Abstract
To investigate the relationship between Leisure time physical activity (LTPA) patterns and PhenoAgeAccel in patients with Type 2 diabetes (T2D), emphasizing the role of regular LTPA in mitigating biological aging. This study utilized data from the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2018, including 4,134 adults with T2D. Multivariable linear regression models and restricted cubic spline (RCS) methods were employed to assess the relationship between LTPA and Phenotypic age acceleration (PhenoAgeAccel), with segmented likelihood ratio tests to detect nonlinear thresholds. Stratified regression and interaction tests were conducted for robust analysis. Compared to individuals with no LTPA patterns, those with regular LTPA patterns had significantly lower PhenoAgeAccel scores (β = -1.164, 95% CI: -1.651 to -0.677, P < 0.0001), while the “Weekend Warrior” and “Inactive-LTPA” patterns showed no significant effects. A nonlinear threshold effect was identified; below 594.57 min of weekly LTPA, there was a significant negative correlation (β = -0.002, 95% CI: -0.003 to -0.001, P = 0.000), with gender-specific effects present. Regular LTPA significantly reduces phenotypic age acceleration in T2D patients, with a nonlinear threshold effect indicating that moderate physical activity is most beneficial. These findings highlight the necessity of personalized physical activity recommendations and provide evidence for public health strategies to promote healthy aging in T2D patients.
Keywords: Leisure-time physical activity, Phenotypic age acceleration, Type 2 diabetes, Biological aging, Public health strategies
Subject terms: Ageing, Endocrine system and metabolic diseases, Disease prevention, Geriatrics, Prognosis, Public health, Quality of life
Introduction
T2D has been widely studied and identified as a major factor accelerating biological aging. Patients with T2D exhibit shortened telomeres, mitochondrial dysfunction, increased inflammation, and cellular senescence, all of which contribute to the acceleration of the aging process1. T2D not only accelerates telomere shortening but also leads to a decline in pancreatic β-cell function and increased insulin resistance, exacerbating the metabolic condition of patients2. Additionally, T2D patients experience a more rapid decline in muscle mass and function compared to non-diabetic individuals, further aggravating biological aging3.
Relevant studies have explored the multifaceted mechanisms of the aging process, particularly the damage to macromolecules within cells and the regulation of the aging clock4. The research indicates that as individuals age, key biomolecules such as DNA and proteins accumulate damage, leading to disruptions in cellular homeostasis and declining functionality, thereby accelerating biological aging. Moreover, environmental factors and lifestyle choices further exacerbate this damage, hastening the aging process4,5. This study provides a critical framework for understanding biological aging and underscores the potential for interventions aimed at delaying the aging process, especially through the regulation of epigenetic mechanisms. Physical activity plays a crucial role in modulating the aging process. Research indicates that regular physical activity can delay aging by maintaining telomere length, reducing oxidative stress, and improving cellular function. Individuals who engage in regular physical activity have longer telomeres compared to their sedentary counterparts, potentially delaying cellular aging6. Furthermore, physical activity is associated with the activation of telomerase, an enzyme that adds nucleotide sequences to telomeres, enhancing telomere stability and cellular longevity7. Physical activity also mitigates aging through the reduction of oxidative stress. Oxidative stress, caused by excessive reactive oxygen species (ROS), plays a significant role in aging and many age-related diseases. Regular exercise has been shown to enhance antioxidant defenses, thereby reducing oxidative damage to cells and tissues8. Despite a transient increase in ROS during intense exercise, the overall effect of regular physical activity is a reduction in oxidative stress and improvement in cellular function9. Moreover, physical activity contributes to improved muscle mass and function, which is vital for maintaining mobility and reducing the risk of frailty in older adults. Longitudinal studies demonstrate that individuals who engage in regular physical activity maintain better muscle mass and metabolic health, mitigating the adverse effects of aging on the musculoskeletal system10. Physical activity also positively impacts cognitive function and brain health by enhancing neurogenesis, reducing neuroinflammation, and improving cerebral blood flow11.
This study aims to explore the impact of different LTPA patterns on biological aging in T2D patients. By conducting a cross-sectional analysis of NHANES data, this study seeks to elucidate the relationship between various LTPA patterns and biological aging in T2D patients and to investigate the underlying mechanisms. We hypothesize that regular LTPA is associated with lower PhenoAgeAccel, thereby providing scientific evidence for targeted public health strategies to mitigate accelerated aging in T2D patients.
Methods
Study population
This study utilized data from the NHANES collected between 1999 and 2018. Overseen by the Centers for Disease Control and Prevention (CDC), NHANES aims to evaluate the health and nutritional status of the non-institutionalized civilian population in the United States. Conducted by the CDC’s National Center for Health Statistics, the survey employs a stratified multistage probability sampling method to assess approximately 10,000 Americans every two years. Data collection involves detailed household interviews and comprehensive physical examinations. The interviews gather extensive demographic, socioeconomic, dietary, and health-related information, while the physical examinations include medical, dental, and a wide array of physiological and biochemical markers.
For this study, a rigorous selection process was implemented, ultimately including 4,134 adults for analysis. This process primarily excluded individuals who did not meet the age criteria or were unable to complete the required survey and physical examinations. Data analysis was conducted using a complete case analysis strategy. The intricate details of the selection process are depicted in Fig. 1. The NHANES datautilized in this study received appropriate ethical approval and adhere to the ethical standards outlined in the Declaration of Helsinki.
Figure 1.
Flow chart.
Definitions of type 2 diabetes diagnosis
A diagnosis of T2D was confirmed if any of the following criteria were satisfied12–14: (1) self-reported diagnosis by a physician; (2) current administration of glucose-lowering medications or insulin therapy; (3) a random blood glucose concentration of at least 11.1 mmol/L; (4) a glycosylated hemoglobin (HbA1c) level of 6.5% or higher; (5) a fasting plasma glucose (FPG) level of at least 7.0 mmol/L; or (6) a 2-hour plasma glucose level of at least 11.1 mmol/L following an oral glucose tolerance test (OGTT).
Definitions of leisure time physical activity
In this study, LTPA was evaluated using the Global Physical Activity Questionnaire (GPAQ). Participants were queried about the frequency (sessions per week) and duration (minutes per session) of their engagement in vigorous and moderate-intensity sports, fitness, and recreational activities lasting at least 10 consecutive minutes per week. Vigorous-intensity activities were defined as those with a metabolic equivalent of task (MET) value of 8.0, leading to significant increases in breathing or heart rate, while moderate-intensity activities had a MET value of 4.0, resulting in moderate increases in breathing or heart rate. The total duration of moderate-to-vigorous physical activity (MVPA) was calculated using the formula: LTPA-MVPA = 2 × Vigorous PA + Moderate PA15,16. This formula reflects physical activity guidelines equating 1 min of vigorous activity to 2 min of moderate activity. The total frequency of MVPA was determined by summing the sessions of both moderate and vigorous activities. Based on the weekly total MVPA duration (threshold of 150 min) and total frequency (threshold of 2 sessions per week), physical activity patterns were categorized into four groups: inactive (no vigorous or moderate PA), insufficiently active (less than 150 min of MVPA per week), weekend warrior (at least 150 min of MVPA per week in 1 or 2 sessions), and regularly active (at least 150 min of MVPA per week spread over 3 or more sessions)17–20.
