Stanford Health Policy's Eran Bendavid and Jay Bhattacharya write in this Wall Street Journal editorial that current estimates about the COVID-19 fatality rate may be too high by orders of magnitude.

"If it’s true that the novel coronavirus would kill millions without shelter-in-place orders and quarantines, then the extraordinary measures being carried out in cities and states around the country are surely justified. But there’s little evidence to confirm that premise—and projections of the death toll could plausibly be orders of magnitude too high.

"Fear of Covid-19 is based on its high estimated case fatality rate — 2% to 4% of people with confirmed Covid-19 have died, according to the World Health Organization and others. So if 100 million Americans ultimately get the disease, 2 million to 4 million could die. We believe that estimate is deeply flawed. The true fatality rate is the portion of those infected who die, not the deaths from identified positive cases."

"The latter rate is misleading because of selection bias in testing. The degree of bias is uncertainbecause available data are limited. But it could make the difference between an epidemic that kills 20,000 and one that kills 2 million. If the number of actual infections is much larger than the number of cases—orders of magnitude larger—then the true fatality rate is much lower as well. That’s not only plausible but likely based on what we know so far."

Read the Editorial 

Controversies over the lack of diagnostic testing for the COVID-19 virus have dominated U.S headlines for weeks. Technical challenges with the first test developed by the Centers for Disease Control and Prevention (CDC) left the nation with minimal diagnostic capacity during the first few weeks of the epidemic, according to a new paper published today in the Journal of the American Medical Association by Michelle Mello, a professor of medicine at Stanford Health Policy and professor of law at Stanford Law School.

On February 29, the Food and Drug Administration (FDA) began allowing high-complexity labs across the country to use tests they developed in-house. On March 5, the Stanford Clinical Virology Lab deployed its own test for patients at Stanford Health Care and Stanford Children’s Health.

We asked Mello to answer some questions about the federal rollout of diagnostic testing.

You write that in the early stages, COVID-19 “spread beyond the nation’s ability to detect it.” Is there anything the U.S. government could or should have done weeks ago to get out ahead of the spread?

Adopting broader testing criteria and allowing use of a wider range of tests would have been helpful in identifying the first U.S. cases and containing the spread. Manufacturing problems like the one that arose with CDC’s test are always a risk, but the fact that CDC put all its eggs in that one basket made the manufacturing snafu highly consequential.

Also, the public messaging from Washington about the seriousness of the problem has been neither consistent nor accurate, and I worry it may have led Americans to take fewer steps to prevent community transmission than we should have. Containment was not “pretty close to airtight.” A vaccine was never going to be ready in “three to four months,” as the Trump administration claimed. The case fatality rate is not “way under 1 percent.” Part of the problem here is that as the stock market continues to plunge, the president and the task force he appointed appear to be more concerned about calming investors than stopping the virus.

We seem to be between a rock and a hard place: You write that remedying gaps in testing is imperative, yet “more testing is not always better.” How do we determine the happy middle ground?

First, the testing criteria have to be calibrated to our actual testing capacity. You can’t announce that any American who wants a coronavirus test can get one and then, within hours, announce that there aren’t enough test kits to make that possible. High priorities for testing include patients with serious, unexplained respiratory illness and contacts of known cases. From there, testing can be expanded, beginning with other high-risk groups, as capacity permits.

Second, we should consider unintended side effects of mass testing. The problem with this virus is that it doesn’t have signature symptoms. It looks like the common cold or the flu. If everyone with a cough or fever, or who has been around someone with a cough or fever, shows up in their doctor’s office demanding a test, it will quickly overwhelm care facilities that should be focusing on patients with a higher likelihood of being infected or and those who are infected and are seriously ill. It may also work against the social distancing measures that public health officials are trying to encourage, because crowded waiting rooms may spread the virus.

The CDC announced Monday it now has the testing capacity in 78 state and local public health labs across 50 states to test for the virus. There are now 75,000 lab kits cumulatively to test for COVID-19 with more coming on board by mid-March. But is there anything we could have done to roll this out earlier?

The alternative would have been to allow laboratories to deploy their own tests from the beginning, using the primers and protocols made publicly available by the World Health Organization. That’s what other countries have done. RT-PCR is a mature technology and high-complexity labs around the country are well-qualified to conduct this type of testing.

