How Screen Time, Diet, Activity and Sleep Shape Physical Fitness in 60,000 Chinese Children

Table of Contents

1. Key Highlights
2. Introduction
3. Study design, sample and what was measured
4. The relationship between an aggregated lifestyle score and fitness: a dose–response
5. Which habit matters most: screen time’s outsized association with fitness
6. Diet, physical activity and sleep: meaningful but distinct contributions
7. Different fitness domains respond differently: nuance in the tests
8. Regional patterns: where lifestyle and fitness diverge across China
9. Practical implications for parents, schools and policymakers
10. Limitations, caveats and what the results do not prove
11. Where research should go next
12. Practical checklist: translating evidence into everyday practice
13. FAQ

Key Highlights

• A nationwide analysis of 60,177 schoolchildren across 27 Chinese provinces found a clear, graded relationship between a composite Healthy Lifestyle Score (HLS) and a multidimensional Physical Fitness Index (PFI); each one-point rise in HLS corresponded to a 0.43-point increase in PFI (95% CI 0.39–0.47).
• Among the four behaviors examined—limited screen time, healthy eating, regular physical activity and adequate sleep—restricted screen exposure showed the largest association with fitness (β ≈ 1.34), followed by healthy eating (β ≈ 0.49), physical activity (β ≈ 0.40) and sleep duration (β ≈ 0.12).
• Fitness improved steadily as children accumulated more healthy behaviors. Only a small fraction met strict screen-time and dietary criteria, signaling opportunities for public-health and school-based interventions.

Introduction

Physical fitness in childhood predicts far more than athletic ability. Cardiorespiratory fitness, muscular strength, flexibility, speed and coordination each track with brain development, academic performance and long-term cardiometabolic health. Despite that, public-health surveillance often treats diet, activity, sleep and screen time in isolation. A nationwide Chinese study conducted in 2015–2016 offers a rare, high-resolution view of how these four behaviors act together to shape a broad spectrum of fitness outcomes across childhood and adolescence.

Researchers assembled a large, geographically diverse sample—60,177 children aged 7–18—assessed seven standardized fitness tests and created a simple Healthy Lifestyle Score (HLS) that counts adherence to four evidence-based behaviors. The results reveal a robust dose–response: more healthy behaviors, better fitness. They also single out screen time as the most powerful individual correlate of fitness, a finding that has practical implications for families, schools and policy makers seeking to improve child health at scale.

The following analysis details the study design, synthesizes the key findings, explores how each behavior contributes to distinct fitness domains, and describes implications for interventions and policy.

Study design, sample and what was measured

The study pooled data collected by provincial teams across 27 provinces in China between May 2015 and May 2016. Investigators used a stratified random cluster sampling approach to capture broad regional and socioeconomic variation. From an initial sample of 77,022 children and adolescents, the analytic dataset retained 60,177 participants after excluding records with missing lifestyle data, implausible anthropometry and extreme fitness values.

Healthy Lifestyle Score (HLS)

• HLS summed four binary indicators (0 or 1) for: healthy eating, adequate sleep, limited screen time, and regular physical activity. Scores ranged 0–4.
• Definitions:
  • Healthy eating: consumption of three or more food groups daily and sugar-sweetened beverages limited to ≤1 per week.
  • Sleep: ≥8 hours per night (lower bound of recommended sleep for predominant ages).
  • Screen time: daily screen exposure within age-appropriate limits (the study used strict cut-offs—generally <1 h/day for minors, sensitivity analyses tested <2 h/day).
  • Physical activity: organized activity at least three times weekly, each session >1 hour (parent-reported).

Physical fitness battery and index

• Seven standardized tests were administered by trained teams: grip strength, standing long jump, 30-second sit-ups, sit-and-reach, 50 m dash, 20-second repeated straddling (agility) and 20 m shuttle run test (SRT, cardiorespiratory fitness).
• To compare across age and sex, each raw test score was converted to a Z-score. The Physical Fitness Index (PFI) was calculated as the sum of six Z-scores minus the Z-score for the 50 m dash (since lower dash times indicate better performance).
• This approach captures multidimensional fitness rather than a single endpoint.

Statistical approach and adjustments

• Multivariable linear regression models assessed associations between HLS (and each component individually) with PFI and with each test’s Z-score.
• Models adjusted for age, sex, province, parental education and occupation, household income and BMI. False discovery rate correction addressed multiple comparisons.
• Restricted cubic spline models examined nonlinear associations; sensitivity analyses included alternative scoring, multiple imputation for missing data, analyses excluding obese participants, and stratification by child versus adolescent groups.

