Page 1
Michael Gordon: In context of Lisa's comments on sustainability, would LOVE to hear about metrics of success...
For any project using PD there should be some “bottom line” measures associated with the core purpose of the project. For example, in the project on childhood nutrition in Vietnam the key measure was the increase in children with healthy height and weight status. One indicator of the sustainability of the change in that project was that when researchers went back to the original villages years later they found those children had maintained their healthy height and weight for years after the project was “over.” Even more exciting was that their younger brothers and sisters who were not even born at the time of the project were ALSO healthy which indicated that the new behaviors that were discovered and adopted through the PD process have become the default behaviors in the community.
In our MRSA project in hospitals in the US the key measure is reduction in infections (and we’re starting to get data to support that). They are also looking at lagging indicators like gown and glove usage and results from hand-washing observation studies. But there are a lot of other ways that we may be able to see things associated with sustainability. Some sites are noticing differences in retention of nurses in units where PD is practiced. There are sites getting higher scores on employee satisfaction. More people in more different roles say something about “infection control” when asked about their job. One Infection Control Practitioner in one of our hospitals said, “It used to be when someone saw me coming down the hall they’d run away because ICPs are looked at as cops in the system that are always looking to catch people doing the wrong thing. Now people come up to me and want to tell me about things that are happening and ask me questions. Something big has changed.”
As an OD practitioner, one of the things I’m very interested in seeing are indicators that some of the PD principles turn up naturally in other projects. For example, one of the participants in one of our training sessions reported that she went back and changed the way she asked questions with a group at a meeting about something different and was very happy with how that changed how that meeting went.
Ted Tschudy: Re Eva's comment...at some point I'd like to hear more about the process/work of "nurturing" positive deviance. Sounds like you are doing some of that now Monique.
That’s really our learning edge, Ted. The PD process is so much about NOT doing many of the things we are in the habit of doing that it’s sometimes hard to define what the role and work of the PD coach really is! Some things I think contribute to the nurturing: Being a mirror to help participants SEE how some seemingly small and insignificant things that happen are important. Helping some of the core team members redefine how they evaluate their success as they move from a do-er role into a more facilitative role. Introducing some new process possibilities (in the hospital environments the death-by-power-point is the norm so options for different kinds of self-organized interactions are not in the repertoire). And, of course, asking good questions.
I think the key for the people in the community who are doing the facilitating is all about listening. Listening to different people, listening more often, listening for possibilities. So, of course, as coaches for the process we need to model the listening behavior (which is why we are always joking with each other about who brought the duct tape for our mouths ‘cause it’s REALLY hard!)
Michael Gordon: sounds like we're largely talking about process improvement, using a straw man that - for one reason or another - appears to work. What am I missing?
Michael, in a way I think you could think of PD as part of the process improvement family. But it differs from some of the other approaches I know about in that it is focused very specifically on the HUMAN part of the process where a lot of process improvement work focuses on making processes “people-proof” so that you don’t have to worry about the variance in human behavior.
For example, at one of our sites they started talking about the problem that nobody used the disposable stethoscopes that were supplied in the isolation carts so that bugs didn’t travel from one patient to the next on the stethoscope. The process improvement work identified the stethoscopes as a vector of transmission and created a system for making disposables available in any hospital room where a patient was identified as needing to be on isolation status. BUT people didn’t use them. They said that the stethoscopes were like from Toys-R-Us so they couldn’t hear through them. So they used their regular stethoscopes instead and just tried to remember to wipe them off between patients. The traditional process improvement work created the process but it couldn’t make everyone use it and it didn’t include mechanisms for finding out about barriers to use.
Through a series of PD dialogues it emerged that one of the several different brands of disposable stethoscopes DID work and, once the supply folks were involved in the dialogue, they changed how disposable stethoscopes were chosen, ordered, deployed, and used.
The people’s whose behavior needed to change (those that needed to use the disposable stethoscopes and those that chose supplies) had to be involved in the process of solving the problem of MRSA bugs traveling on stethoscopes. Of course, there’s no reason that it couldn’t have happened as part of the original process improvement effort … but it didn’t because the traditional way of organizing such processes is to gather selected experts together to figure out the “best” way to solve the problem and then roll it out!
