Monday, June 29, 2020

A graduation address for the COVID-19 era

If you watched your child graduate this year, as I did my older son's junior high school ceremony at the end of May, you most likely did so online. The same for my participation in the virtual graduation ceremony for Georgetown University School of Medicine's Class of 2020, which included 17 family physicians who start their internships in July. Two years ago, I was inspired to write about Dr. Atul Gawande's moving graduation address to UCLA's newest physicians. This year, I turn the clock back to 2012, when Dr. Don Berwick addressed Harvard Medical School's graduating students; his speech later appeared as an essay in JAMA's A Piece of My Mind.

Dr. Berwick, as longtime readers of Common Sense Family Doctor know, is one of my heroes. I finally had the opportunity to meet him in person in early March, when he delivered the Georgetown University School of Nursing & Health Studies Values Based Lecture. He recently captured the essence of that talk about the "Moral Determinants of Health" in another JAMA essay that is well worth reading, but in this post I will focus on "To Isaiah," his Harvard graduation address from eight years ago.

Isaiah is the name of one of Dr. Berwick's past patients, a Black teenager from the Roxbury neighborhood of Boston who developed acute lymphoblastic leukemia and received the "the best of care ... the glory of biomedical science," including chemotherapy and a curative bone marrow transplant. But biomedical science - then and now - proved to be no match for poverty and despair:

Isaiah smoked his first dope at age 5. He got his first gun before 10, and, by 12, he had committed his first armed robbery; he was on crack at 14. Even on chemotherapy, he was in and out of police custody. For months after his transplant he tricked me into extra prescriptions for narcotics, which he hoarded and probably sold. Two of his five brothers were in jail—one for murder; and, two years into Isaiah's treatment, a third brother was shot dead—a gun blast through the front door—in a drug dispute. ...

His world was the street corner and his horizon was only one day away. He hated it, but he saw no way out. He once told me that he thought his leukemia was a blessing, because at least while he was in the hospital, he couldn't be on the streets. And Isaiah died. One night, 18 years after his leukemia was cured, at 37 years of age, they found him on a street corner, breathing but brain-dead from a prolonged convulsion from uncontrolled diabetes and even more uncontrolled despair. ...


Isaiah, my patient. Cured of leukemia. Killed by hopelessness.

Dr. Berwick went on to tell HMS's Class of 2012 that Isaiah's story demonstrated that they had two duties as new physicians. One, to "go to the mat" for their patients, always putting their needs first and advocating for health care to be recognized as a human right in the United States. The second duty was "more subtle - but no less important":

Maybe this second is not a duty that you meant to embrace; you may not welcome it. It is to cure, not only the killer leukemia; it is to cure the killer injustice. ... One million American children are homeless. More people are poor in the United States today than at any other time in our nation's history; 1.5 million American households, with 2.8 million children, live here on less than $2 per person per day.

I am not blind to Isaiah's responsibilities; nor was he. He was embarrassed by his failures; he fought against his addictions, his disorganization, and his temptations. He tried. I know that he tried. To say that the cards were stacked against him is too glib; others might have been able to play his hand better. I know that; and he knew that.

But to ignore Isaiah's condition not of his choosing, the harvest of racism, the frailty of the safety net, the vulnerability of the poor, is simply wrong. His survival depended not just on proper chemotherapy, but, equally, on a compassionate society. ... Isaiah, in his legions, needs those in power—you—to say to others in power that a nation that fails to attend to the needs of those less fortunate among us risks its soul. That is your duty too.


Our nation's health care professionals have been "going to the mat" to treat patients with COVID-19 for the past four months. In doing so, hundreds have already lost their lives. To their credit, the Congress and the President rapidly enacted legislation to reduce obstacles to coronavirus testing and care created by our patchwork health care system, where at least 27 million are uninsured and tens of millions with insurance still cannot afford to see a doctor or pay for essential medications. Thus far, there has been no similar national initiative to eradicate injustice, even as millions have peacefully protested incidents of police violence and people of all races, from all places, have signaled support for the Black Lives Matter movement.

