KERCasts

Using RE-AIM to conduct implementation research

November 23, 2020 KER Unit Season 1 Episode 7
KERCasts
Using RE-AIM to conduct implementation research
Chapters
KERCasts
Using RE-AIM to conduct implementation research
Nov 23, 2020 Season 1 Episode 7
KER Unit

Join us for a conversation with Dr. Russell Glasgow, a world expert on implementation science and the developer of RE-AIM. Interviewed by Dr. Victor Montori, Professor Glasgow describes the trajectory of his career from aerospace engineering, to clinical psychology, to the pragmatic and transparent study of context when considering the effectiveness of interventions. 

Show Notes Transcript

Join us for a conversation with Dr. Russell Glasgow, a world expert on implementation science and the developer of RE-AIM. Interviewed by Dr. Victor Montori, Professor Glasgow describes the trajectory of his career from aerospace engineering, to clinical psychology, to the pragmatic and transparent study of context when considering the effectiveness of interventions. 

Victor Montori:

It's time to start the KERcast, brought to you by the Knowledge and Evaluation Research unit at Mayo Clinic, I am Victor Montori, and I'm hosting today's KERcast. And today, we are in for a treat. With us, Professor Russell Glasgow, Professor Glasgow at University of Colorado is a, one of the leading authorities in dissemination and implementation science. He has been a leader in developing an understanding of the context in which interventions are implemented, and what makes them work in the real world and for the people that they're intended to be used. This initially, not very noteworthy enterprise has become really important, as people have recognized the extent to which the advances of science fail to reach the people they're intended to and we spend more and more resources in developing new interventions, without ensuring that the ones that we already have actually manifest their real value. I'm sure we're going to continue to discuss this during the next hour. Russell, thank you very much for joining the KERcast.

Russell Glasgow:

Well, thanks so much. It's great to be with you and I might just comment, I appreciated your observation there about the field of implementation science just kind of finally coming into its own, if you will, towards the end of my career, I've felt for many years, like you know, beating my head against a cement wall or something, but it is kind of reinforcing to see it get a little more attention now.

Victor Montori:

Well, just like the proverbial water on the rock, hitting your wall, your head against the wall finally has break the wall. You're looking good after the effort, so this good. One of the things that we do here is we try to go through a few questions. But the first one that we normally get into is this idea of, of one's career as a journey and the question I'm intrigued about in relation to that, particularly towards the folks who are joining us who are starting their careers is how does, how does one get to be Russell Glasgow? What is that process is that at the termination from day one, is it, chance? Is it a combo? How do you see that now?

Russell Glasgow:

Good question. Ah, certainly, I'd say circuitous. At a few points, I think I would say planful or purposeful, and also at a number, you know, probably a result a chance or kind of serendipitous events. I, I grew up in a small town, in Iowa, about 10,000 residents went to a pretty small, high school there. I'm the first one, my family to go to college. And I started out I followed one of the what I understood to be the only two paths, according to my high school advisor that worked with people thinking about college is that I could either be a teacher, or an engineer. And so I started down the track of being an aerospace engineer. And then several things happened at that, that that took me different directions.

Victor Montori:

What year are we talking about?

Russell Glasgow:

Uh, let's see. I graduated from high school in 1968.

Victor Montori:

So so that would be what Vietnam or Martin Luther King assassination that era?

Russell Glasgow:

Yes, yeah. The in particular, like my early college years were kind of the Vietnam War era was things were coming to a head, then.

Victor Montori:

And how did that how did that affect teacher versus engineer path?

Russell Glasgow:

Well, fairly markedly actually. I was at a state essentially agriculture and engineering school and I was enjoying the, you know, the physics and the chemistry and the the initial engineering courses, but it just felt less and less relevant to me. There's so much going on out there in the world and things and it just, I it seemed like such at times irrelevant bubble. And at least partially as a result of that I switched to the more liberal arts more, "progressive" University at the University of Iowa.

Victor Montori:

So nothing wrong with that shift. But I want to understand one thing, this is also the this is also the moonshot era, the actual the you know that the literal moonshot, right? I mean, we're getting ready to go to the moon. So for an engineer candidate that would have been like, you know, really, really fascinating. I mean, aerospace engineering, right?

Russell Glasgow:

Yeah, you're right. Um, I'd have to go back Victor. I don't know the exact sequence of years there. But certainly building up to that was, I think, maybe it just seemed so distant, you know, from a first year student, or whatever to...

Victor Montori:

You mean the moon seemed far away?

Russell Glasgow:

One of my lesser qualities is that I'm not a real patient person. So maybe, maybe that had something to do with it.

Victor Montori:

So you were saying that in doing this, the, the Vietnam War and the challenges of what was going on around you made it seem less interesting, less relevant, so you changed directions?

