Study designs of evaluations included in the review
Experimental designs ranging from randomised controlled trials (RCTs) to uncontrolled before-and-after studies were eligible for inclusion.
Specific interventions included in the review
Studies delivering an intervention with exercise as a primary component were eligible for inclusion. The interventions delivered by the included studies were varied and were of an exercise intervention alone or in combination with other components such as social interaction, sleep hygiene measures, light exposure and relaxation. The exercise components of the interventions included walking, exercise to music and seated exercise; in several studies multiple forms of exercise were used. The intensity and duration of the interventions also varied. Where reported, the interventions were implemented mainly by therapists or researchers, but also by volunteers and carers.
Participants included in the review
Studies of participants with some degree of cognitive impairment, a diagnosis of dementia, or for ambulant nursing home residents a score of less than 23 out of 24 on the Mini-Mental State Examination (MMSE), were eligible for inclusion. In the majority of the included studies the participants were exclusively living in care homes; there were also some studies of people living in their own homes and of mixed populations. Where reported, the mean MMSE score ranged from to 7.3 to 23.9 (for intervention or control group or both combined). The comparators included social visits of similar frequency to the exercise intervention, no intervention and an alternative exercise intervention.
Outcomes assessed in the review
Studies assessing behaviour, mood, sleep or functional ability were eligible for inclusion. Details of the specific measures used in the included studies were not provided, but the general outcomes reported included night-time awakening, depression, physical function and activity, mood, group behaviour, walking endurance capacity, mobility, strength and various behavioural problems.
How were decisions on the relevance of primary studies made?
The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.