This large, carefully standardized dataset offers a powerful lens on how combined lifestyle behaviors relate to objective measures of fitness across development.

The relationship between an aggregated lifestyle score and fitness: a dose–response

Across the full sample, the HLS exhibited a clear, graded relationship with overall fitness. Each additional healthy behavior counted in the HLS associated with better fitness: a one-point higher HLS linked to a 0.43-point increase in PFI (95% CI 0.39–0.47). The trend was monotonic—children with more qualifying behaviors performed consistently better on most individual fitness assessments.

Distribution of health behaviors

• Only 1.24% of the sample met all four healthy-behavior criteria.
• Most children met one or fewer criteria: 22.07% scored 0 on the HLS; 42.97% scored 1; 26.83% scored 2; 6.88% scored 3.
• Prevalence of individual behaviors: 53.99% met the physical activity threshold; 24.94% reached the ≥8 h sleep target; 11.44% met the dietary standard; just 9.45% met the strict screen time criterion.

Dose–response at the domain level

• Children with 3–4 healthy behaviors showed larger improvements in PFI, grip strength, standing long jump, sit-ups and 20 m SRT compared to those with only 1–2 behaviors.
• Trend tests across numbers of healthy behaviors were highly significant for most outcomes (P for trend < 0.001).

Nonlinearity and inflection points Restricted cubic spline analysis revealed nonlinearity for some outcomes. PFI and 20 m SRT displayed an inflection point at an HLS of 1—fitness gains accelerated once a child reached at least one healthy behavior. The 20-second repeated straddling (agility) test showed a J-shaped relationship with an inflection at HLS = 1, suggesting complexity in how combinations of behaviors translate into certain motor skills.

These results indicate that improvements in lifestyle behaviors accumulate and interact, producing synergistic benefits for broad measures of physical fitness.

Which habit matters most: screen time’s outsized association with fitness

Among the four behaviors, limited screen time demonstrated the largest association with overall fitness. When examined individually, restricted screen exposure correlated most strongly with higher PFI and with better performance across nearly every fitness domain.

Magnitude of association

• The study reported an estimated β ≈ 1.34 for limited screen time versus higher screen time, a magnitude substantially larger than the coefficients for individual diet, activity or sleep indicators.
• Children who met the strict screen-time criterion had better grip strength, standing long jump, sit-ups, 20 m SRT, agility and flexibility scores compared to peers with greater screen exposure.

Prevalence and public-health concern

• Only 9.45% of participants met the strict screen-time benchmark used in the primary analysis. Sensitivity analyses using a less stringent cutoff (<2 hours per day) lifted the prevalence to roughly 23.85%, closer to other national estimates for Chinese youth.
• Declining adherence to screen-time recommendations has been documented in national surveys, intensifying concern that rising device use could undermine fitness trends.

Why screen time shows the largest effect Several mechanisms help explain the strong relationship:

• Displacement: time spent in front of screens often displaces moderate-to-vigorous physical activity and unstructured play that develop motor skills and cardiorespiratory fitness.
• Sedentary physiology: prolonged sedentary behavior associates with metabolic changes—reduced insulin sensitivity and altered lipid metabolism—that compromise muscular and cardiorespiratory responses.
• Behavioral clustering: high screen use often clusters with other unhealthy behaviors (poor diet, irregular sleep), amplifying negative effects.
• Attention and routine disruption: prolonged screen exposure, especially at night, can fragment sleep and reduce recovery time needed for fitness gains.

Real-world implications A family that enforces a 1–2 hour total recreational screen cap typically creates time and structure for outdoor play, supervised sports, family walks and sleep routines. School-level interventions that limit non-educational device use, plus policies to promote active recess and after-school programs, can reduce screen time while increasing opportunities for movement.

Diet, physical activity and sleep: meaningful but distinct contributions

Screen time drew the largest single association, but diet, physical activity and sleep each contributed to fitness in measurable ways. The study quantified these relationships and showed distinct patterns across fitness domains.

Healthy eating

• Healthy-eating adherence was low (11.44%) by the study’s definition, which required diversity across ≥3 food groups daily and sugar-sweetened beverage consumption ≤1 per week.
• Diet correlated with strength, power, muscular endurance and cardiorespiratory markers—associations consistent with research linking nutrient adequacy and protein quality to muscle development and recovery.
• Effect size for diet on PFI was substantial (β ≈ 0.49 in the abstract’s summary), second only to screen time.

