Discussion

Dr Spence Taylor (Greenville, SC): Congratulations on a wonderful presentation and also congratulations on an incredible effort. This is a real nostalgic presentation for us. In South Carolina, about 15 years ago, the state vascular society attempted this and actually put 3 years of carotid endarterectomy data together, accumulated over 1500 carotid endarterectomies, and we did it almost identical to the way you did it. A lot of the processes that you presented today were the same processes that we used.

There was a striking difference, however; yours worked, ours failed. And it failed and lost momentum after about 3 years for primarily two reasons. We had difficulty maintaining the financial aspect of the registry, which was a major part, and then second, we experienced a lack of perceived value, I believe, from the general practicing private practice surgeon. Time and money, of course, is a big issue with most practitioners, and I think at the end of the day they lost the initiative to participate because of a lack of perception of value.

Even when we, Jay Robison and I, put together the abstract examining our experience and submitted it to the national and regional vascular organizations, they did not accept the abstract. We ended up presenting the data at the Southeastern Surgical meeting and published it in the American Surgeon.

So, I think we may have been a victim of a different time. And I think we clearly see what you are doing with this and it is a little melancholy to see what we could have done. The process improvement potential, as you have shown, may be the greatest part of this initiative; but, unfortunately, we never got there. I congratulate you on an outstanding manuscript and an outstanding effort.

Dr Jack L. Cronenwett: I appreciate your remarks. Why has our registry worked and yours didn’t? There are many reasons why this worked, but I’ll mention three. First, doing this regionally to be able to meet together and develop trust was key. I think that focusing on quality and providing members with feedback that mattered was important. And I think, finally, getting buy-in from hospitals that were familiar with this method based on the previous work of the Northern New England Cardiovascular Disease Study Group and also getting an initial grant from CMS to allow us to begin our central processing of data were all critical in success.

Dr John Hallett (Charleston, SC): I have a couple of comments and one question. The first comment is that this regional quality improvement project would not have happened without Dr Cronenwett’s leadership. Having been an original member of the Study Group at Eastern Maine Medical Center, I saw so clearly the importance of someone with vision and leadership.

Second, your hospital has to invest in a clinical coordinator and data collector who helps your surgeons get the data recorded. And finally and most important, Dr Cronenwett has emphasized that everyone needs to convene twice a year to review the data and select a few critical items for quality improvement.

My question: how can we take this type of regional quality improvement model to a national level where others can use a similar system and then, perhaps, benchmark with this wonderful database that you have in New England?

Dr Cronenwett: As everyone knows, there are other registries. The STS has a registry that has been touted by CMS as being the model for a demonstration project around pay for performance. I met with people at CMS here in Baltimore during this visit, and they believe that this database should qualify, starting in January, for the physician quality reporting initiative.

There are other good databases like the NSQIP, which is being rolled out nationally. The advantage of our database is that it contains a large number of process-specific variables that we think are going to be important going forward as we try to understand not only what the results are but why are they different between hospital A and B. What are the two hospitals doing differently? Then, how can we make improvement?

So I do think that based on this, there may be an opportunity for a future roll out of a national vascular database to share benchmarking data but with regional administration to keep the size manageable and I am looking into that.

Dr John Ricotta (Stony Brook, NY): I’d like to make a comment. As maintenance of certification becomes important for all of us, each of us is going to have to participate in some database that monitors performance, either NSQIP or some other database, and this clearly would qualify. My question relates to cost and where the payment comes from. Many hospitals are already supporting several databases. Please give us an idea of how to get hospitals involved in this type of specialty specific activity.

Dr Cronenwett: I think your comment about maintenance of certification is important. And as you saw in one of the graphics, we have hospitals from 25 beds to 600 beds. So you can imagine that there are many different solutions in each hospital as to how data are collected. These include surgeon entry of data—exclusively in some hospitals—to research personnel to nurses on the floor. Various hospitals have done it different ways. But by and large, the larger hospitals have committed a part of an FTE to do this and they have borne that cost.

Now, fortunately, we have had a grant from CMS to do the central data processing. But I think each hospital has to commit to local support for data collection. And the hook for that, the way to attract them, is to start showing them results, to start showing them how we’re making improvements in length of stay, and how this process going forward can reduce costs.

Dr George Lavenson (Lahaina, Hawaii): Are there any plans to include the outcomes from carotid artery stenting as well as those of carotid endarterectomy? The reason that I ask is that it is important to ensure that any stenting that is done has results that are at least equivalent to very safely performed and available endarterectomy. In a three-step program for stroke prevention that we are presenting tomorrow, consisting first of screening and then diagnosis, the third step is a critical one of assuring that needed intervention is done safely.

Dr Cronenwett: That is a good point. And one of the lessons we have learned from this is don’t try to do too much too fast. But recognizing where carotid stenting was, we actually began almost 2 years ago also collecting appropriate data on carotid stenting. I didn’t present that today, but we now have almost 200 such patients in the registry and we hope to be bringing that forward at another time.

Dr Anil Hingorani (Brooklyn, NY): In one of your slides, there was a 25% incidence of morbidity and mortality on limb loss and amputations in one of the surgeons who is doing lower extremity bypasses. How did you get those types of surgeons to participate and volunteer these data? I don’t see how you would be able to do that.

Dr Cronenwett: Well, one of the things that we learned when we first started presenting these results in a small group setting is that the surgeons are very interested in improving their results. Most of them don’t know the details about their results. And when someone sees a result like that, the response is: “Gee, I see a lot of people have outstanding results, I want to know how you’re getting those.”

That’s been our experience. And I think it is because the surgeons who are doing this are doing it voluntarily, they are committed to improving, based on their participation, and our sense is that a surgeon with what appears to be poor outcome is motivated to do better.

Dr Giovanni Ferrante (Hingham, MA): It sounds like you have got great buy-in from the surgeons who participated and from the institutions. But there are at least two other groups at each institution that may be doing some of these procedures. There should be motivation on the part of the institutions to include the interventional radiologists and the cardiologists. Is there any movement in that direction?

Dr Cronenwett: We have invited anyone who is doing these procedures to participate. And since we started recording carotid stent procedures, for example, we now have some cardiologists who are participating in that phase of the registry. We have a couple neurosurgeons who are participating around carotid endarterectomy. So yes, we have done that.