Friday, September 25, 2020

Coronavirus vaccine is unlikely to be a "magic bullet" that ends the pandemic

Last week, federal health officials announced an ambitious plan to begin free distribution of a vaccine against SARS-CoV-2 within 24 hours of its approval or emergency authorization from the U.S. Food and Drug Administration (FDA). That an efficacious vaccine could, remarkably, become available less than one year after the isolation of the virus that causes COVID-19 is due in part to a public-private program to accelerate vaccine, diagnostic test, and therapy development led by the U.S. Departments of Defense and Health and Human Services known as "Operation Warp Speed." In a New England Journal of Medicine commentary, Drs. Moncef Slaui, Shannon Greene, and Janet Woodcock reviewed the progress of Operation Warp Speed on multiple fronts, including the most promising candidate vaccines in phase 2 and 3 clinical trials. Ultimately, though, the impact of any vaccine on the course of the pandemic in the U.S. and abroad will depend not only on how protective it is against infection and disease transmission, but how the allocation of initially limited supplies is prioritized (e.g., health care and essential workers, vulnerable groups) and what proportion of the population agrees to receive it in the absence of a requirement to do so.

A simulation study in the American Journal of Preventive Medicine estimated the efficacy and percent population coverage that a coronavirus vaccine would need to extinguish the epidemic in the absence of other public health measures such as social distancing and wearing face coverings - in other words, what it would take for a vaccine to allow life to "go back to normal." The effects of a vaccine on productivity losses, hospitalizations, medical costs, and deaths vary depending on what percentage of the population has already been exposed to SARS-CoV-2; one seroprevalence survey from late March to mid-May found a range from 1% to 7% at 10 sites in the U.S., but these figures are likely to be higher four or more months later. To extinguish the epidemic after 5% of the population has been exposed, for example, the AJPM researchers calculated that a vaccine would need to have at least 80% efficacy if administered to 75% of the population.

Unfortunately, expecting any of the coronavirus vaccines in development to have 80% efficacy is unrealistic.  By comparison, the Centers for Disease Control and Prevention (CDC) estimated that the 2019-20 influenza vaccine was only 45% effective, consistent with the 40-60% range in previous years when the available vaccines were antigenically matched to circulating influenza viruses. The highest influenza vaccine coverage was during the 2018-19 season, when 63% of children (state range, 46%-81%) and 45% of adults (state range, 34%-56%) received the vaccine. Although some may be more willing to be vaccinated against SARS-CoV-2 than influenza due to the former's greater morbidity and mortality, influenza vaccine also has a long safety track record that a coronavirus vaccine would not.

It is possible that public health measures in place to slow the spread of COVID-19 will substantially reduce the impact of influenza during the 2020-21 season. A CDC surveillance report showed that U.S. cases of laboratory-confirmed influenza fell sharply after the national COVID-19 emergency declaration on March 1, reflecting not only the natural waning of the flu season but also mitigation interventions implemented by states around this time. In the Southern Hemisphere nations of Australia, Chile, and South Africa, where influenza activity normally peaks in June or July, COVID-19 mitigation was associated with a near-complete suppression of influenza circulation.

For family physicians, preparing for flu season during the COVID-19 pandemic, with a coronavirus vaccine potentially around the corner, means adapting to many uncertainties. What hasn't changed from previous years is that the CDC's Advisory Committee on Immunization Practices continues to recommend influenza vaccination for all people six months and older who do not have contraindications, ideally by the end of October. Whenever it arrives, the coronavirus vaccine is unlikely to be a "magic bullet" that ends the pandemic on its own. So it's critical that clinicians all continue to echo the message about cloth face coverings that CDC Director Robert Redfield, MD delivered at a recent Senate hearing: "These face masks are the most important, powerful public health tool we have" for controlling the pandemic.


The post first appeared on the AFP Community Blog.

Sunday, September 20, 2020

How inequality explains America's uneven experience of the coronavirus

Over the past few weeks, as the number of reported U.S. deaths from COVID-19 approached 200,000, I have puzzled over why the fewer than 3,000 lives initially lost in the terrorist attacks of September 11, 2001 united our country, but the far higher toll of COVID-19 (at one point over the summer, more than 2,000 were dying each day) has only seemed to divide us. One reason is obvious: after 9/11 we quickly identified a clear villain in Al-Qaeda, while - despite attempts to assign responsibility to the Chinese government for its early inaction - rallying Americans against an unthinking viral enemy is more challenging. Another reason is that President Donald Trump is not President George W. Bush. But there's a third reason, too: everyone I know seems to know someone (or know someone who knows someone) who died in the 9/11 attacks. The victims include one of my high school classmates who was working at the Pentagon and a physician who worked at the hospital where I was a family medicine resident at that time. Statistically, it ought to be 67 times as likely that I would know someone who died from COVID-19, but if I wasn't a doctor, that wouldn't be true. (In my DC practice, several patients were hospitalized for in the spring and summer, and the husband of one of my patients died.)