Interpretation and application Diet supports the biological substrate on which training acts. For children, improving dietary diversity and cutting sugar-sweetened beverages is a high-yield, low-cost target. School meal reforms, vending machine policies, and education for caregivers about balanced meals can produce measurable fitness benefits.

Regular physical activity

• Approximately 54% of the sample met the study’s minimum activity criterion (≥3 weekly organized sessions >1 hour). That leaves nearly half of the children below the benchmark.
• The physical-activity indicator correlated with improvements in grip strength, sit-ups, long jump, sit-and-reach, 20 m SRT and speed—consistent with effects of structured exercise on muscular strength, power and cardiorespiratory fitness.
• Effect size for activity on PFI was notable (β ≈ 0.40).

Programmatic strategies

• School-based physical activity interventions have a track record of improving fitness outcomes. Examples include mandatory daily physical-education classes, active commuting campaigns, and after-school sport clubs.
• For younger children, unstructured active play and parental modeling are particularly effective.

Sleep duration

• Sleep showed smaller, yet still significant, associations with several fitness outcomes (β ≈ 0.12 for the sleep indicator in the abstract). Adequate sleep related positively to muscular endurance (sit-ups), flexibility (sit-and-reach) and cardiorespiratory fitness (20 m SRT).
• About 25% of participants met the ≥8-hour threshold.

Why sleep matters Sleep facilitates recovery, hormonal regulation (including growth hormone and testosterone peaks important for muscle development), and neurocognitive functions that support motor learning. Short or fragmented sleep impairs motivation and reduces time available for activity, creating indirect pathways that suppress fitness gains.

Combined effects The study’s most actionable finding is the cumulative benefit: behaviors add together. Each additional healthy behavior produced incremental fitness gains, and children with 3–4 behaviors registered substantially better outcomes than peers with none or one behavior.

Different fitness domains respond differently: nuance in the tests

Not all fitness tests moved in lockstep. The study used a multidimensional battery so researchers could detect domain-specific patterns.

Linear versus nonlinear relationships

• Several tests (grip strength, long jump, sit-ups, sit-and-reach, 50 m dash) showed linear relationships with HLS—fitness improved steadily with each additional healthy behavior.
• Cardiorespiratory fitness (20 m SRT) and overall PFI exhibited nonlinear relationships, with the slope of improvement increasing after a modest HLS threshold (inflection at HLS = 1).

J-shaped associations and counterintuitive findings

• The 20-second repeated straddling test (agility) followed a J-shaped curve. A J-shape can reflect that very low and very high HLS scores link to better agility for different reasons (e.g., children playing many informal games but lacking structured habits, versus highly active children excelling across domains).
• Sit-and-reach (flexibility) and 50 m dash (speed) were reported to be better among children with 1–2 healthy behaviors than those with 3–4 in some subgroup comparisons. Potential explanations include:
  • Sport specialization: children engaged in activities that prioritize strength or endurance may gain less in flexibility.
  • Measurement distribution: small numbers in the high-HLS group (only 1.24% scored 4) can produce noisy estimates.
  • Trade-offs between training modalities: improving one domain (e.g., endurance) does not automatically translate to superior speed or flexibility.

Practical takeaways Fitness is multifaceted. An intervention focused solely on aerobic exercise will push 20 m shuttle run results but may leave strength or flexibility unchanged. Comprehensive programs that include resistance, speed/agility work and stretching maximize broad fitness benefits.

Regional patterns: where lifestyle and fitness diverge across China

Geographic analysis uncovered meaningful regional variation in both lifestyles and fitness.

Highlights

• Provinces with the highest average HLS included Anhui (1.42) and Sichuan (1.22).
• Beijing (PFI = 2.19) and Shanghai (PFI = 1.78) posted the highest mean PFIs.
• Inner Mongolia displayed the strongest statistical association between HLS (β = 1.07) and PFI, and particularly between healthy eating (β = 2.06) and PFI.

Interpreting regional differences

• Urban centers like Beijing and Shanghai often have better sports infrastructure, more organized sport opportunities and higher parental income—factors that support higher measured fitness.
• High HLS in some provinces does not always translate to top PFI, highlighting the influence of school resources, physical-environmental opportunities, cultural norms around sport and leisure, and measurement timing.
• Stronger HLS–PFI associations in provinces such as Inner Mongolia may reflect local environments where modest lifestyle changes produce outsized fitness gains, or where lifestyle measures capture behaviors that directly affect activity opportunities (e.g., diet patterns aligned with regional food systems).