Russell Glasgow:

I did, right. Change directions, institutions and my major to psychology.

Victor Montori:

Hmm. And why psychology?

Russell Glasgow:

Oh, good question. I think I wanted to do something that had a little more direct interaction with, with people, I wanted something that still had science, I always loved the, you know, the STEM curriculum and things, but that dealt a little bit more, I think with, with, with people more directly to...

Victor Montori:

A little warmer. You know it's interesting that I, you know, I'm reflecting here, as you, as you tell this, you know, my, my path to medicine, I wanted to be an electrical engineer. And I was interested in circuits and things and then realized that this whole thing was so cold. And, and so I started asking, you know, what's the warmer version of this, and I went into, I went to thinking of neuroscience, right, the neural circuit in one's head. And then I realized, much to my chagrin, that in order to get into neuroscience and the abnormalities of those circuits, one has to become a physician. So I went into medical school right out of, I mean in Peru you go right out of high school, in pursuit of that warmer version of industrial, electrical engineering. So yeah, and so the psychology of the individual was interesting to you?

Russell Glasgow:

It was, and I have to say, that's maybe one of the more fortuitous, things that happened more, more by chance, in order to put myself through school, I worked, I had work study, aid, and grant and I was fortunate enough to have my work study, be with a really great professor, one of the leading people, young investigators coming out in the field of behavior therapy, Tom Borkovec, and so, you know, that was just wonderful for me, because I could support myself, but then learn all these exciting things, too. And then as a result of working with him for a couple years, I think that put me in good position to get into graduate school, which I did in clinical psychology out at the University of Oregon.

Victor Montori:

And this is, is this the heyday of social psychology as well, where all these experiments with like the Prison Experiment in Stanford and all that stuff.

Russell Glasgow:

It's Yeah, related to that what was more of an influence, where I was, in particular, we had some of that Stanford influence and collaboration, but more probably to some of the leading people there that more directly impacted us were Walter Mischel with a lot of the child and adolescent work, delay of gratification. And then especially Albert Bandura, on his social learning theory, that was kind of the dominant influence, I think, in my program there. And then another one that some of the listeners might be familiar with is, Peter Lewinsohn, done an awful lot of work in depression, as well as neuro psychology. By the way, I mentioned that before. And then my primary mentor, Ed Lichtenstein, who's one of the preeminent smoking cessation researchers in the world.

Victor Montori:

So a lot of individual behavioral modification, experimentation type of thing.

Russell Glasgow:

Yes, exactly.

Victor Montori:

So how does one go from aerospace engineering and individual clinical psychology behavioral modification to implementation science?

Russell Glasgow:

Well, a few different twists and turns, before we, before we get there, the initial one, and maybe one that was purposeful at this stage in my career is I was enjoying the the individual work, and I focused primarily on health behavior change, did a little work on anxiety and depression, particularly, phobia was big in terms of a research paradigm then. But in terms of the health behavior change work and again, I did my dissertation with Ed Lichtenstein on smoking cessation, is I just rapidly became aware of the limits of working one on one with people and just realized, you know, how powerful the social forces, you know, the if you will the social ecologic model, and the different things all the way from policy to family, to community things, like Clean Indoor Air Acts, or the the absence of such and things, and I just became really drawn to our community influence and these these broader, these broader contextual issues.

Victor Montori:

So it seems like it's been content and context and it's been context has been driving your career choices, but then also driving the, your gaze, your attention, you know, from what is happening to in what context it's happening, am I getting that right?

Russell Glasgow:

You are absolutely, I'd say that would probably be one of the themes, that I'm still trying to work on today. I say one other aspect of what I try to work on is, is being pragmatic and trying to not oversimplify things, but to have a relatively straightforward and efficient approach to some really complex issues there and I guess that's the other thing trying to make sense, out of systems and complexity issues and to realize the contextual factors, but to try and figure out where we can about what things in the context can, can we change?

Victor Montori:

Let me take you in a slightly different direction here, just because I'm excited about the apparent contradiction, you just, you just described a couple, a number of things that people are usually trained to simplify in fact. You know, what you mentioned complexity interactions and other things, that's the kind of stuff you control for in doing your, your, your very clean experiment, you try to keep all of those things the same. And in that way, background them, from your attention, so you can focus on the things you're experimenting with. And I will think that in a lot of the psychological experiments, in fact, that's the thing is you control for all of those things, and you put people in controlled environments. I'm just thinking, since you mentioned impulse control, I'm just thinking of the little three year old with a little marshmallow in the, in the chamber with nothing else happening around right. And you're interested in that nothing else happening around in fact, realizing that real life doesn't happen. And so how do you handle that you know, that apparent tension or that real tension really?

