Matthew Crowley: The intervention was initially developed for just poorly controlled type 2 diabetes patients. Once we found it was effective in the research setting, we then had the goal of translating it into practice, so that like many research interventions, it wouldn't kind of sit on the shelf, so to speak. That's something that was effective in the research setting, but can never be delivered in practice.
So the way we managed to start translating into practice was to partner with the VA Office of Rural Health. Rural veterans, rural patients in general have less access to specialty care. Here in the US, much longer distances to travel to get to hospital facilities and doctor's offices, and particularly, endocrinology and other diabetes specialty care. So it seemed like a really logical group to try to roll out the intervention with and practice.
And we were able to partner with the VA Office of Rural Health, which is an operations group within VA. And they supported the initial sort of standing up at the intervention at the sites. The one really important thing about the intervention is that it was specifically developed to use only existing clinical infrastructure, so clinical staffing, existing telehealth equipment that the VA has built up, and then the existing electronic health record.
So translating it into practice with kind of guidance and assistance from the VA Office of Rural Health for these rural veterans was really not only kind of logical from the clinical standpoint, but also this was reasonably straightforward practically.
Interviewer: Did you encounter people for whom telehealth just didn't work, or they just didn't like it?
Matthew Crowley: I don't think there's any single solution for every patient that's going to work universally. There are patients who start up with the best of intentions in the ACDC intervention, and for whatever reason, it's too time intensive or it interferes with their employment or for various other reasons, they just can't maintain it.
In our original study, that was probably somewhere on the order of a quarter of the patients weren't able to sort of persist with the intervention. In practice and in among our rural population, I think it's probably similar numbers. The numbers we report, as far as A1C lowering, include all the patients who start with the intervention, whether or not they stuck with it and how many encountered they completed.
And one thing we found is that for the folks who actually stick with the intervention and get most, if not all, of the intended encounters, they have a even more robust A1C reduction. And then the folks who engage less obviously have less benefit. So the numbers we report are the average numbers across the entire population. Bottom line is that there's people who the approach doesn't work for, but we've found that by far the majority, it does.
Interviewer: Are there people for whom it's not appropriate from a medical point of view?
Matthew Crowley: Well, one nice thing about diabetes care is a lot of it is really amenable to remote delivery. There's a need to do annual assessments for good diabetes care where you can do things like foot exams and other kind of routine care measures. But a lot of it is really amenable to telehealth-based care. And so the typical diabetes patients really does well under a telehealth model, I think.
There are people for whom technology can be an issue-- less technologically savvy, for example. We in this study and in this project used primarily telephone-based care. And most, if not everybody, has access to a telephone and knows how to use it. So the particular model we used here, the technological barriers weren't a real issue.
But I would say that people who are in need of in-person care either because of certain comorbidities or other things going on with their care, or those who are particularly technology phobic may not be ideal. But again, our experience has been most patients do really well under this model.
Interviewer: So you got a really impressive HbA1c reduction in your study. And it was sustained after the main intervention as well. Do you think that's mainly because of the very high starting point, or are there other factors behind that?
Matthew Crowley: I think there's no question that when you work with these really poorly controlled patients, there's greater potential for gain. On the one hand, if you have somebody starting from an A1c of 14 or 15, that there may be too many factors going on. And you may have trouble moving the needle on that group. On the other hand, if you start with somebody who's 7.2, you're not going to be able to see too drastic of an A1c reduction in that population.
I think our average baseline A1c was somewhere around 10, both in our study and then maybe a little lower than that in the implementation population, the rural implementation population we're working with. And that's a really kind of a sweet spot I think where you have substantial potential for improvement, but also kind of the potential for actually addressing the problems that are underlying the patient's poor control.
So we found that's a pretty high yield population to work with. We certainly had a wide range of starting A1c's in our population and have seen great benefits in people with really high baseline A1c's and benefits in people with lower starting A1c's. But I think for sure starting with the poorly controlled population does lead to greater potential for benefit.
