Data and transparency is needed to reduce medical mistakes

An excerpt from On the Mend: Revolutionizing Healthcare to Save Lives and Transform the Industry.

by John Toussaint, MD, and Roger Gerard, PhD

Admitting Error

In a lean environment, doctors and nurses must allow mistakes to be visible in order to perform root-cause analysis and fix the process. But showing mistakes hits most medical providers in a vulnerable place—right in the collective fear of lawsuits and a highly conditioned need to be heroic.

Doctors are deeply reluctant to point out the mistakes of others, much less officially reprimand one another—knowing that they all hide mistakes, and being sympathetic to the pressures faced by colleagues. To issue an official reprimand is to risk destroying the career of a doctor. So errors are not discussed, except in the rumor mill.

Moving doctors from their hard-earned autocracy into becoming team players, where they share responsibility and—to some degree—decision-making is no simple matter. As ThedaCare’s lean initiative spread across hospitals and clinics, leaders of the movement made mistakes and learned a few lessons worth sharing that can be summed up in three words: data, urgency, and trust.

Data Drives the Scientist

Deep within every doctor, a scientist lurks. Trained in data collection and usage, taught to rely on the scientific method, doctors are most comfortable with arguments that include numbers. Unfortunately, the fear of malpractice and damaged reputations has made medical professionals profoundly reluctant to publicly release scores on critical quality markers. That fear must be conquered.

A lean healthcare initiative always begins with data collection and dissemination. What data is collected, and how it is presented, will change over time as an organization’s needs and focus changes, but getting and broadcasting the facts is always necessary because data can cause people to change behavior.

For instance, shortly after that early improvement week in Labor & Delivery—when a young mother helped redesign the birthing process—a group was taking a closer look at the neonatal value stream and noticed that a surprising percentage of babies were delivered earlier than the normal gestation time of 39 or 40 weeks. Preterm birth is defined as occurring at or before 37 weeks and 12.7% of U.S. babies are born preterm, exposing them to medical complications and developmental delays. However, a number of recent studies have shown that babies born even a bit too early—at 37 or 38 weeks—have a greater chance of chronic respiratory disease and learning disorders than children born at 39 weeks or later.27 At ThedaCare, 35% of babies were born during this “early term” period.

An improvement team dug deeper into the data, made additional inquiries and found that many of these babies were delivered early on purpose—by inducing labor at a prearranged date agreed on by mother and doctor. It may have been convenient for physicians and families, but it put those babies at higher risk of complications at birth and often resulted in weeks spent in the neonatal ICU. (ThedaCare tracks babies’ time spent in the neonatal ICU as one measure of the relative health of premature babies.) The team worked with staff and doctors to create new protocols, including setting a 39-week lower limit for inducing labor.28

Adherence to the new protocols was spotty at first. Then, physician performance data was posted on walls in the Obstetrics departments, with every physician’s name over his or her track record, meaning no labor was induced prior to 39 weeks gestation for scheduling convenience. There was 100% compliance on the new protocols within a month. As a result, premature babies requiring intensive care now spend an average of 16 days in the ICU instead of the previous 30.

Doctors are competitive by nature. It is a necessary attribute to getting through medical school and then earning desirable residencies and fellowships. Making data public—if the data is honest and relevant—taps into every doctor’s competitive nature. Presenting unblended physician performance like management did in Labor & Delivery caused some grumbling, but it also ignited a drive to be the best.

Data Can (and Should) Drive the Patient

For the public, straightforward comparative data is difficult to acquire. Even simple statistics such as the rate at which patients are infected during a hospital stay can be difficult to find. Disclosure laws vary state by state and even when hospitals are required to report infection rates to an independent oversight organization, the information does not necessarily get reproduced in a public-friendly way. Data can confound as easily as it can inform, after all.

Data accessibility needs to change—and become standardized—to give healthcare providers impetus to improve. Currently, hospitals with poor quality records can be financial winners, as long as their performance remains unknown. If people are informed as to the quality and safety records of all hospitals, however, the hospital that focuses on improvement should have the advantage.

John Toussaint is President of the ThedaCare Center for Healthcare Value and Roger Gerard is ThedaCare’s chief learning officer. They are the authors of On the Mend: Revolutionizing Healthcare to Save Lives and Transform the Industry.

Submit a guest post and be heard.