PURPOSE:Child and adult muscular power have been shown to associate with contemporary cardiometabolic health. Muscular power typically persists (tracks) between childhood and adulthood. Few studies span childhood to adulthood, so we aimed to identify modifiable and environmental factors associated with the persistence or change in muscular power across the life course. METHODS:Prospective study examining 1938 participants who had their muscular power (standing long jump distance) measured in 1985 as children 7-15 yr old and again 20 yr later in adulthood (26-36 yr old). A selection of objectively measured anthropometric characteristics (adiposity and fat-free mass), cardiorespiratory fitness (CRF), self-reported physical activity, dietary (quality and fruit, vegetable, and protein intake), and sociodemographic data were available at both time points. Muscular power was separated into thirds, and participants were reported as having persistently low, decreasing, persistently moderate, increasing, or persistently high muscular power. RESULTS:Higher adiposity, lower physical activity, diet quality and socioeconomic status (SES) across the life course, and lower adult CRF were associated with persistently low muscular power. Lower adult protein intake and an increase in adiposity over time were associated with decreasing muscular power. An increase in fat-free mass was associated with a reduced probability of decreasing or persistently high muscular power and an increased probability of increasing muscular power. Higher adult fruit intake was associated with increasing muscular power. Lower adiposity across the life course, higher adult CRF and SES, and higher child protein intake were associated with persistently high muscular power. CONCLUSION:Healthy weight, good CRF, greater protein intake, and high SES are important correlates of high muscular power maintained from childhood to adulthood.

It is well established that moderate-vigorous intensity physical activity (MVPA) and resistance training (RT) positively influence muscle capacity and quality and health across the lifespan. Dietary protein intake is also known to influence muscle health. Data suggest that MVPA, RT, and dietary protein intake differ in young men and women, which may impact muscle capacity and quality; however, research in recreationally active young adult cohorts is lacking. PURPOSE: The aim of this study was to determine the associations between dietary protein intake quantity and source and muscle capacity (MC) and quality (MQ), controlling for MVPA and RT, in young men and women. METHODS: Young adults (n=122; 18-22 yr; 54% female) were assessed for a) total, animal and plant protein intakes via 3-day diet recall with analysis by Nutrition Data Systems for Research software; b) body composition via DXA scans, and c) knee extensor muscle strength (MC-S) and power (MC-P) via isokinetic dynamometry and Nottingham leg extensor power rig, respectively. Muscle quality was calculated as strength (MQ-S) and power (MQ-P) relative to leg lean mass. MVPA was determined using accelerometry and RT was determined from questionnaire. RESULTS: Compared to females, males ingested more animal protein when normalized per body weight, as a percentage of total kcal, and as a percentage of total protein intake (all p<.05). Males also had more lean mass and greater MC-S, MC-P, and MQ-S than females (all p<.05); however, there was no sex difference in MVPA, RT, or MQ-P (all p>.05). Additionally, males had a significant relationship between total protein, total animal protein, and total plant protein intake and MC-S whereas females did not. Controlling for sex, MVPA, and RT, higher total dietary protein and total animal protein intake were associated with greater MC-S (r = .29 and .25, respectively, both p<.05). Higher relative animal protein intake was positively related, whereas higher relative plant protein intake was inversely related to MQ-P (both p<.05). CONCLUSIONS: This data suggests that young adult males and females differ in their dietary protein intake patterns. Moreover, dietary protein intake is modestly related to muscle capacity and quality with effects being stronger in males compared to their female counterparts.

To help inform strategies aimed at increasing muscular fitness levels, we examined factors associated with childhood muscular fitness (strength and power) that preceded the recently observed secular decline. Data were available from a nationally representative sample of Australian children aged 7-15 years in 1985 (n = 8469). Muscular fitness measures included strength (right and left grip, shoulder extension and flexion, and leg strength) and power (standing long jump distance). Anthropometric (adiposity, fat-free mass), cardiorespiratory fitness (CRF), flexibility, speed capability, physical activity (individual and parental), dietary quality and intake (fruit, vegetable, protein) and sociodemographic (area-level socioeconomic status (SES), school type) data were available. Statistical analyses included sex-stratified linear regression. Of all examined factors, measures of adiposity, fat-free mass, CRF, flexibility and speed capability were associated with muscular fitness at levels that met Cohen's threshold for important effects (r-squared = 0.02 to 0.28). These findings highlight the multifactorial relationship between muscular fitness and its determinants. Collectively, these factors were powerful in explaining muscular strength (females: r-squared = 0.32; males: r-squared = 0.41) and muscular power (females: r-squared = 0.36; males: r-squared = 0.42). These findings highlight modifiable and environmental factors that could be targeted to increase childhood muscular fitness.

