|Impact of cholinesterase inhibitors on behavioral and psychological symptoms of Alzheimer's disease: a meta-analysis
|Campbell N, Ayub A, Boustani M A, Fox C, Farlow M, Maidment I, Howard R
The authors concluded that cholinesterase inhibitors improved behavioural and psychological symptoms in patients with Alzheimer's disease. However, the review showed that this improvement applied only to patients with mild to moderate Alzheimer's disease, not to those with severe Alzheimer's disease. In addition, the review suffered from a number of methodological limitations, in particular the possibility of publication bias.
To determine the efficacy of cholinesterase inhibitors in improving behavioural and psychological symptoms of dementia in people with Alzheimer's disease.
MEDLINE, CINAHL and the Cochrane Central Register of Controlled Trials were searched from inception to 2007. Search terms were reported. References of included studies were screened. The review was limited to studies published in English.
Randomised controlled trials (RCTs) that compared the cholinesterase inhibitors (donepezil, rivastigmine or galantamine) to placebo in patients with Alzheimer's disease, in any clinical setting, were eligible for inclusion. RCTs had to measure behavioural and psychological symptoms of dementia with the Neuropsychiatric Inventory (an informant-based structured interview with acceptable psychometric properties).
In included trials, Mean Mini Mental State Examination scores ranged from 6.1 to 21. Trials were conducted in outpatient settings, nursing homes, and in community and residential settings combined. Trial duration ranged from three to 12 months. The dose of donepezil was 10 mg/day, galantamine doses ranged from 16 to 24 mg/day and rivastigmine was administered either as a capsule (three to 12 mg/day) or as a skin patch (20 cm2 per day). The mean age of included patients ranged from 72.5 to 85.7 years and the proportion of women ranged from 57 to 85%, where reported. Mean baseline Neuropsychiatric Inventory scores ranged from 9.2 to 23.7.
The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.
Assessment of study quality
Trials were assessed for methodological quality using the Unites States Preventive Services Task Force guidelines. The review was restricted to trials that had attrition rates of less than 40%, concealed measurement of outcomes and used an intention-to-treat analysis.
The authors did not state how many reviewers performed the validity assessment.
Three reviewers independently extracted data on mean total Neuropsychiatric Inventory score, mean change from baseline of the total Neuropsychiatric Inventory score for both treatment and placebo groups, and level of significance for the mean difference in Neuropsychiatric Inventory between the two treatment groups. Authors were contacted for missing data.
Methods of synthesis
Weighted mean differences, the difference in the change in total Neuropsychiatric Inventory score from baseline between treatment and placebo groups, were calculated. Standardised mean differences, the weighted mean difference divided by the pooled standard deviation, together with 95% confidence intervals were also calculated for each trial. The authors stated that data were primarily pooled using random-effects models, due to heterogeneity between trials. Sensitivity analysis was conducted to investigate the effects of Mean Mini Mental State Examination scores, stage of symptoms and age. The authors did not state that heterogeneity was assessed, but details of heterogeneity assessment were included on forest plots which reported data on the χ2 and I2 tests.
Results of the review
Twelve randomised controlled trials (RCTs) were included (n=4,921 patients). Three trials were excluded from the meta-analysis as they did not provide data on the mean difference of the Neuropsychiatric Inventory scores.
Cholinesterase inhibitors lead to greater improvement in Neuropsychiatric Inventory scores than placebo (pooled weighted mean difference -1.38, 95% confidence interval (CI): -2.30 to -0.46, n=3,508 patients).
Restriction of the analysis to the four trials conducted in 1,526 patients with mild to moderate Alzheimer's disease also showed a beneficial effect of cholinesterase inhibitors on Neuropsychiatric Inventory scores (weighted mean difference -1.92, 95% CI: -3.18 to -0.66).
The five trials conducted in 1,982 patients with moderate to severe Alzheimer's disease showed no significant difference in Neuropsychiatric Inventory scores between treatment groups.
There was little heterogeneity between trials for any of the comparisons (p>0.1, I2=34-36%). Results of the other subgroup analyses were not reported. Results for fixed-effect models were also reported and were illustrated using forest plots.
Cholinesterase inhibitors improved behavioural and psychological symptoms of dementia in patients with Alzheimer's disease, but the clinical relevance of this effect was unclear.
The review addressed a focused question supported by clearly defined inclusion criteria. The search was adequate, but the review was limited to published English language studies, so there was a possibility of language and publication bias. Appropriate steps were taken to minimise bias and errors in the extraction of data, but it was unclear whether such steps were also taken for the selection of studies and assessment of trial quality. Trial quality was assessed, but very limited details were reported on the criteria used or the results. Only trials fulfilling certain quality criteria were included in the review. It may have been more informative to have included all trials that fulfilled the inclusion criteria and to then have investigated the effects of trial quality on the results.
Methods of analysis appeared appropriate, but the reporting was somewhat confusing. It was unclear why the authors reported different outcome measures for the fixed-effect and random-effects meta-analysis, rather than focusing on the most informative measure, which, in the case of this review, would have been the weighted mean difference. The authors stated that a number of subgroup analyses were conducted to investigate confounding factors. As the studies included in the review were randomised controlled trials, confounding was unlikely to have been a problem, and such subgroup analyses provided information on differences between trials rather than on confounding. Results were only reported for the subgroup analysis stratified on disease severity, so it was unclear whether results also differed according to the other criteria that the authors stated that they investigated. The review showed that cholinesterase inhibitors lead to improvements in patients with mild to moderate Alzheimer's disease, but not in those with more severe Alzheimer's disease. The authors' conclusions do not take this into account.
Implications of the review for practice and research
Practice: The authors did not state any implications for practice.
Research: The authors stated that there is a need for a more clinically focused evaluation method for behavioural and psychological symptoms of dementia symptoms. This could either be achieved by conducting a new clinical trial or through individual patient data meta-analysis of the trials included in this review. This would allow investigation of the individual domain scores on the Neuropsychiatric Inventory and also a categorical approach to identify responders.
Campbell N, Ayub A, Boustani M A, Fox C, Farlow M, Maidment I, Howard R. Impact of cholinesterase inhibitors on behavioral and psychological symptoms of Alzheimer's disease: a meta-analysis. Clinical Interventions in Aging 2008; 3(4): 719-728
Subject indexing assigned by NLM
Alzheimer Disease /drug therapy /psychology; Behavior /drug effects; Cholinesterase Inhibitors /therapeutic use; Humans; Randomized Controlled Trials as Topic
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.