Definitions of phenotypic age acceleration
This study builds upon previous research on phenotypic age by using an integrative method to determine an individual’s phenotypic age. The algorithm was developed by Dr. Morgan E. Levine and her team at Yale University, based on clinical laboratory blood chemistry parameters from the NHANES III dataset21,22. These parameters include biomarkers such as albumin, creatinine, glucose, total white blood cell count, percentage of lymphocytes, red cell distribution width, mean corpuscular volume, and alkaline phosphatase. These biomarkers were modeled using elastic-net regression to predict an individual’s phenotypic age. Of particular interest is the PhenoAgeAccel measure, which is derived from the residuals of a linear regression between phenotypic age and chronological age. This measure reflects the difference between an individual’s physiological state and their actual chronological age. For instance, two individuals with the same chronological age but different health statuses—one appearing younger due to better health and the other older due to health issues or poor lifestyle choices—will have different PhenoAgeAccel values. Lower PhenoAgeAccel values indicate slower biological aging, suggesting that an individual’s physiological state is better than their chronological age. Previous studies have shown that PhenoAgeAccel is closely associated not only with all-cause mortality but also with a wide range of health outcomes, including cardiovascular disease and cancer23,24. Specifically, PhenoAgeAccel has been demonstrated to accurately predict an individual’s long-term health risks, with higher values associated with a greater risk of disease incidence and faster biological aging. Through large-scale data analysis across diverse populations, Levine and colleagues found that the algorithm demonstrates robust predictive performance across different races, genders, and socioeconomic backgrounds, further enhancing its applicability in various research contexts22,25.
In this study, due to the lack of C-reactive protein (CRP) data from 2011 to 2018, we followed previous research precedents by excluding CRP from our biomarker set. Prior studies have shown that phenotypic age calculations with and without CRP are highly correlated (correlation coefficient of 0.99)16,26,27, indicating that the omission of CRP has minimal impact on phenotypic age calculation.
The specific formula is as follows:
 |
 |
Covariate
This study included various covariates for a comprehensive analysis, covering demographic details, lifestyle factors, and health status indicators. Demographic information comprised age, sex (male, female), ethnic background (categorized as Mexican American, Non-Hispanic Black, Non-Hispanic White, and others), and education levels (below high school, high school, and above high school). Economic status was measured using the Poverty Income Ratio (PIR), which compares household or individual income against annual poverty thresholds, categorized as low income for PIR ≤ 1.3, moderate income for PIR > 1.3 but < 3.5, and high income for PIR ≥ 3.5. Lifestyle factors included marital status (Married/living with partner, Never married, Widowed/Divorced/Separated), Body Mass Index (BMI) categories (< 25, 25-29.9, ≥ 30 kg/m²), smoking status (never, former, current), and alcohol consumption (never, former, mild, moderate, heavy). Sedentary time was measured using the GPAQ, which assessed the amount of time participants spent sitting at various locations and during different activities, excluding sleep. Participants reported their typical daily sitting time, which was quantified in hours and categorized into four groups: <4 h, 4-<6 h, 6–8 h, and > 8 h. Dietary quality was evaluated using the Healthy Eating Index 2015 (HEI-2015), which measures how well an individual’s or group’s diet al.igns with the Dietary Guidelines for Americans. This index includes components representing key dietary aspects such as intake of fruits, vegetables, whole grains, dairy, proteins, the ratio of unsaturated to saturated fats, sodium, and added sugars. The HEI-2015 is scored out of 100, with higher scores indicating better diet quality and adherence to dietary guidelines28,29. Health status indicators included criteria for diagnosing hypertension, hyperlipidemia, and cancer. Hypertension was determined by a physician’s diagnosis, use of antihypertensive medications, and blood pressure readings (systolic ≥ 140 mmHg or diastolic ≥ 90 mmHg). The diagnosis of hyperlipidemia was based on lipid levels (triglycerides, total cholesterol, LDL, and HDL) and the use of lipid-lowering medications. Cancer prevalence was ascertained by diagnosis from a physician or health professional.
Statistical analysis
In this study, the detailed sampling methodology prescribed by NHANES was strictly followed, with complex sampling weights calculated to meet NHANES analytical standards. To accurately reflect the U.S. population, weighted data served as the foundation for all statistical assessments. Continuous variables were presented as means with their respective standard errors (Mean ± SE), while categorical data were described using frequency counts and their corresponding weighted proportions.
To identify group differences, one-way analysis of variance (ANOVA) was employed for continuous variables, and the chi-square test was used for categorical data. Additionally, weighted generalized linear regression models were utilized to investigate the relationship between LTPA and PhenoAgeAccel in individuals with T2D, with further subgroup analyses by gender to examine potential gender-specific effects. A generalized weighted regression prediction model was applied to analyze the continuous values of LTPA and LTPA patterns in T2D patients, exploring their impact on PhenoAgeAccel.
Furthermore, based on the linear regression model, restricted cubic spline plots were used to test for nonlinear trends between variables. For associations exhibiting nonlinear trends, segmented regression combined with likelihood ratio tests was employed to elucidate the threshold effects of LTPA on PhenoAgeAccel in T2D patients. Lastly, to validate the impact of control variables on the relationship between LTPA and PhenoAgeAccel, these variables were incorporated into stratified regression and interaction effect testing models for robust analysis.
All statistical analyses were performed using R Studio (version 4.2.1, USA), with a two-tailed P-value of less than 0.05 considered statistically significant.
Results
Baseline characteristics of the participants
This study included baseline characteristics of 4,134 adults, categorized by their LTPA patterns (Table 1). Compared to other groups, participants lacking physical activity exhibited significantly higher PhenoAgeAccel scores. Significant differences were observed among different LTPA groups in demographic indicators (age, sex, BMI), lifestyle factors (education level, PIR, smoking status, alcohol consumption, HEI scores), and health status indicators (hypertension) (P < 0.05).
Table 1.