Bob Devlin: This is a much customized process - what principles have you found that work? - i.e.: involve broad range of people in dialogue on how they're addressing the issue...
We have a number of “mantras” associated with some principles we’ve found that work. Involving a broad range of people (eventually everybody!) in dialogue is one – we sometimes talk about making sure to include the “unusual suspects.”
Another is “Nothing about me without me.” It’s often the case that groups default into talking about how the problem could be solved if only THEY did something different. We coach groups to listen for when they hear themselves talking about somebody (or some role) who is not present, stop, and find a way to invite that person (or someone from that role) to the conversation.
Another is about the value of ACTION over discussion. When someone starts describing how they do something, whenever possible we try to get them to show us .. right now, right here. Or if they ask “what if” questions about how someone might react (“what if I tell a doctor to wash his hands and he gets mad?”) we might try a role play right there to let different people try out different ways of responding. One of our hospitals has implemented an improv process they are now using in orientation to “play out” different ways of dealing with situations.
Farm Credit Services of America: How is this different from Open Space technology?
Harrison Owen who has been the lead disseminator of Open Space often stresses that it’s really not a meeting design but rather a way that organizations could be all the time. In that sense, PD is very compatible with that way of thinking about how organizations might work. We have introduced Open Space principles to groups using PD as a way of facilitating gatherings of volunteers.
One way that PD is often different from Open Space is in the kinds of problems addressed. PD is very specifically targeted at problems requiring behavioral change. It wouldn’t’ be the process I would choose for a group that wanted to explore redefining its mission, or analyzing its competition, or developing a new product (although I can imagine how PD might help us understand why some units were better at doing that than others). Open Space would be a great process for those things. That said, I could see holding PD dialogues within an Open Space meeting.
I think all the best, effective processes have a lot in common at the level of principle so they can be complementary.
Mary: What are some examples of situations where this would not work?
That’s always a good question, Mary. Some of the situations where PD wouldn’t work would be ones that don’t really involve behavior change (such as “defining the mission”). A situation where people have been “assigned” to participate is not fertile ground. If you are not able to discover ANY deviant behavior that is successful then you can’t use the PD approach either – the “social proof” that a solution is possible is critical.
GaryM_SC: What, then are the systems for capturing and recording ideas and results from these "non-meetings"?
Mark Chapman: so how do you capture the solutions and track them into adoption.
So far, there’s been a lot of variance in strategies for capturing and sharing ideas. As coaches for the process, we focus on stressing the importance of doing that in some way but we don’t have one set way we tell people to do it. We suggest that, wherever possible, having 2 people co-facilitate D&A Dialogues is great both because one can capture while the other pays attention and so they can debrief afterwards. At one site, the facilitator puts action items that emerge on a post it note and hands it to the person who volunteers to do it before they walk out the door. We often help them create venues where the stories can be told so that they begin to “travel.” We have also created communities among hospitals taking this approach and hosted gatherings so that they have the opportunity to tell each other stories.
Finding ways to discover and circulate what’s going on is a key function of the core group of volunteers. They’ve use a diverse set of strategies including – newsletters, bulletin boards, screen saver messages, posters, and ad hoc meetings. At one site, they have a “pay day Friday” meeting every two weeks in the cafeteria for a coffee break where they exchange info about what they’ve seen and heard.
Eva: Why do you think that the positive deviant persons need consultants to generate positive change???
I love that question, Eva. I don’t think positive deviant persons need consultants to help them change because they’ve already found ways to do things differently on their own. But I have found that organizations often need help creating the conditions for diffusion of innovative approaches so that changes that need to be adopted across the whole organization by everyone are made.
Stephen Oyer-Owens: Couldn’t the glove size problem simply be solved by one person asking for a complete stocking of glove sizes throughout the hospital?
