The continuing growth of the international implementation science community is challenged by the lack of agreement on common scientific terms as well as a low level of knowledge about other countries’ health systems, Anne Sales said in the Leonard Davis Institute of Health Economics’ first research seminar of 2021.

Sales, PhD, RN, a Professor at the University of Michigan School of Medicine and Co-Editor-in-Chief of Implementation Science Communications, an international journal, was the main speaker at LDI’s February 26 “Intersecting Implementation Science and Economics: Applying Theory” research seminar, co-sponsored by the Penn Implementation Science Center (PISCE@LDI).

‘A real problem’

“Language is turning into a real problem in implementation science,” Sales told her virtual audience. “We’re using terms in many different ways without necessarily being all that clear about what we mean.” She went on to say that even the 48-word definition routinely cited in journals is inadequate to the task; she replaced it with her own 13-word definition: “Implementation science is the process of planned human behavior change under organizational constraints.”

She also pointed out that there is a low level of understanding among implementation scientists about the health care systems in other countries and how these other systems differ from their own.

I worry that in the space of how we change the behavior within complex organizational constraints, we are already seeing lots of factionalism, lots of chasms that seem to widen over time between individuals doing this kind of work.

Anne Sales

Rinad Beidas, PhD, seminar moderator, LDI Senior Fellow, and Director of the Penn Implementation Science Center, agreed. “Implementation science has grown up differently in different countries; I often get feedback on manuscripts that are viewed as ‘very U.S. centric.’ It’s made me realize that I have a gap in my knowledge about what’s happening in other countries, and it feels like these factions are problematic.”

Key differences

“I worry,” Sales explained, “that in the space of how we change the behavior within complex organizational constraints, we are already seeing lots of factionalism, lots of chasms that seem to widen over time between individuals doing this kind of work. There are more similarities than differences in general, but there are key differences that shouldn’t be ignored.”

Sales called for the creation of “some organizing infrastructure that allows us to interact with each other more readily. I think there are core issues around how systems are configured in different countries and how the differences are not well understood. When I was in Canada, one of the questions I got asked by a Canadian researcher was why is the U.S. not buying into some of these principles?”

“The reality is the U.S. system is fragmented and its incentives are channeled toward individual fees for service,” said Sales. “There is no sense of global capitation and no sense of the things you get when you have the Canadian Health Act or the National Health Service in the U.K., and no sense of the social contracting or the society-wide expectation around health services that such systems offer.”

Understanding other countries’ approaches

“I think all of that defines how the U.S. is different. Canada, in some ways, is more like the U.K. and Australia than it is like the U.S. The Canadian system looks a lot like our system in subtle ways, but it’s not. All of those things matter, and we need to think about them and talk about them and make more attempts to do that in some organized fashion.”

A researcher whose work focuses on theory-based design and the role of social networks in the uptake of evidence-based implementation interventions, Sales discussed theories including prospect, reasoned action, principal-agent, utility, and loss aversion.

The loss aversion relation to implementation science piqued audience interest, particularly in how it impacts the process of building teams in an era when team-based care is increasingly important. Loss aversion, a behavioral concept commonly used in behavioral economics, has not been an area of interest for most implementation scientists. Loss aversion is the idea that people are instinctively more likely to take action to avoid losses than they are to achieve gains. Behavioral scientists have found people can more easily be motivated to alter their behavior to avoid perceived losses.

Loss aversion and implementation

“Starting to factor loss aversion into our expectation of how people might behave could really enrich our thinking about it,” said Sales. “We often think about the expected or planned utility of behavior change—the good things that will come out of it once we get through the hurdle of getting the change done. But in that interim, we don’t focus much on loss. And I think this is where loss aversion as a theory, as a theoretical proposition, really becomes important. For instance, we need to think more about what providers lose when we focus on them as targets of knowledge, translation, or implementation interventions in our strategies.”

“For instance, when people are made part of a team, it involves losing autonomy in their decisionmaking about how they would do things, especially in a very hierarchically formed team,” Sales continued. “I think that may well explain a lot of behavior and also the potential for mitigating some of those behaviors if we can reconfigure teams or design them to get people on board in different ways.”

“There’s a notion of something called the negative opinion leader—the leader of the opposition and the resistance to change. And I think this loss concept could be a big piece of who those negative opinion leaders are, why they have become the leaders of the resistance instead of the champions of change, and how you might be able to convert them. Because if you can focus on negative opinion leaders in the change process situation and flip them, they often become big champions. And because they were so negative before, the fact that they’ve changed their opinion can be very influential.”