This study aimed to assess whether the longitudinal association between childhood muscular fitness and adult measures of glucose homeostasis persist despite changes in muscular fitness across the life course. This prospective longitudinal study included 586 participants who had their muscular power (standing long jump distance), cardiorespiratory fitness (CRF), and waist circumference measured as children (aged 9, 12, 15 years) and again 20 years later as adults. In adulthood, these participants also provided a fasting blood sample which was tested for glucose and insulin. Glucose homeostasis measures including insulin resistance (HOMA2‐IR) and beta cell function (HOMA2‐β) were estimated. Child and adult muscular power levels were separated into thirds, and tracking groups (persistently low, decreasing, persistently moderate, increasing, and persistently high) were created. Sex‐stratified multivariable linear regression models were used to examine the association between muscular power tracking groups and adult measures of glucose homeostasis. Compared with males with persistently high muscular power, males with increasing and persistently low muscular power had higher fasting insulin (increasing: β = 1.12 mU/L, P = .04; persistently low: β = 2.12 mU/L, P = .001) and HOMA2‐β (increasing: β = 8.50%, P = .03; persistently low: β = 11.27%, P = .01) independent of CRF and males with persistently low muscular power had greater fasting insulin (β = 1.22 mU/L, P = .02) and HOMA2‐IR (β = 0.14, P = .02) independent of waist circumference. Non‐significant associations were present for females. For males, maintaining persistently high muscular power between childhood and adulthood could lead to a healthier adult glucose homeostasis profile.

G.M. Frederick1, B.M. Das2, M.V. Fedewa3, R.M. Reed1, R.E. Salyer1, M.D. Schmidt1, E.M. Evans1, FACSM. 1University of Georgia, Athens, GA 30602. 2East Carolina University, Greenville, NC 27858. 3University of Alabama, Tuscaloosa, AL 35487. PURPOSE: Body mass index (BMI) is often used as a surrogate measure of adiposity (%Fat). It is well-established that physical activity (PA) influences body composition. In addition to the well-established sex differences in body composition, college-age males also typically engage in more PA than their female counterparts. This difference in PA could potentially impact the relationship between BMI and %Fat. Therefore, the aim of this study was to determine if PA differentially influences the relationship between BMI and %Fat in college-age males compared to females. METHODS: BMI was calculated from weight and height measured using standard clinical protocols. PA was measured in steps/day using the NL-1000 accelerometer. %Fat was measured via DEXA. RESULTS: Males (N = 124; 18.4 ± 0.5 yrs, 23.2 kg/m2) and females (N = 282; 18.3 ± 0.5 yrs, 23.0 kg/m2) were nearly identical in age and BMI (both p > 0.05). As expected, males were leaner (18.4 ± 5.1 %Fat vs. 32.2 ± 5.7 %Fat, p < 0.001) and accumulated more PA (11,625 ± 2930 vs. 10,866 ± 3467 steps/day, p = 0.03) compared to females. Because of the known sex difference in %Fat, separate linear regression models were evaluated to explore the prediction of %Fat from BMI, PA, and BMI x PA. BMI explained 50.2% of the variance in %Fat among females and only 18.3% of the variance in males (p < 0.001 for both). Adding PA to the model significantly increased the variance in %Fat explained in both females and males (∆R2 = 3.4% and 3.5%, respectively, both p < 0.05). The addition of the BMI x PA interaction term improved the model in females (∆R2 = 1.3%, p = 0.005), but not males (∆R2 = 0.0 %, p = 0.933). CONCLUSION: Weight management is of high public health importance, especially for young adults who have an increasing risk for obesity during this stage of life. While many health promotion efforts focus on weight management with BMI as a primary outcome, it is important to account for sex differences with respect to the relationships among BMI, PA, and %Fat when using BMI for program evaluation in the young adult population.

