Interpretation of Multivariate Association Patterns between Multicollinear Physical Activity Accelerometry Data and Cardiometabolic Health in Children-A Tutorial

Multivariate pattern analysis: a method and software to reveal, quantify, and visualize predictive association patterns in multicollinear data

BACKGROUND

Strongly multicollinear covariates, such as those typically represented in metabolomics applications, represent a challenge for multivariate regression analysis. These challenges are commonly circumvented by reducing the number of covariates to a subset of linearly independent variables, but this strategy may lead to loss of resolution and thus produce models with poorer interpretative potential. The aim of this work was to implement and illustrate a method, multivariate pattern analysis (MVPA), which can handle multivariate covariates without compromising resolution or model quality.

RESULTS

MVPA has been implemented in an open-source R package of the same name, mvpa. To facilitate the usage and interpretation of complex association patterns, mvpa has also been integrated into an R shiny app, mvpaShiny, which can be accessed on www.mvpashiny.org . MVPA utilizes a general projection algorithm that embraces a diversity of possible models. The method handles multicollinear and even linear dependent covariates. MVPA separates the variance in the data into orthogonal parts within the frame of a single joint model: one part describing the relations between covariates, outcome, and explanatory variables and another part describing the "net" predictive association pattern between outcome and explanatory variables. These patterns are visualized and interpreted in variance plots and plots for pattern analysis and ranking according to variable importance. Adjustment for a linear dependent covariate is performed in three steps. First, partial least squares regression with repeated Monte Carlo resampling is used to determine the number of predictive PLS components for a model relating the covariate to the outcome. Second, postprocessing of this PLS model by target projection provided a single component expressing the predictive association pattern between the outcome and the covariate. Third, the outcome and the explanatory variables were adjusted for the covariate by using the target score in the projection algorithm to obtain "net" data. We illustrate the main features of MVPA by investigating the partial mediation of a linearly dependent metabolomics descriptor on the association pattern between a measure of insulin resistance and lifestyle-related factors.

CONCLUSIONS

Our method and implementation in R extend the range of possible analyses and visualizations that can be performed for complex multivariate data structures. The R packages are available on github.com/liningtonlab/mvpa and github.com/liningtonlab/mvpaShiny.

Association between biomarkers of redox status and cytokines with different patterns of habitual physical activity in eutrophic and overweight/obese preschoolers: multivariate analysis of a cross-sectional study

BACKGROUND

Although it is well known that obesity is frequently associated with reduced levels of habitual physical activity (HPA), which contributes to determining severe oxidative stress and inflammatory state, this association is however unknown in preschoolers so far. This study aimed to investigate the association between biomarkers of redox status and cytokines with different patterns of HPA according to the adiposity of preschoolers.

METHODS

A cross-sectional study was conducted in 50 preschoolers (25 overweight/obese, OW/OB and 25 eutrophic, EU), matched for age, sex, economic level, and maternal education. Total antioxidant capacity (TAC), superoxide dismutase (SOD) and catalase (CAT) activities, substances reactive to thiobarbituric acid (TBARS), soluble tumor necrosis factor receptors (sTNFRs), and leptin levels were evaluated. HPA levels were evaluated by accelerometry (ActiGraph GT9X accelerometer). Correlation, multiple linear regression, and partial least squares regression analysis were used to determine the association between redox status biomarkers and cytokines with different patterns of HPA (HPA level, bouts of moderate to vigorous physical activity [MVPA], and multivariate pattern of HPA) in EU and OW/OB preschoolers.

