“This is not normal for him.” A tearful mother pointed to her six-year-old son, lying in the hospital bed. Fevers had been intermittent. He was withdrawn, but not uncommunicative until a few minutes earlier. “My head hurts and my tummy hurts,” he had said before turning away from me and ignoring any further questions. “I know this doesn’t look like much. But you don’t see him every day. I do,” his mother asserted to the team of doctors standing around the bed.

“This is not normal.”


Sign-out that morning had been typical. At 6:30 AM each day, the doctors who worked the night shift reported on the patients to the doctors who would care for them during the day. Julius (not his real name) was presented as a “bit of a mystery.” His pattern of symptoms pointed to a viral illness as the most likely culprit – common in children and often self-resolving. When thinking about more urgent matters, he also seemed OK. He did not appear toxic. His lungs were clear; his heart beat fine; nothing was acutely worrisome. But an incidental finding on routine labs pointed in a very different direction: Julius should be much sicker, or at least show some physical manifestations of what the tests indicated. When repeat testing showed that the abnormality remained, he became a person to watch. His external presentation did not match what the labs told us was happening internally.

The strange part was that his particular set of lab findings does not just come out of nowhere. There were specific causes. Some were scary, some less scary. To rule some out would require invasive, time-consuming, or expensive procedures.

That posed the immediate questions: What to pursue, if anything? And how to decide?


Two months into the wards, and I’ve noticed that experienced doctors seem to have a sixth sense for the subjective question of timing. It is the question of what needs to be done now, what can wait, and for how long. That question is one of the many that is difficult to become competent at by studying, but rather improves through close observation, experience, and immersion.

It's such a critical question because this year is about making decisions. During the first two years of my medical training, the emphasis was: what is going on, and why? We stressed understanding the pathophysiology of a disease process – what was happening in a person’s body to present with these symptoms, how a treatment would work. Now the question posed to us is: what is your plan? It’s an interesting change in approach because even when we are not sure what is going on, the plan is needed anyway. Sometimes that plan involves getting closer to figuring out the underlying process. Sometimes the plan involves managing without an exact understanding of cause. But you always need a plan. The plan is paramount. It’s everyday medicine, and it’s a major focus of what we should be improving upon to advance as doctors.

I have also noticed how personality differences can influence that plan. Different doctors I’ve worked with can approach an identical patient and – depending on whether they are more concerned about missing something or tend to fear more the dangers of over-testing – decide different courses of action.

Sometimes, after ruling out all the scary things, the plan is to wait and see. To repeat lab tests a few hours later, hoping they were an artifact, or that they would self-correct. To check in on the patient and just see how he is progressing. Variations of that same theme can go by many names: Medical observation. Active surveillance. Expectant management.

Watchful waiting.


For Julius the mystery patient, that was the plan. I was instructed to check on him every few hours, but I went back sooner. I believed Mom. Patients and their families, after all, know themselves best. There is objective “normal,” which is usually a range, and then there is normal for you. Doctors can tell you the former, but it is the patient who can be ultimately trusted on the latter. That is, a seemingly insignificant finding on physical exam may still be something of note for an individual. “Listen to your patients,” is not advice only for being compassionate, but for gaining clinical insight as well.

We learned a basic neurological screening test in medical school, and I used it then. Can you please follow my finger with your eyes, without moving your head? Please walk for me. Now can you walk on your tiptoes? Can you touch your pointer finger to your nose? Now can you please touch your pointer finger to my pointer finger?

His coordination was a bit wobbly, and his walk was one of those findings that appeared objectively “normal” but according to his mother was changed completely from his baseline. Was it worrisome enough to do something? To push watchful waiting into some sort of action? I had to know what we were waiting for, after all. Perhaps most obviously, we were continuously monitoring whether something would go wrong, so that if heaven forbid it did we could spring into action and acutely save the day. Otherwise, we were waiting for a clinical finding that would tell us that one path of work-up was a better use of time and resources than another. But I couldn’t tell if this was significant enough to be it.


All the while, nervous patients and their families want answers. You cannot expect to go into a patient’s room every few hours, ask him to perform these seemingly bizarre tasks, and then not expect some kind of response. What did you find? What do you think? When can we go home?

That brings up an interesting ethical dilemma: at what point to give those answers? One option is to be explicit about everything on your differential, from the most benign to the most scary possibility. But I noticed few took this approach. Or if they did, they worded it very, very carefully. On pediatrics, we usually did family centered rounds, which meant each morning we presented the patient’s case in her room, in front of her and her family. I learned early on that my assessment in front of the patient was not always the same as my assessment in the physician’s conference room. For example, cancer might be somewhere on my differential, but if it’s way down on my list, I wasn’t about to proclaim it in the patient’s room.