Baseline characteristics of the participants.
| Characteristic | Overall | No-LTPA | Inactive-LTPA | Regular-LTPA | Weekend Warrior | P-value |
|---|---|---|---|---|---|---|
| N | 4134 | 2597 | 564 | 858 | 115 | |
| PhenoAgeAccel | -0.90(0.14) | -0.30(0.16) | -1.17(0.38) | -2.22(0.21) | -0.44(0.41) | < 0.0001 |
| Sex, n (weighted %) | < 0.0001 | |||||
| Female | 1929(47.42) | 1293(51.09) | 276(50.43) | 337(41.32) | 23(12.03) | |
| Male | 2205(52.58) | 1304(48.91) | 288(49.57) | 521(58.68) | 92(87.97) | |
| Age, years | < 0.0001 | |||||
| 20–39 | 503(14.51) | 239(11.14) | 75(15.55) | 160(19.74) | 29(30.15) | |
| 40–59 | 1381(41.05) | 835(40.78) | 201(40.48) | 303(42.65) | 42(36.60) | |
| ≥60 | 2250(44.44) | 1523(48.08) | 288(43.97) | 395(37.61) | 44(33.25) | |
| Race, n (weighted %) | 0.05 | |||||
| Non-Hispanic White | 1757(68.88) | 1114(67.90) | 245(71.17) | 356(70.15) | 42(65.06) | |
| Mexican American | 796( 9.08) | 534(10.26) | 79( 5.86) | 156( 7.70) | 27(14.71) | |
| Non-Hispanic Black | 777(10.10) | 486(10.67) | 126(10.41) | 146( 8.74) | 19( 8.43) | |
| Other race (including multi-racial and other Hispanic) | 804(11.95) | 463(11.17) | 114(12.57) | 200(13.41) | 27(11.80) | |
| BMI, kg/m2,n (weighted %) | < 0.001 | |||||
| <25 | 617(13.26) | 361(12.02) | 74(12.24) | 169(17.18) | 13(10.89) | |
| 25-29.9 | 1254(28.87) | 742(26.54) | 164(27.40) | 300(34.20) | 48(37.73) | |
| ≥30 | 2263(57.88) | 1494(61.44) | 326(60.36) | 389(48.62) | 54(51.38) | |
| Marital status, n (weighted %) | 0.1 | |||||
| Married/living with partner | 2602(67.80) | 1606(66.36) | 355(68.98) | 558(69.97) | 83(71.12) | |
| Never married | 381( 9.45) | 215( 8.68) | 48( 9.59) | 105(11.41) | 13( 7.74) | |
| Widowed/Divorced/Separated | 1151(22.75) | 776(24.96) | 161(21.43) | 195(18.62) | 19(21.14) | |
| Education, n (weighted %) | < 0.0001 | |||||
| Below | 1285(19.41) | 996(26.48) | 117(11.32) | 150( 9.23) | 22(10.81) | |
| High school | 975(25.85) | 635(27.47) | 130(26.31) | 175(19.71) | 35(41.01) | |
| Above | 1874(54.74) | 966(46.05) | 317(62.37) | 533(71.06) | 58(48.18) | |
| PIR, n (weighted %) | < 0.0001 | |||||
| <1.3 | 1297(19.85) | 956(25.32) | 136(13.96) | 177(11.22) | 28(17.16) | |
| 1.3–3.49 | 1662(36.48) | 1085(41.14) | 214(30.94) | 313(28.94) | 50(38.21) | |
| ≥3.5 | 1175(43.67) | 556(33.54) | 214(55.09) | 368(59.84) | 37(44.63) | |
| Smoke status, n (weighted %) | < 0.0001 | |||||
| Former smoker | 1351(34.32) | 829(31.88) | 189(37.15) | 295(37.26) | 38(41.48) | |
| Nonsmoker | 2073(49.18) | 1266(47.67) | 297(52.02) | 461(52.47) | 49(36.67) | |
| Current smoker | 710(16.50) | 502(20.45) | 78(10.83) | 102(10.27) | 28(21.86) | |
| Alcohol status, n (weighted %) | < 0.0001 | |||||
| Former | 866(17.31) | 634(21.24) | 102(13.89) | 112(11.46) | 18( 8.52) | |
| Never | 658(12.81) | 462(14.57) | 83(12.90) | 101( 9.37) | 12( 6.84) | |
| Mild | 1419(38.77) | 782(34.09) | 218(41.78) | 384(47.42) | 35(41.82) | |
| Moderate | 521(14.86) | 313(15.13) | 76(15.09) | 111(13.96) | 21(15.47) | |
| Heavy | 670(16.25) | 406(14.97) | 85(16.34) | 150(17.79) | 29(27.36) | |
| Healthy eating index, n (weighted %) | < 0.0001 | |||||
| Quantile 1,[12.71,44.47] | 1378(35.12) | 935(39.41) | 169(30.76) | 232(27.78) | 42(35.97) | |
| Quantile 2,(44.47,56.42] | 1378(33.20) | 899(34.53) | 195(35.96) | 243(27.37) | 41(39.40) | |
| Quantile 3,(56.42,95.89] | 1378(31.68) | 763(26.06) | 200(33.28) | 383(44.85) | 32(24.63) | |
| Sedentary behavior, n (hour/day, weighted %) | 0.49 | |||||
| <4 | 1199(23.10) | 751(22.78) | 151(24.24) | 256(22.09) | 41(30.61) | |
| 4-<6 | 1039(24.43) | 631(22.93) | 146(25.08) | 233(27.27) | 29(26.42) | |
| 6–8 | 1192(31.05) | 755(31.45) | 168(29.42) | 239(31.74) | 30(26.87) | |
| >8 | 704(21.43) | 460(22.84) | 99(21.27) | 130(18.89) | 15(16.10) | |
| Hypertension, n (weighted %) | < 0.0001 | |||||
| No | 1462(37.84) | 827(33.65) | 213(40.05) | 369(45.76) | 53(41.78) | |
| Yes | 2672(62.16) | 1770(66.35) | 351(59.95) | 489(54.24) | 62(58.22) | |
| Hyperlipidemia, n (weighted %) | 0.06 | |||||
| No | 629(15.27) | 375(13.90) | 82(15.55) | 148(16.72) | 24(27.72) | |
| Yes | 3505(84.73) | 2222(86.10) | 482(84.45) | 710(83.28) | 91(72.28) | |
| Cancer, n (weighted %) | 0.95 | |||||
| No | 3570(85.13) | 2219(85.03) | 492(85.49) | 756(84.82) | 103(87.53) | |
| Yes | 564(14.87) | 378(14.97) | 72(14.51) | 102(15.18) | 12(12.47) |
Continuous variables are summarized as mean values with standard errors, and categorical variables are presented as counts with their weighted percentages;
One-way ANOVA was utilized for continuous data, while the Chi-square test was employed for categorical data
Figure 2 presents the distribution of PhenoAgeAccel across different LTPA patterns. The combined results of Table 1; Fig. 2 indicate that the No-LTPA group has significantly higher PhenoAgeAccel values, suggesting accelerated biological aging. In contrast, the Regular-LTPA group shows significantly lower PhenoAgeAccel values, indicating that regular physical activity can effectively slow down biological aging. Although the Weekend Warrior group is not as effective as the Regular-LTPA group, it also demonstrates some deceleration in aging. This figure underscores the crucial role of regular physical activity in delaying the aging process.
