 91.1%), a stable but relatively higher walkability trajectory (Stable high, 0.6%), a relatively higher initial neighbourhood walkability which decreased over time (Decreasing, 1.7%), and relatively lower neighbourhood walkability which increased over time (Increasing, 6.5%). Compared to stable high walkability, individuals exposed to stable low, and increasing walkability, had a 5.1% (HR: 1.051; CI: 1.011–1.093) and a 4.9% (HR: 1.049; CI: 1.008–1.092) higher risk of any CVD during follow-up. Similar associations were observed for coronary heart disease and stroke, though not statistically significant. No significant associations were found for heart failure and CVD mortality.

Conclusions:
Adults exposed to stable low or increasing walkability trajectories in residential neighbourhoods had a higher risk of cardiovascular disease.

Practical implications:
The results emphasize the relevance of long-term urban planning considerations for cardiovascular health.

Background: Traditionally, existing studies assessing the health associations of accelerometer-measured movement behaviors have been performed with few averaged values, mainly representing the duration of physical activities and sedentary behaviors. Such averaged values cannot naturally capture the complex interplay between the duration, timing, and patterns of accumulation of movement behaviors. Purpose: To introduce a novel approach to visually represent recorded movement behaviors as images using original accelerometer outputs, allowing the capture of the entire movement behavior profile with information on duration, timing, and patterns in one single image. Methods: Our method involves converting minute-by-minute accelerometer outputs (activity counts) into a 2D image format, capturing the entire spectrum of movement behaviors performed by each participant. By utilizing convolutional autoencoders, we enable the learning of these image-based representations. Subsequently, we apply the K-means algorithm to cluster these learned representations. We used data from 1812 adult (20-65 years) participants in the National Health and Nutrition Examination Survey (NHANES, 2003-2006 cycles) study who worn a hip-worn accelerometer for 7 seven consecutive days and provided valid accelerometer data. Results: Deep convolutional autoencoders were able to learn the image representation, encompassing the entire spectrum of movement behaviors. Cluster analysis based on the learned representations for the movement behavior images, resulted in identification of four distinct movement behavior profiles characterized by varying levels, timing, and patterns of accumulation of movement behaviors. These identified profiles were associated with the markers of cardiometabolic health. Conclusions: Deep learning of movement behavior profiles revealed that, in addition to duration and patterns of movement behaviors, the timing of physical activity may also be crucial for gaining additional health benefits. Practical implications: The timing of physical activity may also be considered for maximizing the health benefits.

Background: Few studies have examined prospective associations between physical activity/sedentary time and cardiometabolic risk markers in youth. Purpose: To do a harmonized meta-analysis of prospective studies with device-measured physical activity and sedentary time in relation to cardiometabolic risk markers in children and adolescents. Methods: We used data from the International Children’s Accelerometry Database supplemented with data from three additional cohorts. All studies measured physical activity/sedentary time by accelerometry at baseline and traditional cardiometabolic risk markers at baseline and subsequently at follow-up. Exposure data were processed using a harmonized approach. Linear regression models (with 95%CI) were used to determine associations with a composite cardiometabolic risk score at follow-up, created by combining standard deviation scores of insulin, glucose, BMI, LDL-cholesterol, triglycerides, and systolic blood pressure. Statistical models were multivariable-adjusted including baseline outcome and repeated with and without adjustment for baseline BMI. Multiple imputation was used to handle missing data. Results: We included a total of 7512 children and adolescents (51% girls) with a mean baseline age of 11.6 (SD: 3.0) years and 2.6 (SD: 0.4) years of follow-up. Moderate-to-vigorous physical activity and vigorous physical activity were not associated with follow-up composite risk score with or without adjustment for baseline BMI (standardized betas with 95% CI, 0.00 [-0.02, 0.03] and 0.01 [-0.01, 0.03]. Insulin and triglycerides were inversely associated with moderate-to-vigorous and vigorous physical activity but substantially attenuated following adjustment for BMI. Higher sedentary time was associated with lower composite risk (-0.04 [-0.07, -0.01] which was not affected by adjustment for baseline BMI. Conclusions: Physical activity and sedentary time may not be major determinants of later cardiometabolic risk markers in population-based children and adolescents. Practical implications: Promotion of physical activity and reductions in sedentary time in young people may primarily be targeted on other outcomes rather than on cardiometabolic risk markers. Funding: None