RESULTS

OW/OB preschoolers had lower CAT activity, higher levels of TAC, TBARS, and cytokines, and similar levels of HPA to EU preschoolers. In EU preschoolers, SOD activity exhibited a stronger negative association with moderate intensity ranges of HPA (R2 = 0.18), and negative correlation with sTNFRs (r = -0.40 to -0.46). TBARS had a stronger positive association with ranges of light intensity in the multivariate pattern of HPA (R2 = 0.10). In OW/OB preschoolers, the HPA multivariate associative pattern was predominantly from vigorous intensity ranges. Thus, SOD activity had a positive association with the multivariate pattern of HPA (R2 = 0.38) and MVPA bouts (β [95% CI] = 0.457 [0.0026. 0.0576]). TAC had a negative association with the multivariate pattern of HPA (R2 = 0.38) and MVPA bouts (β [95% CI] = -0.718 [-0.0025. -0.0003]). Additionally, leptin levels were lower in OW/OB preschoolers engaged in vigorous physical activity (VPA) (8000-9999 counts/min) for longer periods of time.

CONCLUSION

The results of this study indicate that OW/OB preschoolers have higher levels of oxidative stress biomarkers and pro-inflammatory cytokines compared to EU preschoolers. Moreover, VPA may exert antioxidative and anti-inflammatory effects in OW/OB preschoolers.

Multivariate physical activity association patterns for fundamental motor skills and physical fitness in preschool children aged 3-5 years

Physical activity (PA) is important for children's development of fundamental motor skills (FMS) and physical fitness (FIT) but evidence regarding which intensities are associated with these outcomes in early childhood is limited. The aim of this study was to determine the cross-sectional multivariate PA intensity signatures associated with FMS and FIT in children aged 3-5 years. We used a sample of 952 Norwegian preschoolers (4.3 years, 51% boys) who provided data on PA (ActiGraph GT3X+), at least one FMS (locomotor, object control and/or balance skills) or FIT (speed agility, standing long jump, and/or handgrip strength) outcome, body mass index, and socioeconomic status in 2019-2020. We created 17 PA intensity variables (0-99 to ≥15000 counts per minute) from the vertical axis and used multivariate pattern analysis for analyses. The PA intensity spectrum (including sedentary time) was significantly associated with all outcomes. Associations for PA intensities were positive (negative for sedentary time), strongest for moderate and vigorous intensities, and were significant across sex and age groups. Our findings show that the PA intensity spectrum is associated with FMS and FIT in young children and that promotion of PA, in particular moderate- and vigorous-intensity activity, from an early age benefits children's physical development.

Associations between educators' and children's physical activity and sedentary time in Norwegian preschools: A cross-sectional study

Preschool educators may be important role models influencing children's physical activity (PA) behaviours. The main aim of this paper was to examine the relationship between educators' and children's PA and sedentary time (SED) by including 1230 children (mean age 4.8 yrs, 48% girls) and 422 educators (mean age 42.4 yrs, 90% women) from 68 preschools in Western Norway. PA and SED were measured over 10 preschool days using hip worn ActiGraph GT3×+ accelerometers. Associations between child and educator PA and SED during preschool hours, determined by multivariate pattern analyses, provided explained variances (R²) = 2.8-5.2%. Levels of educator moderate PA were positively related to child PA (all intensities) and educator vigorous PA were positively related to child vigorous PA, while educator light PA was negatively associated with PA and positively associated with child SED. Educator SED were positively associated with child vigorous PA and negatively associated to child SED. Association patterns were similar for boys and girls, while educator moderate and vigorous PA were more strongly related to the younger vs. the older children's PA. The positive relationships between educators' moderate and vigorous PA and children's PA found herein suggest educators' PA behaviours should be addressed in future interventions.

