My hospital office is a utility closet. I don’t own the closet but it’s my place of refuge to review patient charts before work, where I can temporarily avoid the hustle of the wards and inevitable interruptions.
After morning teaching, the overnight resident reviews newly admitted patients with our team. Much like the takeoff of an airplane, hearing a patient’s case for the first time is a critical moment in which crucial decisions are made. However, I have to excuse myself midway through the presentation – we have concurrent bullet rounds where I review potential discharges with the allied health team, another important forum in which we ensure patients are ready to go home safely.
We finish assessing the new patients, divide tasks for the day and then I finally have a moment to review lab results. Deep in thought, I am interrupted by the beat of 90s disco music – the ringtone of our team phone – and I liaise with a nurse requesting an order clarification, then field two additional calls over the next 15 minutes. When I glance at the clock, it’s almost noon and I haven’t physically seen most of my patients. I start making my rounds.
‘Deep work’ in medicine
In popular culture, the television drama House, M.D. hinted at the complexity of internal medicine in which our patients have interacting illnesses involving every organ system. Fans of the show would be familiar with the typical arc of each episode in which Dr. House eventually arrives at an epiphany through deep contemplation. In reality clinical medicine is more heuristic-based but this narrative still speaks to the importance of “deep work.” Coined in a bestselling book by Cal Newport, the term refers to the ability to focus without distraction on cognitively demanding work.
All physicians require periods of deep work to advance patient care and this is especially true of internists. However, although still early in my medical career, I have found that our systems and culture of care do not promote deep work, instead favouring constant availability and communication. My time in hospital is spent putting out fires and reacting to issues while most of my proactive deep work is performed before or after the official workday.
A ‘crisis mode’ work climate
Healthcare has embraced the use of novel communication devices over the past decade, progressing from alphanumeric pagers to smartphones. Although this has achieved its goal of increasing physician accessibility, it has collaterally increased disruptions to physician workflow. My personal experiences are not isolated – one study reported that internal medicine physicians are interrupted an average of 3.92 times per hour. Another outlined perceptions of a “crisis mode” work climate as physician teams reported receiving five or more interruptions within a 30-minute period at least twice per day.
Although interruptions are certainly disruptive, they are a side effect of the need for intense communication and teamwork. In the correct context, interruptions can serve important purposes such as proactively identifying problems and preventing medical errors. However, in other cognitively demanding fields such as aviation, there are dedicated processes to reduce interruptions during critical moments. For example, standard operating procedures mandate that pilots have a “sterile cockpit” during takeoff and landing that can only be disturbed for emergencies. If other industries have developed solutions to similar problems, then surely healthcare can do the same?
Triaging interruptions
Dr. Robert Wu, a General Internist at University Health Network and Associate Professor in the Department of Medicine at the University of Toronto, has been researching information technology and communication systems in healthcare for the past 20 years. He acknowledged that while technological advancements have improved “what used to be a painful (communication) process,” they have also “lowered the (threshold) for communicating.”
Accordingly, Wu described how systems to “triage the level of importance of communication” were key enablers to reduce nonurgent interruptions. For instance, guidelines were established with nursing leadership to decrease nonurgent interruptions during specified times such as resident educational hours. However, recognizing the limitation of administrative controls that rely on individual recall and adherence, Wu’s team successfully implemented a complimentary engineering control in which nurses about to send a nonurgent message during educational hours would receive a reminder of the guidelines in addition to an option to send the message later. After implementation, the frequency of interruptions decreased by about 35 per cent from a mean of 0.92 to 0.59 messages per team during each educational hour.
Integrating deep work into healthcare
Dr. Onil Bhattacharyya, a Family Physician at Women’s College Hospital and Senior Scientist at Women’s College Research Institute, said he believes primary care practices are traditionally designed to manage “high volume, high throughput (and) relatively straightforward problems.” However, he noted that there is a subset of patients who benefit from a deep work approach, including those with “medically undiagnosed symptoms” or those with mental health issues who “require a deeper connection to understand what matters to them.”
Interestingly, Bhattacharyya has observed that his ability to care for these individuals has improved with the transition to virtual care during COVID-19. “In virtual care practice, much of which is asynchronous, there’s an opportunity to reflect on things, do a synthesis, email specialists,” he reflected. “There’s no time pressure to come up with something right on the spot and it allows for a deeper type of reflection that … really fits the definition of deep work.” In contrast, in a purely face-to-face practice focused on patient throughput he felt “we didn’t commit the time (for deep work). That wasn’t in the schedule and it was tacked on to the end of the day. We didn’t do it justice.”
