How do you make a life and death decision for a stranger?

When the patient has no family ties, his future lies in the hands of people he never knew.

Meredith Sadler for the boston globe

He came into my life a few weeks into my medical ethics internship at Beth Israel, when my supervisor asked me to write a report on his case. I was a volunteer in my first year out of college. He was a patient, and a puzzle. Some months prior, he’d choked, dialed 911, and collapsed. EMTs resuscitated him, but lack of oxygen to his brain had left him vegetative. His nurses had asked for an ethics consult to answer the question of whether we should prolong his life.

Normally, families make this decision. But this patient hadn’t spoken to his family in many years. He had no one to decide for him. Cases like this rarely happened, and his fascinated me.

In college philosophy classes, my fellow students and I had wrestled with moral dilemmas through thought experiments, but we used them to examine theories, not to make actual decisions. One famous example is the trolley problem. The particulars vary, but the concept is the same: A runaway trolley is speeding toward 10 people, and you can flip a switch to divert it, saving those 10 people but killing one other person instead. What do you do? In the classroom, it didn’t matter what we decided; a discussion of the trolley problem didn’t end with a visit to the tracks. But this patient was real, and the decision wasn’t an exercise. It mattered.

As I skimmed articles and abstracts in my office, in the same building as the patient’s room, the case haunted me. He was Schrodinger’s cat, caught in limbo: Should he be kept alive or would he be better off dead? The uncertainty made me squirm. Was caring for him saving his life or defiling his body? The answer could be in the next paper my research turned up. I kept clicking.

An article in Science questioned whether all patients diagnosed as vegetative are fully unconscious. A team of neuroscientists used functional magnetic resonance imaging to show that the brain of a 23-year-old woman, diagnosed as vegetative, responded to vocal prompts in the same way healthy brains did. Her case wasn’t the same as this patient’s. But still, could he be aware? Was he in pain or bored or lonely or ecstatic?

Before long, I thought of him more and more, at the grocery store, in the car, as I fell asleep at night. I brought him up with my friends while we were out, but asking “Would you want to be kept alive if you became permanently unconscious?” was not a good icebreaker. Even with pre-meds.

For all my interest in the patient, I never actually saw him in person. I could have asked to visit, but he couldn’t consent to a student entering his room. A visit without his permission — would it be for him or for me?

Months passed. I finished my literature review, but every few days, an e-mail popped up with a digest of new articles on the vegetative state or the care of “unbefriended” patients. Each time, I thought of the patient, 11 floors and a wing away. “Unbefriended” felt inaccurate, even vindictive, like an oversimplification of a complicated life. “Unrepresented,” a less popular term, fit better: He had no one who’d known him before he got sick.

Dirty snow covered the curbs, as it had when he came to the hospital the winter before. In January, my supervisor let me know that a court-appointed guardian had received permission to determine the patient’s end-of-life care. His tube feeds were withdrawn. He died a few weeks later.

I attended a memorial service with staff members who had cared for him. Nurses, physicians, social workers, spiritual care workers, and others crowded in. He had lived among them for more than a year. Down the hall, another patient occupied his room. People spoke, some through tears. I didn’t. I had grieved at my desk, alone.

Sarah Ruth Bates is a writer in Boston. Send comments to [e-mail]. Follow us on Twitter @BostonGlobeMag.