Russell Glasgow:

Yeah, it was a real tension. And interesting you a brief aside, you just I kind of had a flashback when you mentioned that control is one of the things early in my career we did was the old Skinnerian rat experiments and things, so I did, I did do some of that, funny, but to more directly

Victor Montori:

I just want to point out that flashbacks is a address your question... symptom of PTSD. So I don't know...

Russell Glasgow:

Well, I have some of that, I'm not sure. Actually I like my rat and I got on quite well. So I don't think that was necessarily bad for either of us. But yes, good, good point taken though. Let's see, I was gonna say this issue, you know, of over simplifying and trying to control everything is actually one of the things that I think throughout my career that I rebelled against, I think and I would say that's something and I'd say, partially as a result of my training, and where I saw the limits there, and what was more or less reified. And I would say, still, today in a number of circles, you know, the classic, highly controlled efficacy, you know, sometimes I like to say triple blind, you know, efficacy study that purposely controls and minimizes contextual factors. You know, that's been one of the things that I guess I have rebelled against. And I like to say, sometimes that what we unlearn, you know, is as important as what we learned.

Victor Montori:

So, you use the expression, triple blind, is that related to something in particular?

Russell Glasgow:

I just made that up, you know, usually the double blind, the double blind study, you know, is, is the, you know, the ultimate or whatever...

Victor Montori:

We wrote a paper with, with colleagues in Canada about the use of the term double blind and that it actually hides who in fact, was blind, but it sounds like in your, in your, in your hands, that blinding is not just to the, you know, whether you got the intervention or the control, but also, you're blind to the, you know, the investigators blind to the context in which the whole thing is happening and the effect of that context on an interaction with the intervention. Just a little aside, in relation to, I'm curious about your opinion about this. So I'm a diabetes doctor, right. And some of the, some people consider that some of our activity has to focus on turning people who have developed, for instance, type two diabetes, turning people away from the habits that may have contributed to the development of that and other chronic conditions, you know, smoking, reduced physical activity, increased consumption of carbohydrates and other things. And, and so behavior modification becomes a part of the, of the activities that we do, along with self management education to help people get further and, and there's a whole lot of work right now to try to use technology to provide surveillance of people's behaviors, and potentially even enact some of that behavior modification through technology, you know, the same way in which our phones and our interactions online are constantly under surveillance by tech giants, who then sell the prediction of our behaviors and behavior modification to those who are interested in that mostly marketing firms, advertising firms. How do you see that evolution from the clinical psychology days, the Bandura days, the experiments of behavior change, with the extent to which that has moved on to now have corporations really making a business case that they can modify our behaviors in directions of those who pay, for instance, in the case of Medicare will be the direction of healthy behaviors?

Russell Glasgow:

Right. A challenging and really complex question I think, as I try to reflect back on myself and the work that I've done in self management, particularly diabetes, I believe that's where maybe you and I first came into contact with each other was some of your diabetes work. I, again, another area in which I was really fortunate to work with one of the real giants, I think, the field was Ed Wagner and his chronic care model. And I was there at that time that I was able to really help Ed and colleagues flush out and try to operationalize the self management, you know, support component of that. And that was kind of another switch for both me and I think the field in terms of doing that in two ways from if you will, the traditional maybe simplistic initial behavior modification to engage much more, you know, the, the patient and the cognitive factors, you know, to switch to cognitive behavioral therapy, there and then again, the context factor I think working a lot with patients and their, their families. First of all, I, you know, I tend, like a lot, I didn't do work with type one. So it was generally older adults with type two. And they're the the family was so central. And that was kind of where I started out focusing on some of these broader factors of, you know, social support. And often, like, well, meaning support that didn't end up being so supportive after all. And then I'm gonna, this is kind of a jump shift. But come back to your technology issue there. We were one of the early adopters, I think, of technology, these were, you know, now really, really rudimentary and crude web based programs, and things, you know, we didn't have the bio sensors and markers and that sort of thing. But we did work with people, you know, to create individualized behavior plans online, that did take into account context, and we felt did a reasonable job anyway, our buzzword then was collaborative goal setting, and to try and do that, you know, in a way that truly was collaborative, not just giving lip service to it, you know, which I think is a challenge. Admittedly, that was pretty challenging, you know, through the, the web based technology. But I'm still a believer, though, I'm very concerned about some of the things that happened, that you've mentioned about, you know, selling all of our information, you know, as well as our genome and things. But I do think that there is a way that technology can really help kind of all sides, both patients and families, but also the healthcare provider system, you know, who has, you know, zero time, time, and sometimes very little training and dealing with behavior change issues. Again, I don't see them, you know, as opposites or either or are trying to replace, you know, patient centered care, but as a way to help inform, you know, and make it more efficient. And I think that's been a theme, both in diabetes, and then some more recent work we've been doing with kind of technology mediated behavior changes to try and inform both, both sides, the patient and providers, and in fact, to use the technology to kind of help prepare both of those, you know, for productive discussions, and things.