Figure 2.
Distribution of PhenoAgeAccel by LTPA Patterns.
Association analysis between LTPA and PhenoAgeAccel in individuals with type 2 diabetes
This study utilized multivariate linear regression models to analyze the relationship between LTPA and PhenoAgeAccel in T2D patients. The results are presented in Figs. 3 and 4, and 5.
Figure 3.
Association analysis between LTPA and PhenoAgeAccel in individuals with Type 2 Diabetes. Crude Model is the unadjusted model. Model 1 adjusted for age, sex, race, PIR, education, BMI, marital status, smoke, alcohol, HEI, sedentary behavior. Model 2 adjusted for age, sex, race, PIR, education, BMI, marital status, smoke, alcohol, HEI, sedentary behavior, hypertension, hyperlipidemia and cancer.
Figure 4.
Association analysis between LTPA and PhenoAgeAccel in individuals with Type 2 Diabetes (Male). Crude Model is the unadjusted model. Model 1 adjusted for age, race, PIR, education, BMI, marital status, smoke, alcohol, HEI, sedentary behavior. Model 2 adjusted for age, race, PIR, education, BMI, marital status, smoke, alcohol, HEI, sedentary behavior, hypertension, hyperlipidemia and cancer
Figure 5.
Association analysis between LTPA and PhenoAgeAccel in individuals with Type 2 Diabetes (Female). Crude Model is the unadjusted model. Model 1 adjusted for age, race, PIR, education, BMI, marital status, smoke, alcohol, HEI, sedentary behavior. Model 2 adjusted for age, race, PIR, education, BMI, marital status, smoke, alcohol, HEI, sedentary behavior, hypertension, hyperlipidemia and cancer
Overall, the analysis revealed a significant inverse association between LTPA levels and PhenoAgeAccel. Higher LTPA levels were associated with significantly lower PhenoAgeAccel scores. In the fully adjusted model, compared to individuals with no LTPA, those with Regular-LTPA patterns exhibited significantly lower PhenoAgeAccel scores (β = -1.164, 95% CI: -1.651 to -0.677, P < 0.0001) (Fig. 3). Additionally, individuals with Inactive-LTPA patterns showed a trend towards reduced PhenoAgeAccel scores, although this was not statistically significant (β = -0.378, 95% CI: -1.076 to 0.321, P = 0.279). The Weekend Warrior group also did not show a significant difference (β = -0.724, 95% CI: -1.571 to 0.122, P = 0.091). RCS analysis indicated a nonlinear threshold effect at 594.57 min per week (P for Nonlinear = 0.0010).
Sex-specific analyses revealed consistent findings. For males, higher LTPA levels were significantly associated with lower PhenoAgeAccel scores. Compared to males with no LTPA, those with Regular-LTPA patterns exhibited significantly reduced PhenoAgeAccel (β = -0.882, 95% CI: -1.409 to -0.355, P = 0.002) (Fig. 4). Males with Inactive-LTPA patterns showed a trend towards lower PhenoAgeAccel, but this was not statistically significant (β = -0.495, 95% CI: -1.313 to 0.323, P = 0.226). Similarly, the Weekend Warrior group did not show a significant difference (β = -0.53, 95% CI: -1.544 to 0.483, P = 0.295). RCS analysis indicated a nonlinear threshold effect for males at 676.58 min per week (P for Nonlinear = 0.0433).
Similarly, for females, there was a significant inverse association between LTPA levels and PhenoAgeAccel. Compared to females with no LTPA, those with Regular-LTPA patterns exhibited significantly lower PhenoAgeAccel (β = -1.637, 95% CI: -2.406 to -0.867, P < 0.001) (Fig. 5). Females with Inactive-LTPA patterns did not show a significant difference (β = -0.397, 95% CI: -1.419 to 0.626, P = 0.435), nor did those in the Weekend Warrior group (β = -1.568, 95% CI: -3.182 to 0.046, P = 0.057). RCS analysis revealed a nonlinear threshold effect for females at 502.31 min per week (P for Nonlinear = 0.0036).
Threshold effect analysis of the relationship between LTPA and PhenoAgeAccel in individuals with type 2 diabetes
This study, using the adjusted Model 2, indicates a nonlinear relationship between LTPA levels and PhenoAgeAccel (Table 2). Specifically, at LTPA levels below 594.57 min per week, the association with PhenoAgeAccel is significant (β = -0.002, 95% CI: -0.003 to -0.001, P = 0.000), while levels above 594.57 min per week show no significant association (β = 0.000, 95% CI: -0.001 to 0.001, P = 0.537). The log-likelihood ratio test strongly supports the statistical significance of the two-piecewise linear regression model (P = 0.005).
Table 2.
Threshold effect analysis of the relationship between LTPA and PhenoAgeAccel in individuals with type 2 diabetes.
| Outcome | Beta | 95%CI | P-value |
|---|---|---|---|
| LTPA | |||
| One-line linear regression model | -0.001 | -0.002,-0.001 | < 0.001 |
| Two-piecewise linear regression model | |||
| Inflection point | 594.57 | ||
| LTPA < 594.57 | -0.002 | -0.003,-0.001 | 0.000 |
| LTPA ≥ 594.57 | 0.000 | -0.001,0.001 | 0.537 |
| P for Log-likelihood ratio test | 0.005 | ||
| LTPA (Male) | |||
| One-line linear regression model | -0.001 | -0.002,0.000 | 0.006 |
| Two-piecewise linear regression model | |||
| Inflection point | 676.58 | ||
| LTPA < 676.58 | -0.002 | -0.003,-0.000 | 0.007 |
| LTPA ≥ 676.58 | 0.000 | -0.001,0.002 | 0.654 |
| P for Log-likelihood ratio test | 0.085 | ||
| LTPA (Female) | |||
| One-line linear regression model | -0.002 | -0.004,-0.001 | 0.001 |
| Two-piecewise linear regression model | |||
| Inflection point | 502.31 | ||
| LTPA < 502.31 | -0.004 | -0.006,-0.002 | 0.000 |
| LTPA ≥ 502.31 | 0.001 | -0.002,0.003 | 0.650 |
| P for Log-likelihood ratio test | 0.013 |
Adjusted for age, sex, race, PIR, education, BMI, marital status, smoke, alcohol, HEI, sedentary behavior, hypertension, hyperlipidemia and cancer.
In the sex subgroup analysis, for males with LTPA levels below 676.58 min per week, the association with PhenoAgeAccel is significant (β = -0.002, 95% CI: -0.003 to -0.000, P = 0.007), whereas levels above 676.58 min per week show no significant association (β = 0.000, 95% CI: -0.001 to 0.002, P = 0.654).For females, the association at LTPA levels below 502.31 min per week is significant (β = -0.004, 95% CI: -0.006 to -0.002, P = 0.000), while levels above 502.31 min per week show no significant association (β = 0.001, 95% CI: -0.002 to 0.003, P = 0.650).