The influence of reliability and variability of objectively measured physical activity on associations with lower body muscle strength in young children

It is not known how extended or multiple monitoring periods affect associations between accelerometer-measured physical activity and outcomes. The aim of this study was to examine how accelerometer monitoring length influenced cross-sectional and prospective associations for physical activity with lower body muscle strength in young children. 176 Norwegian 2-6-year-old children had 3 valid 14-day periods of accelerometer monitoring (ActiGraph GT3×+) between September 2015 and May 2016 (baseline) as well as baseline and 4-year follow-up measurements of standing long jump. We analysed physical activity using a descriptor with 4 intensities using 6 different monitoring lengths both within and across monitoring periods (1 day, 3 days, 1 week, 2 weeks, 3 weeks, 6 weeks) and determined associations with lower body muscle strength using multivariate pattern analysis. We found that the strength of cross-sectional associations with lower body muscle strength improved for longer monitoring periods (explained variances = 7.7%, 9.8%, 11.8%, 15.8%, 18.4% and 22.9% for 1 day, 3 days, 1 week, 2 weeks, 3 weeks and 6 weeks of measurement). Longitudinal associations were weaker and less clear. Our findings suggest that multiple extended physical activity monitoring periods improve the validity of the study findings regarding associations between physical activity and relevant outcomes.

Prospective Association Patterns for the Physical Activity Intensity Spectrum With Body Mass Index and Lower Body Muscle Strength in Norwegian Children Aged 3-9 Years

BACKGROUND

Limited evidence exists regarding prospective associations for physical activity (PA) and sedentary time (SED) with body mass index (BMI) and muscle strength in young children. We aimed to determine prospective associations for PA and SED with change in BMI and standing long jump over 2 and 4 years in children aged 3-5 years at baseline.

METHODS

A sample of 262 Norwegian children (50% girls) was followed from 2015 to 2017 and/or 2019. PA and SED (hip-worn ActiGraph GT3X+) were measured at baseline and BMI and standing long jump at baseline and at follow-ups. Multivariate pattern analysis was used to determine prospective associations between the triaxial PA intensity spectrum (0-99 to ≥15,000 counts per minute) and the change in outcomes.

RESULTS

We found significant prospective associations between the PA intensity spectrum and standing long jump at 2- (explained variance = 5.8%-7.7%) and 4-year (explained variance = 4.8%-5.6%) follow-ups. Associations were negative for SED and positive for all PA intensities. We found no associations between PA/SED and BMI.

CONCLUSIONS

Our findings suggest that PA and SED can predict future lower body muscle strength but not BMI in early childhood.

Accelerometer epoch length influence associations for physical activity intensities with body mass index and locomotor skills in young children

The epoch length is decisive for the capture of physical activity intensities from accelerometry and possibly for associations between physical activity intensities and outcomes in children. The aim of the present study was to determine the multivariate physical activity intensity signatures related to body mass index and locomotor skills in preschool children using different epoch lengths. A sample of 1054 Norwegian children (mean age 4.8 years, 52% boys) from the Sogn og Fjordane Preschool Physical Activity Study provided data on physical activity (ActiGraph GT3X+), body mass index and locomotor skills in 2015-2016. Multivariate pattern analysis was used to determine associations between the physical activity intensity spectrum (0-99 to ≥15000 counts per minute) and the outcomes using files aggregated using 8 different epoch lengths from 1 to 60 seconds. We observed that associations across the intensity spectrum, from sedentary time to vigorous intensities, differed for data derived using different epoch lengths. For both outcomes, associations for moderate intensities became stronger and vigorous intensities became weaker for longer as compared to shorter epoch lengths. Aggregation of accelerometer data using different epoch lengths influences the capture of physical activity intensities and associations between physical activity intensities and related outcomes.

The Multivariate Physical Activity Signatures Associated With Self-Regulation, Executive Function, and Early Academic Learning in 3-5-Year-Old Children