Victor Montori:

I think that's, that's really fascinating. Given that you're an expert in context, I think the the evolution of big tech, and the intrusion of big tech in behavioral modification, I think, is something that is quite a concern. One of the things that we explore here is this issue of what are the values that move people forward in their research? Right, in their career? What would you say is the the primary value that has moved your career forward?

Russell Glasgow:

Um, can I cheat and have two? Yeah, I would say and at least the first one may need a little explanation. But I would say transparency would be number one, and also transparency in the service of the second value. And that's essentially public health impact or public health good. And, and try and flesh that out a little bit. What I mean by that is, I guess I see myself primarily as a methodologist. I no longer do clinical work myself. But I think that a lot of our methods, some of this goes back to the earlier discussion that we had about, you know, control experiments and things like that is often that we, unintentionally, essentially hide things and we hide the context under which we get results. So for example, in our interventions are done, you know, at the world's best tertiary academic medical centers, they're done by the world's experts, you know, in the intervention, they're done with highly selected patients often that have no comorbidities you know, one of the issues near and dear to your heart, I know that you've led us on and they're done with incredibly expensive interventions that very few places would, would be able to, to do or to carry out. And also let me let me state that there's nothing per se the matter with that. And some people feel that's unnecessary stage, my problem with all along has been people jump from there to find something that's efficacious under that very limited, highly controlled context to then expect that that's going to work in the real world, low resource settings with complex patients with all sorts of other issues going on and staff that has no time and settings that have no resources and that sort of thing. So that's kind of what I mean about the transparency and then in both designing and then reporting, in particular, our science, I feel it's more important to, to do that and to focus on just exactly what was the context and and by doing that, I hope I can't prove this yet, but we're focusing more that this can at least be one way to help address our replication crisis, you know, that we have at all levels of science, but I think maybe particularly more as we get to, you know, less laboratory type settings.

Victor Montori:

So and then you said, transparency and public health good?

Russell Glasgow:

Yes.

Victor Montori:

And the public health good is connected to the transparency in that it cannot be realized until we confront the challenges, the barriers and facilitators, so to speak of, of translating those efficacious interventions into the real, real conditions on the ground?

Russell Glasgow:

Absolutely. And then focusing on our, you know, for the last 20 years, I've been with colleagues working on our RE-AIM model that really focuses on the factors or sometimes the steps, if you will, that result in public health impact, in addition to the initial efficacy or effectiveness, what are these other factors, and most of those are contextual and they have to do not only with like, the percentage of the population that can be reached or impacted, but also really importantly, the representativeness and certainly today, I mean, the the equity implications of everything hit us like COVID, of course, is the predominant example that we all think about, you know, 24 hours a day. But, you know, nowhere, I mean, that has just pulled out the, you know, all the equity issues just hit you, you know, right, between the...way over the years, we've been focusing through this representative dimension about all the various ways and things that, you know, inequities can arise and I should say, I think, usually unintentionally, including some of the technology things, I mean, I don't think people set out to develop technology or web based programs to say, you know, hey, we're going to rip people off here, and we don't care about people, you know, that that sort of thing. But I think it often happens unintentionally, particularly if you don't pay, pay attention, you aren't aware of some of these unintended consequences.

Victor Montori:

Yeah. So if you've been, if you, if you don't have the skills or the knowledge or the attitude to use that sort of phrase, to understand the value of context, then you will be blind to it, and you'll miss it. Right. And so that's where the unintended consequences may come in. Well, I would like to come back later to this issue of equity and how you see that playing out. But RE-AIM has been a major contribution, Russ, I mean, you've made several contributions, but sometimes people are known for for the one that has gotten probably most impact and I would put forward that RE-AIM has gotten tremendous impact. Tell us about it. You know about your, I'm asking you about you know, how proud of you are the little kid you know, sort of things. Tell us a little bit more about RE-AIM, how that came to be and I think and what what satisfaction it has given you?

Russell Glasgow:

Yeah, um, good question. I, we have another three hours, right?

Victor Montori:

Yeah, you and I do most other people listening don't, but yeah, but you and I might have, yes, go ahead.