Policy implication Regional heterogeneity argues for tailored interventions. National guidelines set the target; local implementation must adapt to culture, infrastructure and socioeconomic context.

Practical implications for parents, schools and policymakers

The study offers clear guidance that can be translated into interventions at multiple levels. Because behaviors cluster, coordinated strategies that target several habits produce larger fitness returns than single-focus programs.

School-level actions

• Make daily physical education non-negotiable and diversified: include cardiorespiratory sessions, resistance and fundamental-movement skill work, agility/speed drills and flexibility routines.
• Use recess and after-school programs to increase opportunities for unstructured and organized activity.
• Reform school meals to increase food-group diversity and reduce availability of sugar-sweetened beverages.
• Establish screen-free policies during school hours and educate students about responsible device use.

Family-level actions

• Set household screen limits and create screen-free zones and times (especially before bedtime).
• Prioritize regular family meals with diverse food groups and limit sugary drinks.
• Join children in physical activities—family walks, active chores, weekend sports—both to model behavior and to create structure.
• Protect sleep by establishing consistent bedtimes and reducing evening screen exposure.

Community and policy levers

• Invest in safe parks, playgrounds and active-travel infrastructure that make physical activity accessible.
• Promote public-health campaigns emphasizing combined lifestyle benefits for physical fitness and academic performance.
• Incentivize collaboration between public health, education and urban planning to scale evidence-based programs.

Intervention examples that translate findings into action

• A municipal pilot that enforces a 1-hour daily active-play period in primary schools, paired with a healthy-lunch program, can simultaneously address activity and diet.
• After-school clubs that mix aerobic play, skill drills and brief strength circuits provide comprehensive training in limited time.
• Parental workshops and smartphone apps that help families set screen-time limits, plan meals and build sleep routines make household change more achievable.

Limitations, caveats and what the results do not prove

The study’s strengths—large, representative sampling and objective fitness testing—are balanced by important limitations to keep in mind when interpreting the results.

Cross-sectional design

• The study captures associations at a single point in time; causality cannot be inferred. Children with better fitness may adopt healthier behaviors, or unmeasured factors may influence both fitness and lifestyle.

Self-reported lifestyle measures

• Diet, activity frequency and screen time relied on guardian or parent reports. While structured questionnaires reduce noise, measurement error and recall bias remain possible.
• The physical-activity measure emphasized organized exercise and may undercount unstructured play common in younger children.

Data timing

• Data were collected in 2015–2016. Patterns of screen use, device availability and social norms around activity have changed since then. Nonetheless, the fundamental relationships between behavior and fitness likely remain relevant.

Selection and exclusion criteria

• The study excluded participants with missing lifestyle data and extreme fitness or anthropometric values. While necessary for data integrity, this may introduce selection bias.

Residual confounding

• Models adjusted for multiple sociodemographic factors but unmeasured variables (e.g., neighborhood walkability, school sports funding, genetics) could influence results.

Small high-HLS group

• Only a tiny fraction of participants scored 4 on the HLS. Estimates for this subgroup are therefore less stable and warrant cautious interpretation.

These limitations shape how the evidence should inform practice: use the associations as strong indicators of which behaviors matter and prioritize multifaceted, evaluated interventions rather than assuming fixed magnitudes of effect or strict causality.

Where research should go next

Several research streams would strengthen and update these findings.

Longitudinal and interventional designs

• Cohorts that follow children across developmental stages can clarify directionality and the timing of greatest susceptibility to lifestyle change.
• Randomized or quasi-experimental trials that manipulate one or more behaviors (e.g., screen-time reduction plus after-school sports) would test causal effects on fitness and health.

Objective behavioral measurement

• Wearable devices (accelerometers, sleep trackers) and passive screen-use monitoring would reduce reporting bias and reveal dose–response relationships with finer granularity.

Contemporary data collection

• Repeating similar large-scale surveillance with current cohorts would assess how smartphone prevalence, social-media engagement and post-pandemic lifestyles influence associations.

Mechanistic work

• Studies linking diet, sleep and screen exposure with physiological mediators—metabolic markers, sleep architecture, neural development—would explain why combined behaviors affect multiple fitness domains.

Equity and implementation research

• Comparative effectiveness studies across regions and socioeconomic strata can inform which strategies are most scalable and equitable.
• Implementation science should test how to integrate behavior-change programs within school curricula and family routines with high fidelity.

Practical checklist: translating evidence into everyday practice

For caregivers and educators seeking to translate the study’s findings into action, the checklist below focuses on feasible, evidence-aligned steps.

