Here we have assessed the inclusion, planned and actual, of elderly patients in COVID-19 trials on pharmacologic agents and devices. We show that elderly persons are underrepresented and demonstrate that no trial has specifically addressed them. Only one reported clinical outcomes in subgroup analyses of this population, the most frequently and severely affected by the disease. 19 Three of 12 published RCTs actively imposed an upper age limit on participants, as did 200 of 650 interventional trials identified at clinicaltrials.gov.
Some trials essentially impose a functional upper age limit even when no age criterion is explicitly stated. Of the published RCTs, four (42%) excluded patients due to cognitive capacity, eight (67%) due to liver disease, six (50%) because of severe kidney disease and five (42%) because of heart disease or risk of torsade de pointes. These are all conditions that are significantly more prevalent in patients aged ≥ 65 years.
Patients included in published RCTs had an average age of 56.3 years: they were younger than those included in observational studies.6,27 This difference calls into question the RCTs’ external validity, and should remind clinicians to exercise prudence when considering data from such trials for clinical decision-making in the elderly. The majority of RCTs were conducted in China. According to the Wuhan Aging Working Committee Office, by the end of 2019, 14.7% of the population in Wuhan county was ≥ 65 years, and 2.9% were 80 or older. 28 Interestingly, according to the United States Census Bureau, on April 2020, 14.1% of persons in New-York were aged ≥ 65 years. This suggests that cultural, rather than demographic, factors might be driving the younger ages observed in the Chinese trials discussed here.
There are understandable and sound reasons for the exclusion of elderly patients from some trials, particularly those designed for the early development of novel therapeutics. There is often limited experience in elderly populations with the drug of interest. 29 These patients have an increased risk of drug-drug interactions due to potential polypharmacy and age-related physiological changes affecting pharmacokinetics and pharmacodynamics. Remdesivir is contraindicated in patients with creatinine clearance < 30 mL/min and elevated liver function tests. 30,31 Hydroxychloroquine and lopinavir/ritonavir are known to cause QT prolongation with a possible increased risk of torsades de pointes. 32
Yet when drugs of interest are being given off-label to elderly patients essentially en masse, trial protocols should adapt to reflect the larger clinical reality around them, allowing for increased and more equitable representation of this population. Indeed, increased mortality was observed among hospitalized US veterans treated with hydroxychloroquine for COVID-19. This finding highlights the urgent need for RCTs expressly targeting the group most affected by COVID—and most likely to receive the drug off-label anyway. 33 In a similar fashion, Avni et al. showed that elderly patients were often excluded from RCTs assessing bacterial pneumonia and reported that the participants were significantly younger than in observational studies. 11 This serves to remind us the underrepresentation of elderly in RCTs in general. 34 The obvious under-representation of the elderly in COVID-19 trials is an acute manifestation of a larger problem: the elderly tend to be disproportionately excluded from RCTs in all domains. Elderly patients with cognitive, psychiatric or physical comorbidities are largely absent from the RCT “record”, leaving clinicians to rely on data from inferior studies such as retrospective case series and cohorts, which are notoriously unreliable due to confounding by indication and other biases.35
As the aging population continues to grow in size, medical research must better reflect this growing segment of the population. This is especially true regarding COVID-19, which is more common and more severe in elderly, causing devastating effects in nursing homes and LTCFs. Conducting clinical trials in elderly adults should compel the clinician to choose relevant outcomes; when planning an RCT, one must ask: “what is important to the elderly patient?” Such outcomes should include immediate but also long-term outcomes such as deterioration of cognitive and functional capacity, quality-of-life and the need for rehabilitation or LTCF placement. We found no study reporting or intending to report such outcomes. It is well established that severe infections have long-term consequences that continue well beyond the first month. 36 Rahmel at al showed that patients surviving sepsis had a better 5-year survival when benefiting from a rehabilitation program. 37 In an RCT of 72 COVID-19 patients (median age 69), Liu et al reported that a six-week respiratory rehabilitation program could improve pulmonary function and quality of life. 38 Finally, ethical standards should facilitate inclusion of elderly adults with more adapted informed consent, including the possibility to obtain consent by proxy if the patient has diminished capacity.
In conclusion, elderly persons are underrepresented in COVID-19 RCTs, although they are the demographic most frequently and severely affected by the disease. Clinical research including the elderly has never been easy; nevertheless, future trials will need to address this vulnerable and oft-forgotten population, particularly when these individuals are regularly receiving off-label therapies anyway.