Russell Glasgow:

Well, again, to go back to this story about seeing the the limits of the individual level that really was how, how it originated, looking at all these other factors as I started doing more work that was beyond the one to one individual, where we started working in work sites, and then and whole communities and things like that. So I realized some of these other factors. And so all along, RE-AIM was intended to look at the various factors that were important for producing public health or popular population impact. And we started out using these for people behavior change things, as I mentioned health behavior change. But then it has been rewarding, honestly, to see how it's kind of taken hold more broadly and in different fields and things. RE-AIM is, well, this year, I guess it's going on 21 years old...

Victor Montori:

It can start drinking...

Russell Glasgow:

There you go. Good, good, right. I'll use that next time. But like with any framework, I think it, it runs the danger of, of potentially being ossified or stuck where it began. And that's something that I've worked pretty hard on. And there's a number of ways that we don't have time to talk about here. But how RE-AIM has evolved over the time, I think, in general, to take account some of these more contextual factors. But still, one thing that's that's hard for me today, is that often, when you look at how it's characterized, or how grant proposals are reviewed, for example, or publication, they're still using the 1999 version of RE-AIM. You know, and in fact, sometimes I feel really bad for young aspiring scientists who use a correct more up to date more contextual, fuller, I think more sophisticated approach to rename and then they literally get beat up in study section, they say, No, you can't use RE-AIM this way. Despite, you know, a lot of evidence that in fact, you can.

Victor Montori:

Yeah, so people get anchored on the first thing they learned and they haven't kept up with, with the development. What's, everything that gets used a lot because it's very helpful also gets misused. Have you seen any misuses of RE-AIM?

Russell Glasgow:

Yes, yeah, I think we have, um, the most frequent is actually not, not serious, I don't think but the most frequent one is, as you well know, but for some of the listeners, RE-AIM has elements both at the individual level, the reach and effectiveness, and then at the more setting, or broader contextual level or the adoption, implementation, and maintenance. Frequently, people will confuse the two, particularly the reach and the adoption at that level. So that that that's a common issue. The other way is that a number of people feel that RE-AIM can only be used after the fact for post hoc evaluation. And I feel that's one of the really missed opportunities or the area that's quite frustrating, because that's not at all our intent. In fact, I mean, why would you want to wait till you know, the actions all over then to find out that, in fact, you have an intervention that has low reach, low effectiveness, certainly no maintenance and that sort of thing. So so we feel it's really important to use RE-AIM both for planning and more recently, our work has focused on what we're calling iterative use of RE-AIM to make mid course adjustments as your program is going along and being implemented.

Victor Montori:

So almost like the adaptive clinical trial, but in this case, it what is being adapted, he's adapted for implementation optimization,

Russell Glasgow:

Precisely. And as as, as usual, better said then, then I did, I'm gonna be taking some notes here or listing how I could craft, craft this, in my future talks.

Victor Montori:

You're being generous, which, which reminds me that another thing that we like to ask our guests is, you know, our research unit has three values, right? Patient centeredness, integrity, and generosity. Any of these resonate more with you?

Russell Glasgow:

Um, well, they all resonate. This was one of the things your listeners may know that you did kind of prep me for, so I had a chance to think a bit. I don't know that I still have a great answer, but I'm gonna hedge a bit and say they're all relevant. But I think, I think the integrity is the one that resonates the most, if I can just say a word about the generosity, I think, for me has to do with contributing, supporting others, particularly younger investigators, and maybe that's closer to generativity, but at this stage of my career, I get a lot of enjoyment out of that and hopefully, you know, have helped some other people I've always enjoyed like consulting. The patient centeredness I think comes for me from back to my training where I switched away from the, you know, more narrower kind of non cognitive behavior modification to kind of social, cognitive and learning theory and things, and then particular through my work with diabetes and other chronic illnesses on, on self management support. But, but my, if I have to restrict to one answer, I would say, integrity. And that, again, kind of cycles back to this other notion, when you asked me, I said, my own word was transparency, and both thinking about what we're doing in both our science and and clinical community work, but also how we report on it. And again, I think it's just this notion that we often tend, and again, I don't mean to be pejorative, I often come off, as you know, really being seen as overly negative. I don't think this is intentional. But I think often, we will cherry pick the very best settings, the least complex, most motivated patients, the, you know, to do, to do our work in there. And and I almost feel strongly enough that this is a quasi ethical issue in again, not that we do that, that I accept that we need all types of science, but then that we don't in terms of interpreting our results, or what we recommend in how it being used, we don't have the humility to realize that those results were found under that set of conditions. And it's an open question, empirical question, how well that's going to work in other much more challenging situations?

