Emily Damuth and Stephen Trzeciak, critical-care specialists at Cooper University Hospital, would like to be able to tell patients what their lives will be like if they choose to stay on a breathing machine for more than a couple of weeks.
There's just one problem: The doctors don't know.
Trzeciak said intensive-care unit physicians like him rarely see their patients again once they leave the hospital. What happens after the ICU heroics is a "black box" for them, he said.
He and Damuth found that frustrating enough that they led a study that looked at what happens to patients in the United States and other countries who get mechanical ventilation for more than two weeks.
The results were sobering.
About 70 percent of patients left the hospital alive. That's positive news. But, by a year later, about 60 percent of all the patients had died. Only half were able to breathe on their own when they left the hospital.
The research did not address how well the 40 percent who lived were doing, information the doctors said was badly needed.
The analysis included 124 studies from 16 countries but used a smaller number for some questions.
An unintended consequence of medicine's increasing ability to rescue the critically ill, the Cooper team wrote, is a "population of patients who experience protracted survival without recovery."
The study, published this month in Lancet Respiratory Medicine, raises questions on who should be a candidate for such aggressive care and what families should be told on the chances for recovery.
Calling prolonged dependence on mechanical ventilation an "emerging public health problem," the Cooper team said that chronic critical illness, which includes prolonged ventilation, costs the United States $35 billion a year.
Between 5 percent and 10 percent of ventilator patients need the machine for more than two weeks. Damuth said prolonged ventilator use is growing.
Patients in post-acute care - the facilities they went to after the initial hospital - in the United States fared worse on all measures than those in other countries, the study found. That could be because American doctors may be more willing to do a tracheostomy, a step that makes prolonged ventilation possible, in patients with "poor prognosis," Trzeciak said.
Clearly, long-term results can be disappointing, with patients "stuck between life and death," he said.
Trzeciak said some families don't even know their loved ones are likely to have to stay in another hospital - often a long term acute care facility - after they are discharged from the ICU. Only about a fifth of patients can go straight home.
"We don't think it's routine to talk about the risk that you may never get off the ventilator and be dead in a year," he said. "People have to understand what they're signing up for. Our study is just one step toward that."
Jacob Gutsche, a University of Pennsylvania critical-care physician, said that the study highlighted an important issue but that far more data were needed.
"It's difficult to apply the results of the study to any given individual," he said.
There is reasonably good information on how heart patients may do with mechanical breathing, he said, but not nearly enough for other conditions. "If you come in and you're 79 and you get hit by a car and you're on a ventilator for two weeks, no one knows what's going to happen," he said.
He wants a national database tracking critical-care patients.
In general, the doctors said, older ventilator patients with other organs that are failing are at highest risk for doing poorly.
Jeremy Kahn, who studies critical-care medicine and health policy at the University of Pittsburgh, said more information was needed not just about prognosis, but about how to improve survival rates and wean people off the ventilators, a process doctors call "liberation."
"The situation is the real wake-up call - not that outcomes must be improved, but that how to improve them is simply not known," he wrote in a comment about the study in the journal.
In the Cooper analysis, 40 percent of patients survived a year. Many would say yes to those odds. What was unknown was how mentally and physically disabled the survivors were. The quality of the studies done so far did not allow the Cooper team to measure how many patients were ever able to go home or breathe on their own.
Damuth said many might consider life on a ventilator to be of poor quality. You can't talk or eat and you're largely confined to bed.
People differ, though, in what they think is acceptable. Studies have shown, Gutsche said, that people adjust surprisingly well to significant impairment.
"People value life," he said. "It transcends religion. It transcends culture. People want to live."
He has seen patients become more open to the idea of accepting machine assistance to stay alive. Computers have made it easier to feel connected and technology is helping people compensate.
"People are willing to accept a less active quality of life," he said. "It's easier to live with certain disabilities."