There is a public health argument for not going that route: perhaps those labs wouldn’t have done as good a job as CDC’s own lab and the state labs that it handpicked early in the outbreak. What if there were erroneous test results? We could miss cases, or we could put people into isolation, with huge social consequences, based on false-positive results. There is also a worry that some labs aren’t consistent about reporting positive test results to CDC, and underreporting could compromise disease surveillance efforts.

The counterargument is that high-complexity labs have that certification for a reason—they’re good at what they do. And of course, surveillance is also compromised when you miss cases because you don’t test.

You write in your paper that testing for COVID-19 “highlights a controversial area of public policy—the regulation of laboratory-developed tests—in which there has long been tension between the goals of access and quality.” Who should be in charge of regulating these tests?

Laboratory-developed tests are largely unregulated outside of emergencies. The FDA proposed draft guidance in 2014 that, if implemented, would have required labs to make certain showings to FDA about tests they developed in-house, with the particular evidence calibrated to the risks involved in having a wrong test result. Contrary to President Trump’s claim that an Obama-era policy constrained coronavirus testing, the guidance did not relate to emergency situations. During declared emergencies, another statute and set of regulations apply, and the FDA has broad discretion to allow or disallow use of novel diagnostics and therapies as emergency countermeasures.

As a general matter, it makes good sense to require labs to submit evidence that their in-house tests work. It’s odd that laboratory-developed tests are carved out of requirements that apply to other kinds of medical devices. It’s also sensible that our legal framework allows FDA’s regular rules to be relaxed during emergencies so we can tailor our response to the difficult and changing circumstances.

You write that diagnostic testing is critical to an effective response to the novel coronavirus. What sort of policies and guidelines should be put into place to prevent such a sluggish rollout during an emerging epidemic the next time one comes around?

The legal framework for an effective emergency response is in place. Because giving agency heads the discretion to act as potentially unforeseeable circumstances require is a linchpin of this legal framework, it only works if leaders make smart choices. Every emergency is different, and there is a danger of Monday-morning quarterbacking. But we should learn from every misstep we make, and I think the lesson here is to make better use of already developed networks of highly qualified labs to make sure we have adequate testing capacity to isolate cases and trace their contacts very early in an outbreak.

What are some innovative approaches we could be taking to speed up testing for those who really need it?

The South Koreans have set up drive-through testing stations in parking lots to avoid concentrating crowds of people indoors. Of course, that requires that you have plenty of test kits, which we don’t yet – but we should also be thinking about creative ways to address the epidemic. For example, how could video calls be used to monitor the health of people confined at home after being exposed to the virus? How can social media be used to connect neighbors to help one another when some are isolated at home? Hopefully we can find new ways for technology to bring us together when pathogens drive us apart.

More evidence-based research is needed before the U.S. Preventive Services Task Force can recommend that clinicians screen their older patients for cognitive impairment such as dementia and Alzheimer’s disease.

Cognitive impairment is a growing public health dilemma that affects millions of Americans as they age. The Global Burden of Disease study shows that Alzheimer’s rose from the 12th most burdensome disease or injury in the United States in 1990 to the 6th in 2016.

Medical experts who were commissioned to conduct an evidence report for the Task Force projected that the burden of Alzheimer’s disease is expected to grow to 13.8 million U.S. residents by 2050 — or nearly 3.3% of the projected U.S. population by that year.

Their findings, the Task Force recommendation statement and several accompanying editorials were all published Tuesday in the Journal of the American Medical Association.

The symptoms of cognitive impairment can range from problems with memory and language, to learning new things or making decisions that affect everyday life. 

“Early identification of cognitive impairment through screening would ideally allow patients and their families to receive care at an earlier stage in the disease process, potentially facilitating discussions regarding health, financial, and legal decision-making while the patient still retains decision-making capacity,” the authors of the Task Force evidence report wrote.

But after reviewing some 287 studies including more than 285,000 older adults, the Task Force determined there wasn’t sufficient evidence about the benefits or harms of screening adults 65 and older who do not have signs or symptoms. The Task Force also did not find adequate evidence that screening for cognitive impairment improves decision-making or planning by patients, caregivers or doctors.

At the same time, there is little evidence on potential harms of screening, such as depression, anxiety or lower quality of life.