Victor Montori:

Yeah, some people worry that a interpretation of, you know, in that, in that continuum of efficacy to effectiveness of attention to the effect of intervention in the highly controlled circumstances versus the real world effects, so to speak. Some people say that if you were to dismiss, which is not what you're suggesting but the extreme, is to say, oh, okay, we have preliminary evidence. So evidence of efficacy is preliminary evidence of effectiveness, so to speak. And if that is the case, very few things that we know, work under controlled conditions will find their way into practice. In fact, one would say that in in their way of getting into practice, they will have to jump through additional effectiveness study hoops, before we can confidently implement them. And so that's that, we would extend the lag between the discovery to benefit in public health. What's your objection to that line of thinking?

Russell Glasgow:

Well, one, I think if you follow that step by step model, it would take forever. And I think that's part of our problem. It's not only the 17 year translation lag, but I think if you follow that through for doing basic research, then only then, then moving, you know, maybe to human studies, and then to tightly controlled efficacy studies, because usually, at least in my experience, and most everybody, I know, it doesn't move because you don't have 100% success with the first step. So then you cycle back at the first step, go the next little step, then you have to cycle back. And so you know, it literally never gets done when you get out there. So that's one thing that bothers me about that argument a bit. And then secondly, to some extent, not always, and not perfectly, but I do think you can do some of those things simultaneously. And again, one of the design frameworks that's becoming popular and implementation science is called the hybrid effectiveness implementation trial, where you kind of study both of these, but you kind of vary over time how much emphasis there is on the effectiveness versus the implementation. I think strategies like that can move us along, you know, a lot more quickly. The other one that is now that I and others are promoting and trying to do work in here now that's still a little more controversial, is the notion of adaptations. You mentioned adaptive designs before, but like during the course of a program, realizing when things aren't working, and making some again, I would say transparent and documenting what the adaptations are, but it doesn't make sense to me to spend several million dollars doing something that you know, early on isn't, isn't going to work. There are issues here again, technical and complicated issues. But I argue that if you do this thoughtfully, and that you focus on what are the key, some people say the the core components or the principles or theory, I like to say more the functions that are done that you often do need to make adaptations. And then, in fact, in the real world adaptations always happened. It's impossible to do that without and even in many of our, quote pristine, like clinical trials and things like that adaptations get made, but again, often not reported.

Victor Montori:

Yeah, our group works with designers and designers have brought to our work, this notion that as we've developed interventions, which are inherently non pharmacological complex interventions, that we tend to use in practice, that we will develop those interventions in the same context with the same people, and the same pressures and challenges in which they're going to eventually be used if they are found to be effective. So we're dealing with those adaptations even before we are testing for efficacy or effectiveness, but rather in the development process, in fact, our trigger to launch the trial is when we see the the complex intervention work as intended in the context in which is supposed to eventually work. You've been paying attention to the introduction of design in all this, what are your thoughts about it?

Russell Glasgow:

Ah, I've been paying attention and learning, I'm on a learning curve, but I strongly like support what you're doing, we're trying to use, some of those strategies ourselves, but I think it's critical from implementation science perspective, you know, we say it's never too early to start planning for implementation, or dissemination, and sustainability or sustainment, which is another new area that I think is, in some ways, the last frontier, you know, now we are finally starting to get a few things implemented in the real world, they tend to have at least some level of effectiveness, okay, they have some level of generalization, but then getting something that can be sustained, particularly when the level of attention and support and resources and things like that go away, is, is just, you know, really, really, really challenging

Victor Montori:

The maintenance aspect of it over time and, or the normalization, like our colleague Carl May likes to talk about as well. This series is about care that fits and so we're interested in how care fits in the lives of patients and what is the process of fitting that occurs, both in the clinic with the implementation of new interventions, or the adaptation of existing programs of care, and also fitting at home, which connects back to the self management, support of the time where we connected the first time, which by the way, it was the first time I remember, of my mentor, saying, just contact the guy. And the guy answered the contact with with offerings of help, right. To me it open a whole, I mean, I thought some of these famous people that you see in papers are, you know, are heavily protected people who you can't access, you know, unless you know, the secret handshake and the password. But you demonstrated to me a level of generosity that was very enlightening. But this process of fitting is I think, where, where knowing since they happen in the real world, that whatever interventions we develop have to be designed from scratch, to be, to be effective, and to be to reach all the people that need to be reached, to be implemented with some degree of fidelity that makes them effective enough, and to maintained and sustained over time. And this is where I think RE-AIM connects with this series in terms of care that fits in, in the fitting interventions, both at the level of the clinic and at the level of the individual. So there's, there's a very nice, connection, Russell, we have some questions from our audience, and one of them is I'm hearing your answers about this topic of implementation and testing, Something about the usefulness of merging quantitative and qualitative research to lead to development of the best evidence. Can you expand a bit on on how you think we need to leverage mixed methods to best support this kind of work?