These results highlight the importance of considering the nonlinear threshold effects of LTPA levels in predicting PhenoAgeAccel, suggesting that both males and females benefit from LTPA up to specific thresholds, beyond which additional LTPA does not provide further significant reductions in PhenoAgeAccel.
Interaction Effect Test of the relationship between LTPA and PhenoAgeAccel in individuals with Type 2 Diabetes
This study analyzed the relationship between LTPA and PhenoAgeAccel in patients with T2D, examining the interaction effects of sex, age, race, marital status, educational level, PIR, and BMI (Table 3). The analysis revealed that only marital status showed a significant interaction effect with LTPA and PhenoAgeAccel (P for interaction < 0.05). These findings highlight the consistent positive impact of LTPA across various demographic and lifestyle factors in patients with T2D. It is recommended to promote regular leisure-time physical activity in public health strategies as an effective means to slow biological aging.
Table 3.
Interaction Effect Test.
| Characteristic | No-LTPA | Inactive-LTPA | P | Regular-LTPA | P | Weekend Warrior | P | P for trend | P for interaction |
|---|---|---|---|---|---|---|---|---|---|
| Sex | 0.49 | ||||||||
| Female | ref | -0.5(-1.31, 0.32) | 0.23 | -0.88(-1.41,-0.35) | 0.002 | -0.53(-1.54, 0.48) | 0.29 | 0.01 | |
| Male | ref | -0.4(-1.42, 0.63) | 0.43 | -1.64(-2.41,-0.87) | < 0.001 | -1.57(-3.18, 0.05) | 0.06 | < 0.001 | |
| Age, years | 0.1 | ||||||||
| 20–39 | ref | 0.55(-0.82,1.92) | 0.42 | -0.57(-1.70,0.55) | 0.31 | 1.15(-0.41,2.71) | 0.14 | 0.92 | |
| 40–59 | ref | 0.39(-0.68, 1.45) | 0.47 | -0.74(-1.46,-0.02) | 0.04 | -0.43(-2.08, 1.21) | 0.59 | 0.06 | |
| ≥60 | ref | -1.16(-2.23,-0.09) | 0.03 | -1.64(-2.40,-0.88) | < 0.001 | -2.14(-3.70,-0.58) | 0.01 | < 0.0001 | |
| Race | 0.06 | ||||||||
| Non-Hispanic White | ref | -0.67(-1.58, 0.23) | 0.14 | -1.58(-2.23,-0.92) | < 0.0001 | -1.06(-2.22, 0.10) | 0.07 | < 0.0001 | |
| Mexican American | ref | -0.26(-1.35,0.82) | 0.61 | -0.69(-2.02,0.65) | 0.28 | -0.6(-2.48,1.28) | 0.50 | 0.26 | |
| Non-Hispanic Black | ref | 0.56(-0.56, 1.68) | 0.30 | 0.02(-0.85, 0.89) | 0.95 | -1.22(-4.30, 1.86) | 0.41 | 0.79 | |
| Other Race | ref | 0.29(-1.33, 1.91) | 0.72 | -0.4(-1.48, 0.68) | 0.46 | 0.19(-1.83, 2.21) | 0.85 | 0.59 | |
| Education | 0.69 | ||||||||
| Below | ref | -0.55(-1.71, 0.62) | 0.35 | -0.43(-1.59, 0.72) | 0.45 | 1.2(-1.16, 3.56) | 0.31 | 0.72 | |
| High School | ref | -0.74(-2.22, 0.74) | 0.32 | -1.24(-2.27,-0.22) | 0.02 | -0.89(-2.41, 0.63) | 0.24 | 0.02 | |
| Above | ref | -0.29(-1.08, 0.49) | 0.45 | -1.26(-1.92,-0.60) | < 0.001 | -1.36(-2.68,-0.03) | 0.05 | < 0.001 | |
| PIR | 0.45 | ||||||||
| < 1.3 | ref | 0.13(-0.80, 1.06) | 0.78 | -0.51(-1.59, 0.57) | 0.34 | -1.23(-2.96, 0.50) | 0.16 | 0.2 | |
| 1.3–3.49 | ref | -0.88(-1.90, 0.15) | 0.09 | -0.97(-1.80,-0.14) | 0.02 | -0.4(-1.89, 1.09) | 0.59 | 0.03 | |
| ≥ 3.5 | ref | -0.13(-1.17, 0.91) | 0.80 | -1.42(-2.15,-0.68) | < 0.001 | -0.62(-2.40, 1.17) | 0.49 | 0.002 | |
| BMI | 0.11 | ||||||||
| < 25 | ref | 0.37(-1.15, 1.88) | 0.62 | -1.49(-2.58,-0.39) | 0.01 | -0.1(-2.33, 2.13) | 0.93 | 0.02 | |
| 25-29.9 | ref | -0.86(-1.80, 0.08) | 0.07 | -0.64(-1.40, 0.12) | 0.10 | -0.39(-1.85, 1.07) | 0.59 | 0.12 | |
| ≥ 30 | ref | -0.28(-1.17, 0.60) | 0.52 | -1.38(-2.13,-0.62) | < 0.001 | -1.06(-2.28, 0.15) | 0.08 | 0.001 | |
| Marital status | 0.02 | ||||||||
| Widowed/Divorced/Separated | ref | -1.07(-1.90,-0.25) | 0.01 | -1.48(-2.13,-0.83) | < 0.0001 | -0.64(-1.88, 0.59) | 0.30 | < 0.001 | |
| Never married | ref | 1.04(-1.20, 3.28) | 0.35 | 0.2(-0.95, 1.35) | 0.73 | 0.34(-2.05, 2.73) | 0.77 | 0.53 | |
| Married/living with partner | ref | 1.55( 0.37, 2.74) | 0.01 | -0.69(-1.74, 0.35) | 0.19 | -1.54(-3.53, 0.45) | 0.12 | 0.22 | |
| Smoke status | 0.17 | ||||||||
| Former smoker | ref | -0.8(-2.10, 0.49) | 0.22 | -1.56(-2.41,-0.72) | < 0.001 | -2.21(-3.55,-0.88) | 0.002 | < 0.0001 | |
| Nonsmoker | ref | -0.31(-1.16, 0.54) | 0.46 | -0.84(-1.50,-0.17) | 0.02 | 0.56(-1.05, 2.16) | 0.49 | 0.09 | |
| Current smoker | ref | 0.79(-0.81, 2.38) | 0.32 | -1.53(-3.05, 0.00) | 0.05 | -0.38(-2.29, 1.52) | 0.69 | 0.12 | |
| Alcohol status | 0.12 | ||||||||
| Former | ref | 0.11(-1.99, 2.21) | 0.91 | -1.36(-2.74, 0.02) | 0.05 | -2.74(-4.57,-0.91) | 0.01 | 0.04 | |
| Never | ref | -0.6(-1.48, 0.28) | 0.17 | -0.64(-1.96, 0.67) | 0.33 | 1.6(-1.42, 4.62) | 0.29 | 0.49 | |
| Mild | ref | -0.11(-1.04, 0.83) | 0.82 | -1.17(-1.98,-0.35) | 0.01 | -0.26(-1.83, 1.32) | 0.74 | 0.01 | |
| Moderate | ref | -1.6(-3.15,-0.06) | 0.04 | -2.2(-3.32,-1.09) | < 0.001 | -2.97(-5.09,-0.86) | 0.01 | < 0.0001 | |
| Heavy | ref | 0.33(-1.36,2.02) | 0.70 | -0.55(-1.72,0.61) | 0.34 | 0.58(-1.49,2.65) | 0.57 | 0.73 | |
| Sedentary behavior | 0.21 | ||||||||
