Kevin Kerber, MD
On Improving Care and Reducing Cost for BPPV Treatment in the ER
For Kevin Kerber, MD, academic interests have always sprung from real world observations and experiences. After Dr. Kerber tore his ACL during a high school soccer game, he wanted to become an orthopedic surgeon. Later, as he moved through his clinical rotations, he observed inefficiencies in health care delivery first-hand, and became drawn to health services research. Then, while working with Robert W. Baloh at the University of California in Los Angeles, he saw in neuro-otology a field where he could make a real impact on clinical care. Dr. Kerber, assistant professor of neurology at the University of Michigan Health System, has been working to integrate his interests in health services research and neuro-otology ever since.
At the American Neurological Association (ANA) annual meeting last October, Dr. Kerber was honored for these efforts with the prestigious Derek Denny-Brown Neurological Scholar Award, which is given to a newly elected member of the ANA who has achieved a significant stature in neurological research and who promises to continue making major contributions to the field of neurology.
Dr. Kerber received his medical degree from the University of South Carolina School of Medicine in 2000. He then completed an internship in internal medicine at Evanston Hospital in Evanston, Illinois in 2001, and his residency in neurology in 2004 at the University of Michigan. In 2005, Dr. Kerber completed fellowship training in neuro-otology at UCLA, and then received an MS degree in health and health care research from the University of Michigan Rackham Graduate School.
In an interview with Neurology Today, Dr. Kerber discussed his current interventional research in neuro-otology, and offered advice to other neurologists on improving inefficiencies in the health care system.
HOW DID YOU BECOME INTERESTED IN NEURO-OTOLOGY AND HEALTH SERVICES RESEARCH?
After my first year of medical school, I was interested in geriatrics, so I did a research project with Bob Baloh at UCLA. His research focused on balance disorders, and after I spent time seeing patients with him, I became fascinated by the presentation and evaluation of patients with dizziness.
Later, when I went into my clinical rotations and then actually during residency, I saw that many people who were evaluating dizziness were not using the tools that we think that can help people the most. Instead, a lot of people were using methods that were really unlikely to help people. I thought this was a niche area where I had the opportunity to make a difference. That then evolved into my interest in health services research — trying to get at these things from the aggregate level.
WHAT ARE THE SPECIFIC RESEARCH QUESTIONS AND INTERVENTIONS THAT YOU ARE WORKING ON?
We actually just received a new grant to develop an intervention in the emergency room setting, and then to test our intervention. Our goal is to promote the valuable tests for patients with dizziness, as well as reduce those things that are not necessary.
To get to this point, we first had to define routine practice using clinical epidemiology — with both national databases and also with a career development award to do a population-based study based on dizziness in the emergency room. My mentor here at the University of Michigan, Lewis Morgenstern, had done about ten years of epidemiologic work using a population-based study with a specific community [the Brain Attack Surveillance In Corpus Christi, Texas], so I was able to piggyback on top of his project, which has already developed the infrastructure to do this kind of work.
The most common cause of dizziness is BPPV (benign paroxysmal positional vertigo), and research has shown that there are relatively simple, low-cost tests and treatment maneuvers for this condition. One of these maneuvers, called the Epley maneuver, has been extensively studied in both randomized-controlled trials and in systematic reviews. The test for BPPV is the Dix Hallpike test. Both the test and treatment are inexpensive, effective, and easy to use, but in our work, we found that these were very rarely documented in the physician's encounter note, even when the physician diagnosed the patient with BPPV. However, CT scans, expensive but extremely ineffective tests in presentations suggestive of BPPV, were used often. We've done research that patients who have dizziness symptoms who present to the emergency room and get a CT scan are much more likely to have a longer length of stay than patients who don't get a CT scan. A CT scan also exposes the patient to radiation. In fact, patients diagnosed with BPPV in this setting are much more likely to be radiated — that is receive a CT scan — than they are to be rolled over and cured — that is treated with the Epley maneuver.
So, we think [BPPV] is a low hanging fruit in dizziness. The goal of our intervention is to develop tools that support and encourage optimal diagnosis and treatment of BPPV with an emphasis on the Dix-Hallpike test and the Epley maneuver, and a de-emphasis on CT scans or high-cost, low-benefit examinations.
HOW DO YOU DISSEMINATE AND ENSURE THAT AN EFFECTIVE THERAPEUTIC PROTOCOL IS IMPLEMENTED?
That's not easy. Our current study is focused on implementation.
My belief is that we assume that once research is done with randomized controlled trials that everybody should just be able to learn these things. But the problem is that most of these trials that define important interventions are done in expert centers, and people usually fail to then do the next steps of research to actually try to get it into clinical care — and that's what I think has been more of a failure of the academic community to help the frontline providers.
In our study, we're starting out with qualitative research where we are interviewing emergency room doctors to better understand the barriers and facilitators to testing and treating BPPV, and optimal use of neuroimaging. So, that's the first step. Then, we also want to try to understand what is likely to help them in care, and what's likely to change behavior.
We want to create tools that doctors can use at the point of care — quickly and efficiently that get them the information at the time that they need it and in a convenient way so that it actually has a positive impact on their care — and all of those are not easy to do.
First, it's not easy to change physician behavior once they've developed a habit. Then it's also a challenge to get people tools that they use right at the point of care — perhaps they would even be embedded within the electronic medical record — but in a way that it really conveys the critical information within a time frame that does not hinder their ability to use it.
HOW CAN NEUROLOGISTS, PERHAPS IN DIFFERENT FIELDS, CONTRIBUTE TO HEALTH SERVICES RESEARCH AND IMPROVEMENT?
There are many opportunities for people to do this. People can take whatever their area of specialty is and try to observe: what are the things that help people that aren't being used in the real world? And how can we get that information to doctors at the front lines of care who are likely to be seeing these patients as soon as they come in, and in many cases may be the primary managers of a neurologic disorder?
It also takes a tremendous partnership with people with other skills — for example, with behavioral psychologists and media specialists — to develop tools that physicians will actually use – and if you're doing implementation work, with people that are experts in implementation science.
A lot of people are now doing smaller scale studies which are referred to as quality improvement studies, and those are usually focused on single centers. The problem is it's hard to really develop something there that's going to be applicable to other centers as well. The methods that they use to test their interventions are often sub-optimal because they usually use pre-post designs rather than randomizing the intervention; they also tend to focus on surrogate outcome measures, rather than actual clinically meaningful outcome measures.
AT THE ANA, YOU MENTIONED THE RESPONSIBILITY CLINICIANS HAVE TO HELP NOT ONLY THE PATIENTS THEY SEE, BUT ALSO THE ONES THEY DON'T SEE.
I firmly believe that. I think that that's a product of our health care system, which pays doctors only for seeing a patient in the office. I think that this system is essentially tying doctors' arms behind their back and preventing them from doing things that can help people on a larger scale. I think there's a huge bias for doctors to only care about the patients who are directly in front of them. They don't hear the stories of the patients who could never get in to see them because of wait times or blocked access for insurance reasons. Then there's also all the stories of patients who do get access to doctors, but end up getting the wrong care. Those are people we can help too.
TUNE IN, LISTEN UP: Kevin Kerber, MD, assistant professor of neurology at the University of Michigan Health System, was honored with the prestigious Derek Denny-Brown Neurological Scholar Award at the American Neurological Association meeting last October. Here, he discusses his efforts to combine neuro-otology and health services research to produce meaningful outcomes on a larger scale: http://bit.ly/rCBryX.