ABSTRACT
OBJECTIVES: To investigate the inclusion of older people in clinical trials of recently authorized medicines, evaluating adherence to the 20-year-old International Conference of Harmonisation (ICH) guideline on geriatrics (E7). DESIGN: Observational. SETTING: European public assessment reports, published clinical trials, World Health Organization International Clinical Trials Registry Platform. PARTICIPANTS: Individuals aged 65 and older and 75 and older. MEASUREMENTS: Number and proportion of randomized participants and all exclusion criteria of studies involving diseases characteristically associated with aging (venous thromboembolism, osteoporosis, atrial fibrillation) and diseases not unique to older adults (type 2 diabetes mellitus, depression, bipolar disorder, epilepsy). RESULTS: In 114 Phase II and III trials of 12 medicines, 43.1% of participants were aged 65 and older, and 16.1% were aged 75 and older. In trials involving diseases characteristically associated with ageing, 57% were aged 65 and older; 22% were aged 75 and older. In trials involving diseases not unique to old age, 9% were aged 65 and older, and 1% were aged 75 and older. Upper age limits were applied in 30.7% of the trials; the frequency did not differ between larger (sample size ≥ 500) and smaller trials (P = .36), although it was significantly lower in trials involving diseases characteristically associated with aging (18.0%) than in trials of diseases not unique to old age (45.3%; P = .002). Age-sensitive exclusion criteria, based on comorbidity (75.4%), concomitant medication (71.9%), and other criteria correlated with age (60.5%) were applied more frequently in larger trials (P < .02). CONCLUSION: Studies of diseases not uniquely associated with old age included an unacceptably low proportion of older people, contrary to the recommendations of the ICH E7 guideline. Although the proportion of older participants in trials of diseases characteristically associated with aging was appropriate for certain medicines, the use of age-sensitive exclusion criteria limits the representativeness of the trial population for daily practice.
