Wednesday, August 26, 2015

Do We Really Learn From Our Mistakes?

Over two years ago, the folks over at the athenahealth kindly invited me to submit columns to their Health Leadership Forum, and I have done so on an occasional basis since them. As I recently reviewed the columns, I realized that my own thoughts on the topics of leadership and coaching have evolved a bit, and I thought my readers over here at Not Running A Hospital might enjoy witnessing the transition. So for the last several days, I have reprinted the posts from the Forum over here. Comments are welcome at the original site and here. Today's reprint, with added photos, is the last of this series and is from a post dated July 20 2015, "Do We Really Learn From Our Mistakes?"

It’s often said that we learn from our mistakes. Indeed, many a business course in leadership offers that premise as a given. I’ve glibly repeated this often in my classes, speeches, and advisory work.
“You don’t learn from your successes,” I point out, “but rather from your errors.”

But do we really learn from our mistakes as a matter of course?

My friend and colleague Michael Wheeler, in his wonderful book The Art of Negotiation, warns us that it is:
all to easy to be overconfident about our ability to observe and learn. A leader who ruled his country for more than forty years put it well: “The truly strange thing in your lives is that you not only fail, but you fail to learn your lesson . . . No matter how much your beliefs betray you, this is never accepted by you. You are distinguished by your inability to recognize the truth, no matter how irrefutable.”
Wheeler continues:
It one thing to recognize this truth in the abstract, but it’s another to live by it. The writer was the Libyan leader Mu’ammar Gaddafi, who several years later refused political asylum even as his regime was collapsing around him. Gaddafi was captured, beaten, and killed by rebel forces.
Sometimes our inability to be reflective practitioners derives from cognitive errors and biases. Because these failures are cognitive, it is almost impossible to see them happening or, afterwards, to realize that they have occurred.

Cognitive errors show up in many forms. Of the most common are:
  • Anchoring: the tendency for your first observation to carry disproportionate weight in your decision-making.
  • Confirmation bias: often accompanied with anchoring, our confirmation bias values evidence that seems to support our view while discounting evidence that is contrary to your view.
  • Recent experience: Even statistically irrelevant recent events carry more power merely because of their placement in time.
  • Patterning: We are prone, too, to see patterns that don’t exist. Our minds like order, and we will assert the existence of dispositive parameters—even when the actual pattern of events is totally random.
We teach doctors about these cognitive weaknesses — anchoring, confirmation bias, and patterning — but we tell them that they are unlikely to recognize that they are happening. Instead, we need them to buy into systems of group behavior that protect them from themselves.

An illustrative example comes from Joris Lemson, MD PH.D., medical director of the pediatric intensive care unit at Radboud University Nijmegen Medical Centre in the Netherlands. One day, he ordered a dose of strong medicine for a small boy. The nurse obeyed the order, and the boy almost died from the choice of medication.

Later, when the doctor confessed his distress to the nurse, she said, “I wondered about the choice of drugs. If you had been an inexperienced doctor, I would have questioned the order. But I figured, with your experience, you would know what you were doing, and so I didn’t say anything.”

In relating the story to me, he said, “It was at that moment that I realized that I needed to be protected from my own mistakes.” He then instituted a strong training program in Crew Resource Management (CRM). This set of techniques, derived from military aircraft cockpits, offers particular help in hierarchical situations. It empowers subordinate members of the team to interrupt a pilot, doctor, or other chief and help that person from making a serious error.

Joris is honest about the progress of this effort in his PICU. He notes improvement and general compliance with the approach and procedures, but he also notes lapses. For instance, sometimes he as leader will forget to conduct the debriefing. That’s all right, but not if the other crew members forget to remind him when it happens. A tenet of CRM is mutual responsibility and authority: If the chief forgets to carry out part of the protocol, the others are required to point this out.

Oddly, those of us in more office-based leadership positions do not protect ourselves from this kind of error. We might tell people that we want to hear when we are going wrong, but do we behave in such a way that those call-outs are encouraged? Do we greet an interruption or criticism with a gracious smile and a thank-you? Or is our (perhaps unconscious) scowl of displeasure enough to teach subordinates that they are proceeding at their own risk by doing what we think we told them to do?

We need to understand that there is an uneven pattern of power in the boss-subordinate relationship. Our reports, for good reason, have learned over the years that the person who points out that the king has no clothing often ends up on the street or left behind when it comes to promotions or other career advancement. With the scowl, we cement that fear into people’s everyday lives.

Michael Wheeler summarizes the issue by saying, “You have to monitor your own behavior to make sure it aligns with your intentions.”

2 comments:

Carole said...

Topics about truth tend to get the least amount of replies or comments, obviously a touchy subject! Very good post, Mr. Levy.

x said...

Actually no they don't. I have never sent a lawyer to the "monopoly" in town. I have sent in loads of suggestions, questions with medical research, and HIPAA additions to records. What I get is silence or lawyers or risk managers (which is the same thing).

I don't get physicians or process improvers.