Features of the Neighbourhood Built Environment and Their Ability to Predict Fitness in Youth: A Random Forest Approach
Bibliographic record
Abstract
Background: Cardiorespiratory Fitness (CRF) is an important marker of health among youth, strongly linked to improved cardiometabolic outcomes and reduced risk of early-onset cardiovascular disease. Factors at the individual, behavioural, and environmental levels are all implicated in shaping CRF. Notably, favourable Neighbourhood Built Environment (NBE) features are generally recognized as supportive of physical activity and active living; however, the nature of specific features, and the magnitude of their potential contribution to CRF in youth, require further investigation.Objectives: The primary objective of this thesis was to estimate the extent to which NBE features predict CRF in youth. A supplementary analysis examined gender-specific differences in the associations between NBE features and predicted CRF using stratified analyses.Methods: This study used data from the Quebec Adipose and Lifestyle InvesTigation in Youth (QUALITY) cohort. The QUALITY cohort comprised 630 families, including a child aged 8–10 years at baseline, and both biological parents, with at least one parent living with obesity. Data included those collecting during the initial (baseline) visit, and analyses were restricted to participants in the Greater Montreal Area (n = 504). CRF was assessed using peak oxygen volume consumption (VO2peak) during a cycling test on an electromagnetic bike, adjusted for Fat-Free Mass (FFM) as measured by DXA, and expressed as VO2peak/FFM (mL·min⁻¹·kg⁻¹ FFM). Moderate to vigorous physical activity was measured using a uniaxial accelerometer worn for 7 days, with valid wear time defined as a minimum of 10 hours per day on at least 3 weekdays and 1 weekend day. Salient NBE features were captured through a Geographic Information System (GIS) and on-site audits. Random forest models were applied to examine the predictive capability of NBE features in relation to VO2peak/FFM as a proxy for CRF. Bivariate associations between6predictors and predicted CRF were visualized using scatter plots and violin plots, for the full dataset, and stratified by gender.Results: Among 504 children (mean age 9.6 ± 0.9 years; 54% boys) boys exhibited higher mean VO2peak/FFM (71.2 min⁻¹·kg FFM⁻¹, SD = 13.8) compared to girls (63.4 min⁻¹·kg FFM⁻¹, SD = 11.4). The random forest model explained 33% of the variance in CRF, with BMI z-score, sex, and moderate to vigorous physical activity emerging as the top three predictors, followed by several NBE features, including the density of streets with normal traffic, number of intersections, vegetation index (NDVI), land use mix, building density, and signs of social disorder. Sex-specific analyses point to potential gendered-patterns: a modest positive association was observed between NDVI and predicted CRF, while nonlinear trends between number of intersections, land use mix, and predicted CRF among boys; in contrast, associations were largely negligible among girls.Conclusion: This study identified several NBE features with a moderate to weak contribution in predicting CRF, in addition to established sociodemographic factors and levels of physical activity. While the contributions of NBE features were relatively modest, they suggest potential for targets for neighbourhood transformations that may foster healthier lifestyles and improve CRF in youth populations. These findings could be useful to inform urban planning and public health research and policies, and should be extended to address the specific needs of diverse populations more broadly
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.021 | 0.016 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.005 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".