Remembering Seamus Kennedy and my first, major medical mistake.


Photo credit:

I remember the first time I met Seamus Kennedy*. I was rushed, he was not, which is ironic, considering that he was a man facing death, and my only curse was to admit a patient five minutes before I was supposed to leave. Being on the inpatient hematology and oncology service at the private hospital, it wasn’t unusual for patients to come after working a half day or even a whole day, which, unfortunately, meant I often had many admits strolling in the final minutes of my own shift, necessitating me to stay additional fraction or multiples of an hour to complete their admission, depending on how fast or slow I worked.

Like I had done with all others admissions, I reviewed his chart and learned of Seamus’s advanced cancer. I learned how his CNS lymphoma had failed to respond to first-line chemotherapy, radiation, and second-line therapy. I walked briskly into his room, past an empty bed and the dividing curtain, to find Seamus being basked by the bright white of the overhead hospital lights as if to remind those around him of the inevitable. As I approached, Seamus’s bald head reflected the lights, his face hinted at a smile, even though his eyes were closed. I said hello, stated my name, and told him that I would be his doctor. I was automated, proceeding from one step to the next. But before I could, I was enveloped by the presence and personality of the person that was Seamus Kennedy. Seamus was a large person, but his personality was larger. For somebody who had months to live based on the last outpatient note – “will now refer to palliative care since we have no other options” – Seamus seemed to be as relaxed as somebody you would find in the lobby of the Ritz Carlton on South Beach, and just as happy. For every question I asked Seamus, he doubled back, and I patiently answered all of his queries. It became clear that Seamus didn’t come to just get palliative chemo, he was hell bent on living every moment of his life, just as he had done until the moment he was diagnosed with brain lymphoma. I lost track of time and even strayed far away from the usual script of asking about nausea, vomiting, and the other usual cancer questions. I asked him about his Irish heritage, whether he bore any relation to the Kennedy family, even a distant one, even a really distant one, and where his favorite parts in Ireland had been on his many return trips to the Emerald Isle. His responses were never short and not without his own set of questions for me; Seamus’ grandparents had immigrated to America in the late 19th century, he bore no relation to the Kennedys of American politics (not even a really distant one), and his favorite place had been Galway, a part of Ireland I did not have a time to visit during trip there only a few months before I started my internship.

Even though I had changed my tempo during our initial encounter, I regret that I had initially started the interaction so mechanistically. I had not expected to find a friend in that setting or that hour, but I did. I am glad that I was able to appreciate the beauty of the person that was Seamus, something that is rare as so many patients lose their personality to the disease process and we, as physicians, make little effort to uncover the person underneath. William Osler, one of the founding physicians of Johns Hopkins Hospital and perhaps of modern medicine itself, once said, “It is much more important to know what sort of a patient has a disease than what sort of disease a patient has.” The reference to Osler is ironic as Osler established the 24/7 residency program, where interns were interred in the hospital and residents had no other residence. More than a century later, interns and residents are fortunately allowed to have homes and visit them, like myself.

Over the ensuing days, Seamus and I became fast friends. I would spend moments in my day with Seamus, visiting him to fill the void while his wife was at work at the corporate office for a company that made cable boxes for America. During my encounters, I came to learn how real his lymphoma was. In showing me pictures of himself, I saw how much weight he had gained after starting chronic corticosteroid therapy, on which he was now dependent. The corticosteroid – oral dexamethasone every 8 hours – was the difference between him telling me stories of Ireland’s Aran Islands and him drooling on himself, mumbling incoherently, half-asleep. The lymphoma created a lot of brain swelling and the dexamethasone slowed this process down. One morning, I saw the lymphoma beginning to efface the Seamus that I had come to know; I had come to find Seamus slurring his words and drooling. I didn’t know what was happening, but Seamus had mumbled that the nurses didn’t give him the dexamethasone the night before. I chased down the night nurse who thought she had given him the medication but in actuality didn’t because she had gotten confused with another patient. A medical mistake that paled in comparison to the gaffe to be committed later. Shortly after getting the medication, Seamus started to return to his usual self.

