• Innovate or Standardize?

    Oct 31, 2012

    By James L. Reinertsen, MDpg43_column-jim-reinertsen

    Quality improvement calls for a continuous interplay between these dueling approaches.


    "If we standardize this so that everyone does it the same way, it will inhibit creativity and innovation."

    This is a common response from physicians when leaders ask them to adopt a standardized care process. It is especially common in academic health centers, where the faculty will add, "Besides, it's our mission to teach and to innovate. How can our students learn and how can we bring new ideas into practice if we all do things the same way every time?"

    Leaders need to know that this question poses a false choice. It's not either standardize or innovate. It's both/and.

    First Standardize

    The physicians are right when they say that, to improve care, they must try new methods. But how will they learn whether the new method is actually better than the old? That's where innovation and standardization come together.

    Let's take standardization first. In far too many clinics and hospitals, the "old method" against which any innovation might be compared is actually a chaotic hodge-podge of physicians' orders based on a combination of long-standing habits, patient preferences, and individual physicians' interpretations of medical evidence. In most cases, the physicians' choices are within the evidence-based goal posts, which are often very wide, allowing a lot of room for judgment.

    The core processes of care in such a system are highly variable, from physician to physician and from day to day. So the first task in setting up a system that supports learning and improvement is to standardize core processes of practice so that if someone wants to test a new method, there is a reasonably well-defined system against which to compare.

    Then Innovate

    Once such a "protocol-driven" stable system is in place, it's possible to start testing innovations, not as a series of bespoke individual experiments in which every physician gets to try out whatever method he or she wants, but as a disciplined test of change. A test of change doesn't have to be a formal research protocol on a drug or technology. It might be a simple test of a new way to reduce healthcare-acquired infections or to prevent readmissions. This is the essence of the science of quality improvement.

    One of the best historical examples is that of pediatric cancer. Three generations ago, a diagnosis of childhood cancer was a death sentence. But physicians started standardizing chemotherapy, radiation treatments, and other methods, and placed virtually all U.S. children with cancer on protocols-either the current best practice or someone's idea of the next breakthrough. They then compared the outcomes for children under the old protocol and the new one. If the new protocol seemed better, that became the standard, then someone came up with yet another idea for the next advance, and so on. Today, well over 90 percent of pediatric cancer is cured.


    To meet the challenges of accountable care, we're going to have to create and test many new ideas. But if we simply encourage everyone to innovate, without the accompanying discipline of standardization, we will fail to learn in a systematic way, and progress will be slow. And health care will continue to be too costly because of pointless variation in care. And worst of all, our patients will be put at significant risk for safety problems because all that individualized innovation makes our systems much more complex than they need to be. Sound familiar?

    We need both innovation and standardization.

    James L. Reinertsen, MD, is CEO, The Reinertsen Group, and a senior fellow, Institute for Health Care Improvement (jim@reinertsengroup.com).