Effects of nonpharmacological interventions on functioning of people living with dementia at home: A systematic review of randomised controlled trials.

<bold>Objective: </bold>Slowing functional decline could enable people living with dementia to live for longer and more independently in their own homes. We aimed to update previous syntheses examining the effectiveness of nonpharmacological interventions in reducing functional decline (activities of daily living, activity-specific physical functioning, or function-specific goal attainment) in people living in their own homes with dementia.<bold>Methods: </bold>We systematically searched electronic databases from January 2012 to May 2018; two researchers independently rated risk of bias of randomised controlled trials (RCTs) fitting predetermined inclusion criteria using a checklist; we narratively synthesised findings, prioritising studies judged to have a lower risk of bias.<bold>Results: </bold>Twenty-nine papers (describing 26 RCTs) met eligibility criteria, of which we judged 13 RCTs to have a lower risk of bias. Study interventions were evaluated in four groups: physical exercise, occupational, multicomponent, and cognition-oriented interventions. Four out of 13 RCTs reported functional ability as a primary outcome. In studies judged to have a lower risk of bias, in-home tailored exercise, individualised cognitive rehabilitation, and in-home activities-focussed occupational therapy significantly reduced functional decline relative to control groups in individual studies. There was consistent evidence from studies at low risk of bias that group-based exercise and reminiscence therapies were ineffective at reducing functional decline.<bold>Conclusion: </bold>We found no replicated evidence of intervention effectiveness in decreasing functional decline. Interventions associated with slower functional decline in individual trials have been individually delivered and tailored to the needs of the person with dementia. This is consistent with previous findings. Future intervention trials should prioritise these approaches. [ABSTRACT FROM AUTHOR]