What color is your cloud? Study finds large variability in resident workloads

Photo by Scott Schiller
Photo by Scott Schiller

For decades medical residents have put themselves into two camps: “black clouds” and “white clouds.” Black-cloud residents carry with them the bad luck of consistently getting a patient load that requires more work; the perceived workload intensity and stress may keep them pacing the halls at night, while their white-cloud counterparts are likely to sleep peacefully while on call.

Does this cloud status actually exist, though? Adam Was, MD, fourth year Stanford resident of pediatrics and anesthesia, decided to find out. The results of his study were just published in Pediatrics.

“The study was inspired by my late-night argument with other interns about our workloads,” said Was. “We commonly discuss what type of cloud we have, meaning what kind of workload. So one of the interns said his workload was really high, but someone else argued that we all have the same workload and he was just complaining about it more. I realized that we could do an objective, rigorous study of actual workloads to get a real answer.”

With the help of KT Park, MD, assistant professor of pediatric gastroenterology and senior author of the study, Was measured the workload of twenty-six pediatric residents during the six core inpatient rotations of their intern year — to make sure they were comparing “like to like.” Using the Stanford Children’s Health research database, they quantified the workload intensity of each of the residents based on the number of electronic notes and orders that they wrote while in the hospital. Was explained:

“ We wanted to focus on objective data that described the work done at the hospital, as opposed to just the number of hours spent there. Residents do a lot of things that aren’t captured in electronic notes and orders, but we found this data to be the most robust and representative.”

And the outcome? The differences are real. The researchers found a very significant variability of workload intensity between the residents. High-workload residents wrote 91 percent more orders and 19 percent more notes than low-workload residents. Here’s Park:

“I really thought that we were going to conclusively lay to rest this idea that there is statistically significant workload variability between residents. I was very surprised. We did sophisticated mathematical models and there is no way around it — there are high-workload and low-workload residents. There is no ethological explanation right now, and it remains a big question mark especially for program directors.”

Thinking through the study’s implications from a program director’s point of view was the main role of the third author, Becky Blankenburg, MD, clinical associate professor of pediatrics and pediatric residency program director, who thinks the results can guide residency directors. “This data provides more information for resident assessments and will allow us to better individualize the residents’ curriculum based on what they’ve really been exposed to,” she said.

Determining the root causes behind this workload variability is beyond the scope of their study. However, the authors have a few of their own theories.

One belief: high-workload or black cloud residents behave differently than their white cloud colleagues. For example, some black cloud residents may be inefficient, while others may create extra work for themselves. And some white cloud residents may need to be more vigilant.

“I would like to get into the heads of the residents in real time,” Park said. “As they put in that note or order or take that phone call, what is the impetus? From my observation, anxiety and perfectionistic tendencies drive them to do more than what’s necessary for effective patient care.”

Blankenburg agreed, “Some residents early on learn to look at the big picture and some see only the trees without seeing the forest. Another important factor is how comfortable people are with ambiguity. If you’re able to deal with ambiguity better, you might not order as many tests.”

The researchers are contemplating how best to use this information and how to design a follow-up study to understand the root causes of resident workload variability. One idea is to somehow incorporate peer evaluations, since their study found self-assessments to be inaccurate. “I think peers would do the best job of picking up on cloud status or workload intensity,” Blankenburg said.

Although successful, did the study settle the late night argument that inspired it?

“The study data was annonymized, so we don’t know who was who,” said Was. “So I never got to settle my original argument of whether I was doing more or less work. Before the study, I thought I was a black cloud. Afterwards, I feel like I’m a confused and possibly grey cloud.”

This is a reposting of my Scope blog story, courtesy of Stanford School of Medicine.

Enlisting artificial intelligence to assist radiologists

Photo by Gerd Leonhard
Photo by Gerd Leonhard

Specialized electronic circuits called graphic processing units, or GPUs, are at the heart of modern mobile phones, personal computers and gaming consoles. By combining multiple GPUs in concert, researchers can now solve previously elusive image processing problems. For example, Google and Facebook have both developed extremely accurate facial recognition software using these new techniques.

GPUs are also crucial to radiologists, because they can rapidly process large medical imaging datasets from CT, MRI, ultrasound and even conventional x-rays.

Now some radiology groups and technology companies are combining multiple GPUs with artificial intelligence (AI) algorithms to help improve radiology care. Simply put, an AI computer program can do tasks normally performed by intelligent people. In this case, AI algorithms can be trained to recognize and interpret subtle differences in medical images.

Stanford researchers have used machine learning for many years to look at medical images and computationally extract the features used to predict something about the patient, much as a radiologist would. However, the use of artificial intelligence, or deep learning algorithms, is new. Sandy Napel, PhD, a professor of radiology, explained:

“These deep learning paradigms are a deeply layered set of connections, not unlike the human brain, that are trained by giving them a massive amount of data with known truth. They basically iterate on the strength of the connections until they are able to predict the known truth very accurately.”

