As we’ve seen, changing healthcare is very challenging. Part of the reason that I think it’s challenging has to do with the way we’re prepared as health professionals. We’re prepared to work as soloists. But the reality is that we work in systems and every system is perfectly designed to get the results that it gets. So if we want different results, we really don’t have any choice but to work at the underlying design of the system that’s producing those results. Every time we make a change, we in fact employ multiple knowledge systems. We bring generalisable science together with particular context with the expectation that if we’ve made improvements we should be able to measure it.
Now, if we push into those elements in that simple logic formula, we see that we build the knowledge of generalisable science by controlling context out as a variable. When that knowledge is built, it lives in journals and hard drives and we build it by keeping context out as a variable so that we can apply it in multiple settings and contexts. But once it’s built it’s kind of dead on arrival. It lives in journals and hard drives and books and doesn’t go anywhere by itself. The second knowledge that we bring together when we try and make a change is knowledge of the particular context.
Here, we’re interested in trying to understand the habits and the traditions, the processes and systems, the things that people avoid and the things that people seek out about a particular setting– the things that give the setting its identity. But you see that building knowledge of generalisable science and building knowledge of the particular context are built in very different ways. Now, the third type of knowledge has to do with knowledge of measurement. It isn’t that we’re using a different kind of number system, but rather, the variables that are necessary when you’re trying to measure for improvement.
You really want to measure over time and you often are measuring different kinds of aspects of the result– cost or the biology or the satisfaction, for example. But it isn’t just the words in the boxes in this little logic formula. It’s also the symbols, the plus sign. So it’s the plan that we have for hooking up the generalisable science in the particular context, and it’s the arrow, it’s making it happen, it’s the knowledge of execution. In this particular setting this is what it takes to bring about this kind of combination of science and the features of this particular setting. Now, health professionals are very busy people and so we use habits to get our way through the day.
And a habit starts with some kind of signal or a cue that we respond to and we recognise that cue and then engage in certain routines or actions that, in fact, to us make sense to the cues we’ve recognised. And this cue recognition and action routine is then recognised or rewarded in some way, and that feeds subsequent situations that we face allowing us to move expeditiously through the day. Another feature of health professional work is that we work with smart people. And when we want to invite others into this process of change, we have to begin with the behaviour that we’re interested in changing and then try to understand what precedes that behaviour.
There’s some structure of interpretation which to us makes this behaviour a sensible kind of action to take. We engage in the structure of interpretation by observing and testing and imagining and a variety of sensemaking strategies and what feeds that structure of interpretation is the language and the frames of thinking that we’re in the habit of using and the practises and habits. And so coaching others into this new behaviour involves recognising the importance of working on the structure of interpretation that that individual or group of individuals may be using as they try to engage in the behaviours that make sense to them.
Professor Susan Michie and her colleagues have suggested that if we explore the evidence about behaviour change approaches and elements, we find that this combination of self-monitoring, goal setting, action planning, when put together, actually leads to much more effective behaviour change. One of the trickier aspects of this has to do with the monitoring and measurement. And our colleagues and I at Dartmouth tried to conceptualise some tips that might be useful to someone trying to engage in that self-monitoring or measuring. And the tips are pretty straightforward– keep it simple, seek usefulness rather than perfection, take the discipline of writing out the definition of what you’re actually trying to measure. Use balanced measures.
Build the process of measurement into daily work so it’s not always something new or extra to be done– useful with qualitative and quantitative data. Use available data if it’s possible. Otherwise, create small and good samples. Display the data and trends over time, and establish a clear ownership for the process of measurement. But as Professor Michie has suggested, it’s not just about measurement and monitoring. We have to put these things together. We have to build new knowledge and so for me, the work of Clarence Lewis, Walter Shewhart, and W Edwards Deming over the last century putting a simple system together of plan, do, study, act, or PDSA, has been a very helpful way to take action.
And the consultants and associates in Process Improvement have suggested three prologue questions that might be helpful. What are we trying to accomplish? How will we know the change is an improvement? And what changes can we make that will lead to improvement? This model for improvement was adapted by the Institute for Healthcare Improvement nearly 40 years ago and has been the mainstay of trying to create a learning path for change as we go forward trying to make these changes actually happen. So if we believe that every system is perfectly designed to get the results it gets, we recognise that if we want different results we have no choice but to engage and redesign the system that is producing those results.
I think it helps to engage that redesign process using a simple model for improvement, as I’ve suggested. Thank you.