The evidence regarding associations between intensity-specific physical activity and cognitive and learning outcomes in preschoolers is inconsistent and limited by low sample sizes and analytical approaches that cannot handle the multicollinearity among multiple physical activity intensity variables. We aimed to determine the multivariate physical activity intensity signatures associated with self-regulation, executive function, and early academic learning in preschool children aged 3–5 years. A 711 Norwegian preschool children (mean age 4.6 years, 52% boys) provided valid data on physical activity (ActiGraph GT3X+), self-regulation, executive function, and early academic learning during 2019–2020. Multivariate pattern analysis was used to determine associations between uniaxial and triaxial intensity spectra (time spent in intensities from 0–99 to ≥15,000 counts per minute) and the outcomes in the total sample and in subgroups split by sex and age (median split). Uniaxial data led to the highest explained variances (R²) and were reported as the primary findings. We found significant association patterns between physical activity and numeracy (R² = 4.28%) and inhibition (R² = 1.48%) in the total sample. The associations with numeracy were negative for time spent sedentary (0–99 counts per minute) and positive for time spent in moderate to vigorous intensities (≥ 1,000 counts per minute). The associations with inhibition were positive for time spent sedentary (0–99 counts per minute) and in vigorous intensities (≥ 8,500 counts per minute) and negative for time spent in low to moderate intensities (100–3,499 counts per minute). Associations with numeracy were stronger in boys (R² = 5.58%) and older children (R² = 7.27%), and associations with inhibition were stronger in girls (R² = 3.12%) and older children (R² = 3.33%). In conclusion, we found weak associations with numeracy and inhibition across the physical activity intensity spectrum in preschool children.

Physical activity intensity profiles associated with cardiometabolic risk in middle-aged to older men and women

Accelerometers provide detailed data about physical activity (PA) across the full intensity spectrum. However, when examining associations with health, results are often aggregated to only a few summary measures [e.g. time spent "sedentary" or "moderate-to-vigorous" intensity PA]. Using multivariate pattern analysis, which can handle collinear exposure variables, we examined associations between the full PA intensity spectrum and cardiometabolic risk (CMR) in a population-based sample of middle-aged to older adults. Participants (n = 3660; mean ± SD age = 69 ± 8y and BMI = 26.7 ± 4.2 kg/m²; 55% female) from the EPIC-Norfolk study (UK) with valid accelerometry (ActiGraph-GT1M) data were included. We used multivariate pattern analysis with partial least squares regression to examine cross-sectional multivariate associations (r) across the full PA intensity spectrum [minutes/day at 0-5000 counts-per-minute (cpm); 5 s epoch] with a continuous CMR score (reflecting waist, blood pressure, lipid, and glucose metabolism). Models were sex-stratified and adjusted for potential confounders. There was a positive (detrimental) association between PA and CMR at 0-12 cpm (maximally-adjusted r = 0.08 (95%CI 0.06-0.10). PA was negatively (favourably) associated with CMR at all intensities above 13 cpm ranging between r = -0.09 (0.07-0.12) at 800-999 cpm and r = -0.14 (0.11-0.16) at 75-99 and 4000-4999 cpm. The strongest favourable associations were from 50 to 800 cpm (r = 0.10-0.12) in men, but from ≥2500 cpm (r = 0.18-0.20) in women; with higher proportions of model explained variance for women (R² = 7.4% vs. 2.3%). Most of the PA intensity spectrum was beneficially associated with CMR in middle-aged to older adults, even at intensities lower than what has traditionally been considered "sedentary" or "light-intensity" activity. This supports encouragement of PA at almost any intensity in this age-group.

Physical activity spectrum discriminant analysis-A method to compare detailed patterns between groups

Investigating physical activity (PA) patterns as a detailed intensity spectrum instead of crude intensity categories have improved the ability to analyze the relationship between measured PA and health variables. The aim of this methodological study was to introduce and investigate the utility of using detailed PA intensity spectrum compared to crude PA intensity categories for comparison of PA between groups and between repeated measures. The study sample consisted of two groups of children, where one group was scheduled for extended physical education (PE) by daily classes while the other group followed usual PE schedule. Accelerometer data was processed into traditional crude PA intensity categories and into detailed PA intensity spectrum. Multivariate partial least squares regression for discriminant analysis (PLS-DA) was applied for PA intensity spectrum group comparison and compared to traditional univariate statistical analysis. Repeated measures were investigated using independent PLS-DA as well as multilevel PLS-DA for paired analysis. While traditional analysis of crude PA intensity categories was unable to find any group differences, multivariate analysis of the PA intensity spectrum identified statistically significant differences. By the extension of multilevel PLS-DA for paired comparison, a clear difference in the PA intensity spectrum was demonstrated between repeated measures. In conclusion, analysis of detailed PA intensity spectrum demonstrates utility for comparing detailed PA data between groups and between repeated measures in interventional and observational research.