After a few days of chemotherapy, Seamus was done and it was time for him to go home. On the day of his discharge, our team rounded on him in the morning and said their goodbyes. His presence and now departure was palpable among the team. We had walked out of the room, and well out of earshot, our attending had begun talking about Seamus’ care and prognosis going forward. It was dismal. Months. At best. I took the moment to write down his discharge medications on the prescription pad that I kept in my pocket. As the attending stopped rounds to teach about therapies for CNS lymphoma, I wrote down all of his medications on small prescription pages the size of index cards, including his dexamethasone and an antimicrobial – Bactrim – to prevent a type of pneumonia that can occur in patients taking dexamethasone for long periods of time, as Seamus had been doing. After rounds, I put the prescriptions in the patient’s chart, just as I had been instructed, for the nurse to photocopy, document, and ultimately give to the patient.

A couple weeks after Seamus had been discharged, I was nearing the end of my rotation in the hematology and oncology service at the cancer hospital when I found out that Seamus was returning with same type of pneumonia that I had hoped he would never get. If he had been taking the antimicrobial I had prescribed for him, how did he get the pneumonia? There was no question that Seamus had the type of pneumonia I had dreaded: Pneumocystis jiroveci pneumonia. Seamus was short of breath, the oxygen saturation in his blood was low, his x-ray showed random patches and the CT was even worse. This type of pneumonia occurs typically in people with advanced immunosuppression, like patients with HIV or taking medications that can lower the immune system’s ability to fight infections, like dexamethasone. In an attempt to thwart the fungus from ever gaining a foothold in patients who may be susceptible, physicians give Bactrim (also known as TMP-SMX or trimethoprim-sulfamethoxazole). Seamus, struggling to breathe, told me he had never taken the Bactrim. In fact, he never saw the prescription that I had written. Being responsible for his prescriptions and his care, I felt that I had erred gravely. I felt like all of my training and my efforts were worthless.

Worse, Seamus had been feeling short of breath for more than a week and had been trying to get our attention. He and his wife had called the hematologist’s office several times to state their concern for Seamus’ worsening shortness of breath. The first time they could not get through and left a voicemail and nobody called back. The second time Seamus’ wife called and got through to a nurse who informed her that the doctor had been out and she couldn’t take a message for liability reasons and to try calling again later. The third time she called she was repeatedly put on hold and could never talk to a physician. The fourth time she called she was told that Seamus had a follow up appointment the following week with his hematologist and not to worry. When Seamus finally arrived in the hematologist’s office, Seamus could barely stand up before feeling like he was going to pass out. While walking, his oxygen saturation plummeted to the 70th percentiles, by comparison anyone less than 88% percent should be on continuous oxygen. Seamus was skirting the line of needing to be intubated and a breathing machine for life support. As if Seamus’s medical care had not experienced enough problems, Seamus came in minutes before shift change. My resident shot one look at the clock and, without giving it a second thought, told me to “handle it” and walked out.

As a new intern, handling a respiratory emergency was like tasking a monkey to land the space shuttle. She didn’t even come to see Seamus. When I first saw Seamus, he looked like a goldfish that had been plopped out of the fishbowl. My plan had been to give oxygen and he was already on it. I needed help. I knew that if I went straight to the day attending, I might be ridiculed for my lack of knowledge, be viewed as the intern that throws his resident under the bus, or experience retribution from my resident. I skirted the issue entirely, I paged the night attending, who happened to be a specialist in critical care. I told him the whole story in painful detail. Without judgement or complaint, he walked me through Seamus’ management like a dad holding his hands around a child learning how to ride a bike. Seamus did well overnight and slowly recovered over the ensuing days.

What didn’t improve was how I felt about the whole situation. Seamus recovered, but I shaved two, maybe three, weeks off his remaining few months left, precious weeks that could have been spent living his life at home. Adding the time he spent in the hospital, the time he felt sick before arriving, and the time he spent recuperating at home afterwards, the total amount of time likely could have been three weeks. And, I had no way to give him those three weeks back.