For families

• Limit recreational screen time to under 2 hours per day for school-age children; tighter limits (<1 hour) yield larger benefits where feasible.
• Offer at least three different food groups at each main meal and reduce sugar-sweetened beverages to special occasions.
• Facilitate at least 60 minutes of daily movement combining active play, sport or structured activity; encourage organized sessions when possible.
• Maintain consistent sleep schedules tailored to age, targeting eight or more hours per night for most school-aged children.
• Model behavior: parents who reduce screen use, eat diversified meals and prioritize activity set a powerful example.

For schools

• Schedule daily physical education and create active recess that emphasizes movement skills and fun.
• Review school meal standards and vending policies to limit sugary drinks.
• Provide after-school and intramural options across a variety of activities so children of differing interests can participate.
• Integrate sleep and screen hygiene education into health classes and communicate simple home strategies to families.

For policymakers and community planners

• Fund safe outdoor spaces and active-transport infrastructure.
• Support public-health campaigns that pair screen-reduction messages with practical alternatives (structured play, family activities).
• Incentivize schools to adopt evidence-based physical-activity standards and meal improvements.

FAQ

Q: What exactly is the Healthy Lifestyle Score (HLS)? A: HLS is a simple index that counts adherence to four healthy behaviors: eating from at least three food groups daily while limiting sugary drinks, sleeping ≥8 hours per night, limiting recreational screen time within age-appropriate bounds, and participating in organized physical activity at least three times weekly for one hour or more. Scores range from 0 (none) to 4 (all).

Q: How strong is the link between HLS and physical fitness? A: Each additional healthy behavior in HLS associated with an average 0.43-point increase in the study’s Physical Fitness Index. The relationship was dose-dependent—more healthy behaviors yielded progressively better fitness across most domains.

Q: Does screen time really matter more than exercise? A: In this dataset, restricted screen time showed the largest individual association with PFI and with many fitness tests. That does not imply screen time is more important than exercise in all contexts, but it highlights screen reduction as a high-impact, underutilized lever that often co-occurs with increases in activity and improvements in sleep and diet.

Q: Can changing one behavior make a big difference? A: The spline analyses showed fitness gains accelerate once children adopt at least one healthy behavior, suggesting even single, achievable changes—cutting screen time, improving sleep consistency or enhancing diet—can produce measurable benefits and open the door to further improvements.

Q: Are these findings causal? A: The study is cross-sectional; it shows strong associations but cannot prove causation. Longitudinal and intervention research is needed to confirm directionality. Still, the associations align with experimental evidence that physical activity, diet quality, sleep and reduced sedentary time each support fitness and health.

Q: Do the results apply outside China or to today’s device-heavy environment? A: The biological and behavioral relationships between diet, activity, sleep and fitness generalize broadly. Regional and temporal contexts vary—the data were collected in 2015–2016—so absolute prevalence estimates or effect magnitudes may differ elsewhere or now. The central message—that combined healthy behaviors produce greater fitness—remains relevant globally.

Q: How should schools prioritize limited resources? A: Prioritize comprehensive approaches: daily physical education, active recess, healthy meals, and education on sleep and screen habits. Combining small, low-cost changes (recess quality, water access, limits on vending machines and classroom screen policies) yields greater returns than isolated, high-investment actions.

Q: What are the next practical steps for parents worried about screen time? A: Create a family media plan: define screen-free mealtimes, set maximum daily recreational screen limits, remove screens from bedrooms at night, and replace screen hours with active alternatives (walking, games, organized sport). Communicate these limits clearly and provide consistent routines.

Q: Is there an “optimal” combination of behaviors? A: The study shows additive benefits—the more healthy behaviors, the better the fitness profile. There is no single optimal pattern; instead, aim for a balanced mix: daily movement, diverse diet, limited screens and adequate sleep.

Q: Where can policymakers have the largest impact? A: Investing in school-based programs, public outdoor infrastructure, and community campaigns to reduce recreational screen time and improve diet quality offers scalable returns. Policies that make physical activity accessible and safe for all children address structural barriers to healthy behaviors.

Adopting a handful of focused, evidence-informed practices—reducing recreational screen time, improving dietary diversity, increasing regular physical activity and protecting sleep—translates into measurable, multidimensional fitness benefits for children. The data from over 60,000 Chinese schoolchildren underscore that these behaviors do not operate in isolation. They cluster, interact and add together. For families, schools and communities, the message is straightforward: multiple modest changes produce cumulative gains that matter for strength, speed, endurance, flexibility and cardiometabolic resilience as children grow.