Researchers from Leiden University Medical Centre analyzed 87,975 adults in the UK Biobank [average age 63, 57% female] with no history of diabetes. Participants wore wrist accelerometers for 7 days to measure moderate-to-vigorous physical activity [MVPA] — activity intense enough to leave you slightly out of breath, like brisk walking, running or cycling. They were followed for a median of 7.7 years, during which 2,140 people [2.4%] developed type 2 diabetes.[T2D]
1. Amount of activity matters most
- People were split into lower vs higher activity groups within their age band.
- Lower group: median 121 min MVPA/week — below the 150 min/week guideline.
- Higher group: median 409 min/week.
- Overall, the lower group had twice the relative risk of developing T2D compared to the higher group, after adjusting for age, sex, smoking, family history and other factors.
2. Relative vs absolute benefit — the key distinction
- Younger adults [<63 years]: Lower activity was linked to >2x higher relative risk vs higher activity. The effect was strongest here.
- Older adults [≥63 years]: Lower activity was linked to 65% higher relative risk.
However, because older adults have a much higher overall risk of T2D to begin with, the absolute difference in cases was larger in older groups. In public health terms, getting more older adults active would prevent more actual cases.
The same pattern was seen for prediabetes status: the relative benefit of activity was stronger in people without prediabetes [>2x risk if inactive] than in those with prediabetes [54% higher risk if inactive].
3. Timing of activity
People who did most of their MVPA later in the day [1pm to midnight] had a 10% lower relative risk than those who exercised mostly earlier [5am to before 1pm]. The timing effect was more pronounced in younger adults — 23% lower risk for late-day exercise vs 4% in older adults. Researchers say this does not support different timing recommendations for different groups.
What the authors conclude:
“All adults should be encouraged to be as active as they can… at the population level, we may see the greatest reduction in the number of new cases among older adults.” — Dr Diana van Heemst
And that both relative and absolute risk need to be looked at together to understand impact.
Limitations noted: UK Biobank participants tend to be healthier, wealthier and better educated than the general UK population [97% white in this sample]; prediabetes status was measured several years before activity; and 7-day accelerometer wear may not capture long-term habits.