| <4 | ref | -0.96(-2.15, 0.24) | 0.11 | -0.58(-1.44, 0.28) | 0.18 | -0.92(-2.49, 0.64) | 0.24 | 0.1 | |
| 4-<6 | ref | 0.28(-1.16, 1.73) | 0.69 | -1.03(-1.93,-0.13) | 0.03 | 0.47(-0.95, 1.88) | 0.51 | 0.1 | |
| 6–8 | ref | -0.64(-1.66, 0.39) | 0.21 | -1.06(-2.16, 0.03) | 0.06 | 0.25(-1.90, 2.40) | 0.81 | 0.13 | |
| >8 | -0.31(-1.80, 1.18) | 0.67 | -1.94(-3.11,-0.76) | 0.002 | -3.3(-5.74,-0.87) | 0.01 | < 0.001 | ||
| Healthy eating index | 0.85 | ||||||||
| Quantile 1 | ref | -0.03(-1.14, 1.09) | 0.96 | -1.02(-1.74,-0.31) | 0.01 | -0.82(-2.18, 0.53) | 0.22 | 0.004 | |
| Quantile 2 | ref | -0.72(-1.74, 0.29) | 0.16 | -1.32(-2.20,-0.44) | 0.005 | -0.43(-2.21, 1.34) | 0.62 | 0.02 | |
| Quantile 3 | ref | -0.47(-1.64, 0.70) | 0.42 | -1.16(-2.00,-0.33) | 0.01 | -0.89(-3.03, 1.24) | 0.40 | 0.01 | |
| Hyperlipidemia | 0.38 | ||||||||
| Yes | ref | -0.51(-1.27, 0.25) | 0.18 | -1.19(-1.73,-0.65) | < 0.0001 | -1.29(-2.27,-0.32) | 0.01 | < 0.0001 | |
| No | ref | 0.16(-1.32, 1.65) | 0.82 | -1(-1.99, 0.00) | 0.05 | 0.48(-1.64, 2.61) | 0.64 | 0.34 | |
| Hypertension | 0.2 | ||||||||
| Yes | ref | -0.54(-1.51, 0.43) | 0.27 | -0.93(-1.51,-0.35) | 0.003 | -1.08(-2.46, 0.31) | 0.12 | 0.001 | |
| No | ref | -0.06(-0.86, 0.74) | 0.88 | -1.66(-2.37,-0.94) | < 0.0001 | -0.2(-1.38, 0.98) | 0.73 | < 0.001 | |
| Cancer | 0.38 | ||||||||
| Yes | ref | -0.19(-0.97, 0.59) | 0.62 | -1.13(-1.68,-0.58) | < 0.001 | -0.76(-1.53, 0.02) | 0.06 | < 0.001 | |
| No | ref | -0.95(-2.42, 0.51) | 0.19 | -1.12(-2.74, 0.51) | 0.17 | 0.09(-4.03, 4.22) | 0.96 | 0.23 |
Discussion
This study systematically explored the relationship between different LTPA patterns and PhenoAgeAccel in T2D patients, revealing several key findings. Firstly, the analysis results showed significant differences in PhenoAgeAccel across different LTPA patterns. Specifically, compared to the No-LTPA group, the Regular-LTPA pattern significantly reduced PhenoAgeAccel, whereas the “Weekend Warrior” and “Inactive-LTPA” patterns did not exhibit significant anti-aging effects. This indicates that regular physical activity has a significant impact on mitigating biological aging in T2D patients. Additionally, the study suggests gender-specific effects. For males, LTPA below 676.58 min per week significantly reduced PhenoAgeAccel, but beyond this threshold, the association was no longer significant. Similarly, for females, LTPA below 502.31 min per week significantly reduced PhenoAgeAccel, but the association became non-significant beyond this threshold. These results highlight the universal effect of moderate and consistent physical activity in delaying biological aging. In summary, this study reveals the complex relationship between LTPA and PhenoAgeAccel in T2D patients. While increasing physical activity helps delay aging, excessive physical activity does not confer additional anti-aging benefits.
Consistent with the findings of this study, existing literature indicates that regular physical activity has significant health benefits for T2D patients. Colberg et al29. found that regular physical activity improves glycemic control, prevents and delays the onset of T2D, and positively affects lipid profiles, blood pressure, cardiovascular events, mortality, and quality of life. Arem et al30. reported that regular moderate-intensity physical activity (such as at least 150 min per week) significantly reduces overall mortality. Additionally, Bassuk and Manson31 emphasized that moderate physical activity can reduce insulin resistance and inflammation, enhance insulin sensitivity and glycemic control, and significantly lower the risk of T2D and cardiovascular diseases. These findings further support the results of this study, highlighting the important role of regular physical activity in mitigating biological aging in T2D patients.
Regarding physiological mechanisms, studies have shown that physical activity counteracts biological aging through multiple pathways. Sailani et al32. found that lifelong regular physical activity is associated with hypomethylation of promoter regions in skeletal muscle genes, which may improve insulin sensitivity and enhance energy metabolism, myogenesis, contractile properties, and oxidative stress resistance. Additionally, physical activity improves mitochondrial function and increases GLUT4 expression, significantly enhancing glucose uptake in skeletal muscle and thereby improving insulin resistance33. Teixeira-Lemos et al34. noted that regular moderate-intensity physical activity significantly improves glycemic control, dyslipidemia, and hypertension in T2D patients through its antioxidant and anti-inflammatory properties, thereby reducing cardiovascular risk. These mechanisms collectively explain the significant impact of regular physical activity in mitigating biological aging in T2D patients.
