File
Lee H1,2, Smith T1, Mistry D3, Lamb S1
1University of Oxford, Nuffield Department of Orthopaedics Rheumatology and Musculoskeletal Sciences, Oxford, United Kingdom, 2University of Newcastle, School of Medicine and Public Health, Newcastle, Australia, 3University of Warwick, Warwick Clinical Trials Unit, Coventry, United Kingdom
Background: Dementia affects almost 48 million people worldwide. Because there is no effective cure for dementia at present, therapies often aim to alleviate the symptoms of dementia and to prevent cognitive decline. In 2012, the UK National Institute for Health Research commissioned a randomised trial to investigate the potential benefit of exercise in people with dementia (DAPA trial). In 2018, the DAPA trial concluded that a combination of a moderate to high intensity aerobic and strength exercise programme in addition to usual care was no more beneficial than usual care alone for slowing cognitive impairment in people with dementia. The trial also found that the intervention did not alleviate the burden from primary carers who were associated with people involved in the trial. Although the DAPA intervention was designed to target plausible mechanistic pathways in vascular and Alzheimer's type dementia, it was unclear whether the intervention successfully modulated such mechanisms, and whether modulating these mechanisms would lead to better patient and carer outcomes.
Purpose: To understand why the DAPA intervention did not slow cognitive impairment or alleviate carer burden. That is, to understand where the hypothesised mechanisms of action broke down in the DAPA trial.
Methods: The DAPA trial was a multicentre, pragmatic randomised trial of 494 participants with mild to moderate dementia. 329 participants were assigned to an aerobic and strength exercise programme plus usual care, and 165 were assigned to usual care alone. The intervention consisted of four months of supervised exercise and ongoing support for physical activity, plus usual care. The primary patient-outcome was the score on the Alzheimer's disease assessment scale-cognitive subscale at 12-months. The primary carer-outcome was the score on the Zarit Burden Interview at 12-months. To investigate the mechanistic pathways of the DAPA intervention, we selected plausible social, physical, and behavioural factors measured at 6-months, and used causal mediation analysis to estimate indirect effects of the intervention and its path-specific effects.
Results: The intervention did not produce a meaningful effect on patient cognition and carer burden at 12-months, and did not have an effect on the selected mediators at 6-months. The social, physical, and behavioural mediators were not associated with patient cognition. However, these mediators were significantly associated with carer burden: social, B = -0.71 [95% CI = -1.33 to -0.09]; physical, B = 1.93 [1.05 to 2.81]; behavioural, B = 0.21 [0.09 to 0.34] (B represents unstandardized regression coefficients).
Conclusion(s): An exercise programme for people with dementia targeted mechanisms that were not associated with the primary patient outcome - cognition. Although, social, physical and behavioural mechanisms were associated with carer burden, the intervention did not cause change in these mechanisms.
Implications: These findings can inform the design of future interventions for this population. If the aim of the intervention is to alleviate carer burden, social, physical, and behavioural factors could be worthwhile intervention targets. However, if the aim is to prevent cognitive decline, these mechanisms may not be worthwhile targets for people with dementia.
Keywords: Dementia, Exercise, Mechanism
Funding acknowledgements: Funded by the National Institute for Health Research (09/80/04) with additional support from NIHR Oxford CLAHRC and Biomedical Research Centre.
Purpose: To understand why the DAPA intervention did not slow cognitive impairment or alleviate carer burden. That is, to understand where the hypothesised mechanisms of action broke down in the DAPA trial.
Methods: The DAPA trial was a multicentre, pragmatic randomised trial of 494 participants with mild to moderate dementia. 329 participants were assigned to an aerobic and strength exercise programme plus usual care, and 165 were assigned to usual care alone. The intervention consisted of four months of supervised exercise and ongoing support for physical activity, plus usual care. The primary patient-outcome was the score on the Alzheimer's disease assessment scale-cognitive subscale at 12-months. The primary carer-outcome was the score on the Zarit Burden Interview at 12-months. To investigate the mechanistic pathways of the DAPA intervention, we selected plausible social, physical, and behavioural factors measured at 6-months, and used causal mediation analysis to estimate indirect effects of the intervention and its path-specific effects.
Results: The intervention did not produce a meaningful effect on patient cognition and carer burden at 12-months, and did not have an effect on the selected mediators at 6-months. The social, physical, and behavioural mediators were not associated with patient cognition. However, these mediators were significantly associated with carer burden: social, B = -0.71 [95% CI = -1.33 to -0.09]; physical, B = 1.93 [1.05 to 2.81]; behavioural, B = 0.21 [0.09 to 0.34] (B represents unstandardized regression coefficients).
Conclusion(s): An exercise programme for people with dementia targeted mechanisms that were not associated with the primary patient outcome - cognition. Although, social, physical and behavioural mechanisms were associated with carer burden, the intervention did not cause change in these mechanisms.
Implications: These findings can inform the design of future interventions for this population. If the aim of the intervention is to alleviate carer burden, social, physical, and behavioural factors could be worthwhile intervention targets. However, if the aim is to prevent cognitive decline, these mechanisms may not be worthwhile targets for people with dementia.
Keywords: Dementia, Exercise, Mechanism
Funding acknowledgements: Funded by the National Institute for Health Research (09/80/04) with additional support from NIHR Oxford CLAHRC and Biomedical Research Centre.
Topic: Mental health; Neurology; Research methodology & knowledge translation
Ethics approval required: Yes
Institution: NHS Health Research Authority
Ethics committee: UK NHS national ethics review committees
Ethics number: 11/SW/0232
All authors, affiliations and abstracts have been published as submitted.
