Leinonen_2015_PLoS.One_10_e0124500

OBJECTIVE: To determine whether there are differences in age and sex distribution and presence of comorbidities between participants included in randomized controlled trials of acetylcholinesterase inhibitors and nationwide cohort of persons with Alzheimer's disease.
METHODS: PubMed, Scopus and Cochrane Library databases were searched for original articles from their inception to January 4, 2015. Double-blind randomized controlled trials with donepezil, rivastigmine or galantamine compared to placebo in participants with Alzheimer's disease were included. Data from a nationwide cohort of persons with clinically verified diagnoses of Alzheimer's disease was defined as a reference population.
RESULTS: 128 full-text articles were assessed for eligibility, 31 of them fulfilled criteria. Mean age of participants in randomized controlled trials (n = 15,032) was 5.8 years lower (95% CI 5.7 to 5.9, P < 0.001), compared to the mean age of 79.7 years in the reference population with Alzheimer's disease (n = 28,093). Most of the articles did not report age distribution of participants. The proportion of women was 63.2% (9,475/14,991) in randomized controlled trials and 67.8% (19,043/28,093) (P < 0.001) in the reference population. Information on comorbidities and use of concomitant drugs were lacking or poorly reported in most articles.
CONCLUSIONS: There is a discrepancy between participants in randomized controlled trials of acetylcholinesterase inhibitors and real-life population with Alzheimer's disease. Participants in randomized controlled trials were significantly younger. Further, more detailed reporting of age distribution, comorbidities and concomitant drugs would be important information for clinicians when evaluating conclusions from randomized controlled trials to real-life practice. The existing recommendations of inclusion of older people should be followed to ensure safe pharmacotherapy for older people.