While this study offers several important strengths, it is also accompanied by a number of limitations that should be acknowledged. Firstly, it leverages the large sample size and broad representativeness of NHANES data, which includes individuals of different ages, genders, races, and socioeconomic backgrounds, thus ensuring high external validity and generalizability of the findings. Additionally, the study employs various statistical analysis methods, including multiple linear regression models and RCS analysis. These methods not only reveal linear and nonlinear relationships between LTPA and PhenoAgeAccel but also identify specific threshold effects of different LTPA patterns and gender in this relationship, providing a more comprehensive perspective on the impact of physical activity on biological aging. The study systematically explores, for the first time, the effects of different LTPA patterns on biological aging in T2D patients, highlighting the crucial role of regular physical activity in mitigating biological aging in this population. This finding is significant for public health strategies, emphasizing the promotion of moderate and regular physical activity as an effective measure to delay biological aging. However, this study also has several limitations that need to be addressed. Firstly, it employs a cross-sectional design, which limits causal inference. Although it reveals an association between LTPA and PhenoAgeAccel, it cannot establish a causal relationship between physical activity and the mitigation of biological aging. To better capture the long-term effects of physical activity on biological age acceleration, longitudinal designs should be adopted. Additionally, the study relies on self-reported physical activity data, which may introduce recall bias and social desirability bias, potentially affecting the accuracy of physical activity measurement and, consequently, the reliability of the results. To improve data accuracy, future research could consider using objective measurement tools such as accelerometers or other wearable devices. Although the analysis models in this study adjust for various confounding factors, including demographic information, lifestyle factors, and health status indicators, unmeasured confounders may still exist. Factors such as genetic predisposition, psychological stress, and social support might influence the relationship between LTPA and PhenoAgeAccel. To minimize bias, future studies should aim to include more potential confounders. Finally, this study’s sample is limited to American adults, which may pose certain limitations to the external validity of the findings and their generalizability across global populations. Cultural, socioeconomic backgrounds, lifestyle factors, and access to healthcare resources vary significantly across different countries and regions, which in turn influence the patterns and intensity of physical activity. These differences could potentially alter the relationship between LTPA and PhenoAgeAccel. For instance, in low-income regions or areas with limited healthcare resources, physical activity types and frequency may differ markedly from those in the U.S., particularly under the influence of work-related stress and environmental constraints. Such differences could impact the generalizability of our findings. Therefore, future research should be conducted across populations with diverse cultural and socioeconomic backgrounds to validate whether the relationship between LTPA and biological aging holds universally. Additionally, lifestyle factors, dietary habits, and access to physical activity resources, which may vary across regions, should be considered as potential moderators of LTPA’s effects. Expanding research into global, diverse populations would not only broaden the applicability of the findings but also provide a foundation for designing tailored public health interventions suited to the characteristics of different populations.
Conclusion
This study systematically explored the relationship between different LTPA patterns and PhenoAgeAccel in T2D patients, revealing the significant role of regular leisure-time physical activity in mitigating biological aging in this population. The study elucidated both linear and nonlinear relationships between LTPA and PhenoAgeAccel, identifying specific threshold effects of different LTPA patterns and gender in this relationship. The results demonstrated that regular moderate-intensity leisure-time physical activity significantly reduces PhenoAgeAccel in T2D patients, while the “Weekend Warrior” and “Inactive-LTPA” patterns did not show significant anti-aging effects. Additionally, the gender-specific analysis further emphasized the universal effectiveness of moderate and regular physical activity in delaying biological aging. This study has important public health implications, highlighting the promotion of moderate and regular physical activity as an effective strategy to delay biological aging in T2D patients.
Acknowledgements
The research team extends its gratitude to the contributors of the NHANES datasets for their invaluable participation.
Abbreviations
- LTPA
Leisure-Time Physical Activity
- PhenoAgeAccel
Phenotypic Age Acceleration
- T2D
Type 2 Diabetes
- NHANES
National Health and Nutrition Examination Survey
- CDC
Centers for Disease Control and Prevention
- PIR
Poverty Income Ratio
- GPAQ
Global Physical Activity Questionnaire
- MET
Metabolic Equivalent of Task
- MVPA
Moderate-to-Vigorous Physical Activity
- FPG
Fasting Plasma Glucose
- HbA1c
Glycosylated Hemoglobin
- OGTT
Oral Glucose Tolerance Test
- CRP
C-Reactive Protein
- BMI
Body Mass Index
- HEI
Healthy Eating Index
- ANOVA
Analysis of Variance
- RCS
Restricted Cubic Spline
- ROS
Reactive Oxygen Species
- GLUT4
Glucose Transporter Type 4
Author contributions
Conceptualization: Dongzhe Wu, Mingyu Shang. Data curation: Dongzhe Wu. Formal analysis: Dongzhe Wu, Yujia Liu, Pengxuan Li, Xiang Pan. Methodology: Yishuai Jia, Yujia Liu, Xiang Pan.Project administration: Dongzhe Wu, Mingyu Shang, Pengxuan Li. Visualization: Dongzhe Wu. Writing-original draft: Dongzhe Wu. Writing-review & editing: Dongzhe Wu, Yishuai Jia.
Funding
Our study greatly benefitted from the data provided by the National Health and Nutrition Examination Survey (NHANES), and we extend our heartfelt thanks to the National Center for Health Statistics (NCHS) for their invaluable contribution.
Data availability
The dataset(s) supporting the conclusions of this article is(are) available in the NHANES website (https://wwwn.cdc.gov/nchs/nhanes/).
Declarations
Ethics approval and consent to participate
The National Center for Health Statistics of the Center for Disease Control and Prevention Institutional Review Board approved the protocol. The research program complied with the basic elements of the Declaration of Helsinki. All participants signed an informed consent form while participating in NHANES. The NHANES study protocols were approved under the following protocol numbers: NHANES 1999–2004 (protocol #98 − 12), NHANES 2005–2006 (protocol #2005-06), NHANES 2007–2008 (continuation of protocol #2005-06), NHANES 2009–2010 (continuation of protocol #2005-06), NHANES 2011–2012 (protocol #2011-17), NHANES 2013–2014 (continuation of protocol #2011-17), NHANES 2015–2016 (continuation of protocol #2011-17), and NHANES 2017–2018 (continuation of protocol #2011-17).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Dongzhe Wu and Yishuai Jia contributed equally to this work.
Contributor Information
Pengxuan Li, Email: lipengxuan123@gmail.com.
Mingyu Shang, Email: 2023110079@bsu.edu.cn.
