Jonathan Welch with his mother. Following her death, Welch and his family sought reassurances that others would not be treated as she was. (Family photo/Courtesy Jonathan Welch)

Back when I was training to become an emergency physician, I’d worry about the day I’d be involved in a medical error. It seemed inevitable. With land mines everywhere — the possibilities of missed diagnoses, delayed treatments, miscommunication— it felt as if almost anything could lead to catastrophe. I imagined attending the in-house case review afterward, chastened as my hospital colleagues dissected my decisions. Yet I also thought that something positive would come from the process, that lessons from an error would sharpen my clinical skills and improve care in the hospital.

But when I was entangled in my first medical error, I played an unexpected but common role: I was a 33-year-old son trying to save my mother’s life.

The call came to me in Boston out of the blue one morning in September 2010. On the line was an emergency physician in the Wisconsin town where I had grown up, telling me my mother was sick. He sounded harried, and I heard papers rustling in the background.

My mother was already fighting breast cancer, but he was calling about a new development. During the preceding year, my mother had been getting chemotherapy for a recurrence of breast cancer, for which she’d treated in 2008. The chemotherapy had been working: My mother’s cancer was stable.

But she’d suddenly worsened on the morning of the phone call. My dad found her barely responsive in a chair at home, and he called 911. When she arrived at emergency department of the local hospital, she had a high fever, her heart was racing and her white blood cell count— critical for fighting infection — was extremely low. I knew what was wrong: neutropenic sepsis.

I often diagnose and treat cancer patients with this condition. It meant that my mother had a serious systemic infection complicated by her body’s severe shortage of infection-fighting cells, which had been nearly eliminated during chemotherapy. The condition is well known, easily diagnosed and has a clear and standard treatment protocol.

I also knew we were on the clock. The first 24 hours of my mother’s hospitalization would be critical to saving her life. Studies of sepsis have shown that early and aggressive treatments during that time can make the difference between life and death. The needed interventions include continuously monitoring vital signs, and giving antibiotics and lots of fluids. Very sick patients need to be treated in an intensive care unit with a special central-line IV that delivers powerful medications.

My mother was in serious trouble. I caught the next flight home.

The clock is ticking

Once home, I hurried to my mother’s hospital bed, where the rest of my family already was gathered. I arrived about 9 p.m. “How are you doing?” I asked her. “Fair,” she replied, her voice guttural and weak.

Sequestered during my flight, I’d been cut off from updates. But I’d kept doing the math. The hospital now was 12 hours into its critical opportunity to halt her systemic infection. I was eager to know my mother’s heart rate and blood pressure, two basic indicators of whether she was getting better. I peered above her hospital bed, looking for the cardiac monitor providing this information. It wasn’t there.

Confused, I approached my mother’s nurse, thinking the monitor might be near her work station. But it wasn’t there, either. This could be dangerous: Without a monitor, her doctors and nurses could miss sudden changes in her vital signs that would require swift action.

I leaned over and took a look at my mother’s medical chart. Some infrequent vital signs had been recorded. And I saw a clear, terrifying picture. My mother’s blood pressures had crashed during the day. Her numbers now were half of what they’d been at her arrival in the emergency department. My mother’s emergency physician and oncologist had done few if any of the essential and obvious interventions needed to save her life. The nurse seemed calm, as if everything was normal.

What was their problem? Was I missing something? I felt trapped in an alternate reality where the medical rules were the opposite of everything I’d learned and practiced. By now it was past 10 p.m., and my mother’s oncologist, who was in charge of her care, had left the hospital for the day. I couldn’t figure out what he’d been thinking.

I did know, however, what my mother needed: to be in the hospital’s ICU, where the standard treatments for a systemic infection could he started. “I want my mother transferred to the ICU right away,” I told the nurse. “She’s getting worse.” When the transfer was approved I felt relieved that my mother would finally get the care she needed.

Getting her to the ICU

My mother was moved to the ICU around midnight, 15 hours after she’d arrived at the hospital. I figured I’d get a bit of rest once her central-line IV and other treatments were started.

But during the next hour, I slowly realized little was changing. I didn’t see an intensive-care doctor, or any doctor for that matter. No central-line IV or the needed powerful medications were started. What was going on?

By 1 a.m. I was panicking. When I asked the nurse about the treatment plan, her response was a not-so-veiled criticism of my mother’s doctor. “We do have a sepsis treatment protocol,” she said, “but your mother’s oncologist hasn’t ordered it.”

“Her oncologist?” I was surprised. At the hospital where I work, critical-care doctors are the specialists who treat ICU patients with systemic infections. Wasn’t that the case here? “Your mother’s oncologist hasn’t consulted the intensive-care specialist,” she told me.

My mother now had been in the hospital for 16 hours without receiving the sepsis protocol — the standard group of treatments and actions — that she needed to save her life. The clock kept on ticking toward 2 a.m.