“You can give it 10,000 images of colon cancer. It will find the common features across those images automatically,” said Garry Choy, MD, a staff radiologist and assistant chief medical information officer at Massachusetts General Hospital, in a recent Diagnostic Imaging article. “If there are large data sets, it can teach itself what to look for.”

A major challenge is that these AI algorithms may require thousands of annotated radiology images to train them. So Stanford researchers are creating a database containing millions of de-identified radiology studies, including billions of images, totaling about a half million gigabytes. Each study in the database is associated with the de‐identified report that was created by the radiologist when the images were originally used for patient care.

“To enable our deep learning research, we are also applying machine learning methods to our large database of narrative radiology reports,” said Curtis Langlotz, MD, PhD, a Stanford professor of radiology and biomedical informatics. “We use natural language processing methods to extract discrete concepts, such as anatomy and pathology, from the radiology reports. This discrete data can then be used to train AI systems to recognize the abnormalities shown on the images themselves.”

Potential applications include using AI systems to help radiologists more quickly identify intracranial hemorrhages or more effectively detect malignant lung nodules. Deep learning systems are also being developed to perform triage — looking through all incoming cases and prioritizing the most critical ones to the top of the radiologist’s work queue.

However, the potential clinical applications have not been validated yet, according to Langlotz:

“We’re cautious about automated detection of abnormalities like lung nodules and colon polyps. Even with high sensitivity, these systems can distract radiologists with numerous false positives. And radiology images are significantly more complex than photos from the web or even other medical images. Few deep learning results of clinical relevance have been published or peer-reviewed yet.”

Researchers say the goal is to improve patient care and workflow, not replace doctors with intelligent computers.

“Reading about these advances in the news, and seeing demonstrations at meetings, some radiologists have become concerned that their jobs are at risk,” said Langlotz. “I disagree. Instead, radiologists will benefit from even more sophisticated electronic tools that focus on assistance with repetitive tasks, rare conditions, or meticulous exhaustive search — things that most humans aren’t very good at anyway.”

Napel concluded:

“At the end of the day, what matters to physicians is whether or not they can trust the information a diagnostic device, whether it be based in AI or something else, gives them. It doesn’t matter whether the opinion comes from a human or a machine. … Some day we may believe in the accuracy of these deep learning algorithms, when given the right kind of data, to create useful information for patient management. We’re just not there yet.”

This is a reposting of my Scope blog story, courtesy of Stanford School of Medicine.

Becoming Doctors: a podcast featuring Stanford medical students is revived

Photo by Patrick Breitenbach

I admit it: I’m an NPR and podcast junky. I tune in every Saturday to NPR’s Wait Wait… Don’t Tell Me and This American Life, and I can’t wait for the third season of Serial. So I was extremely excited to hear that Stanford Medicine’s podcast, Becoming Doctors, was being relaunched.

A School of Medicine graduate, Danica Lomeli, MD, originally started the podcast series to document and share the intense clinical experiences of her classmates as they trained to become physicians. After making several episodes, Lomeli passed the project on to Stanford medical student Emily Lines — a perfect fit, since Lines was a DJ at college radio stations throughout undergraduate and graduate school.

Initially Lines, shown here, used Becoming Doctors to share the stories of pre-clerkship students in their first two years of medical school, as described previously in Scope. Lines is now graduating and moving on to a residency in family medicine at the University of Colorado, Denver. Before she leaves, she relaunched her podcast to document the stories of a few of her fellow medical students who are interested in primary care and community medicine.

Emily Lines on match day (courtesy of Emily Lines)
Emily Lines on match day (courtesy of Emily Lines)

There are three new episodes available and a few more are on their way. Each episode is between 10 to 20 minutes long.

For instance, in the new “Adding Layers” podcast, Paula Trepman shares why she is interested in family medicine — based on her experiences as an undergraduate working abroad at rural primary care centers and her clinical experiences through Stanford at the Pacific Free Clinic and Mayview Community Health Center.

Lines told me about her new podcasts in recent emails:

What inspired you to relaunch Becoming Doctors with a new focus on primary care?

As I was interviewing for residency positions in family medicine, I met so many extraordinary people from other medical schools. I was really inspired by the types of things they were doing to promote primary care and family medicine in their programs. In turn, I was also eager to share with them what was happening at Stanford — many people were surprised to learn about the growing presence of the field at a research university. It seemed like a natural transition to use my podcast to seek out the stories of those who are championing primary care at Stanford, and to share them.

There’s a really incredible group of people at Stanford who are doing academic research in primary care, advocating for primary care in medical education, and starting grassroots organizations in local communities. Others are just delighting in taking advantage of the clinical educational opportunities in primary care here at Stanford. I wanted to give voice to our inspiring community!