I struggled similarly when the differential was less clear. However hard it is to hear a scary diagnosis on the differential, it must be perplexing and disappointing to hear that your trusted team of professionals is not sure what is going on.

There is also the powerful pressure of wanting to feel like something is getting accomplished. Some may feel unsatisfied at the end of a doctor’s appointment if the doctor “didn’t do anything.” The pressure from the patient’s side to act can be significant, and it can influence decision-making. Perhaps the smartest option at the moment is to do nothing, but that means fielding the inevitable question: “What are we going to do next?”


Eventually we did call a neurology consult. As the doctor explained to the family why, I was impressed with how she communicated. She was honest about what we were thinking, and it was persuasive. It made sense why we were seemingly doing nothing at the moment. While listening to her, I saw understanding on the parent’s face, and had I not been sold on the plan before, I was then.

It was a day of conflicting opinions from experts, all actually quite convincing, of why Julius' findings were/were not primarily neurological or were/were not stemming from elsewhere. It was a day of observing, from the closeness of the patient’s bedside to the distance of tracking vitals on a computer monitor. It was a day of waiting.


At 7:30 pm, we changed teams again. The night shift doctor who had admitted Julius and presented him to us as a “bit of a mystery” came back that evening with a well-rested look I knew would appear quite different 12 hours later. “I’ve been thinking about Julius all day,” he said. “What did you end up doing?”

I thought about visiting him every few hours, asking him to walk back and forth and to touch his finger to his nose. I thought about plotting fever patterns and listening to his heart and lungs and hearing that they sounded OK. I thought about talking to experts and about consoling Mom when she was upset – or rather trying to, more often struggling to answer her questions because in actuality no answers were known. I thought about all this as the day team intern answered:

“Nothing. We didn’t do anything.”


Seventy two hours after a six-year-old presented with fevers, lethargy, a changed gait, and that frustrating incidental finding on labs, he was walking around, had energy, and was talkative. Feeling better could be explained. Dropping fevers could be explained. But, most surprisingly of all, his labs re-entered the normal range. When they stayed that way upon repeat testing, it was decided: he was free to go home.

As we wrote out the discharge summaries, the day team intern complimented the night team: “You cured Julius!” He was clearly joking. But I wondered if there was any truth to it.


Over the next few days I thought about fever patterns and repeat blood tests. I thought about taking action versus waiting. Some things self-resolve, and that’s known. Parents might ask for antibiotics, for example, but their child’s picture instead fits a viral illness, which her healthy immune system will fight off in a matter of days. Self-resolution for many clinical pictures is well known and well documented.

What was odd was that Julius’ particular picture is not one that should not have ostensibly self-resolved.

And yet it did, without us even being able to pinpoint what had caused it in the first place. A patient got sick and got better, all under our watchful eyes, without intervention.

With this in mind I read with interest Dr. Danielle Ofri’s thoughtful piece in the New York Times about uncertainty in medicine. “Much of the time… we function in an ambiguous zone, without clear-cut answers,” she contemplates. “We have piles of data at our disposal, but how to use this information is not always clear-cut.”

I get that. Still, I used to envision the decision making tree in medicine as more active. Should we try option A or option B? But lurking in the background is always option C: the do nothing option. So far, I guess I have just been surprised by how many times it has ended up being the best option.


Medicine is often about outcomes. If the patient is doing well, it is technically okay if you cannot explain the pathophysiology of what happened in the first place. Doctors work in a system of many patients limited by the constant constraints of time and efficiency, and I understand that there are times when it just makes sense to say OK: I accept it, even if I can’t explain it. We are not actually sure why your son presented the way he did. But we are glad he’s back to normal.

Still, I am afraid of being complacent. Weeks later, I continue to think about Julius, in the back of mind at least, as I read more and see more, hoping something will fall into place that did not before. I realize that striking a balance between critical thinking and efficiency will be on ongoing challenge, with no clear guidelines and requiring persistent judgment calls. I am not sure how this will play out in the long run.

What I do know is that watchful waiting can be an effective diagnostic or therapeutic choice when used in the right setting. There’s plenty of precedent. Call back if any of X, Y, or Z scary things happen, we say in the outpatient setting. Call back if things don’t get better in X number of days.

And, watchful waiting does not mean doing nothing. You have to know what changes would inspire you to act. You are hoping for the best but always thinking about – and always being prepared for – the worst. That is a skill – one that we need to develop intuition for, and one that is far from nothing.

Dr. Ofri writes that uncertainty is a fixed part of medicine. I would assert that waiting is, too. It’s always there, lurking on the list of potential diagnostic and treatment options. Patients – and perhaps even providers – might be uncomfortable with this. Recognizing it as an option should not mean becoming complacent with not understanding. Sometimes, with the information at hand, it's a choice that just makes sense.

(Certain details of this story have been omitted or modified slightly to protect the privacy of the patient.)