NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Hansen RA, Gartlehner G, Kaufer DJ, et al. Drug Class Review: Alzheimer's Drugs: Final Report [Internet]. Portland (OR): Oregon Health & Science University; 2006 Jun.

Cover of Drug Class Review: Alzheimer's Drugs

Drug Class Review: Alzheimer's Drugs: Final Report [Internet].

Show details


A. Literature search

We searched MEDLINE, Embase, The Cochrane Library, and the International Pharmaceutical Abstracts to identify articles relevant to each key question. We used either Medical Subject Headings (MeSH or MH) as search terms when available or key words when appropriate. We combined terms for the selected indication (Alzheimer's disease), drug interactions, and adverse events with a list of five specific Alzheimer's drugs (donepezil, galantamine, rivastigmine, tacrine, and memantine); extended release dosage formulations were included in this search. We limited the electronic searches to "human" and "English language", and searched sources from 1980 to 2005 (December) to identify literature relevant to the scope of our topic. We used the National Library of Medicine publication type tags to identify reviews, randomized controlled trials (RCTs), and meta-analyses. We manually searched reference lists of pertinent and relevant review articles and letters to the editor. All citations were imported into an electronic database (EndNote 8.0).

Additionally, we hand-searched the Center for Drug Evaluation and Research (CDER) database to identify unpublished research submitted to the FDA. Finally, the Center for Evidence-based Policy at the Oregon Health and Science University (OHSU) contacted pharmaceutical manufacturers and invited them to submit dossiers, including citations, using a protocol available at We received dossiers from two pharmaceutical companies.

Our searches found 1,109 citations, unduplicated across databases. We found an additional 58 articles from manually reviewing the reference lists of pertinent review articles. We included no studies originating from pharmaceutical dossiers; all studies submitted from pharmaceutical dossiers were present in our other searches. The total number of citations included in the database was 1,167.

B. Study selection

Two persons independently reviewed abstracts; if both reviewers agreed that the trial did not meet eligibility criteria we excluded it; we obtained the full text of all remaining articles. Records were considered for exclusion if they did not meet pre-established eligibility criteria with respect to study design or duration, patient population, interventions, outcomes, and comparisons to Alzheimer's medications outside our scope of interest.

For this review, results from well-conducted, head-to-head trials provide the strongest evidence to compare drugs with respect to efficacy, effectiveness, and adverse events. We defined head-to-head trials as those comparing one Alzheimer's drug with another. Included studies were RCTs lasting at least 12 weeks that had an outpatient study population with a total sample size greater than 100 participants.

If we could not find sufficient evidence of efficacy or effectiveness from at least one randomized, double-blinded, head-to-head trial, we reviewed randomized, controlled, open-label trials. For comparing different drugs, however, the strength of evidence must be rated lower for these results than for results from blinded trials.

If no head-to-head evidence was published, we reviewed placebo-controlled trials. We reviewed all placebo-controlled trials to provide an overview of efficacy without taking drug equivalency into account. Compared to placebo and all other things equal, higher dosages may yield greater treatment effects than do low or medium dosages. For that reason, we did not evaluate the dosage of one drug relative to the dosage of an alternative drug in a different trial. In addition, heterogeneity among study populations and placebo groups demand caution in making comparative judgments about treatment effects across trials.

We examined adverse events in both experimental and observational studies. For observational studies we included those with large sample sizes (> 100 patients) that lasted at least 1 year and reported an outcome of interest.

We initially reviewed studies with health outcomes as the primary outcome measures. Outcomes were institutionalizations, behavioral symptoms (e.g., aggression, agitation, mood disorders, psychosis), discontinuation effects, mortality, and changes in the rate of decline in day-to-day functioning and activities of daily living. Because health outcomes often were not reported, we also included intermediate outcomes (e.g., cognition, global assessment). Safety parameters included overall and specific adverse events (e.g., hepatotoxicity, weight loss, and gastrointestinal symptoms), withdrawals due to adverse events, discontinuation effects, and drug interactions.

We included meta-analyses in our evidence report if we found them to be relevant for a key question and methodologically sound (based on the QUORUM17 statement); we did not review individual studies if they had been included in a high-quality meta-analysis. We excluded meta-analyses that were not based on a comprehensive systematic literature search or did not maintain the units of the studies in their statistical analyses. We included recent pooled analyses of RCTs if they covered all published trials and their methods were sound. We checked our database to ensure that our literature search had identified trials included in any meta-analyses that we discarded; we then obtained any missing articles so that all constituent studies would be represented in this review.

If we could not find sufficient evidence of efficacy or effectiveness from at least one randomized, double-blinded, head-to-head trial, we reviewed placebo-controlled trials and randomized, controlled, open-label trials. For comparing different drugs, however, the strength of evidence must be rated lower for these results than for results from the preferred type of trial. Findings of placebo-controlled trials are hard to compare across studies because disparate populations may respond differently.

Overall, we reviewed 1,167 article abstracts and retrieved 206 of those as full text articles for background information or to be reviewed for inclusion into the evidence report.

C. Data abstraction

We designed and used a structured data abstraction form to ensure consistency in appraising each study. Trained reviewers abstracted data from each study and assigned an initial quality rating; a senior reviewer read each abstracted article, evaluated the completeness of the data abstraction, and confirmed the quality rating. We abstracted the following data from included trials: study design, eligibility criteria, intervention (drugs, dose, duration), additional medications allowed, methods of outcome assessment, population characteristics, sample size, loss to follow-up, withdrawals attributed to adverse events, results, and adverse events reported. We recorded intention-to-treat (ITT) results if available.

D. Quality assessment

We assessed the internal validity (quality) of trials based on predefined criteria (Appendix C). These criteria are based on those developed by the US Preventive Services Task Force (ratings: good, fair, or poor)18 and the National Health Service Centre for Reviews and Dissemination.19 We assessed external validity (generalizability) and reported on it, but these assessments did not influence our quality ratings.

Two independent reviewers assigned quality ratings; they resolved any disagreements by discussion and consensus or by consulting a third independent party. Elements of internal validity assessment included, among others, randomization and allocation concealment, similarity of compared groups at baseline, use of ITT analysis, and overall and differential loss to follow-up.

Loss to follow-up was defined as the number of persons randomized who did not reach the endpoint of the study,20 independent of the reason and the use of ITT analysis. We adopted an overall loss to follow-up of 40% as a cut-off point for poor quality.

We rated trials that had a fatal flaw in one or more categories as poor quality; we did not include them in this analysis. We rated trials that met all criteria as good quality. The majority of trials received a quality rating of fair. This includes studies that presumably fulfilled all quality criteria but did not report their methodologies to an extent that answered all our questions. Thus, the "fair quality" category includes trials with quite different strengths and weaknesses and a range of validity.

The last observation carried forward (LOCF) method of data analysis is a particular issue in Alzheimer's disease. The reason is that the natural course of the disease leads to a gradual decline in cognition and daily functioning over time. Particularly in longer studies measurements carried forward can bias results towards an overestimation of the treatment effect. We took this potential bias into consideration when we appraised each study and highlighted possible bias in the text whenever appropriate.

Copyright © 2006, Oregon Health & Science University, Portland, Oregon.
Bookshelf ID: NBK10315


Other titles in this collection

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...