That’s a great question and it gets right at the heart of the PD process. Of course, there is already some process in place in the hospital that is supposed to result in there being all needed equipment, including gloves, throughout the hospital. There are “par levels” for stocking and people accountable for tracking stock and processes defined for monitoring etc. But these are processes that have been defined at the system level and they are processes that are designed to be applied house wide. It often turns out that when you get down from the system level to a particular room in a particular unit on a particular floor – the process doesn’t deliver as planned. The gloves are stored in an opaque cupboard so it’s not easy to see when the supply gets low or empty. There are significantly more nurses with small hands in this unit than “average” units so they run out faster than average. The up tick in compliance with glove wearing that has resulted from increased understanding of the importance of these measures means that the number of gloves that was “enough” during this month last year is not enough this year. It turns out that for a half dozen “best practices” that have been identified as important to preventing infection transmission there are thousands of local, idiosyncratic, accommodations that need to be discovered or invented to make it possible for a individual in a specific place to do that practice – so just TELLING them to “keep all those gloves stocked” doesn’t’ work. Instead, the glove wearer and glove stockers and glove order-ers need to get together with others in the community (including the unusual suspects that haven’t previously been seen as having anything to do with gloves and their availability) and invent a process that works for that particular situation.
Charlotte Miller: So, I'm also interested in Sue P's question - what are the specific actions you take to begin working in a system? If these are all volunteers, how do they know there is an opportunity to solve? Chris van Bergeijk: are local people the ones to define that there is a problem at the beginning?
One criterion for what makes a problem amenable to a PD approach is that there needs to be some agreement that there IS a problem. As is true of many organizational/community change projects, it usually starts when a core group of people share a concern and want to make something happen. Typically, one or more of them have heard about PD and invite us in to talk about it. We introduce them to the PD approach as a possibility and leave it to them to decide whether it makes sense to them that this could be a way to work on the problem.
In the case of the MRSA projects, the hospitals involved had already identified that has a big problem they wanted to solve. But often, DEFINING the problem in a way that can be approached with PD is the biggest challenge and we spend a lot of time on that phase. In the example Monique described at Waterbury hospital they started out talking about the problem as about communication but they needed to find something much more specific and focused to work on (in that case, problems patients had with taking medications appropriately after leaving the hospital). Our perspective that this is a complex system helps us get that by working on the specific problem we are actually having an impact on the larger system.
Once a problem is defined, what we typically do next is engage a core group in learning how to facilitate what we call Discovery & Action Dialogues (they sometimes call them something else). We find that the sooner they get out into the organization and have the opportunity to engage a lot of people in a lot of these dialogues, the better because that triggers an emergent process where stuff starts happening and leading to more stuff happening.
Elena Papavero: Any link here to tempered radicals? Are positive deviants found at all levels in the org?
There are definitely people at all levels of the organization who have discovered better – positively deviant – ways of doing things. One subtle but important perspective in the PD work is that it is NOT about the individual deviant (or radical or maverick or creative etc.) but rather about their behaviors that all of us can choose to adopt. I’ve read many accounts of how extraordinary individuals have found ways to do things outside the organizational norms and managed to make a difference, get something done, etc. There are leadership texts about ways leaders can identify and support these individuals which is a good thing to do. There are books about how you can become one of these cool people! But this rarely changes the organization itself. The PD approach is about engaging everyone in a different way of interacting around problems. We do highlight the creative ideas of particular people and make sure they are spoken of via the grapevine but it’s not about “being like” Sally but rather finding ways to do what Sally does.
Denise: Would like to know how PD is different from participatory action research (PAR) approaches. Thanks.
I think the Positive Deviance approach fits well in the family of approaches that includes PAR. (from Wikipedia: PAR proceeds through repeated cycles, in which researchers and the community start with the identification of major issues, concerns and problems, initiate research, originate action, learn about this action and proceed to a new research and action cycle.) There are some nuances that may be important or at least interesting distinctions in how it’s applied. One is that a lot of the PD work is focused on many many “micro” changes in behavior where much PAR work addresses things at a policy or larger group level. I can imagine situations in which a PAR approach could serve to identify the ‘evidence-based’ practices that are important to the change (analogous to the research showing that hand-washing is critical in the health care community). We could think of this as the WHAT part of the equation. PD serves to help people in different parts of the community figure out the HOW part of the equation that is essential to successful implementation of the WHAT.