Deep work by design: Dr. House clinic
Dr. Barry Kassen, a General Internist and former Division Head of the University of British Columbia (UBC) Department of Medicine, runs an outpatient practice focused on medically complex and undiagnosed patients. In contrast to traditional outpatient practices, Kassen noted that in his clinic “time (isn’t) a factor so I’m not really solving a problem in half an hour or an hour … it’s however long it takes and I’m prepared to do that.”
In keeping with other healthcare providers, Kassen said “most of my (deep) work is done outside of the clinic” such as seeking in-person consultations with other specialists in challenging cases. He also emphasized the importance of taking time for deep reflection, especially listening to a patient’s story and avoiding the tendency to “hear but not listen.”
“If I’ve had any solutions or breakthroughs, it’s usually because I’ve heard the story a bit differently than someone else,” he said. “Or I found something on examination, or seen something that no one else paid attention to.”
As this clinic does not have an official name, the UBC residents proudly refer to it as the Dr. House clinic, a nod to Kassen’s skills as a diagnostician.
Shallow work drives physician burnout
Recent evidence demonstrates that internal medicine interns spend a minority of their time on direct patient care (13 per cent) compared to indirect care such as using electronic medical records (EMRs) (43 per cent) or communicating with other healthcare providers (39 per cent). This burden of “shallow work,” underscored by the significant time dedicated to navigating unintuitive EMRs, has been identified as a significant factor in physician burnout.
Recognizing this problem, Wu emphasized the importance of “trying to offload … tasks that traditionally have been put on the teaching units.” For instance, his institution established dedicated care coordinators with primary responsibility over patient flow such as mobilizing community resources to prepare patients for discharge.
In my experience, working with care coordinators has been incredibly helpful as they can proactively anticipate and address barriers to discharge. However, Bhattacharyya noted that even “relatively shallow work … still requires some knowledge of the patient” and is often “not purely clerical.” For example, he pointed out that while disability application forms may appear relatively straightforward, they require significant time and detailed medical knowledge to complete and may ultimately have a “huge impact on someone’s financial situation.”
Conclusion
Although the sterile cockpit approach may seem like a silver bullet for medicine, it is ultimately an imperfect analogy. One fundamental difference between these industries is that healthcare relies on dozens of independently functioning providers – imagine an airplane with a dozen co-pilots who need to be in constant communication. In addition, each patient’s hospital course can be highly variable and unpredictable.
Unlike aviation, Wu described caring for patients on an internal medicine team as having innumerable “different, unplanned trajectories.” Because of the “competing tensions” to balance deep work and intense communication in healthcare, Wu postulated that eliminating nonurgent interruptions may ultimately be a “wicked problem” or, in other words, a potentially unsolvable issue.
Bhattacharyya arrived at a similar conclusion, commenting that the need for deep work ultimately depends on the fundamental role of the physician. “Is our job to manage many small problems quickly or is our job to solve a couple of … hard problems? I think for a clinician, certainly primary care, our job is probably to solve a lot of little things quickly. But (there are also) really big important ones that we can’t miss … so we need to strike the right balance, right?” Bhattacharyya said, again underlining the competing tensions in the role of the physician.
Kassen highlighted the work culture of inpatient medicine as a barrier to deep work. With the multitude of distinct teams and healthcare providers involved in patient care, he noted “everybody sees their part as the most important” and as a result it is “other people (who) are defining your time.” Additionally, he described how the flow of information can be like a game of broken telephone, with accuracy distorted as it passes between care providers. Because of this, Kassen emphasized the importance of critically processing information. “The hardest thing in inpatient medicine is to step back and not use somebody else’s lens to decide what the problem is, but to stop and use your own lens … which is really the lens of the patient,” he said.
These observations parallel my experiences as an early-career physician. Most of our daily work requires us to solve problems rapidly and efficiently through mental heuristics honed by years of experience. However, there are moments that require us to slow down and think deeply to solve complex problems and I fear we will fall short of these challenges in our “crisis mode” work climate that increasingly pulls us in multiple directions.
There is inherent tension between the interwoven responsibilities of physicians: rapidly responding to urgent requests, quickly resolving large numbers of simple problems, all while vigilantly solving a small number of complex problems. The relative balance of these needs in different healthcare settings, from hospitals to outpatient clinics to home care, will ultimately determine how we can re-engineer our practice environments to best support patient care.