I wish I’d done more at that point — insisted on waking both my mother’s oncologist and the hospital’s intensive-care doctor at home, demanded that they come to the hospital. Instead, by that point I felt lost and powerless. I’d already insisted that my mother be moved to the ICU. What would happen if I made additional demands? Would the ICU nurse start avoiding my mother’s room? If I criticized my mother’s oncologist, what would happen to their relationship? I knew there could be a downside to being too demanding in a hospital.

I was losing my own confidence as a doctor, becoming instead the helpless son. Every 10 minutes or so, my mother, uncomfortable in the stiff bed, asked me to get her up. “Please,” she begged. I couldn’t do that with her blood pressures so low; I could only help her change position.

By 3 a.m., I’d given up on the hospital. My single thought was: “We’ve got to get her out of here.” I began making plans to move her to another hospital in the morning. The transfer request forced my mother’s oncologist to relinquish primary responsibility to an intensive-care doctor who provided the right treatments. In the end, she stayed put.

But by the time the sepsis protocol was finally put in place, 23 hours had passed since my mother had entered the hospital. She still had a chance to survive, but because of the squandered opportunity, it was now a very small one.

During the following days, my mother battled hard. But her infection and the delay in treatments proved too much. Toward the end, in a final moment of lucidity, she opened her eyes and whispered, “I never got to say goodbye.” She was dead by the end of the week.

‘Do you want to sue?’

In the midst of grieving and preparing for my mother’s funeral, I was also lost in a haze of disbelief. Why had so many things gone wrong?

When I discussed my mother’s case with my Boston colleagues, they were incredulous. “Do you want to sue?” they asked. I felt angry and abandoned by my mother’s oncologist, the emergency physician and the nurses who remained quiet when the right treatment wasn’t used. But as the weeks progressed, any thought of a lawsuit didn’t sit well. The prospect of lengthy legal proceedings seemed counterproductive. And I came to believe that suing the hospital and doctors would interfere with our family’s accepting my mother’s death and the normal grieving process.

As a doctor in a high-risk field, I also considered how destructive a malpractice suit could be. If we sued, my mother’s oncologist would be a central figure in the case, as the debacle fell squarely in his lap. He was imperfect, but he’d treated my mother’s cancer, both 14 years ago when it first appeared and during her recent recurrence, and she’d respected him. He was nearing retirement, and a lawsuit would be a terrible way to end his career. I didn’t want to do that to him, even though he’d never explained his inaction. A lawsuit just wasn’t in line with our Midwestern-family mind-set.

An explanation or an apology

Mostly, we wanted the hospital to provide an explanation for what happened. And we wanted an apology. Additionally, because the hospital had wasted opportunities to save my mother, we wanted assurances that it would never do that to anyone again.

“If you want to improve quality,” a colleague suggested, “why don’t you write a letter?” Send it to the hospital’s administrators, he suggested, describing my mother’s experience and suggesting specific ideas for improving how systemic infections were treated.

My family crafted our letter carefully and positively. In it, we suggested specific ways that the hospital could achieve national standards for treating systemic infection. Given the technical nature of our concerns, we asked that I be contacted as our family’s spokesperson. We mailed the letter to the hospital president, chief medical officer and chair of emergency medicine, and then we held our collective breath.

Within a week, I received a voice-mail message from the director of the hospital’s emergency department. “Hi, Dr. Welch, I did receive your letter today. . . . I have already had discussions with the chief of the intensive care unit, and we are planning to review this case. . . . I wanted to let you know and touch base with you to see if you wanted to have any further discussion right now.”

Finally, I thought, we’re getting somewhere. I returned the call and left a message, asking to be called back, and waited for a response. But there wasn’t one. In fact, there wasn’t a return phone call, or any communication from the hospital, for four months.

Initially I thought the administrators there might still be reviewing the case. But as the months ticked by, I began to think our letter had been tossed aside. Angry and reconsidering a lawsuit, I called the director of the emergency department again. This time my call was returned by one of the hospital’s top administrators.

The administrator, who happened to be an intensive-care physician, said he’d reviewed the case with other doctors, and, he admitted, the hospital’s actions didn’t reflect “the degree of urgency” required. A sanitized, verbal admission of error, but an admission nonetheless.

During the next month, some of our letter’s concerns were addressed, and changes were made: The emergency department was beginning to enact new guidelines for treating systemic infection. But other changes were dismissed: The hospital wouldn’t have intensive-care physicians assume immediate responsibility for all new ICU patients. The administrator said the hospital had tried to move toward that model earlier but there were “political barriers” to doing so.

He also said that the hospital hadn’t responded for four months because he was waiting for improvements to be made, and he apologized for not updating us sooner. “I’m sorry,” he said.

It was the only time our family heard those words.

Welch is an instructor in medicine at Harvard Medical School and an emergency physician at Brigham and Women’s Hospital in Boston. This article was excerpted from the Narrative Matters section of the journal Health Affairs, which can be read at