Deep learning of movement behavior profiles and their association with markers of cardiometabolic 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, that altogether may be codependently related to health outcomes in adults. In this study, we introduce a novel approach to visually represent recorded movement behaviors as images using original accelerometer outputs. Subsequently, we utilize these images for cluster analysis employing deep convolutional autoencoders. 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. The images were encoded into 32 latent variables, and cluster analysis based on these learned representations for the movement behavior images resulted in the identification of four distinct movement behavior profiles characterized by varying levels, timing, and patterns of accumulation of movement behaviors. After adjusting for potential covariates, the movement behavior profile characterized as "Early-morning movers" and the profile characterized as "Highest activity" both had lower levels of insulin (P < 0.01 for both), triglycerides (P < 0.05 and P < 0.01, respectively), HOMA-IR (P < 0.01 for both), and plasma glucose (P < 0.05 and P < 0.1, respectively) compared to the "Lowest activity" profile. No significant differences were observed for the "Least sedentary movers" profile compared to the "Lowest activity" profile. 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.

Associations between insomnia and pregnancy and perinatal outcomes: Evidence from mendelian randomization and multivariable regression analyses.

BACKGROUND: Insomnia is common and associated with adverse pregnancy and perinatal outcomes in observational studies. However, those associations could be vulnerable to residual confounding or reverse causality. Our aim was to estimate the association of insomnia with stillbirth, miscarriage, gestational diabetes (GD), hypertensive disorders of pregnancy (HDP), perinatal depression, preterm birth (PTB), and low/high offspring birthweight (LBW/HBW). METHODS AND FINDINGS: We used 2-sample mendelian randomization (MR) with 81 single-nucleotide polymorphisms (SNPs) instrumenting for a lifelong predisposition to insomnia. Our outcomes included ever experiencing stillbirth, ever experiencing miscarriage, GD, HDP, perinatal depression, PTB (gestational age <37 completed weeks), LBW (<2,500 grams), and HBW (>4,500 grams). We used data from women of European descent (N = 356,069, mean ages at delivery 25.5 to 30.0 years) from UK Biobank (UKB), FinnGen, Avon Longitudinal Study of Parents and Children (ALSPAC), Born in Bradford (BiB), and the Norwegian Mother, Father and Child Cohort (MoBa). Main MR analyses used inverse variance weighting (IVW), with weighted median and MR-Egger as sensitivity analyses. We compared MR estimates with multivariable regression of insomnia in pregnancy on outcomes in ALSPAC (N = 11,745). IVW showed evidence of an association of genetic susceptibility to insomnia with miscarriage (odds ratio (OR): 1.60, 95% confidence interval (CI): 1.18, 2.17, p = 0.002), perinatal depression (OR 3.56, 95% CI: 1.49, 8.54, p = 0.004), and LBW (OR 3.17, 95% CI: 1.69, 5.96, p < 0.001). IVW results did not support associations of insomnia with stillbirth, GD, HDP, PTB, and HBW, with wide CIs including the null. Associations of genetic susceptibility to insomnia with miscarriage, perinatal depression, and LBW were not observed in weighted median or MR-Egger analyses. Results from these sensitivity analyses were directionally consistent with IVW results for all outcomes, with the exception of GD, perinatal depression, and PTB in MR-Egger. Multivariable regression showed associations of insomnia at 18 weeks of gestation with perinatal depression (OR 2.96, 95% CI: 2.42, 3.63, p < 0.001), but not with LBW (OR 0.92, 95% CI: 0.69, 1.24, p = 0.60). Multivariable regression with miscarriage and stillbirth was not possible due to small numbers in index pregnancies. Key limitations are potential horizontal pleiotropy (particularly for perinatal depression) and low statistical power in MR, and residual confounding in multivariable regression. CONCLUSIONS: In this study, we observed some evidence in support of a possible causal relationship between genetically predicted insomnia and miscarriage, perinatal depression, and LBW. Our study also found observational evidence in support of an association between insomnia in pregnancy and perinatal depression, with no clear multivariable evidence of an association with LBW. Our findings highlight the importance of healthy sleep in women of reproductive age, though replication in larger studies, including with genetic instruments specific to insomnia in pregnancy are important.

Independent associations of sleep timing, duration and quality with adiposity and weight status in a national sample of adolescents: The UK Millennium Cohort Study.

