Excluding Elderly from Clinical Trials Makes Real-World Practice of Medicine Harder

    • March 21, 2011

By some estimates, spending on clinical trials of new medicines and devices in the United States exceeds $25 billion a year. But a new article in the Journal of General Internal Medicine, co-authored by two Robert Wood Johnson Foundation (RWJF) Clinical Scholars at the University of Michigan and the Department of Veterans Affairs, reveals that many of those studies are omitting one of the largest and most rapidly growing segments of the population: the elderly.

Donna Zulman, M.D., and Jeremy Sussman, M.D., reviewed the published results of 109 clinical trials from 2007. They found that a fifth excluded patients above a specified age, and that almost half of the remaining studies used criteria likely to exclude the elderly disproportionately—frailty or impaired cognition, for example. Those exclusions mean “that doctors cannot be confident that clinical trial results apply to their older patients,” says Zulman, the lead author. “That poses a real challenge for doctors taking care of these patients. Given the lack of evidence relevant to older adults, doctors must rely on art rather than science. And while there is always some art involved in medicine, it’s obviously not ideal to have doctors working off the cuff, when they should instead be able to rely on actual data.”

The exclusion of seniors has implications for the health care system as a whole, as well. About one in eight Americans is over the age of 65 and, by 2030, that share is expected to almost double. Seniors account for 34 percent of personal health care expenditures, with the majority of spending attributed to individuals with chronic diseases. Those are the very patients being excluded from clinical trials, according to Zulman and Sussman.

Many clinical trials in the United States are conducted by drug companies seeking to gain approval of a medication. For the companies, it might make short-term economic sense to limit the pool of patients in the study to those who are healthy, apart from the condition being targeted by the drug or device being tested. But Zulman observes that such reasoning leaves doctors without the data they need to make good decisions. “We pour all this money into the trials, but then we can’t be certain if the results apply to many of our older patients—individuals with multiple medical problems,” she says. “It would be better to invest resources up front, designing clinical trials to give us information on how to use drugs and devices in this population.”

As an example of the benefits of including seniors in such trials, Zulman points to the 2008 findings from a landmark study of treatments for patients with type 2 diabetes. Unlike past studies on the subject, this one made special efforts to include older patients and those with longstanding diabetes. The study found that the then-common treatment approach of lowering patients’ blood sugar to near-normal levels did not reduce the risk of heart attacks and strokes among patients who were at especially high risk for such events. That finding surprised doctors, who had relied for many years on data from previous studies of younger diabetic patients that had found just the opposite. The new findings have caused doctors to reexamine their treatment approach, Zulman says, and many now recommend less stringent blood sugar goals for their older, diabetic patients.

Zulman and Sussman also found that clinical trials rarely assess how treatments affect physical function and quality of life—outcomes that are of great importance to many older patients. Only one in four clinical trials examined these types of outcomes. Zulman notes that, here again, such data-collection practices deny physicians and their elderly patients information invaluable to making choices from among various treatment options.

One way for researchers conducting clinical trials to help develop such knowledge with minimal additional expense is to capture and analyze data about specific subgroups of patients in their studies. This practice can provide information about differential treatment effects in older adults, or patient populations with certain characteristics. Indeed, the majority of the studies that Zulman and Sussman examined looked at subgroups of patients, but only three in five of these studies conducted an age-specific subgroup analysis. Furthermore, when trials did include subgroup analyses by age, researchers frequently did not use rigorous methods of data analysis, thus decreasing the utility of results.


The two scholars offer a number of recommendations. First, as Zulman says, “There is rarely a good reason to exclude older patients from trials on the basis of their age alone, unless the disease in question doesn’t occur in older patients.” More generally, they say researchers conducting clinical trials should seek pools of study participants that reflect the populations that would be treated with the devices or drugs being tested. In particular, if a tested drug is aimed at conditions common to older adults—heart disease, for example—older patients should be included in the trial. They note that the Food and Drug Administration has begun taking steps to accomplish that goal. “Researchers are now encouraged [by the FDA] to report clinical trial data by age and to identify differences in safety or effectiveness associated with age,” they write. “These guidelines could be strengthened further by requesting that trials report how well their study sample represents the population of interest, and by encouraging clinical justification for upper age limits and eligibility criteria that might disproportionately affect older individuals.”

The authors also argue for trials that include patients with multiple conditions, not just the one targeted by the tested device or drug. Such patients are typical in the real world where many seniors have multiple chronic conditions, but are rarely included in trials. In addition, they urge researchers to monitor trial participants for “outcomes that have greater relevance to older individuals, including quality of life, health status, and physical function, as well as side effects of special concern to these older patients, such as the exacerbation of geriatric conditions like incontinence, dementia, falls and adverse polypharmacy effects.”

Together with more rigorous data analysis of age-specific subgroups in trials, such steps could create vital knowledge that would help doctors treat the growing number of elderly patients, including those with multiple conditions. “Our population is aging,” Zulman notes, “and many older patients have multiple medical problems. These patients are complicated to treat and are expensive to the health care system. So it is really critical that we understand how drugs and interventions work in this population.”

RWJF Clinical Scholars Program Plays Key Role

Zulman credits her participation in the RWJF Clinical Scholars program with helping make her work in this area possible. “The program offers exceptional mentorship and coursework, as well as financial support for our projects,” she says. “What is most valuable, however, is the opportunity to interact with researchers and policy-makers who shape health care delivery and policy in this country. Through these interactions, we are encouraged to identify important problems facing our health care system, and we learn how to frame our research such that we can have an impact and help move health care in a positive direction.”

For more than three decades, the Robert Wood Johnson Foundation Clinical Scholars® program has fostered the development of physicians who are leading the transformation of health care in the United States through positions in academic medicine, public health and other leadership roles. Through the program, future leaders learn to conduct innovative research and work with communities, organizations, practitioners and policy-makers on issues important to the health and well-being of all Americans.