How can that be? It got me thinking about the spectrum of people who have died from COVID-19 compared to the immediate casualties of 9/11. Although one might assume that workers in the Twin Towers, the Pentagon, and United Airlines Flight 93 would tilt white and upper-class, victims included not only stockbrokers and investment bankers but office assistants, cafeteria workers, maintenance workers, and janitors - people from all walks of life. In comparison, about half of those who died from COVID-19 in the U.S. were residents or employees of nursing homes. Of the remainder, Black, Hispanic, and American Indian / Alaska Native persons are  far more likely to have been affected; according to data from the CDC, a member of those groups is 3 times as likely to have been infected, 5 times as likely to have been hospitalized, and up to twice as likely to have died from COVID-19. So if your immediate social circle includes few people over 65 or people of color, there's a reasonable chance that you don't know anyone who's become severely ill or died from the infection.

For the past several years, one of my Georgetown colleagues has shown this map in the first lecture of my medical school course "Patients, Populations and Policy." The brightly colored lines trace the paths of Washington, DC Metro lines; the numbers are life expectancy at birth in years. At first glance, the take home point might appear to be that people live longer in the suburbs than in the inner city. But that's not quite right, since life expectancy east of DC, in Maryland's Prince Georges County, is 78 years, nearly the same as life expectancy of someone living near DC's Metro Center, where three of the subway lines come together. What's different about PG County compared to neighboring Montgomery County and Fairfax and Arlington Counties in suburban northern Virginia? PG County is 63% African American, while the corresponding percentages for DC, Montgomery, Fairfax, and Arlington are 47%, 20%, 11%, and 10%.

As stark as the disparity in longevity appears in this map from 2013, it has actually worsened since then. A recent analysis in the scientific journal Nature found that a non-Hispanic White male resident of DC in 2016 had a life expectancy of 86 years, while a non-Hispanic Black male could expect to live to 68. Among female DC residents, the gap was a smaller but still startling 12 years (89 vs. 77). Deconstructing these life expectancy gaps, the researchers found that heart disease, cancer, and homicide accounted for about half of the gap among men, while heart disease, cancer, and unintentional injuries accounted for more than half of the gap among women. Some of the widening gap is no doubt related to gentrification, residential segregation, and migration of higher-income Black persons from DC to PG County (where the COVID-19 death rate has actually been higher than that in DC itself). Disparities in overall health have thus far been closely tied to COVID-19's impact: deaths in the DC's primarily White and Asian populated Wards have been much lower than those in Wards with larger numbers of Black and Hispanic residents.

A note of caution: although age and race inequalities largely explain America's uneven experience of COVID-19 to date, that is no assurance that it will stay that way. HIV/AIDS was a viral disease that only affected urban gay men and intravenous drug users - until it wasn't. Consider Utah, the youngest state in the nation, where 4 out of 5 residents are White, which has largely been spared from COVID-19 compared to older and more diverse states such as California, Florida, and New York. After averaging 300-400 cases per day during late summer, the state's case count has surged above 1000 for each of the past two days, and though much of this increase is occurring in young adults (age 18-39), hospitalizations are starting to rise, too. As far as this virus is concerned, anyone sick enough to require hospitalization is sick enough to die. This health crisis remains as urgent today as it was in early March, and our best tools remain those of public health, not medicine. Just as America united against Al-Qaeda after 9/11, a unified national response to the coronavirus, which has thus far been sorely lacking, is desperately needed.

Wednesday, September 16, 2020

Preparticipation physical exam and return to sports during the COVID-19 pandemic

For primary care practices that care for children, the preparticipation physical exam (PPE) is an annual rite. My residency program usually sets aside two full days in late summer where the residents and attending physicians do nothing but "sports physicals" in order to meet the demand for these exams from prospective high school athletes. Although the utility and effectiveness of this traditional evaluation has long been debated - a 2019 American Family Physician article on Right Care for Children included the PPE in its list of overused interventions - many clinicians also use the time to address non-sport related issues, making the visit "a potential preventive care entry point and an opportunity to provide routine immunizations, screen for other conditions, and provide anticipatory guidance." Last year, the American Academy of Family Physicians, the American Academy of Pediatrics, and several sports medical societies published the 5th edition of the Preparticipation Physical Evaluation monograph, which was summarized in the June 1 issue of AFP.

The COVID-19 pandemic closed schools and cancelled youth sports nationwide beginning in early March. As schools are now reopening in virtual, in-person, and hybrid models, some students are also returning to competitive sports. To address the medical needs of these athletes, the American Medical Society for Sports Medicine (AMSSM) recently released Interim Guidance on the Preparticipation Physical Exam for Athletes "to provide clinicians with a clinical framework to return athletes of all levels to training and competition during the pandemic."