Short sleep appears to elevate obesity risk in youth; however, sleep is a multidimensional construct, and few studies have investigated parameters beyond duration. The objective of this study was to investigate if sleep onset time, duration, latency and night waking frequency are independently associated with adiposity and weight status in UK adolescents. This was a cross-sectional observational study of 10,619, 13-15 years olds. Adjusted linear and logistic regressions were used to investigate associations of self-reported sleep characteristics with adiposity markers (body mass index z-score and percent body fat) and weight status. Compared with a sleep onset before 10pm, later sleep timing was associated with higher adiposity and higher likelihood of overweight and obesity in boys (after midnight, odds ratio [95% confidence interval]: 1.76 [1.19-2.60]) and girls (between 11pm and 11:59pm: 1.36 [1.17-1.65]). Sleeping ≤ 8 hr, compared with > 9-10 hr, was associated with higher odds of overweight and obesity in both sexes (boys: 1.80 [1.38-2.35]; girls: 1.38 [1.06-1.79]), and so too was sleeping > 10 hr in girls (1.31 [1.06-1.62]), indicating evidence for a U-shaped association. Also in girls, compared to a sleep latency of 16-30 min, sleep latencies ≥ 46 min were associated with higher adiposity (46-60 min, beta coefficient [95% confidence interval], percent body fat: 1.47 [0.57-2.36]) and higher likelihood of overweight and obesity (46-60 min: 1.39 [1.05-1.83]), and often as opposed to never waking in the night was associated with higher adiposity (body mass index z-score: 0.24 [0.08-0.41]; percent body fat: 1.44 [0.44-2.44]). Sleep duration and timing in both sexes, and sleep quality in girls, appear to be independently associated with adiposity and weight status in adolescence, and may be important targets for obesity prevention.

Device-measured physical activity and sedentary time in a national sample of Luxembourg residents: the ORISCAV-LUX 2 study.

BACKGROUND: Existing information about population physical activity (PA) levels and sedentary time in Luxembourg are based on self-reported data. METHODS: This observational study included Luxembourg residents aged 18-79y who each provided ≥4 valid days of triaxial accelerometry in 2016-18 (n=1122). Compliance with the current international PA guideline (≥150 min moderate-to-vigorous PA (MVPA) per week, irrespective of bout length) was quantified and variability in average 24h acceleration (indicative of PA volume), awake-time PA levels, sedentary time and accumulation pattern were analysed by linear regression. Data were weighted to be nationally representative. RESULTS: Participants spent 51% of daily time sedentary (mean (95% confidence interval (CI)): 12.1 (12.0 to 12.2) h/day), 11% in light PA (2.7 (2.6 to 2.8) h/day), 6% in MVPA (1.5 (1.4 to 1.5) h/day), and remaining time asleep (7.7 (7.6 to 7.7) h/day). Adherence to the PA guideline was high (98.1%). Average 24h acceleration and light PA were higher in women than men, but men achieved higher average accelerations across the most active periods of the day. Women performed less sedentary time and shorter sedentary bouts. Older participants (aged ≥55y) registered a lower average 24h acceleration and engaged in less MVPA, more sedentary time and longer sedentary bouts. Average 24h acceleration was higher in participants of lower educational attainment, who also performed less sedentary time, shorter bouts, and fewer bouts of prolonged sedentariness. Average 24h acceleration and levels of PA were higher in participants with standing and manual occupations than a sedentary work type, but manual workers registered lower average accelerations across the most active periods of the day. Standing and manual workers accumulated less sedentary time and fewer bouts of prolonged sedentariness than sedentary workers. Active commuting to work was associated with higher average 24h acceleration and MVPA, both of which were lower in participants of poorer self-rated health and higher weight status. Obesity was associated with less light PA, more sedentary time and longer sedentary bouts. CONCLUSIONS: Adherence to recommended PA is high in Luxembourg, but half of daily time is spent sedentary. Specific population subgroups will benefit from targeted efforts to replace sedentary time with PA.

Physical activity can attenuate, but not eliminate, the negative relationships of high TV viewing with some chronic diseases: findings from a cohort of 60 202 Brazilian adults.

BACKGROUND: This study examined the joint associations of leisure time physical activity and television (TV) viewing time with the prevalence of chronic diseases among Brazilian adults. METHODS: Data from the Brazilian Health Survey, a nationally representative survey conducted in 2013 (n = 60 202; ≥18 years), were used. Time spent in TV viewing and leisure physical activity, physician diagnoses of diabetes, hypertension and heart disease and information on co-variables (chronological age, education, ethnicity, candies/sweets consumption, sodium intake and tobacco smoking) were collected via interview. Descriptive statistics (mean and 95% confidence interval) and logistic regression models were used for etiological analyses. RESULTS: Physical activity attenuated but did not eliminate the risk associated with high TV viewing for at least one chronic disease in the general population [odds ratio [OR]: 1.29 (1.11-1.50)] and among women [OR: 1.31 (1.09-1.60)], adults [OR: 1.24 (1.05-1.46)] and older adults [OR: 1.63 (1.05-2.53)]. On the other hand, physical activity eliminated the risk associated with high TV viewing for at least one chronic disease among men [OR: 1.24 (0.98-1.58)]. CONCLUSIONS: We conclude that physical activity can attenuate but not eliminate the negative effects of high TV viewing on chronic disease among subgroups of Brazilian adults.

Objectively-measured sedentary time and physical activity in a bi-ethnic sample of young children: variation by socio-demographic, temporal and perinatal factors.

BACKGROUND: Evidence suggests that South Asian school-aged children and adults are less active compared to the white British population. It is unknown if this generalises to young children. We aimed to describe variability in levels of physical activity and sedentary time in a bi-ethnic sample of young children from a deprived location. METHODS: This observational study included 202 South Asian and 140 white British children aged 1.5 to 5y, who provided 3181 valid days of triaxial accelerometry (Actigraph GT3X+). Variability in sedentary time and physical activity levels were analysed by linear multilevel modelling. Logistic multilevel regression was used to identify factors associated with physical inactivity (failing to perform ≥180 min of total physical activity including ≥60 min moderate-to-vigorous physical activity (MVPA) per day). RESULTS: There were no significant ethnic differences in the overall levels of behaviours; South Asian and white British children spent half of daily time sedentary, just over 40% in light physical activity, and the remaining 7.5 to 8% of time in MVPA. Sedentary time was lower and physical activity levels were higher in older children, and levels of MVPA and vector magnitude counts per minute (CPM) were higher on weekends compared to weekdays. In South Asian children, sedentary time was lower on weekends. Sedentary time was lower and physical activity levels were higher in spring compared to winter in white British children, and in all seasons compared to winter in South Asian children. South Asian children born at high birth weight performed more MVPA, and in both ethnicities there was some evidence that children with older mothers were more sedentary and less active. Sedentary time was higher and light physical activity was lower in South Asian children in the highest compared to the lowest income families. South Asian girls performed less MVPA, registered fewer vector magnitude CPM, and were 3.5 times more likely to be physically inactive than South Asian boys. CONCLUSIONS: Sedentary time and physical activity levels vary by socio-demographic, temporal and perinatal characteristics in young children from a deprived location. South Asian girls have the most to gain from efforts to increase physical activity levels. TRIAL REGISTRATION: The Pre-schoolers in the Playground (PiP) pilot randomized controlled trial is registered with the ISRCTN (ISRCTN54165860; http://www.isrctn.com).

Physical activity maintenance and metabolic risk in adolescents.

Aim: Examine the association between child and adolescent physical activity maintenance categories and metabolic profile in adolescence. Methods: This cross-sectional study was conducted with 1152 adolescents (57.4% female) aged 10-16 years from Londrina, Brazil. Physical activity was self-reported in childhood (7-10 years old, retrospective data) and adolescence through questionnaires. Cardiorespiratory fitness (20 m shuttle-run test), body fat (skinfolds), waist circumference, blood pressure (automatic instrument) and blood variables (fasting glucose, HDL-cholesterol and triglycerides) were measured at adolescence. Results: Frequency of physical activity in childhood and adolescence was 50.3 and 17.2%, respectively, and only 25.7% of boys and 10.9% of girls were active at both ages. Adolescents who were physically active in childhood alone were less likely [OR = 0.71 (95% CI: 0.52-0.97)] to present low cardiorespiratory fitness in adolescence compared to those who were non-active in childhood. Regardless of controlled, actives in childhood and adolescence were less likely to present low cardiorespiratory fitness [OR = 0.50 (95% CI: 0.34-0.73)], high blood pressure [OR = 0.52 (95% CI: 0.32-0.85)] and high metabolic risk score [OR = 0.44 (95% CI: 0.22-0.90)] compared to the non-actives at both ages. Conclusions: Actives through childhood to adolescence are less likely to present low cardiorespiratory fitness, high blood pressure and high metabolic risk.

Prevalence, trajectories, and determinants of television viewing time in an ethnically diverse sample of young children from the UK.

