Study designs of evaluations included in the review
The majority of the studies selected were within-patient (repeated measures) designs; these included comparisons of pre- and post- intervention performance. Studies were excluded if they were theoretical or philosophical papers, or were case studies of one or two individuals. Only articles published in refereed journals and written in English were reviewed.
Specific interventions included in the review
Music and music therapy. Studies which did not include a control condition were excluded. Studies where music was used in the context of testing an alternate hypothesis, such as the effect of intergenerational interaction or comparative recall of song lyrics and spoken material, were also excluded.
Specific therapeutic techniques used in the included studies were: singing; instrument playing; dance or movement; listening to music (live or taped); musical games; and composing or improvising.
Participants included in the review
Participants with dementia were included. Studies evaluating the caregivers rather than the participants were excluded. The diagnostic test used to diagnose dementia varied between the included studies and was not specified in seven studies. The specific tests used were: Reisberg (Global Deterioration Scale, scores ranged from 5 to 7); Mini-Mental State Examination (scores ranged from less than 14 to less than); Clinical Dementia Rating (score of 2 or 3); and the Multidimensional Dementia Assessment (score of 3).
Two studies also based the inclusion of patients on behavioural aspects: one on the presence of disruptive vocalisations and the other on the presence of agitated behaviours.
Outcomes assessed in the review
The included studies assessed a variety of different outcomes. Studies that did not provide quantification of the dependent variable were excluded. The specific outcomes (dependent variables) measured in the included studies were: agitation; face-name recognition; disruptive vocalising; alert responses; participation; seating, wandering; sleeping; interaction, mood; sitting; social behaviour; amount of food consumed, irritability, fear-panic, depression; time with dinner; cooperative behaviour; general mini-mental status examination; reality orientation; and relaxation behaviour, pulse. The authors also examined the effect of the length of treatment on its effectiveness.
How were decisions on the relevance of primary studies made?
The authors do not state how the papers were selected for the review, or how many of the reviewers performed the selection.