Perfect Is the Enemy of the Good
In one of my former incarnations, I was CEO of a medical imaging company. We designed and built a system for magnetic resonance imaging based around a novel magnet design-an MRI machine with the world's most powerful non-superconducting magnet. Systems like the ones we had designed are today known as "open-MRI" systems because of the much wider bore in which the patient is placed during image acquisition.
But this column is not about MRI systems. Rather, it is about an important lesson I learned from that entrepreneurial exposure. You see, to create this novel MRI system, we recruited a superbly talented engineering team and turned them loose to design a magnet that many "experts" had deemed impossible to build. And design it they did. The only problem was, this engineering group of racing thoroughbreds could never quite reach the finish line. Despite deadlines and our best intentions to get into production, there were always aspects of the magnet that could be improved. And engineers being engineers, they engineer. Never ask an engineer if something can be made better, unless you have some time to find out. At some point, however, someone has to say, "Enough is enough-let's get on with it!"
It turns out that Hopkins doctors and nurses, though they might not know or appreciate the analogy, are "clinical engineers" at heart. They want to do the very best for our patients. And when redesigning systems for clinical care delivery, they want to be sure that they have optimized every step of the process.
The problem is when perfect, being the enemy of the good, freezes us into inaction. A decision to change preprinted orders in the cardiac surgery intensive care unit was made by our patient innovations team, only to get bogged down in nine separate steps of review to be sure there wasn't an uncrossed t or undotted i. More than a month's delay ensued before we could call off the engineers and implement the newer, safer orders. Similarly, a decision by our intensivists to use antibiotic catheters in certain high-risk patients has been held up for nine months in an intellectual debate about whether such a move will actually provide a cost-effective reduction in serious blood-borne infections in open-heart surgical patients. This debate is somewhat akin to trying to predict whether the light in the refrigerator goes off when the door is closed. It's so much faster to put on your longjohns and climb into the fridge to see for yourself!
Quality improvement is not about a single, billion-dollar fix. It is about a billion one-dollar changes. Think of them as micro-experiments in patient care, done in rapid succession with very short cycle times.
The importance of speed is that the shorter the time between idea generation and implementation, the more innovation is encouraged. Long cycle times allow ideas to get bogged down in the bureaucracy, either by over-engineering the change in a futile attempt to get to perfection, or by excessive mental gymnastics attempting to predict the outcome in advance of doing the experiment. In this case, a little less perfection can bring a whole lot more good.