References
- 1.Aguayo-Mazzucato, C. et al. Acceleration of β cell aging determines diabetes and Senolysis improves Disease outcomes. Cell Metabol. 30, 129–142e4 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gunasekaran, U. & Gannon, M. Type 2 diabetes and the aging pancreatic Beta cell. Aging. 3, 565–575 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Leenders, M. et al. Patients with type 2 diabetes show a Greater decline in muscle Mass, muscle strength, and functional capacity with aging. J. Am. Med. Dir. Assoc. 14, 585–592 (2013). [DOI] [PubMed] [Google Scholar]
- 4.Rando, T. A. & Chang, H. Y. Aging, rejuvenation, and epigenetic reprogramming: resetting the aging clock. Cell. 148, 46–57 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wu, D. et al. Better Life’s essential 8 contributes to slowing the biological aging process: a cross-sectional study based on NHANES 2007–2010 data. Front. Public. Health. 12, 1295477 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Savela, S. et al. Physical activity in midlife and telomere length measured in old age. Exp. Gerontol. 48, 81–84 (2013). [DOI] [PubMed] [Google Scholar]
- 7.Semeraro, M. D. et al. Physical activity, a modulator of aging through effects on telomere biology. Aging. 12, 13803–13823 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Polidori, M. C., Mecocci, P., Cherubini, A. & Senin, U. Physical activity and oxidative stress during aging. Int. J. Sports Med. 21, 154–157 (2000). [DOI] [PubMed] [Google Scholar]
- 9.Ji, L. L. Exercise at Old Age: does it increase or alleviate oxidative stress? Ann. N. Y. Acad. Sci. 928, 236–247 (2001). [DOI] [PubMed] [Google Scholar]
- 10.Ubaida-Mohien, C. et al. Physical Activity Associated Proteomics of Skeletal Muscle: Being Physically Active in Daily Life May Protect Skeletal Muscle From Aging. Front Physiol.10, 312 (2019). [DOI] [PMC free article] [PubMed]
- 11.Barha, C. K., Hsu, C.-L., Ten Brinke, L. & Liu-Ambrose, T. Biological Sex: A Potential Moderator of Physical Activity Efficacy on Brain Health. Front Aging Neurosci.11, 329 (2019). [DOI] [PMC free article] [PubMed]
- 12.Wang, L. et al. Trends in Prevalence of diabetes and control of risk factors in diabetes among US adults, 1999–2018. JAMA. 326, 1–13 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Mathew, T. K., Zubair, M. & Tadi, P. Blood Glucose Monitoring. In StatPearls [Internet] (StatPearls Publishing, 2023). [PubMed] [Google Scholar]
- 14.Zhou, H., Li, T., Li, J., Zhuang, X. & Yang, J. The association between visceral adiposity index and risk of type 2 diabetes mellitus. Sci. Rep. 14, 16634 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Liao, J. et al. Association of daily sitting time and leisure-time physical activity with body fat among U.S. adults. J. Sport Health Sci. 13, 195–203 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wu, D. et al. Dose-response relationship between leisure-time physical activity patterns and phenotypic age acceleration in American adults: a cross-sectional analysis. J. Exerc. Sci. Fit. 22, 445–455 (2024). [Google Scholar]
- 17.Wang, K., Xia, F., Li, Q., Luo, X. & Wu, J. The associations of Weekend Warrior activity patterns with the visceral Adiposity Index in US adults: repeated cross-sectional study. JMIR Public. Health Surveill. 9, e41973 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Shiroma, E. J., Lee, I. M., Schepps, M. A., Kamada, M. & Harris, T. B. Physical activity patterns and mortality: the Weekend Warrior and Activity bouts. Med. Sci. Sports Exerc. 51, 35–40 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Li, G. et al. Association between physical activity and depression in adult prescription opioid users: a cross-sectional analysis based on NHANES 2007–2018. Gen. Hosp. Psychiatry. 89, 1–7 (2024). [DOI] [PubMed] [Google Scholar]
- 20.Liang, J. H. et al. Whether weekend warrior activity and other leisure-time physical activity pattern reduce the risk of depression symptom in the representative adults? A population-based analysis of NHANES 2007–2020. J. Affect. Disord. 340, 329–339 (2023). [DOI] [PubMed] [Google Scholar]
- 21.He, H., Chen, X., Ding, Y., Chen, X. & He, X. Composite dietary antioxidant index associated with delayed biological aging: a population-based study. Aging (Albany NY). 16, 15–27 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Levine, M. E. et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging (Albany NY). 10, 573–591 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sun, X. et al. Associations of epigenetic age acceleration with CVD risks across the Lifespan: the Bogalusa Heart Study. JACC Basic. Transl Sci. 9, 577–590 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Sun, L., Huo, X., Jia, S., Sun, J. & Wang, C. PhenoAgeAccel mediates the association between circadian syndrome and all-cause mortality, as evidenced by NHANES. J. Nutr. Health Aging. 28, 100269 (2024). [DOI] [PubMed] [Google Scholar]
- 25.Liu, Z. et al. A new aging measure captures morbidity and mortality risk across diverse subpopulations from NHANES IV: A cohort study. PLoS Med.15, e1002718 (2018). [DOI] [PMC free article] [PubMed]
- 26.Liu, W. et al. Oxidative Stress Factors Mediate the Association between Life’s essential 8 and accelerated phenotypic aging: NHANES 2005–2018. J. Gerontol. Biol. Sci. Med. Sci. 79, glad240 (2024). [DOI] [PubMed] [Google Scholar]
- 27.Thomas, A., Belsky, D. W. & Gu, Y. Healthy Lifestyle Behaviors and Biological Aging in the U.S. National Health and Nutrition examination surveys 1999–2018. J. Gerontol. Biol. Sci. Med. Sci. 78, 1535–1542 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Reedy, J. et al. Healthy eating Index Protocol: review, update, and development process to reflect Dietary Guidance across the Lifespan. Curr. Dev. Nutr. 5, 447 (2021). [Google Scholar]
- 29.Davis, T. A. 68 Nutritional Importance of Animal-Sourced foods in a healthy Diet. J. Anim. Sci. 100, 30–31 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Arem, H. et al. Leisure time physical activity and mortality: a detailed pooled analysis of the dose-response relationship. JAMA Intern. Med. 175, 959–967 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bassuk, S. S. & Manson, J. E. Epidemiological evidence for the role of physical activity in reducing risk of type 2 diabetes and cardiovascular disease. J. Appl. Physiol. 99, 1193–1204 (2005). [DOI] [PubMed] [Google Scholar]
- 32.Sailani, M. R. et al. Lifelong physical activity is associated with promoter hypomethylation of genes involved in metabolism, myogenesis, contractile properties and oxidative stress resistance in aged human skeletal muscle. Sci. Rep. 9, 3272 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Santos, J. M., dos, Moreli, M. L., Tewari, S. & Benite-Ribeiro, S. A. The effect of exercise on skeletal muscle glucose uptake in type 2 diabetes: an epigenetic perspective. Metabolism - Clin. Experimental. 64, 1619–1628 (2015). [DOI] [PubMed] [Google Scholar]
- 34.Teixeira-Lemos, E., Nunes, S., Teixeira, F. & Reis, F. Regular physical exercise training assists in preventing type 2 diabetes development: focus on its antioxidant and anti-inflammatory properties. Cardiovasc. Diabetol. 10, 12 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The dataset(s) supporting the conclusions of this article is(are) available in the NHANES website (https://wwwn.cdc.gov/nchs/nhanes/).