Doctors have participated in many protests, leading some to label them hypocrites because of the real possibility that the protests could accelerate COVID-19 community spread (though early findings from Washington State suggest that their contribution has been minor as compared to indoor social gatherings without masks). Thus far, the Washington, DC metro area is one of few in the country that has seen a sustained decline in cases despite large protests and tear gassing of protestors. But as imposing a public health problem COVID-19 remains, it pales in comparison to the morbidity and mortality toll of racism, social injustice, institutionalized inequality, and poverty - all factors which have fed the pandemic and contributed to the disproportionate devastation the virus has caused in communities of color. To respond to Dr. Berwick's "second duty," it is absolutely right and appropriate for all physicians to take a knee against injustice, and to use the power of our medical degrees to make our political representatives "go to the mat" for the most vulnerable Americans.

Wednesday, June 24, 2020

Artificial intelligence in primary care: progress and challenges

As applications of artificial intelligence (AI) in health care multiply, AI-enabled clinical decision support is coming to primary care. For example, a recent article in the Journal of Family Practice discussed applications of machine learning (ML) software to screening for diabetic retinopathy (DR) and colorectal cancer, and a study in the Journal of the American Board of Family Medicine utilized ML to create a new clinical prediction tool for unhealthy drinking in adults. Although research on primary care AI remains limited in scope and diversity of authorship, Drs. Winston Liaw and Ioannis Kakadiaris argued in a Family Medicine commentary that appropriately guided, such research could help preserve the parts of primary care that physicians and patients value most:

The digital future is not a passing trend. We will not return to paper charts. The volume of information we are expected to manage will not decline. Without a strategy for our digital present and future, our specialty risks being paralyzed by data, overwhelmed by measures, and more burned out than we already are.

We can define our future, by embracing AI and using it to preserve our most precious resource—time with patients. Adaptation to this new reality is key for our continued evolution, and AI has the potential to make us better family physicians. ... For AI to elevate the practice of family medicine, family medicine needs to participate in relevant design, policy, payment, research, and delivery decisions.

Evaluation and implementation of AI-based clinical approaches is challenging. In addition to being externally validated and corrected for biases, ML models should be transparent about data sources and assumptions and quantify and communicate uncertainty. In addition, involvement of clinicians in model building and adoption into clinical decision support systems is essential.

In the Diagnostic Tests feature in the March 1 issue of American Family Physician, Dr. Margot Savoy reviewed an application that adheres to all of the best practices for AI in primary care. IDx-DR, a software program that uses AI to analyze retinal images from an automated nonmydriatic camera, is approved by the U.S. Food and Drug Administration for DR screening in adults 22 years and older. In a prospective study of 819 adults with diabetes recruited from 10 primary care practices, IDx-DR correctly identified 173 of the 198 patients with more than minimal DR according to the reference standard.

In a separate project, Google Health researchers evaluated the implementation of a deep learning algorithm for DR detection in 11 clinics in Thailand, a country with low screening and early treatment rates due to a shortage of ophthalmologists. Unexpected issues arose, according to an article in the MIT Technology Review:

When it worked well, the AI did speed things up. But it sometimes failed to give a result at all. Like most image recognition systems, the deep-learning model had been trained on high-quality scans; to ensure accuracy, it was designed to reject images that fell below a certain threshold of quality. With nurses scanning dozens of patients an hour and often taking the photos in poor lighting conditions, more than a fifth of the images were rejected.

Patients whose images were kicked out of the system were told they would have to visit a specialist at another clinic on another day. If they found it hard to take time off work or did not have a car, this was obviously inconvenient. Nurses felt frustrated, especially when they believed the rejected scans showed no signs of disease and the follow-up appointments were unnecessary.