My first portion of the podcast series focused on pre-clerkship students. This relaunch addresses students from all stages who are interested in primary care and community medicine. There’s value to looking at why students are drawn to primary care at the start of medical school, as well as how their clinical experiences shape this interest.

Why do you want to document the stories of medical students?

My podcast has always been centered on the idea that storytelling is an incredible tool for medical students. By telling our stories, we can develop a practice of introspection and mindfulness about our challenging career. It allows us to stop and think about what we are seeing and experiencing and to decide how we want to assign meaning to these experiences. By listening to these stories, we also learn to see the world from another perspective, which is ever valuable as a clinician.

In the era of the Affordable Care Act and healthcare reform, our eyes are on primary care as one of the main channels to improve the health of our country. I’m eager to see the role Stanford will play in this process, both from a research and clinical perspective. I thought it would be timely to document some student experiences in this changing era.

What is your favorite podcast (besides Becoming Doctors)?

NPR’s This American Life inspired this podcast — it used to be called This Medical Student Life! However, my personal favorite is a podcast called the Dirtbag Diaries. I’m a rock climber and general lover of the outdoors and it’s a podcast about great people who do great things outside!

This is a reposting of my Scope blog post, courtesy of Stanford School of Medicine.

Genetics of sea creatures: One researcher uses her science training to help the environment

Photo courtesy of Lauren Liddell

Lauren Liddell, PhD, developed a passion for genetics early at a girls science day at her Michigan middle school, when she extracted the DNA of a banana.

Nearly two decades later, Liddell now works as a postdoctoral research fellow in genetics at Stanford School of Medicine. Unlike most of her departmental colleagues who study health sciences, Liddell is applying her molecular genetics expertise to one of the most critical environmental challenges that we face today: climate change. I recently spoke with Liddell about her research and her participation in the Rising Environmental Leadership Program, a year-round program that helps graduate students and postdoctoral fellows hone their leadership and communications skills.

How did you end up studying sea anemones at Stanford School of Medicine?

As a freshly minted PhD studying molecular genetics, I approached John Pringle, PhD, about working as a postoc in his lab. … Several years ago, John’s passion for scuba diving and overall curiosity led him to shift his research to tackle environmental problems. Specifically, we’re trying to understand sea anemone-algae symbiosis, in the hopes of discovering things that may be useful for coral conservation.

Coral reefs are a poster child for climate change right now, because coral is dying — about 35 percent of the Great Barrier Reef off the coast of Australia is already dead or dying through a process called bleaching. Bleaching is caused by the loss of the symbiotic algae that live in the guts of coral. Normally the gut algae collect energy from the sun and turn it into food that supports the life of the coral host. As ocean temperatures rise and the ocean acidifies, the algae leave the coral host and the coral starves and bleaches — bleached coral reefs are basically the skeletons.

So we use sea anemones in the lab to study coral, similar to how scientists use mice to study human processes. We’re studying Aiptasia sea anemones as a model for coral reef bleaching, because sea anemones are easier to work with in the lab and they have the same gut algae, Symbiodinium, as coral reefs. We want to understand what goes wrong with symbiosis when ocean temperatures and acidity increase.

What have you found?

We’re trying various genetic methods to identify the genes that are important for this symbiosis. We’re also investigating how some corals are able to survive bleaching, whereas others die off. We have two main strains of sea anemones and multiple “flavors” of Symbiodinium algae that we use to test how the different environmental stressors, like heat and acidity, affect symbiosis.

Surprisingly, we’ve found that the Hawaiian sea anemone is less tolerant to heat stress than the Floridian strain. And even more exciting, we’ve found that the Symbiodinium “flavor” can affect the ability of the sea anemone host to resist heat!

Describe your experience with the Rising Environmental Leadership Program?

The Rising Environmental Leadership Program (RELP) is an exciting program for people who are passionate about making a real impact on society. The program included a week-long boot camp in Washington D.C., where we met with Congress, nonprofit organizations like the Nature Conservancy, and governmental agencies like the Environmental Protection Agency and the Department of Energy. … We really got to see firsthand how science research directly informs science policy.

After going through the RELP Boot Camp, what is your dream job?

Originally I wanted to be a liberal arts professor, because I love teaching and getting people excited about science. But moving to the Bay Area really opened my eyes to many other opportunities to make an impact. For instance, companies like 23andMe can help people understand their genetics and what that means for their health.

I’m currently looking for careers in biotech. Once I’ve gained some business skills though, I plan to apply for an AAAS science and technology policy fellowship to get more firsthand experience with policymaking. My RELP experience made it blatantly clear that we need to train the politicians about science, so they can make informed decisions that impact our future.

This is a reposting of my Scope blog post, courtesy of Stanford School of Medicine.

%d bloggers like this: