The first step is to acknowledge it as an inherent part of clinical practice
By Stuart Foxman
What was going on with the woman who just arrived in emergency? She had given birth two weeks earlier, and was tearful and clearly stressed. One of the nurses on duty suspected postpartum depression. By the time Dr. Pat Croskerry saw the woman, that idea had been planted in his head — “an unfortunate case of priming,” he says.
If there’s one thing he doesn’t trust, it’s being too sure of yourself. “Life doesn’t work like that. I’m suspicious of reflexive, shoot-from-the-hip responses,” says Dr. Croskerry.
Along with being an ER doctor and cognitive psychologist in Halifax (he’s now retired), Dr. Croskerry authored the book, The Cognitive Autopsy: A Root Case Analysis of Medical Decision Making. He’s an expert in diagnostic errors and notes that one of the leading culprits is the tendency to jump to conclusions. One reason for that is unease with uncertainty.
Dr. Croskerry values diagnostic uncertainty. It’s a guard against jumping to conclusions. After reviewing the woman’s story and examining her, he felt postpartum depression was a possibility, but wasn’t convinced. Something about the patient’s reflexes were off — it was subtle, but worried him.
The ER was closing soon and the patient was waiting for an answer. Dr. Croskerry didn’t have one yet. Her presentation reminded him of a paper he had just read about Guillain-Barré syndrome, a rare but serious autoimmune disorder. It can cause muscle weakness, numbness or reflex loss and, when acute, can be life-threatening. Maybe he was only considering that possibility because the article was fresh in his mind. He called a neurologist who felt postpartum depression was likely, but agreed to see the woman.
“I said ‘I’m not sure what’s going on, but there’s a couple of things I wouldn’t mind ruling out.’ You’re being truthful enough to acknowledge uncertainty,” says Dr. Croskerry.
By midnight, the woman was intubated and Guillain-Barré syndrome was confirmed. “If I had sent her home, she would have died,” he says.
William Osler, the pioneering Canadian physician, called medicine “a science of uncertainty and an art of probability.” Navigating that can be a struggle. What are the repercussions of diagnostic uncertainty, and how can it be managed to promote patient and clinician well-being?
Knowledge isn’t infinite and estimates are imprecise
Society tends to associate certain qualities to being a good doctor, like decisiveness and confidence, and that’s also what we often attribute to an expert diagnostician, says Dr. Janice Kwan, a staff physician (general internal medicine) at Mount Sinai Hospital in Toronto, and an Assistant Professor, Department of Medicine, University of Toronto.
“This flies in the face of what we know about diagnosis, which requires a willingness to engage with evolving information,” she says. “Diagnostic uncertainty is inherent to the practice of medicine.”
Uncertainty can flow from inadequate data, atypical presentations, insufficient history-taking, not knowing the most efficient way to diagnose, unawareness of the patient’s goals of care, and more. Dr. Maria Dahm, a Research Fellow at the Institute for Communication in Health Care, Australian National University, breaks it down into three dimensions: personal (individual knowledge gaps); scientific (the limits of biomedical knowledge); and probabilistic (imprecise estimates of risk or prognosis).
Whatever the reasons, uncertainty can have consequences for clinical outcomes, patient experiences, resource use and physician wellness.
“If I had sent her home, she would have died”
Missed, delayed or incorrect diagnoses can lead to unnecessary surgeries and hospitalizations, avoidance of care, more referrals, suboptimal response to treatment, and a reluctance to withdraw from certain therapies.
Much of the uneasiness revolves around perceptions and assumptions, i.e., how doctors internalize uncertainty and worry about what their patients or colleagues think.
“Uncertainty is a big stressor,” says Dr. Dahm.
She says doctors who can’t explain what causes a symptom might see that as a professional failure — or even a personal inadequacy. In the absence of a concrete diagnosis, some patients may believe their symptoms are being dismissed. Or they might begin to mistrust the competence of their doctor.
The absence of a clear path to proceed can trigger emotional distress for doctors and patients alike. But managing uncertainty isn’t solely about reducing anxiety. Dr. Kwan’s academic area of focus is how to improve diagnostic safety. She links that, in part, to ease with uncertainty.
“In my practice, we manage patients who are very complex. It’s not unusual to have patients who have symptoms that are unexplained. In many ways, we have to develop a comfort with discomfort. That’s part of being a safe clinician.”
Dr. Croskerry says medical training reinforces the idea that if patients have faith the doctor is capable of fixing their problem, that bolsters the curative effect.
“So, reassurance and certainty are a good therapeutic start,” he says. “Doctors come to realize that if they’re hesitant or equivocate, they don’t seem to do as well. Most of us learn to bring in equivocation and caveats later, not as an opening gambit.”
Accepting uncertainty can be part of the journey to arrive at a diagnosis. Yet the urge to be certain can also lead to wasteful or even harmful interventions, says Dr. Wendy Levinson, a Professor of Medicine at University of Toronto, and Chair of Choosing Wisely Canada, the national voice for reducing unnecessary tests and treatments.
“The culture of medicine is to leave no stone unturned. Good doctors think of everything it could be — this is the fallacy,” says Dr. Levinson. “People call it being thorough, but it’s almost an obsession with eliminating uncertainty. I remember a family doctor saying he wanted to retire because he couldn’t stand worrying all the time that he missed something. His need to know made it very uncomfortable to him.”
That can take a toll. Dr. Dahm says diagnostic uncertainty can make doctors feel spent, disconnected and lacking, all of which can contribute to burnout. She says early-career clinicians are at particular risk, but any doctor can be vulnerable.
One study in Frontiers in Psychiatry found GPs who reported greater diagnostic uncertainty had higher levels of emotional exhaustion, job dissatisfaction and turnover intention.
Uncertainty can have consequences for clinical outcomes, patient experiences, resource use and physician wellness
“It’s not the uncertainty that leads to burnout, it’s the discomfort with uncertainty,” says Dr. Joel Katz, an internist at Brigham and Women’s Hospital in Boston. “Accepting that we don’t have all the answers is an antidote to burnout.”
How to communicate and cope
Dr. Katz is an advocate of visual thinking strategies (VTS) to improve diagnoses. VTS has been used to guide a group of medical students in analyzing a piece of art in a museum: what do they see, what makes them say that, and what else can they find?
Part of the idea is to slow down and train the eye to notice details. But it’s not just about observational skills. The exercise is also about speculative thinking, collaborating and developing a tolerance for something that has many interpretations. That’s beneficial in a clinical setting. “It’s perfect for things that are ambiguous,” says Dr. Katz.
Even if clinicians don’t embrace uncertainty, it’s important to understand how to cope with and communicate it.
- Reframe the nature of diagnostic uncertainty. It’s not a threat to your authority, but integral to the process, says Dr. Dahm. She says clinicians who openly encourage and engage in discussions of uncertainty, without blame or penalty, model excellent diagnostic processes.
- Be transparent. Part of being patient-centred is openness, says Dr. Levinson. You won’t always have a definitive diagnosis, but can always explain how the diagnostic process works. “Transparency is best for everybody — including the physician. If you don’t know, say ‘I don’t know for sure, I think its two or three possibilities, and this is what’s most probable,’” she says.
- Negotiate expectations. “Be explicit: ‘When you came here today what was your goal?’ It’s about validating the patient as well — validating their feeling of being unwell and acknowledging their emotions,” says Dr. Dahm.
- Lead with honesty. Recognize that patients react differently to uncertainty. Some might expect their doctor to have all the answers right away, others can wait for a label. All can appreciate directness, says Dr. Katz. “Some studies show that patients want an unambiguous plan. But that tends to be in areas like cancer. In general, from the studies I’ve seen, the more overriding feeling is that in the long run of a therapeutic relationship, patients respect honesty and clarity more than they do a single answer. My hypothesis is that people who can accept uncertainty and learn to communicate it, will be better doctors.”
- Consider what matters most to patients. “Physicians think patients want tests and treatments. Patients want to feel heard. There’s a big disconnect,” says Dr. Levinson. “Patients don’t expect their doctors to be perfect. They can live with uncertainty, if they feel the doctor listened to them, paid attention to their concerns, reassured them and did their best.”
- Don’t get lost in translation. What clinicians say and patients hear can be very different, says Dr. Dahm. Terms that may seem clear, like “occasionally,” “rarely,” “suspicious,” “suggestive of,” etc., may actually cause confusion. She says other seemingly straightforward statements can be misinterpreted. “I’d like to follow up with you next week” can mean you’re unsure of a diagnosis and taking a watch-and-wait approach. To the patient, it may come across as a routine appointment. Be clear about what you know, what you don’t and what comes next.
- Make patients part of the process. Dr. Kwan says outcomes can also improve when doctors are clear with patients about where they are in the diagnostic process, and when patients are an active part of it. She calls diagnosis a team sport, which includes the patient at the centre of the team.
- Create “diagnostic safety nets.” “When diagnosis is uncertain, inform patients and their families what red flags they should look for after visits, how they should seek help if symptoms change or worsen, and what to expect regarding the time course of their symptoms. This creates conditions for detecting missed diagnoses earlier,” says a 2021 article in Patient Education and Counseling.
In many studies on the impact of communicating uncertainty, the findings raise uncertainty themselves.
A 2019 paper in the journal Diagnosis noted one study where the parents of paediatric patients felt doctors who expressed uncertainty were less competent and less trustworthy, resulting in lower adherence to recommendations. Yet in another study, patient satisfaction increased when doctors explicitly discussed uncertainty. However, the same study found satisfaction fell when that uncertainty was communicated poorly.
Given the complexity, the Diagnosis paper stated, “it is not surprising that physicians often hesitate to admit diagnostic uncertainty.”
The absence of a clear path to proceed can trigger emotional distress for doctors and patients alike
Dr. Kwan would love to see comfort with the idea of uncertainty integrated much earlier in medical education curriculum. Right now, if that comfort comes at all, it’s often through on-the-job experience, she says.
It’s ironic that uncertainty — the very thing that drives curiosity in medical research and inquiries in practice — can also cause such discomfort. The logic of welcoming it should be self-evident. “But it’s not,” says Dr. Dahm.
Living with it is a necessity, and knowing how to communicate it is a science that goes beyond the science of diagnoses. She says part of becoming more comfortable with the diagnostic uncertainty is “unlearning that there is a direct answer to every question you face.”
“It’s not unique to us,” says Dr. Kwan. “We have uncertainty in medicine like we do in life. The first step is to embrace uncertainty and make it part of our clinical practice.”
