Physical Activity Variety Independently Predicts Longevity—Beyond Total Volume
We know exercise reduces mortality. The data are overwhelming.
But most studies examine total physical activity volume. Few investigate which specific activities matter most. Even fewer ask whether engaging in multiple types offers advantages beyond simply accumulating more total exercise.
Han et al. just published pooled data from two large prospective cohorts addressing exactly these questions. The findings challenge how we think about exercise prescription.
The Study Design
Researchers analyzed data from the Nurses' Health Study (1986-2018) and Health Professionals Follow-Up Study (1986-2020). 111,467 participants free of diabetes, cardiovascular disease, cancer, respiratory disease, and neurological disease at baseline. Physical activity assessed biennially using validated questionnaires.
Nine specific activities tracked: walking, jogging, running, bicycling, swimming, tennis/squash/racquetball, climbing stairs, rowing/callisthenics, and weight training. Up to 15 repeated assessments per participant over 30+ years.
2,431,318 person-years of follow-up. 38,847 deaths recorded: 9,901 from cardiovascular disease, 10,719 from cancer, 3,159 from respiratory disease.
The researchers examined two separate questions. First: which individual activities are associated with lower mortality, and what are their dose-response patterns? Second: does physical activity variety—independent of total volume—predict mortality?
Individual Activities Show Distinct Patterns
Most activities were associated with lower all-cause mortality in non-linear dose-response relations.
Comparing highest versus lowest engagement categories, the hazard ratios were:
Walking: 0.83 (17% lower risk)
Tennis/squash: 0.85 (15% lower)
Rowing/callisthenics: 0.86 (14% lower)
Running: 0.87 (13% lower)
Weight training: 0.87 (13% lower)
Jogging: 0.89 (11% lower)
Climbing stairs: 0.90 (10% lower)
Bicycling: 0.96 (4% lower)
Swimming: 1.01 (no association)
But the dose-response curves varied dramatically across activities.
Walking benefits plateaued around 7.5 MET-hours per week. Stair climbing at 0.75 MET-hours per week. Tennis at 5 MET-hours per week. Weight training at 7.5 MET-hours per week.
For swimming, bicycling, and jogging, inverse associations with mortality were apparent up to specific thresholds (2.5, 7.5, and 9 MET-hours per week respectively), but no longer significant beyond these levels.
Running and rowing showed sharp reductions in mortality risk up to about 3 and 2.5 MET-hours per week, followed by continued but slower decline with higher volumes.
The clinical implication: more isn't always better. Each activity has an optimal dose range. Beyond certain thresholds, additional volume may not provide proportional benefit.
The Variety Finding Changes Everything
The researchers created a physical activity variety score. They summed the number of individual activities in which participants consistently engaged, counting each activity as one point if it met predefined thresholds (5 flights/day for stairs, 20 min/week for other activities).
The score was derived from updated measures at each follow-up cycle, then cumulatively averaged to represent long-term physical activity variety.
Highest variety (group 5) compared to lowest (group 1) was associated with 19% lower all-cause mortality (HR 0.81, 95% CI 0.78-0.85). Also 13-41% lower mortality from cardiovascular disease, cancer, respiratory disease, and other causes.
This is after adjusting for ethnic group, family history, baseline BMI, postmenopausal hormone use, smoking status, alcohol intake, total energy intake, diet quality, social integration, and baseline hypertension/hypercholesterolemia.
Most importantly: this association remained significant after further adjusting for total physical activity levels. The model including variety had better fit than the model with total activity alone.
Participants ranked highest for both total physical activity and variety had 21% lower mortality compared to the reference group. Higher variety scores within each category of total physical activity were associated with lower mortality.
The variety effect was independent of total volume.
Why Variety Matters
The mechanism appears to be complementary physiological effects.
Short-term intervention data show aerobic exercise improves cardiorespiratory fitness by increasing peak oxygen consumption but has minimal effect on muscular strength. Resistance training enhances strength without substantially changing VO2 max. Combining both modalities yields improvements in both domains.
Different activities produce distinct effects on body composition, cardiorespiratory fitness, metabolic profiles, and bone strength. Bicycling and running have different ventilatory responses, blood flow patterns, skeletal muscle oxidative capacity, and central/peripheral innervation.
The researchers propose that individuals who engage in multiple physical activities may maximize health benefits by engaging with each activity within its optimal threshold. Most individual activities showed plateauing benefits. Diversifying across multiple activities allows you to capture the unique benefits of each without excessive volume in any single modality.
The variety approach also doesn't specify fixed combinations. Individuals can choose different activities over time according to personal preferences and physical condition.
Clinical Applications
The findings support promoting engagement in multiple types of physical activity alongside increasing total volume.
Don't just accumulate volume in one activity. Diversify your physical activity portfolio.
Combine activities with complementary effects. Pair aerobic and resistance training. Include activities that challenge different movement patterns, energy systems, and physical capacities.
Recognize that most activities show diminishing returns beyond specific thresholds. You don't need endless volume in a single modality.
For cardiovascular and respiratory disease mortality specifically, running, tennis, climbing stairs, and weight training showed more linear associations. These may be particularly important for cardiorespiratory fitness and metabolic health.
For cancer mortality, most activities except swimming showed significant non-linear associations.
Limitations and Nuance
Physical activity was self-reported. MET scores assume active engagement, so lack of intensity information (particularly for swimming) may cause misclassification of true energy expenditure.
Not all activities were assessed in baseline questionnaires for both cohorts. Weight training data were added later. This limits sample sizes and comparability for some activities.
The cohorts are predominantly white health professionals, limiting generalizability.
Reverse causation remains possible despite lag analyses and sensitivity analyses, particularly for respiratory disease.
The swimming finding is notable. Higher levels showed no association with lower mortality. Self-reported duration may correspond to wide ranges of actual energy expenditure due to variation in intensity. Someone swimming casually for 30 minutes reports the same duration as someone swimming vigorously. This misclassification could bias results toward the null.
Bicycling showed controversy. A meta-analysis of 17 studies found linear association with 9% lower risk per 5 MET-hours/week increase. But updated evidence indicates either non-linear relation or no association. The inability to differentiate active commuting from recreational cycling, or capture variations during follow-up, contributes to inconsistencies. This study found reduction in mortality before 7.5 MET-hours/week (64 min/week), with lowest risk around 2.5 MET-hours/week. But cohort-specific analysis showed downward trend in men and suggestive upward trend in women, requiring further investigation.
The Bottom Line
The evidence supports long-term engagement in multiple types of physical activity for longevity.
Most individual activities reduce mortality in non-linear patterns with plateauing benefits.
Physical activity variety independently predicts lower mortality beyond total volume.
The clinical application is straightforward: diversify your training. Combine activities with complementary physiological effects. Maximize benefits from each activity within its optimal dose range rather than excessive volume in a single modality.
The data support what many coaches have known empirically. Variety isn't just for adherence or injury prevention. It's an independent predictor of longevity.
Reference: Han H, Hu J, Lee DH, et al. Physical activity types, variety, and mortality: results from two prospective cohort studies. BMJ Medicine. 2026;5:e001513. doi:10.1136/bmjmed-2025-001513
