Exercise and dementia: what should we be recommending?

Purpose Exercise has the potential to provide benefits for people living with dementia, yet the balance of evidence is uncertain. This paper aims to provide an evidence synthesis to determine whether exercise improves their health and well-being and what exercise should be recommended. Design/methodology/approach Structured search for existing literature reviews on exercise for dementia. Relevant articles were selected and critically appraised against systematic criteria. The findings from 15 high quality reviews were collated by using a best evidence synthesis approach. Findings The evidence is convincing for improving physical health, promising for cognitive benefits, mixed for psychological benefits and limited for behavioural outcomes. No evidence of harm was found. Overall, exercise can improve physical and mental health for people living with dementia: there is sufficient evidence to recommend multimodal exercise. Social implications The potential beneficial outcomes are of significant importance both for people with dementia and their caregivers. In the absence of more specific findings, the current recommendation for older adults in general is pragmatically justified - some activity is better than none, more activity provides greater benefits. Adding social interaction may be important for psychological and behavioural outcomes Originality/value To the best of the authors' knowledge, this paper is the first to encapsulate the literature to date on exercise for dementia. Combining the findings from previous reviews enabled a novel synthesis across the range of relevant interventions and outcomes.