We treated older coronavirus patients. Here's how to save more of them.

We are emergency and intensive-care doctors who have worked in three hothouses of the COVID-19 pandemic: northern Italy, New York City and Miami. Treating scores of critically ill patients, we all observed similar patterns: Many of the patients we saw in our emergency rooms had advanced cases of COVID-19 pneumonia when they arrived — and many of those critically ill patients came from nursing homes.

More often than not, these older pneumonia patients wound up on ventilators. This is almost always a bad outcome. In New York City, an astronomical 80% of patients who required a ventilator at the height of this pandemic died, according to city and state officials. (In our experience, the death rate among patients not requiring a ventilator has been relatively low.) Similarly, we see evidence that the incidence of blood clots and renal failure in patients on ventilators is significantly greater than in patients who were less sick and didn't need a ventilator.

This data does not mean that the machines themselves are killing people, just that by the time those patients are being hooked up to ventilators, they are already in dire condition.

How then do we identify patients with COVID-19 pneumonia earlier so that they can be treated before requiring a ventilator? As one of us, Dr. Levitan, noted in an earlier commentary ("What I learned during 10 days of treating COVID pneumonia," StarTribune.com, April 22), clinicians have a universally available, quick and remarkably effective tool to detect the attack on the lungs caused by COVID-19 pneumonia: pulse oximetry.

The pulse oximeter is a small device that attaches to the tip of a finger, and in 15 seconds measures oxygen saturation of the blood. Invented by Takuo Aoyagi and Michio Kishi in 1974, it is now considered one of the vital signs in medicine (along with pulse rate, respiratory rate, temperature and blood pressure). Mr. Aoyagi, who died on April 18, has had an incalculable impact on patient safety worldwide — and his contribution is especially significant in this pandemic.

In an analysis of more than 4,000 COVID-19 patients evaluated between March 1 and April 7 at NYU Langone Health facilities, one of the strongest predictors of critical illness — defined as involving ICU care or mechanical ventilation — was the patient's oxygen saturation on arrival at the hospital.

It is time, then, for the federal government, led by the Centers for Disease Control and Prevention, to mandate that all nursing homes and long-term-care facilities — tied to a third of the COVID-19 deaths — do pulse oximetry monitoring at least daily. In facilities with known coronavirus infections, we suggest this be checked twice a day.

A majority of COVID-19 pneumonia patients that we treated or observed during the surges in New York City and Italy had severe lung injury on first presentation. They were, in other words, arriving at the hospital too late, and many were winding up on ventilators.

In all medicine — whether in patients who have traumatic injury, cancer, diabetes or an infectious disease — earlier identification and treatment leads to better outcomes. COVID-19 is no different. We must continue improving the ICU care of patients with advanced COVID-19 pneumonia. But we will have the greatest public health impact if we prevent it from occurring in the first place.

The value of early detection has become apparent in northern Italy, once the epicenter of the pandemic. The surge there has abated, and patients are no longer afraid to come to the emergency department. That means patients with symptoms of COVID-19, such as fever, muscle aches and cough, are coming to the hospital earlier — and their illness is less severe.

There is some heartening evidence — admittedly inconclusive — that earlier treatment makes a difference. In a small pilot study of 250 COVID-19 cases conducted by Dr. Cosentini in Italy, half were found to have mild pneumonia — but their oxygen saturation was not yet compromised. All of these patients were able to be discharged from the ER, and they were sent home with pulse oximeters. Only 5% returned and were hospitalized when their oxygen saturation levels declined slightly. None of these patients required a ventilator. And none of these 250 patients died.

While there is no specific cure for COVID-19 and we have nothing that directly kills the virus, we do have treatments that help patients and prevent the need for a ventilator. These include various noninvasive methods of delivering oxygen, patient positioning maneuvers that open up parts of the lungs, and close monitoring and treatment of inflammation. There is no panacea; some patients will still have the disease worsen, and there are some patients who will still have serious injury from COVID-19 unrelated to the lungs.

For our country to benefit from this strategy, we need to completely change public health messaging and create a new standard of care — and that messaging must come from the federal government. If the CDC leads, health agencies around the world will follow.