For now, the responsibility to dive deep on complex problems hinges on the willpower of individual doctors but I envision a future healthcare system that creates space for deep work – a physical and cultural space more aspirational than a 4×4 closet. With simultaneously increasing patient volumes and complexity in healthcare, the tension between efficiency and deep work has never been greater and the time to act is now.
The comments section is closed.
Nicely articulated. Re “tension between…rapidly responding to urgent requests, quickly resolving large numbers of simple problems, all while vigilantly solving a small number of complex problems…” Some thoughts: Batching non-urgent task requests vs. serial interruptions.
Creating “protected time” for thinking through complex patient problems.
Cost effective deployment of NP or PA to manage straightforward tasks?
Remuneration models that support these.
An insightful article and an issue that has increasingly concerned me. Not sure how we can change the system, but it needs to be changed. A very recent and personal example of this … my sister started the process to find the cause of her symptoms 3 years ago. Test referrals were “lost” and cavalier attention paid to getting timely assessment by her family physician. Multiple requests denied for additional evaluation as symptoms escalated. It was obvious to me that surgical intervention was indicated. We advised her to “hang out” in the ER until someone paid attention. The ER physician said “there is nothing we can do for you. You will just have to wait for your scheduled appointment with the specialist” (one month out after waiting 1 year, including a one and a half month reschedule due to MD vacation). I believe my sister would not be with us now had she waited one more month for a consultation based on a CT scan done 1 year earlier. Meanwhile, in the 3 previous months, she was rapidly losing all ability to care for herself and had lost 35 lbs – a current CT scan would have shown a tumour in her neck that had disintegrated two of her cervical vertebrae. There is more to this story and gaps in the narrative here, of course, but the dive would not have been too deep to see that the patient needed a more thorough evaluation, especially in the latter 3 months. Covid virtual visit weaknesses may not have picked up that she was emaciated and had the “look” of a cancer patient. Thankfully, a second visit to the ER two days later had a physician and nurses more willing to do deep work. Emergency surgery was performed within 5 days. She is slowly regaining some nerve function and is able to walk short distances with a walker. She can feed herself with assistive devices. Her tumour is malignant (they couldn’t get it all) and has multiple points of metastasis, primary unknown. Her primary and ER physicians and their staff were dismissive, disrespectful and neglectful. Her new team is collaborative, attentive and patient centred. Her journey is far from over, but she now feels heard with compassion. The reason for her “shallow” care should be explored.
Thank you Brandon for your insightful review of the competing pressures Physicians face both in and out of acute care. At times it does feel that we just need to step into your closet to think quietly. New models of team based care and new models of compensation to facilitate time for the deep work will require ongoing dialogue between physicians doing the work and those responsible for administering our precious healthcare dollars. You are off to a great start. Doctors of BC and all PTMAs across Canada are actively engaged in this work and we appreciate the voice of residents and new to practice physicians in this work. We are Better Together and will only succeed with collaborative leadership across silos.
Wow, this encapsulates my residency experience in a nutshell. Thanks for verbalizing these thoughts! Must-read for any medical students and trainees interested in the healthcare system and the chaos of trying to get things done.
“There is inherent tension between the interwoven responsibilities of physicians: rapidly responding to urgent requests, quickly resolving large numbers of simple problems, all while vigilantly solving a small number of complex problems.”
I have always thought about this concept. As an RN who sees my nursing colleagues page MDs multiple times in a day, sometimes even per hour, I’ve sometimes thought, “wow, how does anyone get any work done? Wouldn’t it make more sense to help work flow to batch our questions/ comments together?” But after being in this line of work for as long as I have, I’ve seen so many colleagues page throughout the day, almost as a stream of consciousness, and so it’s been normalized. I find it’s especially normalized in academic settings (versus community healthcare settings) when working with resident physicians. It’s interesting to see this perspective, thanks for sharing, but I guess my question is, what should we do then? Should we create physical and cultural spaces for deep work?
Doctors don’t work with automation the way pilots do.
In as much as the aviation industry complains about distractions, I would argue that pilots are good at multitasking precisely because they make due despite the distractions. The order of priority is a very dynamic thing and therefore the necessity to adapt becomes a requirement.
There is a thing called “task saturation” and it is usually where the flight becomes or can potentially enter an “undesirable state” and ultimately that can lead to an accident. That’s when “deep work” just doesn’t cut it anymore. Most aviation accidents happen during this time of “deep work” also called the approach and the landing phase, but they also happen when people are tired at the end of a long day or a long flight.
We humans have limitations (and different kinds) and all industries have to learn what they are and how to place proper boundaries on them. The ability to absorb distractions and reprioritize in a timely and relevant manner is one of them.