Background The independent associations of life-long leisure time physical activity (PA) and current participation in PA with cardiometabolic risk at the beginning of late adulthood are unknown. Purpose To investigate the associations of PA trajectories across midlife and current PA with cardiometabolic risk at age 61. Methods Data came from the JYLS study (N=159, 52% women). Leisure-time PA frequency was assessed at ages 27, 42, 50, and 61 with a single question. Current PA at age 61 included self-reported weekly vigorous PA, weekly strength training, regular active commuting, and occupational PA (yes/no). Cardiometabolic risk factors at age 61 included waist circumference, blood pressure, HDL cholesterol, triglycerides, and glucose. PA trajectories were conducted using k means for longitudinal data. Data were analyzed with general and generalized linear models, adjusted for gender and medication. Results Of the three PA trajectories found, consistently active (N=67) had better cardiometabolic health compared to increasingly active (N=58), and consistently inactive (N=34). They had smaller waist circumference (mean difference = -5.2, 95% confidence interval [-9.9, -0.5] and -9.2 [-14.8, -3.5]), higher HDL cholesterol (0.22 [0.06, 0.38] and 0.22 [0.02, 0.42]), and lower triglyceride levels (-0.35 [-0.82, -0.08] and -0.51 [-0.88, -0.14]). When current PA was included, these differences did not remain statistically significant. Strength training was associated with smaller waist circumference (-4.67 [-8.2, -1.1]) and higher HDL (0.18 [0.02, 0.34]), and active commuting with higher HDL (0.15 [0.01, 0.29]). PA was not associated with blood pressure or glucose. Conclusions Being physically active since young adulthood is beneficial for body composition and blood lipids at the beginning of late adulthood, but current PA may be even more important. Practical implications It is not too late to start exercising at age 61. Identifying PA trajectories across midlife may help to target PA promotion. Funding Research Council of Finland (323541)

Purpose: To estimate the impact of reallocating sedentary leisure time to other movement behaviours on associations with incident cardiometabolic diseases. Methods: A prospective observational study of 462,370 Chinese adults (mean age 51y; 59% female) who were free from diabetes and cardiovascular diseases at baseline. Isotemporal substitution Cox regression models were used to estimate the influence of replacing sedentary leisure time with more sleep, housework, Taichi, or conventional exercise on the rates of incident diabetes (ICD-10: E10-14), myocardial infarction (MI; I21-23), and stroke (I60-I61 & I63-I64). The results are reported as adjusted hazard ratios and 95% confidence intervals per 20-minute time exchanges. Potential impact fractions were calculated to estimate the proportional changes in incident disease cases associated with time substitutions, assuming causality. Results: During >5.25 million person-years of follow-up there were 19,738 incident diabetes, 6,767 MI, and 51,460 stroke cases. Replacing sedentary leisure time with sleep (diabetes: 0.94 (0.90-0.97), MI: 0.93 (0.89-0.99), stroke: 0.96 (0.94-0.98)), housework (diabetes: 0.95 (0.93-0.96), MI: 0.93 (0.91-0.96), stroke: 0.98 (0.97-0.99)) or Taichi (diabetes: 0.95 (0.91-0.98), MI: 0.89 (0.84-0.95), stroke: 0.97 (0.95-0.99)) was associated with lower disease risks. The more substantial associations were obtained when replacing sedentary leisure time with conventional exercise (diabetes: 0.94 (0.89-0.99), MI: 0.84 (0.76-0.92), stroke: 0.93 (0.91-0.96)). Potential impact fractions ranged from 3.5% (sedentary leisure time to housework for incident stroke) to 16.5% (sedentary leisure time to conventional exercise for incident MI). Conclusions: Replacing sedentary leisure time with conventional exercise or non-exercise behaviours is associated with lower rates of cardiometabolic diseases. Practical implications: Substantial public health benefits could be gained by replacing sedentary leisure time with small amounts of exercise, including feasible non-exercise behaviours. Funding: The China Kadoorie Biobank has received funding from numerous bodies (https://www.ckbiobank.org/about-us/funding); this analysis was supported by a Health and Medical Research Fund (HMRF) Research Fellowship (grant no: 06200087).