Is the practice of medicine more of an art or an applied science?

It’s a debate patients may not even know is taking place. But the way your primary care physician, surgeon or hospital answers this question may determine whether you live or die.

Doctors on the far “art” side of the spectrum maintain that every patient and physician is different. Therefore, they believe there is no one right way to treat a patient — labeling adherence to a common solution as “cookbook medicine.” They conclude that when two doctors approach the same problem in different ways, it’s simply a matter of each doctor finding their own best solution for the patient’s needs. They believe personal judgment and doctor’s intuition are keys to highly personalized care.

This point of view doesn’t fly with doctors who view medical care as an “applied science.” They disagree in large part with those who say there’s no one way to treat a patient. Most variations in care delivery, they argue, are relics from the days before sophisticated information technologies and advanced diagnostic tools. They rely on research and scientific evidence to shape guidelines for improving clinical outcomes. They believe in most cases, variation reflects out of date thinking and leads to poorer patient outcomes.

As with all worthwhile debates, both sides make valid points. But which side is saving more lives? Let’s examine three life-threatening conditions to help separate medical myth from reality.

The American Stroke Association (ASA) reports nearly 800,000 Americans suffer a stroke each year. It is the #4 cause of death in the U.S. and a leading cause of disability.

Recognizing that the clinical management of stroke patients varied greatly throughout the U.S., the ASA joined forces with the American Heart Association (AHA) to publish nationally accepted standards for testing and treatment.

Together, they found that when blood flow to the brain is blocked, every passing second increases the risk of impairment or death. They believe that following AHA/ASA guidelines could greatly reduce that risk, whereas taking a case-by-case approach greatly increases that risk.

A recent Journal of the American Medical Association article found that “implementation of a national quality improvement initiative” improved not only the timeliness of procedures, but also lowered rates of hemorrhaging and in-hospital deaths.

Further research shows the best hospitals are the ones using AHA/ASA recommendations to standardize admitting orders for stroke patients. These hospitals embedded procedures in a common electronic medical record (EMR), which is proven to lower rates of hospital-acquired pneumonia and death.

Simply put: Hospitals using a standardized approach are the ones lowering the chances of death and disability the most. Hospitals that allow physicians to base their admitting orders on personal preferences don’t fare so well — and neither do their patients.

Preventing recurrence of heart attacks

According to the American Heart Association, over 700,000 people in the U.S. have heart attacks each year. About 1 in 3 Americans with a first-time heart attack suffer a subsequent heart attack.

Research shows that administering beta-blockers, ACE inhibitors, aspirin and statins after a heart attack can substantially reduce the risk of future cardiovascular events.

The best physicians prescribe these medications more than 98 percent of the time, according to the National Committee for Quality Assurance (NCQA). It is easy to do, relatively inexpensive and linked to very few complications.

But according to the Dartmouth Institute, the use of both beta-blockers and statins is highly varied across geographies in the U.S.

As one might expect, regions with more consistent use of these medications “have been shown to reduce the risk of future cardiovascular events among those who survive an initial heart attack.”

Once again, science beats intuition and consistency of practice trumps individual physician preference.

Setting protocols for the operating room

In November 1999, the Institute of Medicine published a report “To Err Is Human,” revealing that up to 98,000 people die in hospitals from preventable medical error each year. A subsequent report, “Crossing the Quality Chasm,” helped to further raise awareness of these tragedies, known as “never events.”

The Joint Commission, which accredits and certifies health care organizations, has since published a set of guidelines to help prevent medical errors, including those stemming from “wrong site, wrong procedure and wrong person surgery.”

These guidelines include a number of verification processes to follow before a patient enters an operating room or goes under the knife. Protocols include a surgical safety checklist and a surgical team “time out” to verify all information related to the patient is accurate.

In the New England Journal of Medicine, Atul Gawande showed that when physicians rigorously adhere to this set of protocols in the operating room, medical errors were cut nearly in half (from 1.5 percent down to 0.8 percent).

Still, some physicians don’t subscribe to these approaches. As a result, patients suffer more often than necessary.

Are there artistic exceptions to the rule?

Let’s return to the viewpoint that every patient is different.

Patients are inclined to think so. After all, no two people have the exact same genetic makeup or opinions or life experiences. So, surely our medical problems must be unique, as well, right? Wrong.

The truth is most medical conditions have precedent. The right radiology test to order or the right medication to prescribe has a scientific basis — not an artistic one. And in the 21st century, that scientific basis can be defined and embedded into decision-support tools for doctors. They serve to define the best way to care for a particular patient and when these guidelines are followed, more often than not, they improve the quality of medical care.

Make no mistake, medical practice is a skilled profession. And there is art to that skill, particularly in building good doctor-patient relationships and eliciting a clear patient history. But today’s doctors simply have more certainty than they did in the past in how best to diagnose and treat specific medical problems.

Even with highly complex problems like cancer — where a person’s unique genetic makeup requires a “personalized medicine” solution — the best course of care has a scientific basis.

Until we accept that our personal medical issues are rarely unique — and until we recognize that variation in physician practices actually lowers quality, not raises it – we’ll only perpetuate the belief that medical practice is an art which can’t be codified and, as a result, we’ll harm more patients than we should.

Convincing physicians to choose science

Doctors can no longer blame the lack of clinical information as the reason they need to rely on their personal experience and preferences. The Internet has made the best evidenced-based guidelines accessible through nearly all mobile devices.

The problem is not that doctors can’t access these evidence based guidelines. They just choose to ignore them.

Let’s face it. None of us likes being told what to do, particularly when it comes to our areas of expertise. Doctors are no exception.

In the end, this debate is less about “art” vs. “science.” It’s about doctors valuing their own intuition over scientific evidence.

We can predict that doctors who today refuse to follow the national recommendations for treating patients with strokes, heart attacks and a variety of other medical problems will be hard to convert. But we must change their behavior. The health of their patients and our nation depends on it.

Robert Pearl is a physician and CEO, The Permanente Medical Group. This article originally appeared on Forbes.com.