Like all primary care tools, the way that AI-enabled decision support is implemented in real life will contribute as much to its success or failure as test results under optimal conditions.

**

This post first appeared on the AFP Community Blog.

Thursday, June 18, 2020

Less medicine, more health? COVID-19 pandemic provides a natural experiment

One of my favorite health care books of 2015 was Less Medicine, More Health: 7 Assumptions That Drive Too Much Medical Care, by Dr. Gil Welch, a general internist who was then a health services researcher at Dartmouth. After resigning his position at the college in 2018 in the wake of an at-best-questionable accusation of plagiarism, he joined the Center for Surgery and Public Health at Brigham and Women's Hospital and has continued to produce excellent work, including an insightful analysis of prostate cancer mortality trends and more recently, a CNN opinion piece with Dr. Vinay Prasad about "the unexpected side effect of COVID-19." In their CNN piece, Welch and Prasad pointed out that the much lower utilization of the U.S. health care system during the pandemic has provided a unique opportunity to for researchers to examine if certain high-volume medical interventions (e.g., physicals, cancer screenings, orthopedic surgery, and surgery for low-risk cancers) actually improve health outcomes:

After Covid-19, if we dare imagine the day, it will be important to ask who was harmed by delayed and forgone medical care. But the severe financial strains on individuals and public budgets make it just as essential to ask who benefited from avoiding interventions with no salutatory effect. ... Covid-19 provides a once in a lifetime opportunity to study what happens when the well-oiled machine of medical care downshifts from high to low volume in order to focus on acutely ill patients. It will be comfortable for physician researchers to study what was lost. It will be courageous for them to study what was gained.

If a physical necessarily includes laying hands on patients, I haven't done one in more than 3 months, as my practice has shifted almost entirely to telehealth. To tell the truth, I haven't missed these physicals much. I've long been a skeptic of the value of a general health checkup in adults, even though in our fee-for-service payment system they are essential to primary care practices paying the bills (and their absence has led to many practices facing bankruptcy as the pandemic grinds on). It's not just that physicals are often a waste of time and money, though - sometimes, they can even be harmful.

In a 2014 JAMA essay that was re-published in this year's theme issue marking 40 years of of "A Piece of My Mind," Dr. Michael Rothberg recounted the near-fatal cascade of testing that began with his 85 year-old father's checkup with a new primary care physician. This doctor, who felt a possible aortic aneurysm on the abdominal examination, ordered an ultrasound scan (which revealed no aneurysm, but saw something in the pancreas), followed by a CT scan (which revealed a normal pancreas, but saw a lesion on the liver), followed by a liver biopsy that revealed a hemangioma, a benign but extremely vascular tumor that bled profusely and required the transfusion of 10 units of blood. He spent a painful week in the hospital not being able to urinate without a catheter and received a $50,000 bill for his troubles - inspiring the essay's title, "The $50,000 Physical."

I could tell you several similar stories of testing cascades that went wrong from my own decade-and-a-half in practice; probably every general internist or family physician who isn't fresh out of residency has at least one. Of course, no organization recommends screening for an aortic aneurysm in a healthy 85 year-old man, but no one recommends ordering a chest x-ray or electrocardiogram at a physical either, or doing a Pap smear in a woman younger than 21 or older than 69, and these "low value" (really, "no value") screening tests still happen far too often. A retrospective cohort study published last week in JAMA Internal Medicine found that adults in Ontario who received those 3 particular tests ended up having significantly more subspecialist visits and subsequent (likely unnecessary) diagnostic tests or procedures in the next 90 days.

"De-implementation" is the formal health services research term for "figuring out how to do fewer things to patients that do more harm than good." Just as COVID-19 has accelerated an overdue transition to providing more health care virtually, I hope that it will also inspire researchers to "study what was gained" from postponed or cancelled appointments with family doctors and surgeons. If the pandemic has a silver lining, this might be it.