High-intensity activity is more strongly associated with metabolic health in children compared to sedentary time: a cross-sectional study of the I.Family cohort

Background

Physical activity (PA) during childhood is important for preventing future metabolic syndrome (MetS). To examine the relationship between PA and MetS in more detail, accurate measures of PA are needed. Previous studies have only utilized a small part of the information available from accelerometer measured PA. This study investigated the association between measured PA and MetS in children with a new method for data processing and analyses that enable more detailed interpretation of PA intensity level.

Methods

The association between PA pattern and risk factors related to MetS was investigated in a cross- sectional sample of children (n = 2592, mean age 10.9 years, 49.4% male) participating in the European multicenter I. Family study. The risk factors examined include body mass index, blood pressure, high-density lipoprotein cholesterol, insulin resistance and a combined risk factor score (MetS score). PA was measured by triaxial accelerometers and raw data was processed using the 10 Hz frequency extended method (FEM). The PA output was divided into an intensity spectrum and the association with MetS risk factors was analyzed by partial least squares regression.

Results

PA patterns differed between the European countries investigated, with Swedish children being most active and Italian children least active. Moderate intensity physical activity was associated with lower insulin resistance (R² = 2.8%), while vigorous intensity physical activity was associated with lower body mass index (R² = 3.6%), MetS score (R² = 3.1%) and higher high-density lipoprotein cholesterol (R² = 2.3%). PA of all intensities was associated with lower systolic- and diastolic blood pressure, although the associations were weaker than for the other risk factors (R² = 1.5% and R² = 1.4%). However, the multivariate analysis implies that the entire PA pattern must be considered. The main difference in PA was observed between normal weight and overweight children.

Conclusions

The present study suggests a greater importance of more PA corresponding to an intensity of at least brisk walking with inclusion of high-intense exercise, rather than a limited time spent sedentary, in the association to metabolic health in children. The methods of data processing and statistical analysis enabled accurate analysis and interpretation of the health benefits of high intensity PA that have not been shown previously.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12966-021-01156-1.

GRANADA consensus on analytical approaches to assess associations with accelerometer-determined physical behaviours (physical activity, sedentary behaviour and sleep) in epidemiological studies

The inter-relationship between physical activity, sedentary behaviour and sleep (collectively defined as physical behaviours) is of interest to researchers from different fields. Each of these physical behaviours has been investigated in epidemiological studies, yet their codependency and interactions need to be further explored and accounted for in data analysis. Modern accelerometers capture continuous movement through the day, which presents the challenge of how to best use the richness of these data. In recent years, analytical approaches first applied in other scientific fields have been applied to physical behaviour epidemiology (eg, isotemporal substitution models, compositional data analysis, multivariate pattern analysis, functional data analysis and machine learning). A comprehensive description, discussion, and consensus on the strengths and limitations of these analytical approaches will help researchers decide which approach to use in different situations. In this context, a scientific workshop and meeting were held in Granada to discuss: (1) analytical approaches currently used in the scientific literature on physical behaviour, highlighting strengths and limitations, providing practical recommendations on their use and including a decision tree for assisting researchers’ decision-making; and (2) current gaps and future research directions around the analysis and use of accelerometer data. Advances in analytical approaches to accelerometer-determined physical behaviours in epidemiological studies are expected to influence the interpretation of current and future evidence, and ultimately impact on future physical behaviour guidelines.

The multivariate physical activity signature associated with body mass index in young children

The evidence regarding associations between intensity-specific physical activity and adiposity in young children is conflicting. Moreover, the evidence is limited by analytical approaches that cannot handle the multicollinearity among multiple variables across the entire intensity spectrum. We aimed to determine the multivariate physical activity intensity signature associated with body mass index in a large sample of preschool children aged 3-6 years. 1182 Norwegian preschool children (mean age 4.7 years, 51% boys) provided data on physical activity (ActiGraph GT3X+) and body mass index during 2015-2016. Multivariate pattern analysis was used to determine associations between the entire triaxial intensity spectra (time spent in intensities from 0-99 to ≥ 15000 counts per minute) and body mass index in the total sample and in subgroups split by sex and age (median split). The association patterns were comparable across the three axes. For the vertical axis, associations were negative for time spent sedentary (0-99 counts per minute), positive for time spent in lower intensities (100-2999 counts per minute), and negative for time spent in vigorous intensities (4000-12,999 counts per minute). Associations were stronger in older than in younger children and no associations were observed for vigorous intensities among younger children. Association patterns were comparable for boys and girls. In conclusion, we found clear associations with body mass index across the physical activity intensity spectrum in preschool children. However, the age-specific association patterns suggest negative (unfavorable) associations with vigorous physical activity intensities develop around 5-6 years of age.

Time-Based Data in Occupational Studies: The Whys, the Hows, and Some Remaining Challenges in Compositional Data Analysis (CoDA)

Data on the use of time in different exposures, behaviors, and work tasks are common in occupational research. Such data are most often expressed in hours, minutes, or percentage of work time. Thus, they are constrained or ‘compositional’, in that they add up to a finite sum (e.g. 8 h of work or 100% work time). Due to their properties, compositional data need to be processed and analyzed using specifically adapted methods. Compositional data analysis (CoDA) has become a particularly established framework to handle such data in various scientific fields such as nutritional epidemiology, geology, and chemistry, but has only recently gained attention in public and occupational health sciences. In this paper, we introduce the reader to CoDA by explaining why CoDA should be used when dealing with compositional time-use data, showing how to perform CoDA, including a worked example, and pointing at some remaining challenges in CoDA. The paper concludes by emphasizing that CoDA in occupational research is still in its infancy, and stresses the need for further development and experience in the use of CoDA for time-based occupational exposures. We hope that the paper will encourage researchers to adopt and apply CoDA in studies of work exposures and health.

The multivariate physical activity signature associated with metabolic health in children and youth: An International Children's Accelerometry Database (ICAD) analysis

There is solid evidence for an association between physical activity and metabolic health outcomes in children and youth, but for methodological reasons most studies describe the intensity spectrum using only a few summary measures. We aimed to determine the multivariate physical activity intensity signature associated with metabolic health in a large and diverse sample of children and youth, by investigating the association pattern for the entire physical intensity spectrum. We used pooled data from 11 studies and 11,853 participants aged 5.8-18.4 years included in the International Children's Accelerometry Database. We derived 14 accelerometry-derived (ActiGraph) physical activity variables covering the intensity spectrum (from 0-99 to ≥8000 counts per minute). To handle the multicollinearity among these variables, we used multivariate pattern analysis to establish the associations with indices of metabolic health (abdominal fatness, insulin sensitivity, lipid metabolism, blood pressure). A composite metabolic health score was used as the main outcome variable. Associations with the composite metabolic health score were weak for sedentary time and light physical activity, but gradually strengthened with increasing time spent in moderate and vigorous intensities (up to 4000-5000 counts per minute). Association patterns were fairly consistent across sex and age groups, but varied across different metabolic health outcomes. This novel analytic approach suggests that vigorous intensity, rather than less intense activities or sedentary behavior, are related to metabolic health in children and youth.