Russell Glasgow:

Yes. First of all, I couldn't agree more on the importance of it. Secondly, I am learning myself more and more about qualitative research. I was a bit late comer to the game again, because of this earlier, kind of what was then viewed as science perspective, the quantitative one, but I just appreciate it more and more and I've been fortunate to work with some really great qualitative and mixed methods, researchers. So most directly, I think, in RE-AIM, we now really call for and recommend mixed methods, approaches, in all RE-AIM studies, and it all of the RE-AIM, the five, RE-AIM dimensions. And the way in which I think that's so important is that we can do you know, the world's best job of documenting, like, knowing the exact numerator and denominator who participated in things, but without knowing why they participated, and why they or why they didn't, without knowing the how, and in general, at least in the areas I work into, the quantitative things just aren't at that level, and I'm not sure that they ever will be or should be, you know, to understand that. So I think it's that notion, you know, to move ahead, and to help us understand the reasons why, you know, both conceptually and pragmatically, is is just really, really important. I'm intrigued. And again, another area where I still trying to learn on is this the kind of interface now between quantitative and qualitative around configural, or configurational analysis, things like qualitative comparative analysis, you know, which really is, from my perspective, kind of a blend, where you take the often sometimes qualitative features, but then put those to look at, again, from a complexity science perspective, kind of what's the constellation of factors that, that can produce a successful outcome and again, what I like about it, is it's not just a reductionistic, additive, you know, type approach, and it's not a there's only one method, you know, to get there, it's you can this equifinality principle, you can get the same place through different ways.

Victor Montori:

You could equifinality principle?

Russell Glasgow:

Yes, yeah,

Victor Montori:

I never heard of it. That's, so is, what is it, what is, what does that mean?

Russell Glasgow:

It just just what I said I probably oversimplified it. But the notion is that there are different pathways or often, if you will, different mechanisms of change that may be, let's just take a simple example, to control your diabetes, okay? It may not be that everybody, the only way that you need to get there is to intensify your therapy, you know, and get there for other people, so in different conditions, you can you can do this via different, different pathways.

Victor Montori:

But so different pathways to achieve the same end, very interesting. Let's go back to this equity issue that you raised earlier. Many might think that, and by many, I mean, mostly me. Might think that, that when you're talking about equity, you're talking about reach, and you're talking about reaching people who might identify themselves as poor versus rich, or black versus white, you know, or multimorbid versus single morbid, or, and, and there's gonna be a statistic a table one describing their prevalence in your population and if you have enough diversity, you can check the box and move on. Right? Is that it?

Russell Glasgow:

Well, I certainly am not an expert. This is another area which I'm just it's so rewarding to see the work exploding and things that again, finally getting some real attention. To me focusing on if you will, the underlying factors, again, more at a population basis, the social determinants of health that I think drive a lot of these things is important. But in RE-AIM, this issue of equity, or we get at it through this representativeness dimension, I think it is clearly important at each of these levels. So starting out, let me just give a quick example and I kind of wished you and I talked, I decided not to use any slides, but I almost wish this one I had this slide that I have that I call the health equity slide, which shows all these RE-AIM dimensions. But if you just think the story, that if we talk about what settings are going to be able to deliver, you know, new interventions, like will this work, you know, like in tribal communities, okay, in federally qualified health centers, you know, as well as academic centers. So that's what we call adoption. Then we get as you well articulated the reach at the individual level, who who has access who can participate who can take the time off work who has the doesn't have the transportation challenges, then we go to effectiveness the E. And that is fortunately something we're getting more and more attention to. But the notion is, you know, subgroup effects or heterogeneity of effects. Very importantly, the implementation dimension of the I, that's the one that we've kind of expanded the most over the years. But that notion about both with what subsets of individuals and what subsets of staff, okay can successfully deliver, as you said, with fidelity, and not only deliver with fidelity to the key principles, but be sufficiently sensitive enough clinically, that they can appropriately adapt, you know, the intervention. And then, of course, the maintenance thing. And often that maintenance, a big factor is resources to be able to sustain. So these equity issues come up at each step. And the little story I have that I try to show this through a graph is even if you start out with an intervention that in its basic effectiveness research looks to be completely equitable, you know, whatever group that you want to talk about the most socially, determinants challenged. So 100%, if you have a difference of continued across these dimensions, at the other ones, you end up where you go through the mental maths exercise, where the less advantaged group only gets, like one quarter or less of the benefit, even though the publication value says equal, you know, equal effectiveness.