Nowadays, mistakes like this are entirely averted by the use of electronic medical records and electronic prescribing, where physicians can electronically send prescriptions to a patients pharmacy as easy as an email. In my attempt to understand why Seamus got all of his other medication prescriptions and not the fateful Bactrim, I reanalyzed all the surrounding events. I even went to medical records to look for the photocopies of the prescription. I even doubted that I had ever written the Bactrim in the first place, or if I had written it, it never made it into the chart. In the bewildering repository that is medical records, I was never able to find the photocopy of any of his prescriptions. Even now, I wonder if I actually wrote the prescription or thought I wrote the prescription. I had no proof to show that I wrote it and no way of obtaining it.

Worse, I didn’t know who was responsible and who was going to apologize. Being the closest member of the medical team to Seamus during his first admission, I felt responsible for the mistake and for the requisite apology. But for what was I apologizing? For not handing the prescription to him directly? For it being lost? For him being unable to get through to us when he was trying ask about his shortness of breath? My attending did not apologize, nor did the hospital, my resident, the nurses, or anybody else. In studies, only one third of physicians explicitly apologized in a way that acknowledged the error, most just detachedly express regret, as everyone else had done with Seamus1; “oh, you didn’t get the medication. That’s too bad.”

Despite medicine’s increasing commercialization, we are still the face of the care we provide. Beneath our white coats, beneath our degrees, we are still humans. We have made mistakes and will continue to make mistakes, but we also have the capacity to acknowledge the mistake and apologize. The apology is so important; 98% of patients surveyed “desired or expected the physician’s active acknowledgement of an error”2.

The words did not come easily. In the days after his re-admission, my heart grew so heavy that I felt like I could not enter his room. I was ashamed to see him, wearing the oxygen mask, gasping between words. The encounters were distinctly less often and less enjoyable. Being genuinely good people, Seamus and his wife never complained about the serial mishaps or noticed my withering spirit. The fact that they did not notice made it at all the more difficult to break the ice, to form an eloquent apology and explanation, and to actually voice the words with my own voice. It felt like I was bringing out an issue that had conveniently been swept under the rug. With each passing day, I felt the pressure to apologize.

My last day arrived on the rotation. It was a Sunday, and Seamus was still requiring oxygen to breathe, albeit much less than before. Even during the day, I delayed the inevitable. Finally, late in the afternoon, I walked into his room, much different than the very first time I walked into his room. I wasn’t walking in like a robot; I was walking in as a human, a very small human. Feeling the pressure that we as humans – humans who are entrusted with the care of other humans – owed an apology to Seamus, I apologized. For all of the mistakes, mistakes that were mine, could have been mine, and definitely were not mine. Fearing anger, yelling, disparagement, and tears – mine or theirs – I voiced everything and took responsibility. The Kennedy’s accepted my apology and understood my explanation. I gave them my personal cell phone number for emergencies. They thanked me for the apology and for providing Seamus with the care that I had given. There was no yelling. There was no blaming. There were almost no tears. Despite being delayed, the apology could not have been better. I hugged them both, and I left.

I never saw Seamus again. I knew that his time was limited. I had promised to follow up with them but I never did. The speed of intern year – new patients, new medications, new diagnoses – distracted me from the incident. Periodically, I would google the local newspaper obituaries for Seamus Kennedy’s name. Finally, one day, I found it. I was heartbroken. A part of me wanted to believe that Seamus would never die, that I would never find his obituary, but I did. Since that experience, I remember three cardinal rules that were once bestowed upon me by a surgical oncology fellow when I was a medical student, but had not realized their value until this incident. The rules are written to minimize the number of medical mistakes, even though it maximizes our own paranoia. The rules:

  1. Trust no one.
  2. Expect sabotage.
  3. Do it yourself

The rules are overly cynical but are necessary if we are to be proactive about medical mistakes. When I have medical students and interns of my own, I now impart these rules to them. Now that most prescriptions are now electronic, I rarely find myself telling people to hand prescriptions in person. I hope that others can learn from this mistake, even if the mistake was mine or not, and also learn the importance of the apology. And, of course, we can remember Seamus Kennedy for his personality, and not his lymphoma.

*Patient’s name has been changed to preserve anonymity and is completely unrelated to Seamus Kennedy, the musician. Other pieces of this story may have also been changed to preserve anonymity.