Doing more of the same may not be the best recipe for a longer life. That is the central finding of a landmark study from Harvard’s T.H. Chan School of Public Health, which followed more than 111,000 American men and women for as long as three decades and concluded that people who mixed up the types of exercise they did tended to live longer than those who simply did more of a single activity.
The research, published in the open access journal BMJ Medicine in January 2026, upends a familiar assumption about physical activity. For years, the public-health message has been a simple one: move more. The new findings suggest that message may need a subtle but important addition — vary the way you move, not just the amount.
A Study Three Decades in the Making
The work rests on an unusual depth of data. The researchers drew on two of the longest-running health studies in the world — the Nurses’ Health Study, which has followed 70,725 women since 1986, and the Health Professionals Follow-Up Study, which has tracked 40,742 men over the same period. Participants were free of diabetes, cardiovascular disease, cancer, respiratory disease and neurological disease when the analysis began, and they updated their leisure-time physical activity every two years using validated questionnaires.
Across more than 38 million person-years of follow-up, the team recorded 38,247 deaths — 9,901 from cardiovascular disease, 10,719 from cancer and 3,159 from respiratory disease. With a dataset of that scale, the researchers were able to ask a question most exercise studies cannot: not just how much people moved, but how varied that movement was.
Variety Beats Repetition
The answer was striking. Participants who reported the widest variety of physical activities had a 19 percent lower risk of dying from any cause compared with those whose routines were the most repetitive. The reduction held across the board: the highest-variety group had a 13 to 41 percent lower risk of death from cardiovascular disease, cancer, respiratory disease and other causes, with the strongest single effect seen for respiratory disease, where mortality fell by 41 percent.
Crucially, the benefit was independent of how much exercise people did in total. After the researchers adjusted for overall physical activity levels, variety still mattered. In other words, two people who burn the same number of calories a week may face different risks depending on whether that energy comes from one repetitive pursuit or from a mix of activities.
A Threshold, Not a Straight Line
The relationship was also not linear, the researchers noted. The gains from adding variety appeared to reach a plateau, suggesting there may be an optimal threshold beyond which diversifying further adds little. The finding echoes a broader pattern emerging across exercise science: when it comes to physical activity, there are diminishing returns — and the sweet spot may be smaller than many people assume.
Most individual activities were associated with a lower risk of death, with one notable exception. Swimming did not show the same protective association in this analysis, a result the authors attributed to the difficulty of isolating its effects — swimming is often done alongside other activities, and it may also be chosen by people with specific health profiles. Even so, the takeaway was clear: the benefits of exercise do not belong to any single sport.
The Strength-Training Sweet Spot
The variety study arrived alongside a companion finding that reinforced its practical message. In a separate analysis of nearly 147,000 people followed for up to 30 years, researchers identified a specific dose of resistance training linked to the greatest longevity benefit: roughly 90 to 120 minutes a week. At that level, the risk of death from any cause fell by about 13 percent, while deaths from cardiovascular disease fell by 19 percent and deaths from neurological diseases, including dementia, fell by 27 percent.
Beyond 120 minutes a week, the extra benefit largely disappeared. Like variety, strength training appears to follow a curve that flattens — a reassuring message for anyone who has assumed that a longer life requires ever-longer hours in the gym.
Why Variety Might Matter
The mechanisms behind the effect remain partly speculative, but researchers and physiologists point to several plausible explanations. Different activities stress different systems: aerobic exercise strengthens the heart and lungs, resistance training builds muscle and bone, and balance or flexibility work protects against falls and frailty in later life. A varied routine, in this view, is a form of full-body insurance — it ensures that no single system is neglected.
There is also a behavioural dimension. People who do many different activities may simply be more likely to stay active over a lifetime, avoiding the boredom and repetitive-strain injuries that can derail a monotonous routine. Variety, in other words, may be both a physiological strategy and a motivational one.
What It Means for Everyday Life
For most people, the practical implications are modest and encouraging. The findings do not demand a punishing schedule; they suggest the opposite. Walking, cycling, swimming, strength work, dancing, gardening and team sports all count, and combining several of them — rather than perfecting one — may deliver the greatest return.
The study also offers a useful corrective to the extremes of modern fitness culture, which often pushes specialization and intensity above all else. Its message is quieter but more durable: an active life is important, but a varied active life appears to be even better. The simplest way to act on the research, experts say, is to keep moving, and to make a point of changing how.
The Harvard findings land at a moment when longevity has become a national obsession, fuelled by wearables, apps and a flood of often contradictory advice. Against that noise, the study offers something rare: a conclusion grounded in three decades of data and more than a hundred thousand lives. The path to a longer life may not run through a single perfect workout. It may run through many imperfect ones — as long as they are not all the same.