|Efficacy and feasibility of nonpharmacological interventions for neuropsychiatric symptoms of dementia in long term care: a systematic review
|Seitz DP, Brisbin S, Herrmann N, Rapoport MJ, Wilson K, Gill SS, Rines J, Le Clair K, Conn D
The review found that several non-pharmacological interventions may improve neuropsychiatric symptoms of dementia in individuals in long-term care, including staff training, mental health consultation, psychosocial activities and sensory stimulation. It also found that insufficient high-quality evidence was available and that the feasibility of some interventions appeared limited. Despite some limitations in the review, these conclusions appear reliable.
To evaluate the efficacy and feasibility of non-pharmacological interventions for neuropsychiatric symptoms of dementia in individuals in long-term care.
MEDLINE, EMBASE, PsycINFO, The Cochrane Library and Google Scholar were searched from 1980 to 2010. Search terms were reported. The reference lists of retrieved studies, key articles, relevant reviews and guidelines were checked. The search was restricted to studies in English.
Eligible studies were parallel-group randomised controlled trials (RCTs) that compared non-pharmacological interventions versus usual care, medication or an alternative non-pharmacological intervention for individuals with neuropsychiatric symptoms of dementia. At least 50% of participants in included studies were required to be receiving long-term care.
Most participants in the review had relatively advanced dementia, with mean Mini-Mental State Examination scores (where reported) that ranged from 5 to 10. In most studies, neuropsychiatric symptoms ranged from relatively mild to moderately severe. The median mean participant age was 84 years. in studies that reported gender distribution, most participants (78%) were women. Types of intervention included training of staff, individualised geriatric mental health assessment, psychosocial activities, exercise, music therapy, and other types of sensory stimulation. In most studies specialised staff external to the long-term care setting were required to deliver staff training or implement the intervention. Most studies measured outcomes with neuropsychiatric symptom scales, and 20 different scales were used. Median study duration was 12 weeks (range one to 52). Most studies were set in Europe or North America (where reported).
Two reviewers selected the studies, resolving disagreements by discussion.
Assessment of study quality
The Cochrane Risk of Bias Tool was used to assess the following aspects of study quality: random sequence generation; allocation concealment; blinding; completeness of outcome data; selective outcome reporting and other potential forms of bias (including funding source). Risk of bias was rated as low, unclear or high for each domain.
Two reviewers independently assessed study quality.
For continuous outcomes, mean differences from baseline in each study group were extracted or calculated, with standard deviations and (in some cases) p values for differences between the groups. For binary outcomes, the proportion of participants with the outcome in each group was reported. The potential feasibility of interventions was rated as high, medium, low or unclear using a tool designed by the authors. The tool assessed three items: time required to train staff or implement the intervention; requirement for access to specialised mental health services; and direct monetary costs for equipment. Where no primary outcome was reported, the change in total neuropsychiatric scores was used (if possible). Where outcomes were measured at multiple time points, the first endpoint after the intervention was reported in the review.
Two reviewers independently extracted the data and reviewed it for consistency.
Methods of synthesis
Studies were organised by type of intervention. Data were not statistically combined due to clinical and methodological heterogeneity between the studies. The studies were combined in a brief narrative synthesis using a "vote-counting" approach, and individual study findings for efficacy and feasibility were presented in online tables.
Results of the review
Forty studies were included, with over 3,500 participants (range 16 to 306): 13/40 studies were at low risk of bias for sequence generation, 1/40 for allocation concealment, 23/40 for blinding, 30/40 for completeness of outcome data, 37/40 for selective reporting and 39/40 for other forms of bias. Study reporting was often too poor to allow an adequate assessment of risk of bias
Sixteen out of 40 studies reported that the intervention was associated with a significant improvement in neuropsychiatric symptoms according to at least one measure. Improvement was usually of small magnitude and only two studies reported a clinically significant benefit from the intervention. There was no significant difference between the groups in the other 24 studies. No studies compared non pharmacological versus pharmacological interventions
Feasibility was rated as low or medium with respect to staff training requirements for 39/40 interventions, with respect to direct costs for 14/40 interventions and respect to time requirements for 19/40 interventions.
Several non-pharmacological interventions may improve neuropsychiatric symptoms of dementia in individuals in long-term care. Interventions included staff training, mental health consultation, psychosocial activities and sensory stimulation. However, insufficient high-quality evidence was available and the feasibility of some interventions appeared limited.
The objectives and inclusion criteria of the review were clear and relevant sources were searched for studies, but restriction by language meant that studies may have been missed. It was unclear whether the search was also limited by publication status, and the risk of publication bias was not discussed. Appropriate steps were taken to reduce the risk of reviewer bias and error in conducting review processes. Few details were reported about the interventions or the comparators (commonly "usual care").
As the authors stated, the studies were methodologically and clinically too different to combine, and many studies were limited by failure to specify a primary outcome, reporting of multiple outcomes and small size. It was unclear how the review authors determined which of the statistically significant study findings were likely to be clinically meaningful. The assessment of feasibility appeared reasonable but as most trials were done in North America the generalisability of the findings to other settings was uncertain. Despite some limitations in the review, the authors conclusions appear reliable.
Implications of the review for practice and research
Practice: Authors did not state any implications for practice.
Research: Authors stated that studies were needed to assess the effectiveness of non pharmacological interventions for neuropsychiatric symptoms of dementia implemented as part of routine care by long-term care staff, and to develop pragmatic strategies to manage such symptoms.
Canadian Institutes of Health Research Knowledge Synthesis Grant.
Seitz DP, Brisbin S, Herrmann N, Rapoport MJ, Wilson K, Gill SS, Rines J, Le Clair K, Conn D. Efficacy and feasibility of nonpharmacological interventions for neuropsychiatric symptoms of dementia in long term care: a systematic review. Journal of the American Medical Directors Association 2012; 13(6): 503-506.e2
Subject indexing assigned by NLM
Aged; Behavior Therapy; Caregivers /education; Clinical Trials as Topic; Dementia /psychology /therapy; Evidence-Based Medicine; Exercise Therapy; Humans; Music Therapy; Psychomotor Agitation /psychology /therapy
Date bibliographic record published
Date abstract record published
This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.