Victor Montori:

Yeah, I there is a there's quite a bit to do, it seems to me on this issue, because, you know, while in general, in evidence based practice, we tend to dismiss treatment heterogeneity, and subgroup effects because they're rare, rare on biological variables on biographical variables on contextual variables, subgroup effects are much more likely to be present. And I think that's something that in general, evidence based practice, for instance, has, for the most part continued or has ignored and continues to ignore. And, and I think we would like to see more subgroup analysis, quantitative and qualitative subgroup analysis on the basis of biographical and contextual subgroups, which I think go beyond simple labels to really understand how people get in the way.

Russell Glasgow:

I'm sorry to interrupt, but just a comment. I couldn't agree more. And again, I think one of a huge opportunity that we're starting to see some work on is at the, at the more basic fundamental determinants, you know, the the social determinants of issues and how those play out there. And I think, fortunately, we're starting to see some really great examples of pragmatic ways to address those issues, you know, food scarcity, inability to afford medication, transportation, family violence, you know, on and on those key issues. I think both through patient reported outcomes that are feasible to collect, you know, in real world settings, either prior to or at the visit, or whatever and and follow up on and/or combined with some geospatial measures, I think. So again, another area where I think there's great promise to bring together different methodological and conceptual approaches to, to deal with those complex issues.

Victor Montori:

Particularly relevant when you have to do research on what Tudor Hart called the deep end, right, or where people are, are struggling with resources to live and, or to take care of themselves. And on top of that, you're inviting them to participate in research, and they may not have the capacity to do so. And then, you know, we have this almost self fulfilling prophecy of research that fails to help them because they can't participate. And we don't develop research that is contextually sensible, to enable their participation. Right. So there's a meta RE-AIM there in, in their interest in terms of the RE-AIM of the research process itself if I may.

Russell Glasgow:

Absolutely, yeah, absolutely.

Victor Montori:

Russell, we're coming to the end of our time, and I have two more questions to ask you what, what is what's what's the meaning of your work?

Russell Glasgow:

The meaning of my work, I didn't reflect on on that one. Um, I think I'm gonna go back a bit of a broken record. I think it's, it's to this transparency notion there and again, that that things are complex and contextual. But it's possible to, to address those, I think I'd say that's the meaning of my work if you can through RE-AIM and some other approaches, trying to, if you will, you know, shine the light are some real challenges that to help translate this more basic research, you know, into something that is going to make a difference in the real world. This is a, you know, a continuing journey, we don't have it worked out yet. And again, we're trying now with a new extension, or what we're calling RE-AIM 2.0, to try and focus specifically on some contextual factors, you know, that may be especially related to some of the different RE-AIM dimensions. The problem with context, is we all kind of intuitively understand that it's important, but some people have even defined context as being everything, you know, everything outside the intervention, so then it becomes totally, you know, unwieldy and how do you begin to study everything in the world, so not that we haven't worked out but a work in progress is through this extension of RE-AIM 2.0 to try and focus on some specific limited set of factors.

Victor Montori:

Yeah I have to agree with you. I think, I think the work so far, recognizing there's more and more to come, but the work so far, I think has enabled us to see that context, that context that has been shaping your career, but also shaping your view of science and the application of science to it, that context has your work has made managing that, dealing with that, addressing that feasible. And, and through the lenses that you've been offering us generously and rigorously like RE-AIM, you've helped us understand the real challenges and opportunities of taking what we think might work and, and really deriving public health value as a result of that work. So no, I it's a, it's been a great, great privilege to have you here and, but also to have you in our career, and then to have your contribution embedded in the science that we've been trying to do at our at our KERunit. So I have to thank you in that way. But also thank you for for being with us. And I have my last question, Russ and that is what's next for Russell Glasgow?

Russell Glasgow:

Um, I think right now, it's we have a new center on implementation science from the Cancer Institute. And we're looking at trying to integrate different disciplines around our theme is like a paragraph long, but it has to do with pragmatic implementation science approaches to value and values in rural primary care. So it's this notion, again, coming back to this mixing, a hopefully sincere and somewhat in depth understanding of patient values and preferences with a more traditional economic and cost approach to value, you know, the more hardcore issues, but that, that integration, and it's been really rewarding to see several other centers around the country, starting, you know, to address this interface and I think this this issue of cost, and I don't mean that just narrowly terms of only numerical fiduciary costs, but burden to on both staff and patients and how that impacts sustainment. So that's, that's where we're going and again, if it's like anything else, my approach to that will take another turn or two in terms of getting to that, but we're having fun.

Victor Montori:

Eventually, we'll get to the moon. Russell Glasgow, what a pleasure to chat with you today. You are generous, and also a phenomenal teacher because you've been able to tell us a lot about these concepts, and no slides. But thank you so much for being with us and for everyone that has joined us. Thanks for joining the KERcast from the KERunit and I hope that you can join us on our next episode. Until then, please take care. Thanks again Russell.