A major part of value-based health care is the measurement and dissemination of outcomes using comparative effectiveness research. However, there hasn’t been much research done in that area. Dr. Michael Recht, chair of the department of radiology at NYU Langone Medical Center, took some time to speak with HealthCare Business News about his new collaborative training program in biomedical big data and comparative effectiveness research for the field of radiology.

HCB News: Can you explain what comparative effectiveness research (CER) is?
Dr. Michael Recht:What we are really doing is looking to demonstrate the value of imaging — when we should do imaging and what imaging is appropriate. Rather than doing it anecdotally, we need to do statistically powerful studies to find out which imaging test makes a difference in the outcomes of patients.

We’re trying to look at the comparative effectiveness of different treatment modalities. For example, if someone comes into the emergency room and you think they may have appendicitis, do you want to do any imaging at all? If you want to do imaging, is the right imaging ultrasound, CT or MR? You really decide, not anecdotally or by individual experience, but with a large study with statistical power to decide what is the appropriate way to work-up that patient — which has the most positive impact on the outcome of the patient as well as being cost effective?

HCB News: Why is CER training only available to a very small subset of medical imagers?
MR: It’s not part of the traditional curriculum for radiology residents and imagers. It’s expensive and it requires a lot of training and a lot of people to teach you. For any individual program to do that, it would be incredibly difficult. The best that an individual residency could do is to give a few introductory lectures, but you’re not going to get intense training. There are some master’s programs or more intense courses, but they tend to be very expensive.

Here at NYU we have a master’s program that a few of our faculty have attended, but it’s an expensive program and most radiology programs really don’t have the resources to do that. We’re also lucky that NYU has such a great program — we have a number of people in the medical school who really work on comparative effectiveness. But many programs don’t have that resource at their medical schools, so it would mean going someplace else to do the program.

What we decided was that in order to train imagers so that they can actually do comparative effectiveness research on imaging, we would try to develop a collaborative course that would be available to all imagers. Although we are starting with radiology, we really plan on expanding this to other imagers as well, such as cardiologists. It’s important not just for the imaging that is now within a radiology department, but with any type of imaging.

HCB News: How can big data analytic skills and comparative effectiveness techniques demonstrate the value and critical role of imaging?
MR: You need to look at a large amount of data. Imaging is big data in and of itself — we’re talking about a significant amount of data when you’re looking at an MR scan, a PET/CT scan or a PET/MR scan. PET/MR is a new hybrid modality wherein the equipment is very expensive and the question everyone is asking is, “Does it make a difference?” Should we be buying this equipment? If we buy this equipment, what are the applications that we should be using it for? What are the conditions we are using it for?

You need to do a large amount of analysis of the imaging data, and in addition you have to look at data that’s in the electronic medical record. Imaging tends to be an indirect effect, so it’s not like surgery, where you go to surgery and immediately know whether the patient does better or not. In imaging, you do an imaging test but then it takes a long time to learn the outcome of the patient. That imaging test might lead to surgery, medical care or drug therapy. There are a number of different outcomes that are possible, based on the imaging exam. You need to be able to get all of that data, collate it, curate it and analyze it so you can actually come out with definitive answers as to which type of imaging really has the most beneficial effect on the patient’s outcome.

HCB News: Can you briefly describe what your two programs will involve?
MR: We created a two-tiered program. The first tier is an introduction to big data and comparative effectiveness and we wanted to make sure that every resident had a chance to learn the basics. The way we accomplish that is to use an American Institute of Radiologic Pathology course, which is run by the ACR. Essentially, every resident in America goes to this four week course and almost all of the residents in Canada go, as well as several residents from outside of North America.

We are going to be giving eight hours of introductory lectures on big data and comparative effectiveness to all of the residents that go to the AIRP course. We know that these eight hours aren’t going to teach people the skills so they can go out and do comparative effectiveness research or analyze big data, but what we are trying to do in a sense is whet their appetite and introduce them to these subjects. The second tier is geared toward junior and mid-career faculty. This set of courses is geared to giving imagers tools to enable them to do research — not to become experts in the field — but to really be able to start doing some research.

This second tier is a two-year program. The first year is going to encompass five separate courses — two are on big data analytics skills and three are on comparative effectiveness. In the second year we are going to pair people up with a mentor so people will have a research project and they will be able to work on it with their mentor over the next year. At the end of that we are hoping that everyone will present the results of their research at a symposium at a national conference.



HCB News: You said you are doing this to develop the next generation of radiologist-investigators. What will that involve?
MR: Up until now, if you look at the radiology literature, people are doing a number of observational studies. The next step has to determine the value of doing an imaging study. Does the study allow the surgeons to do more appropriate surgery? Does it allow clinicians to determine which patients actually need the surgery or not? Is imaging really altering the management of patients?

We have to make sure that we are integrating with our clinical referring physicians so that our research is not only proving that we can do something, but what we do really makes a difference for the patient. That is what I think is the next step, and to accomplish this we have to do much larger studies than we have done traditionally in imaging.