In addition to the physical risks inherent in playing a sport, student athletes now must also be concerned about minimizing their risk of contracting SARS-CoV-2 where maintaining physical distancing is not possible. The AMSSM notes that unlike professional teams, high school and most college teams will not have the resources to perform testing, contact training, and quarantine. It advises discussing COVID-19 risks in detail with the patient and family at the time of the PPE, and considering factors "such as the disease burden in the community, the overall health of the athlete, the living environment, [and] each athlete's network of friends and family members who have have comorbid conditions" in the decision to play.

Athletes who have apparently recovered from COVID-19 "may have silent clinical pathology in any organ, including the heart," and consequently "should be evaluated in their medical home prior to resuming physical activity and organized sports." Table 2 in the guidance document (p. 27) outlines the recommended cardiopulmonary evaluation in athletes with prior COVID-19 infection, depending on the specific clinical scenario.

The AMSSM also provides guidance on specific conditions that may pose an increased risk for severe COVID-19, including pregnancy, diabetes, hypertension, asthma, and severe obesity. Although athletes with sickle cell trait are not at higher risk for adverse outcomes in general, they may have an increased risk of hypercoagulability complications for several months after recovery.


This post first appeared on the AFP Community Blog.

Friday, September 4, 2020

Maryland's Primary Care Program: incremental progress or breakthrough?

Much has changed in the past six years since our last Health Policy Journal Club at Georgetown. Our residency, formerly a collaboration with Providence Hospital, is now known as the Medstar Health/Georgetown-Washington Hospital Center Family Medicine Residency Program. I stepped down as director of the Robert L. Phillips, Jr. Health Policy Fellowship three years ago, though I still enjoy working alongside these talented family physicians in clinic, such as Dr. Brian Antono, who recently blogged about his fellowship experiences for Harvard Medical School's Center for Primary Care. And this academic year I am not only working remotely due to COVID-19, I am more than 2,000 miles off campus as a visiting professor at the University of Utah in Salt Lake City.

What hasn't changed is that our family medicine residents remain excited about health policy and advocacy. Since their continuity clinic continues to be located in Maryland, we decided that a great topic to revive this series of seminars was the Maryland Primary Care Program (MDPCP), which was recently featured in a Milbank Memorial Fund Issue Brief.

MDPCP is a multi-payer "advanced primary care" program modeled after previous patient-centered medical home projects such as the Center for Medicare and Medicaid Innovation's (CMMI) Comprehensive Primary Care initiative. CMMI partnered with Maryland's Department of Health to launch MDPCP last year with Medicare as the first participating payer. (CareFirst Blue Cross Blue Shield joined the program in 2020.) With 476 participating primary care practices, MDPCP provides prospective, non-visit based payments known as "care management fees" and operational support from a program management office and Care Transformation Organizations (CTO). According to MedChi, the average practice received $176,000 in care management fees in 2019.

Interestingly, Medstar not only participates through its network of Medstar Medical Group practices, but is also a CTO serving Medstar and non-Medstar practices throughout the state. MDPCP practices must implement "data-driven, risk-stratified care management," integrate behavioral health services, screen patients for social needs, convene a patient advisory council, and use health information technology for continuous quality improvement.

We spent some time discussing one unique aspect of MDPCP, a tool to reduce avoidable health services developed by the University of Maryland's Hilltop Institute. This electronic tool uses artificial intelligence to sift through patients' demographics, claims, and other data to produce a list of those with the greatest likelihood of an emergency department visit or hospitalization, theoretically allowing primary care physicians to intervene to prevent the event and its associated medical expenses. However, it wasn't clear to us how easy it would be to apply this information, given that we usually need to prioritize patients on the schedule for that day.

Another feature of the program allows MDPCP practices to identify "high-volume, high-cost specialists" in order to "focus attention on the relative costs between specialists and to have providers engage specialists in conversations and cooperative agreements about creating value." First, though it may be helpful to know which subspecialists are more likely to prescribe (possibly inappropriate) expensive tests or procedures, the tool does not measure excellence in other areas, such as patient satisfaction and quality of communication with primary care physicians. Also, patients may not have a choice of specialists, depending on the insurer's network. Finally, it seems awkward and unrealistic for a family doctor to tell a specialist that his or her practice style is too aggressive, even if there's good data to back it up.

MDPCP promises to narrow the primary care-subspecialist reimbursement gap and provide opportunities to improve patient care in the short term. However, expecting primary care practices to bend the health care cost curve on their own, even with additional funding and support, may backfire in the long run. Whether MDPCP represents incremental progress in primary care, or a genuine breakthrough, remains to be seen.


This post first appeared on The Health Policy Exchange.