The 34-year-old man lay in his ICU bed for over a month — dependent on a breathing tube and artificial respirator to stay alive. The patient knew his life hung in the balance, as he was a physician himself. Some days the suffering was so intense that he contemplated ways he could unplug the machine on his own.

Now, nearly 50 years later, that patient, Edward Viner, an oncologist who served as chief of the Department of Medicine at Cooper University Health Care in New Jersey for more than two decades, reflects on how he was able to survive such a harrowing experience.

It was the nurses he calls his “angels.” But it was not all of his nurses. When it was time for shift change in the ICU, Viner says he felt that he could detect almost immediately if the nurse coming on duty truly cared. He could tell some nurses cared deeply, but some did not.

“When my nurses cared,” he distinctly remembers, “I knew that shift would be a positive experience and that their compassion would help me fight on and help save me.”

With all of Viner’s knowledge from a lifetime of treating patients, does he really believe that his nurses’ compassion changed his outcome? Is there data to back up the claim that caring can make a difference and that health-care outcomes are not just dependent on how much health-care providers know, but rather how much they care? We do not raise this consideration on ethical or emotional grounds, but rather on the basis of medical science.

Research shows there has been an erosion of the relationship between those who provide health care and the patients they treat, and specifically an erosion of compassion. Nearly half of Americans believe the U.S. health-care system and health-care providers are not compassionate, one survey found. Numerous studies have reported that physicians miss the majority of opportunities to respond to patients with compassion. Research on the burnout epidemic in health care finds that 35 percent of physicians are so burned out that they have an inability to make a personal connection with patients. This can result in callous or uncaring behavior.

And now, with nearly universal implementation of electronic medical records, health-care providers spend more time staring into computer screens than looking their patients in the eyes. So long as computers are substituted for conversations, doctors are hard-pressed to know their patients as people — much less offer compassion.

So the evidence shows that health care is in the midst of a compassion crisis. But this raises a pivotal question: Does compassion really matter?

It is axiomatic that health-care providers ought to treat patients with compassion. It’s a moral imperative that is vital to the “art” of medicine. But is compassion just in the art of medicine? Or are there also measurable beneficial effects belonging in the science of medicine?

To answer this question, we embarked on an eye-opening, two-year journey through biomedical literature, curating data from hundreds of published scientific research studies. We report the results in our new book, “Compassionomics: The Revolutionary Scientific Evidence That Caring Makes a Difference.”

But what is compassion, specifically? Scientists define compassion as the emotional response to another’s pain or suffering involving an authentic desire to help. A closely related term is empathy — which is the first step of feeling and understanding another’s emotions — but compassion goes beyond empathy because it also involves taking action (You can think of it like this: empathy + action = compassion).

Our aim was to systematically analyze the published scientific data to focus on the effects of compassion for patients, patient care and those who care for patients.

After curating all the scientific evidence, we found undeniable signals in the data that compassionate care is associated with vast benefits for patients across a wide variety of physical conditions, such as chronic low back pain, diabetes and even recovery from the common cold.

Compassion is also associated with better psychological outcomes for patients, including relief from depression, anxiety and post-traumatic stress disorder. One mechanism by which compassion for patients can improve patients’ health is through better patient self-care, such as better patient adherence to prescribed therapy. When health-care providers care deeply about patients, and the patients feel that, research shows they are more likely to take their medicine.

Patient adherence to prescribed medication does not just improve outcomes, it also helps chip away at the avoidable downstream health-care costs that pile up when chronic diseases go unchecked — estimated at $100 billion to $290 billion annually in the United States. Considering the United States spent 17.9 percent of its gross domestic product on health care in 2016 ($3.3 trillion), and it is expected to grow to 19.4 percent in 2027, it is important to think about any interventions that can help bend the cost curve.

Care that is compassionate and patient-centered is also associated with lower health-care resource use and lower overall health-care costs. It appears that health-care providers who do not have a strong connection with their patients may be more reliant on (expensive) testing and technology. Furthermore, physicians who are viewed as caring less about their patients are more likely to find themselves at the center of a medical malpractice case.

There is also emerging evidence that more human connection in health care could be an effective way to combat the burnout epidemic that is rampant among those who provide care. Compassionate care can be a fulfilling experience for health-care providers that helps build resilience and resistance to burnout. In other words, compassion is good for both the giver and the receiver.

On functional magnetic resonance imaging of those providing compassion, researchers can see it activates the areas associated with reward and positive emotion. Burnout does not occur from getting too close to patients, it is actually the opposite. Having better relationships with patients is actually protective against burnout.

But is there a “cost” to compassion? For health-care providers to show compassion to patients, doesn’t it take a lot more time? Actually, no (or at least not much). Research shows that it takes only 40 seconds for a health-care provider to communicate compassion. — “I know this is a tough experience to go through,” “I am here with you” and “We will go through this together.”

The science is clear: Compassion matters — in not only meaningful but also measurable ways. In addition to compassion being the right thing to do, science shows it is also the smart thing to do. There is science in the art of medicine, and the science is strong.

As the national dialogue about health care continues over the 2020 election cycle, there will be intense debate on how to improve the U.S. health-care system and how to pay for it. We interject into this debate a vital consideration: Science points to compassionate care as a powerful, evidence-based therapy that is essentially cost free.

Trzeciak practices critical care medicine at Cooper University Hospital in Camden, N.J., and is chief of medicine at Cooper University Health Care (CUHC). Mazzarelli practices emergency medicine there and is co-president of CUHC.