ABSTRACT Purpose Muscle cross-sectional area (MCSA) is often used as a surrogate for the forces applied to bones during physical activity. Although MCSA is a strong predictor of cortical bone status, its use makes assumptions about the relationship between muscle size and force that are inaccurate. Furthermore, to measure MCSA and other muscle force surrogates typically requires expensive and/or radiative laboratory equipment. Thus, this study aimed to determine whether clinical laboratory– and field-based methodologies for measuring muscular force capacity accounted for similar variance in diaphyseal cortical bone status as a commonly used muscular force surrogate, MCSA, at the midtibia in young men and women. Methods Healthy young adults ( n = 142, 19.7 ± 0.7 yr old, 52.8% female) were assessed via peripheral quantitative computed tomography at the midtibia for cortical bone status and MCSA. Muscle force capacity was measured via Biodex dynamometer, Nottingham leg extensor power rig, and Vertec vertical jump. Regression analysis compared the independent variance predicted by each muscle force measure with that of MCSA, accounting for relevant confounders. Results MCSA, knee extension peak torque, and peak anaerobic power from vertical jump were independent predictors of select cortical structural outcomes (cortical thickness and area, periosteal and endosteal circumference, and estimated strength) accounting for up to 78.4% of the variance explained (all P < 0.05). However, cortical volumetric bone mineral density was unrelated to any measure or surrogate of muscle force capacity. Conclusions MCSA is a strong independent predictor of cortical bone structure; however, both laboratory- and field-based measures of peak torque and/or peak anaerobic power are promising alternatives, explaining similar and sometimes greater variance than MCSA.

BACKGROUND:Recess and physical education time continue to diminish, creating a need for additional physical activity opportunities within the school environment. The use of school gardens as a teaching tool in elementary science and math classes has the potential to increase the proportion of time spent active throughout the school day.METHODS:Teachers from 4 elementary schools agreed to teach 1 math or science lesson per week in the school garden. Student physical activity time was measured with ActiGraph GT3X accelerometers on 3 garden days and 3 no-garden days at each school. Direct observation was used to quantify the specific garden-related tasks during class. The proportion of time spent active and sedentary was compared on garden and no-garden days.RESULTS:Seventy-four children wore accelerometers, and 75 were observed (86% participation). Children spent a significantly larger proportion of time active on garden days than no-garden days at 3 of the 4 schools. The proportion of time spent sedentary and active differed significantly across the 4 schools.CONCLUSIONS:Teaching lessons in the school garden may increase children's physical activity and decrease sedentary time throughout the school day and may be a strategy to promote both health and learning.

We present a case study of the performance testing of a commercially engineered genetic programming algorithm applied to the automated modeling of industrial machine learning problems. This paper summarizes some of what has been learned over the past five years of working with a large number of industrial machine learning challenges in a commercial or enterprise setting. Automation and parallelism via cloud computing is used to reduce test time. Two frameworks for conducting performance tests are discussed, highlighting the advantages of collecting statistics throughout the search. A performance test suite of industrial machine learning problems is described, and examples of performance test results are shown. Finally, a summary of challenges and open questions is provided.

BACKGROUND:Sprint interval cycle training is a contemporary popular mode of training but its relative efficacy, under conditions of matched energy expenditure, to reduce risk factors for cardiometabolic disease is incompletely characterised, especially in young women. The purpose of this investigation was to determine the relative efficacy of six weeks of moderate-intensity cycling (MOD-C) and vigorous sprint-interval cycling (VIG-SIC) on lipid profile, insulin (INS) and insulin resistance using the homeostatic model assessment (HOMA-IR) and C-reactive protein (CRP) in inactive, overweight/obese (OW/OB) young women. METHODS:Participants (BMI ≥25 kg/m2, waist circumference ≥88 cm) were randomly assigned to MOD-C (20-30 min at 60-70% of heart rate reserve(HRR)) or VIG-SIC (5-7 repeated bouts 30-second maximal effort sprints, followed by four minutes of active recovery) supervised training three days/week for six weeks, with each group matched on energy expenditure. Adiposity (%Fat) was measured using dual x-ray absorptiometry. RESULTS:Forty-four participants (20.4 ± 1.6 years, 65.9% Caucasian, 29.8 ± 4.1 kg/m2) were included in the analysis. The improvement in CRP observed in the MOD-C group was larger than the VIG-C group (p = .034). Overall, high-density lipoprotein (HDL-C) and low-density lipoprotein (LDL-C) levels improved following training (p < .05); however, total cholesterol, triglyceride, INS and HOMA-IR did not improve (p > .05). CONCLUSION:These results indicate MOD-C training may be more effective in reducing CRP than VIG-SIC.

The purpose of the study was to examine the effects of sprint interval training (SIT) and moderate-intensity continuous cycle training (MICT), with equal estimated energy expenditure during training on body composition and aerobic capacity. Body composition measured via dual-energy X-ray absorptiometry and aerobic capacity were assessed following 6 weeks of training in previously inactive overweight/obese young women (n = 52; age, 20.4 ± 1.5 years; body mass index, 30.3 ± 4.5 kg·m-2, 67.3% white). Training was performed in a group-exercise format that mimicked cycling classes offered by commercial fitness facilities, and included 3 weekly sessions of either 30-s "all-out" sprints followed by 4 min of active recovery (SIT), or continuous cycling at 60%-70% heart rate reserve to expend a similar amount of energy. Participants were randomized to SIT or MICT, attended a similar number of sessions (15.0 ± 1.5 sessions vs. 15.8 ± 1.9 sessions, P = 0.097) and expended a similar amount of energy (541.8 ± 104.6 kJ·session-1 vs. 553.5 ± 138.1 kJ·session-1, P = 0.250). Without significant changes in body mass (P > 0.05), greater relative reductions occurred in SIT than in MICT in total fat mass (3.6% ± 5.6% vs. 0.6% ± 3.9%, P = 0.007), and android fat mass (6.6% ± 6.9% vs. 0.7% ± 6.5%, P = 0.002). Aerobic capacity (mL·kg-1·min-1) increased significantly following both interventions (P < 0.05), but the relative increase was 2-fold greater in SIT than in MICT (14.09% ± 10.31% vs. 7.06% ± 7.81%, P < 0.001). In conclusion, sprint-interval cycling reduces adiposity and increases aerobic capacity more than continuous moderate-intensity cycling of equal estimated energy expenditure in overweight/obese young women.

Objective. The aims of this study were to characterize the dose-response relationship between moderate-to-vigorous intensity physical activity (MVPA), and light-intensity activity with HbA1c in adults at low, moderate, and high risks of type 2 diabetes, and to compare the relationship of short (1 to 9 min) versus long (10 + min) bouts of MVPA with HbA1c.Methods. Data from 2707 participants from the 2003-2006 National Health And Nutrition Examination Survey were analyzed in 2014-2015. Type 2 diabetes risk was classified into three groups based upon age (<40 years; >= 40 years) and BMI (<30; >= 30). The relationship between HbA1c and accelerometer-based physical activity variables was assessed using multiple regression models.Results. There was a curvilinear dose-response relationship between HbA1c with total activity and MVPA in adults at moderate or high risk for type 2 diabetes: higher amounts of physical activity were associated with lower HbA1c. The association of physical activity on HbA1c was stronger at lower levels of physical activity. There was no dose-response relationship in adults at low risk for type 2 diabetes. The relationship between short bouts with HbA1c was stronger than for bouts >= 10 min.Conclusions. In adults at risk for type 2 diabetes, there is a dose-response relationship between physical activity and HbA1c levels such that the relationship: (1) is curvilinear; (2) is stronger when a higher percent of total activity comes from MVPA; and (3) is more potent with short bouts of MVPA. Fractionalized physical activity of at least moderate-intensity may contribute to long-term glucose control. (C) 2016 Elsevier Inc. All rights reserved.

It is well established that exercise interventions that reduce adiposity and improve cardiorespiratory fitness (CRF), are of primary importance for cardiometabolic health. The contemporary sprint interval cycling, a high-intensity, low-volume exercise modality is popular and time efficient; however, it’s effectiveness to reduce adiposity and improve CRF has not been completely characterized in comparison to moderate-intensity continuous exercise due to lack of control of exercise energy expenditure in the research protocol. PURPOSE: The aim of this study was to compare the relative effects of high-intensity/vigorous, low-volume sprint interval cycling (VIG-SIC) and moderate-intensity continuous cycle training (MOD-C) of equal energy expenditure on adiposity and CRF. METHODS: Body composition (dual-energy x-ray absorptiometry) and CRF (VO2peak) were measured in response to 6 weeks of training in previously inactive overweight/obese young-adult females (n=52, 20.4±1.5 yo, 30.3±4.5 kg/m2, 43.5±4.9 %Fat, 27.9±4.7 ml/kg/min) blocked on BMI and randomized to VIG-SIC (n=23) or MOD-C (n=29). Energy expenditure during training sessions was estimated by Keiser 3m cycles (Keiser, Keiser M3 Indoor Cycle, Fresno, California). Habitual energy expenditure was measured objectively using the Actiheart monitor (CamNtech, USA), and energy intake was measured using the Automated Self-Administered 24-hour Recall (ASA24) for a 3-day period pre- and post-training. RESULTS: Group changes in body mass, reported energy intake or habitual energy expenditure did not differ (all P>0.05). Regarding total and central adiposity, greater reductions occurred in VIG-SIC (both P<0.05) compared to MOD-C (both P>0.05) in total fat mass (-1.2±1.9 vs. -0.2±1.4 kg, GROUP*TIME P=0.021) and central fat mass (-0.8±1.2 vs. -0.1±0.8 kg, GROUP*TIME P=0.018). CRF (ml/kg/min) improved for both training groups (both P<0.001); however, a significant GROUP*TIME interaction (P=.003) indicated a 2-fold greater increase in VIG-SIC (14.09±10.31%) compared to MOD-C (7.06±7.81%). CONCLUSIONS: Relatively short-term high-intensity low-volume interval training reduces adiposity and increases CRF more than continuous moderate-intensity cycling of equal energy expenditure in overweight/obese young-adult females.

OBJECTIVE: The objective of this study was to examine whether childhood cardiorespiratory fitness attenuates or modifies the long-term cardiometabolic risks associated with childhood obesity.DESIGN AND METHODS: The study consisted of a 20-year follow-up of 1792 adults who participated in the 1985 Australian Schools Health and Fitness Survey when they were 7-15 years of age. Baseline measures included a 1.6-km run to assess cardiorespiratory fitness and waist circumference to assess abdominal adiposity. At follow-up, participants attended study clinics where indicators of Metabolic Syndrome (MetS) (waist circumference, blood pressure, fasting blood glucose and lipids) were measured and cardiorespiratory fitness was reassessed using a submaximal graded exercise test.RESULTS: Both high waist circumference and low cardiorespiratory fitness in childhood were significant independent predictors of MetS in early adulthood. The mutually adjusted relative risk of adult MetS was 3.00 (95% confidence interval: 1.85-4.89) for children in the highest (vs lowest) third of waist circumference and 0.64 (95% confidence interval: 0.43-0.96) for children with high (vs low) cardiorespiratory fitness. No significant interaction between waist circumference and fitness was observed, with higher levels of childhood fitness associated with lower risks of adult MetS among those with either low or high childhood waist circumference values. Participants who had both high waist circumference and low cardiorespiratory fitness in childhood were 8.5 times more likely to have MetS in adulthood than those who had low waist circumference and high cardiorespiratory fitness in childhood. Regardless of childhood obesity status, participants with low childhood fitness who increased their relative fitness by adulthood had a substantially lower prevalence of MetS than those who remained low fit.CONCLUSIONS: Childhood waist circumference and cardiorespiratory fitness are both strongly associated with cardiometabolic health in later life. Higher levels of cardiorespiratory fitness substantially reduce the risk of adult MetS, even among those with abdominal obesity in childhood.

Health related quality of life (HR-QOL) is an active area of research in contemporary physical activity and obesity literatures. However, the majority of the research has targeted either older adults or individuals afflicted with chronic disease. The influence of physical activity (PA) and adiposity (%Fat) on the mental and physical aspects of health related quality of life (HR-QOL) in younger individuals, especially those in life transition stages has not been investigated. PURPOSE: The aim of this study was to evaluate the independent effects of PA and %Fat on HR-QOL in college freshman females. METHODS: Females (n=280, 18.3±.5 yr) were assessed objectively for both PA, using NL1000 accelerometers (average steps/day), and %Fat via DXA. HR-QOL was assessed using the SF-36 and the following subscales were evaluated [1. physical function (FUNC), 2. role physical (PHYS), 3. bodily pain (BP), 4. general health (GH), 5. physical health component (PHC), 6. vitality (VIT), 7. social functioning (SOC), 8. role emotional (EMO), 9. mental health component score (MHC)]. RESULTS: As expected, PA (10,754 ± 3578 steps/day) and %Fat (33.4 ± .06%) were related (r=−.26, p<.01). PA and %Fat were not independent predictors for FUNC, PHYS, BP, PHC, SOC, or EMO (all p>.05). However, PA was a significant independent predictor of GH, VIT, MH, MHC explaining 5.1%, 6.4%, 2.6%, and 4.9%, respectively (all p<.05). Adiposity independently predicted 4.1% of GH. CONCLUSIONS: PA appears to be more important for QOL, especially mental health and vitality than %Fat. However, the relative and independent effects of PA and %Fat on HR-QOL in college-aged females, especially among those with more diverse PA levels and adiposity levels, warrants further study using prospective designs.