BACKGROUND: Excessive screen viewing in early childhood is associated with poor physical and psycho-social health and poor cognitive development. This study aimed to understand the prevalence, trajectory and determinants of television viewing time in early childhood to inform intervention development. METHODS: In this prospective longitudinal study, mothers of 1558 children (589 white British, 757 Pakistani heritage, 212 other ethnicities) completed questionnaires when their children were approximately 6, 12, 18, 24 and 36 months old. Mothers answered questions about their own and their child's TV-time. TV-time trajectories were estimated by linear longitudinal multilevel modeling, potential determinants were considered in models. RESULTS: The modelled trajectory estimated that 75% of children aged 12 months exceeded guidelines of zero screen-time. At 12 months of age an accelerated increase in TV-time was observed (<1 h/day at 14 months, >2 h/day by 30 months old). For every hour of mothers' TV-time and every hour the TV was on in the home, children's TV-time was 8 min and 1 min higher respectively at 6 months old (P < 0.05), and 15 min and 3 min higher respectively at 36 months old (P < 0.05). Children whose mothers did not agree that it was important their child did not watch too much TV, had 17 min more TV-time than their counterparts (P < 0.05). Children of first time mothers had 6 min more TV-time (P < 0.05). At 12 months of age, children of mothers experiencing stress watched 8 min more TV (P < 0.05). By 36 months, children of Pakistani heritage mothers had 22 min more TV-time than those of white British mothers (P < 0.05), and an additional 35 min of TV-time if their mother was not born in the UK (P < 0.05). CONCLUSIONS: High levels of TV-time were prevalent. Intervention developers should consider targeting interventions before 12 months of age. Modifiable determinants included mothers' own TV-time, the time the television is on in the home and mothers' attitude towards child TV-time. These behaviours may be key components to address in interventions for parents. Mothers experiencing stress, first time mothers, and Pakistani heritage mothers (particularly those born outside of the UK), may be priority groups for intervention.

Family history of cardiovascular disease and parental lifestyle behaviors are associated with offspring cardiovascular disease risk markers in childhood.

INTRODUCTION: Cardiovascular disease risk markers have become more prevalent in childhood. To provide increased understanding of the etiology of this public health issue, we investigated associations between family characteristics with cardiovascular disease risk markers in adolescents from a developing nation. METHODS: In this cross-sectional study data for fasting glucose, lipoproteins (LDL-C and HDL-C), triglycerides, and total cholesterol were collected from 991 adolescents aged 10-17 who were recruited from public schools in Londrina city, Southern Brazil. Family history of cardiovascular disease and parental engagement in risk behavior (alcohol consumption and tobacco smoking) were collected using a self-reported parental questionnaire. Socioeconomic status, adolescent physical activity (estimated by Baecke questionnaire), self-reported tobacco and alcohol intake, and somatic maturation (age at peak height velocity) were used as covariates. Logistic regression was used for the main analyses. RESULTS: Independent of adolescent lifestyle behaviors, associations (ORadj [95% CI]) were found between: (1) paternal family history of cardiovascular disease with increased likelihood of high adolescent offspring BMI (1.53 [1.01 to 2.32]) and high triglycerides (2.93 [1.04 to 8.27]); (2) maternal family history of cardiovascular disease with heightened odds of high adolescent offspring triglycerides (2.84 [1.02 to 7.91]); (3) maternal cardiovascular disease with higher odds of high fasting glucose (2.16 [1.13 to 4.14]), and (4) maternal smoking with increased odds of high LDL-C (1.78 [1.14 to 2.79]) and high total cholesterol (1.77 [1.01 to 3.10]) in adolescent offspring. CONCLUSION: Family history of cardiovascular disease and maternal tobacco smoking are related to increased cardiovascular risk in adolescents, potentially independent of their own lifestyle behaviors.

Cardiorespiratory fitness effect may be under-estimated in 'fat but fit' hypothesis studies.

BACKGROUND: Both cardiorespiratory fitness and body fat have been independently related to metabolic syndrome in adolescents; however, the strength of these relationships seems to be dependent on the outcome composition. AIM: To analyse the relationship between cardiorespiratory fitness and body fat combined with different indicators of metabolic risk in adolescents. SUBJECTS AND METHODS: The sample was composed of 957 adolescents (58.7% girls). Cardiorespiratory fitness was obtained using the 20-metre shuttle run test and skinfold thickness was collected for body fat estimation. Metabolic risk score was calculated from waist circumference, systolic and diastolic blood pressure, glucose, HDL cholesterol and triglycerides measurements and an alternative outcome without the central obesity indicator was adopted. Chronological age and somatic maturity were used as covariates. RESULTS: Higher metabolic risk was observed in the highest fat/lowest fit adolescents (p < .05), regardless of sex and outcome. In the regression models, for full metabolic risk score, body fat presented higher coefficients compared to cardiorespiratory fitness in both sexes (boys: 0.501 vs -0.097; girls: 0.485 vs -0.087); however, in the metabolic risk without waist circumference, the coefficients became closer (boys: 0.290 vs -0.146; girls: 0.265 vs -0.120), with a concomitant decrease in body fat and increase in cardiorespiratory fitness coefficients. CONCLUSION: These findings suggest that body fat is strongly related to cardiovascular risk, but, when the outcome is calculated without the central obesity indicator, cardiorespiratory fitness becomes more related to metabolic risk.