Background: the prevalence of sarcopenia increases with age. Physical activity might slow the rate of muscle loss and therewith the incidence of sarcopenia.

Objective: to examine the association of physical activity with incident sarcopenia over a 5-year period.

Design: data from the population-based Age, Gene/Environment, Susceptibility–Reykjavik Study were used.

Setting: people residing in the Reykjavik area at the start of the study.

Subjects: the study included people aged 66–93 years (n = 2309).

Methods: the amount of moderate–vigorous physical activity (MVPA) was assessed by a self-reported questionnaire. Sarcopenia was identified using the European Working Group on Sarcopenia in Older People algorithm, including muscle mass (computed tomography imaging), grip strength (computerised dynamometer) and gait speed (6 m).

Results: mean age of the participants was 74.9 ± 4.7 years. The prevalence of sarcopenia was 7.3% at baseline and 16.8% at follow-up. The incidence proportion of sarcopenia over 5 years was 14.8% in the least-active individuals and 9.0% in the most-active individuals. Compared with the least-active participants, those reporting a moderate–high amount of MVPA had a significantly lower likelihood of incident sarcopenia (OR = 0.64, 95% CI 0.45–0.91). Participants with a high amount of MVPA had higher baseline levels of muscle mass, strength and walking speed, but baseline MVPA was not associated with the rate of muscle loss.

Conclusion: a higher amount of MVPA seems to contribute to counteracting the development of sarcopenia. To delay the onset of sarcopenia and its potential adverse outcomes, attention should be paid to increasing physical activity levels in older adults.