Hospitalists are the de facto geriatricians of the inpatient setting, and therefore, must be adequately trained—and sensitive—to the needs of older patients. We spoke with Ethan Cumbler, MD, FACP, director of acute care for the elderly at the University of Colorado in Denver, about what it will take to improve the care of geriatric patients in the hospital setting—now, and in the future.
How did you become interested in the care of elderly patients?
Before I came to the university, I worked at a private hospital, and one of our roles was to co-manage geriatric trauma patients. Working in that role and being the primary admitting hospitalist group for a large geriatric clinic gave me a lot of exposure and interest to the issues that are unique to the elderly.
So in coming to the University of Colorado, where my role was not only a clinician but also an educator, I had a chance to sit down with the head of the hospitalists, Jeff Glasheen, MD, who had been reviewing the data on the knowledge deficits that exist between what we are teaching hospitalists and what hospitalist are doing in practice. And one of the greatest gaps was in geriatric medicine; that created an opportunity to create a new curriculum and new way of teaching future hospitalists how to care for elderly inpatients.
What is the hospitalist’s role in the treatment of elderly patients, and how has it evolved as the number of geriatric specialists has declined?
Like it or not, the hospitalist is the de facto geriatrician of the inpatient setting. There simply are not enough geriatricians to go around. And that would be less of a problem if hospitalists were well-trained in the care of elderly patients and if hospital systems were well-designed to deliver that care.
Being a good physician for elderly patients requires more than being a knowledgeable clinician with regard to geriatric physiology and issues. To truly provide quality care for elderly patients, you have to work within a system that is designed to care for that vulnerable population. Because the care of the elderly happens at the intersection of multiple disciplines—not just the physician one-on-one, but nursing care the patient receives, the physical environment of the hospital, the attention that goes into their social support, and their transition of care. And to a large extent, to get all of those pieces right takes much more than just one clinician trying to take good care of elderly patients. It requires a system that’s designed to consistently deliver good care.
One of the biggest challenges in treating elderly patients is in assessing cognitive function. How is that being addressed?
At University of Colorado, we do a standardized brief practical geriatric assessment for which we put together a couple of tools that allow us to rapidly get a sense of how the patient is functioning—cognitively, socially, and physically. For example, most hospitalists have probably been trained with the Folstein Mini-Mental State Exam (FMMSE), which is a thorough and validated screening tool for cognitive impairment. The problem with that test is that it takes a fair amount of time to complete, and answering a long battery of cognitive screening questions is low on the priority list for both the hospitalist and patient. So what we do is the Mini-Cog test, which is a simple three-item recall and a clock-drawing test that actually functions as well or better than the FMMSE, but provides the information much faster and in a way that is tolerable to the patient and fits into the workflow of a busy hospitalist.
Similarly, we do a test called the ‘get up and go test,’ which is as simple as having a patient stand up and walk 10 feet, turn around, and come back and sit back down. You can learn a great deal in 30 seconds about their strength, mobility, balance, and fall risk, and frankly, about their likelihood of being able to return home without additional support in a very short period of time.
In terms of discharge, what can be done to make care less fragmented and facilitate better communication between providers, especially when dealing with elderly patients?
This is an issue that I think every hospital and every hospitalist struggles with, regardless of the age of the patient. But the challenges in transitions are magnified in the elderly, because the complexity of their medical problems and the number of medications involved make this a target-rich environment for errors to occur. At University of Colorado, we mandate that the primary care physician (PCP) be contacted twice on every admission—once when the patient arrives at the hospital, and again when the patient leaves the hospital. The idea is to change the dynamic between the hospitalist and the PCP from a one-time handoff to a dialogue about the patient’s care. And I found this to be helpful, not only for the outpatient clinician, but also for the inpatient provider, in developing good care and transition plans.
With the elderly population, it is important to recognize that some of these patients have functional and cognitive barriers that make access to medications and follow-up care more problematic. So when we get patients who have functional limitations and limited social support, we might send that patient home from the hospital with the actual medication rather than with a prescription.
What have you found in your research of the use of vulnerability assessments on admission?
There is an adage in medicine that says discharge planning begins on admission, but how to operationalize this has never been well explained. What we looked at was using the Vulnerable Elders Survey (VES), which measures some activities of daily living and independent activities, along with age and self-reported health status, to try to predict who would be able to return home and who would need to go to a skilled nursing facility on discharge. And what we found was that patients with a lot of functional limitations prior to admission are very likely to end up in a skilled nursing facility. If a patient has a high score on the VES-13 indicating functional limitations, and our plan is to send them home, then we know that we need to make sure the patient has good social support. And if they don’t, we need to think about how we can make that plan better.
This is one of the other concepts that make geriatric medicine unique. In general internal medicine, most of the time, the relationship is between you and your patient, whereas in geriatric medicine—especially as you reach the extreme stage—you are often dealing with the patient in the context of their social support structure. And in coming up with good care plans for an elderly patient, you have to take into account their social support.
What do you think we can expect in terms of future research in the field of geriatric care?
There is a lot of work to be done in the inpatient setting with regard to geriatric patients. We all struggle with how to, as a system, identify delirium accurately; we struggle even more with how to create systems of care which are capable of preventing delirium, which goes beyond the individual clinician. We all struggle with how to prevent falls within the hospital, and the research is just beginning on how to accomplish those goals.
An area that I find particularly interesting is the issue of how to bring patients more fully into the process of their medical care. How do we move beyond the concept of patients as recipients of care, and into a concept of patients as partners in their care? I’ve done some research looking at whether patients understand what their medications are, and other researchers have looked at patients’ understanding of their care plan in the hospital, and we’ve found huge gaps. When we examine how often elderly patients are mobile within the hospital, we find that for the most part, elderly patients spend their time sedentary. If we were talking about younger patients, that might not make a difference, but deconditioning, which happens in the hospital, is one of the key hazards of hospitalization for this vulnerable population.
How do we create systems that nurture healthy behavior by elderly patients? How do we create safe systems that nurture appropriate provider behavior for this population? This is an area of research that I think is very interesting.