Stop Trying to Change Physician Behavior

"The best way to help individuals align their behavior with their deepest motives was to stop trying to control their thoughts and behaviors. …others reach their own conclusions about the values that are most important to them and the changes that might be required for them to live according to their values."

from: Influencer: The New Science of Leading Change (2013)

Healthcare institutions are replete with marvelous machines and treatment technologies, and many more are being developed as a result of the ongoing digital revolution. These tools are transforming care delivery, and will render the practice of medicine more efficient, more personal and over time, much more effective.

However, the implementation of technology without process and behavioral change is often counterproductive. We have done this experiment many times before in medicine and other industries - adding new tools into workflow without changing behaviors that then leads to decreased rather than improved productivity. The replacement of steam engines with electric dynamos in factories, typewriters by computers in business settings, and recently in healthcare, the replacement of paper with electronic medical records are all objective examples. This phenomenon has been termed the “Productivity Paradox”.

One of my previous jobs “inside” healthcare was as a medical school and health system department chair. In that role I was frequently asked by the administrative executives how I could “change physician behavior” – usually in an effort to improve clinical documentation, or to encourage activities that would augment patient satisfaction. In my current roles as a large healthcare company executive, and an advisor to growth stage companies, I curiously find myself frequently fielding the same question. In a recent meeting while attending the 2017 J.P. Morgan healthcare conference, and speaking to a large health system CMIO about a new cloud-based analytics platform, he said that “this technology sounds good, but how can you help us change the behavior of our physicians once your analytics solution provides them recommendations?”

At its core, changing healthcare delivery requires two groups of people – providers and patients, behave differently. A great deal has been written about behavior change among patients, but what about providers themselves? How do you “change the behavior” of physicians? I begin with use of the actual terms “changing physician behavior”. 

Current training and licensure requires most physicians are of adult age, and adults usually have to change their own behaviors, not have them “changed." So how do you convince grown up professionals to change their own behavior? Let's start with what has not been consistently shown to be effective:

Money

Greg Oldham and J. Richard Hackman pioneering work on professional motivation were published in a 1980 book, Work Redesign, and later featured in Daniel Pink’s 2009 book Drive. I would argue healthcare delivery is more cognitive than assembly line. If you agree, their findings that skill variety, task identity (my work accomplishes something), task significance (my work impacts others), feedback (you tell me how I am doing, and complement me when I am doing well) and autonomy are significantly more motivating than money should ring true.

Data

I was recently asked as a panel discussant if doctors were suffering from “data fatigue”. I responded, “no, they’re suffering from insight incompetence”. I don’t imply that doctors don’t have insight, but that insights based on data evaluation are often not provided in a useful way to them. Using data to compare physician productivity measures may impact behavior, but what I am referring to here is providing them with more specific clinical and economic information indicating a clear need for a change in their current processes and behaviors. Rather than a list of patients with elevated HgA1c values, how about enumerating those patients actually requiring intervention, the best intervention options, and how those can be delivered such that caring for the other few thousand patients on the panel will not be rendered more difficult as a result? Physicians don’t just want data - they want insights, recommendations and at times operational support. Delivering acute care is a full time job at best, and these new “opportunities” to improve care delivery require extra time and effort when capacity is already limited.

From my own personal experiences, as well as some recent insights from behavioral science, here are a few suggestions of what may actually work :

Absolute Transparency

Answer the question “what’s in it for me?” as honestly as possible. If this means some short term sacrifice for a longer term gain, make sure you share both of these realities, and provide details. If the sacrifices are easy to articulate but not the benefits, think twice about what you are asking them to do. Medicine is an altruistic career choice, not a masochistic one.

Instilling Confidence

Leaders seeking change in behaviors mistakenly assume content knowledge and skill are one and the same. Telling someone she needs to high jump seven feet might be aspirational, but not realistic. The clinical culture rejects and abhors incompetence. Over-invest in training and environmental support to ensure there is confidence the new behaviors can be performed leveraging new tools and approaches - don’t just “implement” them.

Appealing to the “Higher Order Values” of Clinicians

The business ethic has seeped deeply into the fabric of healthcare, and productivity metrics have largely crowded out the empathetic and society-serving motivators that encouraged most physicians to aspire to a medical career. Telling them to be “more productive”, even if linked with monetary incentives will not have a lasting impact on behavior. However, proving to them the new behavior will allow them to deliver better patient care, and and accomplish more of their original goals of “helping people” - as trite as this sounds - can be powerful in this era of 8 minute primary care visits. 

A few years ago, I struggled for several weeks to encourage a group of primary care physicians to help launch a CMS Shared Savings Accountable Care Organization. I enumerated all the new technology and approaches that would be available, but wasn’t getting any traction. After several meetings I finally asked the group if they were happy with the clinical outcomes that their patients were achieving – the “help” that they were actually providing to them in the current structure of care delivery. The consensus answer from the group was “no”, which then led to the collective question, “how can we do better?”