We were really impressed with the maintenance of the benefit we saw in the interventions, so people lowered their A1c's by about 1.4% at six months, which is the intended duration of the intervention. And we had developed a maintenance protocol as part of our implementation work because a lot of patients got to the end of that six months and said, wait, no more intervention, I'm done?
So we kind of out of necessity developed a lighter touch maintenance protocol that patients could stay in per the local site team's discretion. And I think that probably led to some of the maintenance of the effect, too.
Interviewer: Yeah. Yep. So now the service has been up and running for a while. Are there any improvements that you think you could make?
Matthew Crowley: We're always looking to constantly improve the intervention in rehab. I think the opportunity to do so over the four years that some of our sites have been delivering it in practice, our Asheville, North Carolina site started in late 2016, I believe, and has been continuing to deliver the intervention since then.
There are definitely tweaks we've made along the way to materials and adding certain self-management content to the self-management component of the intervention. We've tried to take steps to make the work load a little lower for the providers to make it easier to sustain. Developing the maintenance protocol that I mentioned just a second ago was a big change we made, and that was in response to the demand from our patient population, as well as our site teams, kind of thinking, like, well, what do we do with these folks after six months?
So certainly, we've made a lot of tweaks along the way. And that's one of the fun things about taking a intervention that's developed in the research setting and actually putting it into practice and continuing to deliver it indefinitely, as you continue to find ways to make it better.
Interviewer: If someone's looking at your study and thinking, well, I can't do that because I don't have a telehealth service like you have access to, what would you say to convince them to make the investment?
Matthew Crowley: Well, it's a great question. And we're actually looking at-- in addition to being at the Durham, North Carolina VA as a researcher and endocrinologist, I also work at Duke University and a faculty member there. And one of the big questions I've had is, OK, so this works really well within the VA system because there's this infrastructure that we can tap into. How do we make it work at other places where there's not such a robust telehealth infrastructure to start with?
And I think there are certainly ways to do it. We're actually working on a study right now. We're trying to get the study funded, where we're using clinic staff, kind of repurposing clinic staff to move from sort of in-person care to more remote care. So it's kind of duplicating or replicating the infrastructure that VA has using clinical staff. There are ways to make that sustainable and reimbursable that are certainly more complicated in the fee for service side of the system than in integrated VA side.
But we're finding ways to make that work. The real reason to figure out how to make it work and if I were trying to convince somebody to do so, it's that we have patients who just are not doing well enough with primarily clinic-based care. And that's because type 2 diabetes and diabetes in general is this pervasive disease, where the self-management burden is just incredibly high and incredibly complex for patients.
The type of support that is available through clinic-based care is just not sufficient to adequately care for a lot of these really complex patients who are dealing with a lot of other things. And again, we found that by bringing care to them in a more accessible, more frequent fashion, it really does help them make the changes they need in order to improve. And that's what it's all about in the end, is trying to get these folks to get the best outcomes they can.
So it is a bit more of an uphill climb in other systems without the kind of baseline telehealth infrastructure. But there are ways to do it. We're looking at a lot of those right now. And the juice is worth the squeeze, so to speak, because it really is a better way to achieve good outcomes for a lot of these folks.
Interviewer: And finally, would you care to comment on the pertinence of this to the current pandemic situation?
Matthew Crowley: Sure I would. Yeah, the pandemic was certainly a major disruptor for everybody. And we've actually found that at every one of our implementation sites, delivery of ACDC has basically continued unimpeded. There's been some pull of some of our local site teams in different directions, and people have been asked to follow and monitor COVID patients. And that's put a little bit of effort away from caring for the diabetes folks.
But really, I've been pretty blown away by how well our sites have maintained delivery of ACDC through the pandemic. And the truth is it makes a lot of sense why that's the case. We have at Duke and within the VA and at other centers I know across the country and probably across the world, a lot of people have been moving from clinic-based care to telehealth specifically because it's safer in the setting of pandemic. You're not having as much exposure of patients to staff and each other and travel.
And the general shift towards telehealth that's happened just out of necessity during the pandemic I think just illustrates why things like ACDC are really useful and just, in a lot of ways, a better way to take care of diabetes.