Improvement on Ice: Curling and Quality Improvement Science

Submitted by Mary Smillie on February 5, 2012 - 10:15

Mary Smillie, Health Quality Consultant, Saskatchewan

Ever heard of Tim Horton’s Brier?  The Scotties Tournament of Hearts? No?  Well, I’m guessing you’re not Canadian. They’re Canada’s national curling championships, and I don’t mean hairdressing. If you’re into quality improvement in health care, you’ve got something to learn from curling.

Curling is a frozen cousin of lawn bowling. The game is played on ice. Players slide 40 pound polished circular stones down a sheet of ice 170 feet long to a set of 4 concentric rings spanning 12 feet in diameter (the house). Teams of four have eight rocks each per end (an end is like an inning in baseball).  There are 3 basic shots: a draw to a specific spot in the house; a guard that protects a stone from being removed by the opponent; or a takeout, which knocks an opponent’s stone out of play.  At the highest level it is a game of extreme precision. A team scores points for each stone closer to the center of the rings (‘the button’) than any of the opponents’ stones after all 16 shots have been made.

Though the game was not invented by the Institute for Healthcare Improvement, it is a veritable poster child for the Model for Improvement and Plan, Do Study, Act Cycles.   In the 1980s and 90’s Associates for Process Improvement (API) worked closely with W. Edwards Deming, the grandfather of quality improvement to learn his techniques. API developed the Model’s framework to synthesize Deming’s concepts for developing, testing and implementing changes that result in improvement. Using the Model involves answers to three fundamental questions:

1. What are you trying to accomplish?

2. How will you know a change is an improvement?

3. What changes can you make that will result in improvement?

 Underlying all of three questions is the Plan, Do, Study, Act (PDSA) Cycle.

Curling can help us learn how to apply this model. The initial question forces health care teams to think deeply about what outcomes for patient populations they’re seek to achieve. In curling it’s easy: win the game. In health care answering this question is surprisingly difficult. If we want to improve  health care performance, we have to be clear about the end game for patients/families.

The second question requires data to inform the team whether they are making progress towards the aim from question #1. Again in curling the answer is straightforward:  the scores after each end are perfect indicators of whether our changes are leading to improvement. In health care deciding which  data best indicate  whether we are getting closer to or further from our aim is more challenging, though just as important as tracking the score of a curling match.  .

The third question is an invitation to try, test and learn about which changes actually improve performance.  It is this final question, “what change can we make that will result in improvement?” where curling really shines as a teacher of improvement. The PDSA cycle becomes the team’s mechanism for testing different ideas to see which contribute to better outcomes, and which do not.

The vast majority of curlers have never heard of the Model for Improvement or PDSA cycles but they perform them the entire game. All four team members continuously make changes based on the results of their previous shots (PDSA cycles). Did the stone curl more or less than predicted?  (Making explicit predictions about what you think will happen is an often overlooked key learning aspect of PDSA cycles.)  Is this particular game’s ice surface fast or slow?  Did the team accurately direct the stone at the broom (target), and were errors systematic or random?   Based on this learning, the skip will modify the call for the next rock. At the same time, the opposing team is running its own PDSA cycle.  Both skips and teams will continuously learn from each other’s success and failures to improve their chances of scoring in each end, and ultimately winning the game.

Every stone thrown presents an opportunity for testing and learning as does the overall strategy for each end. The team with the last rock advantage will try and set up the end to score at least two points.  The opposition will work to cover up the center of the rings to counteract the last rock advantage and ideally ‘steal’ one or two points through a defensive strategy. The skip for each team will have a plan (and prediction) at the start of the end based on whether or not they have last stone. If, as the end progresses, the skip recognizes the strategy will not work as planned, he/she will modify the strategy based on  the opposition’s performance and the strengths and weaknesses of the other team members. It is a perfect incarnation of continuous improvement theory and practice in real time.

Contrast this rapid learning and improvement approach with the conventional approach to planning and change in health care. A team is created. They spend months in discussions and planning. They impleCurling pantsment a suite of directives. Everything is implemented in one fell swoop, and not  evaluated until after full implementation.

When you compare the traditional approach to change with the Model for Improvement approach, you can appreciate why health care practitioners find the Model for Improvement so foreign. Short and simple are revolutionary concepts in health care.

Healthcare teams could learn about improvement science from watching curling. They just have to look past those terrible pants from Norway.


Commenting on this Blog entry is closed.