“Given the burden of dementia and the intense public interest in preventing cognitive impairment, the lack of progress is disheartening,” Carol Brayne, MD, with the Department of Public Health and Primary Care at the University of Cambridge in the UK, wrote in an accompanying JAMA editorial to the Task Force evidence report.

But, she added, “Political considerations and pressure from commercial interests and patient advocacy groups notwithstanding, public policies for dementia screening should be supported by evidence.”

The Task Force — an independent panel of national experts in prevention and evidence-based medicine — encourages clinicians to remain alert for early signs of symptoms of cognitive impairment, while calling for more research on the detection of dementia.

“Research is especially needed on whether screening and early detection of cognitive impairment helps patients, caregivers, and doctors make decisions about health care or plan for the future,” said Douglas K. Owens, chair of the Task Force and the director of Stanford Health Policy.  “We share the frustration of clinicians who want to offer something that could help patients prevent cognitive impairment. We hope that additional research will enable us to know whether that’s possible.”

The most commonly used screening tests include the Mini-Mental State Examination as well as the clock-drawing test. Screening tests involve asking patients to perform a series of tasks that asses one or more aspects of cognitive functions. The USPSTF concluded that more research is needed to know whether such screening tests can lead to interventions that help prevent or improve cognitive impairment. 

Elderly patients hospitalized with congestive heart failure have a poor prognosis and high risk of death and hospital readmission. So, their post-discharge care can strongly influence their outcomes.

Yet despite data showing that transitional care interventions, such as home visits by nurses, can reduce death rates and hospital readmission by more than 30%, many health systems have not implemented such programs. Health policy experts say this is due in part to cost concerns and doubts about the effectiveness of these delivery services.

 Now, a team of Stanford Medicine and Veterans Affairs researchers has sought to assess whether transitional care interventions provide good value and better outcomes, as there are 5 million people living with congestive heart failure in the United States and 500,000 new cases diagnosed each year. CHF is the stage of chronic heart disease in which fluids build up around the heart, causing it to pump inefficiently.

The researchers updated a 2017 study on the impact of transitional care intervention with four years of additional data. They then used it to compare standard post-discharge management with three post-discharge regimes for patients 75 or older that they found to be most effective: disease management clinics, nurse home visits and nurse case management.

All three transitional care interventions delivered appreciable health benefits to the patient population, said Jeremy Goldhaber-Fiebert, PhD, associate professor of medicine at the Stanford School of Medicine and core faculty member of Stanford Health Policy.

The findings were published in the Annals of International Medicine. Goldhaber-Fiebert is the senior author. The lead authors are Manuel R. Blum, MD, MS in Epidemiology & Clinical Research at Stanford in 2019 and now at the Department of General Internal Medicine at the University Hospital of Bern; Henning Øien, PhD, Norwegian Institute of Public Health, Oslo; and Harris L. Carmichael, MD, a Stanford/Intermountain Fellow in Population Health, Delivery Science, and Primary Care

“Transitional care interventions for older individuals with congestive heart failure — particularly nurse home visits — offer a high-value care alternative that could improve the health and longevity of millions of Americans,” he said.

The researchers said these transitional care services should become the standard of care for post-discharge management of patients with heart failure.

Heart failure causes 1 in 8 deaths nationwide

The prevalence of heart failure is estimated to be 26 million people worldwide and growing. In the United states, heart 5.7 million adults have been diagnosed with HF, with an estimated annual direct cost of $39.2 billion to $60 billion. Total heart failure costs in the United States are expected to exceed $70 billion by 2030, the authors wrote. According to the Centers for Disease Control and Prevention, heart disease costs the United States about $219 billion each year from health-care services, medicines and lost productivity.

Of the 15 million Americans in their mid-70s to 80s today, about 1 million suffer heart failure.

“So population gains from more effective post-discharge care would be hundreds of thousands of life years,” Goldhaber-Fiebert said. “Likewise, tens of thousands of costly rehospitalizations could be prevented each year if these interventions were delivered successfully.”

Heart failure primarily affects older people and is the second-most common inpatient diagnosis billed to Medicare. Yet the authors cite a recent study of 18 million Medicaid charges which found that only 7% of eligible patients at risk of rehospitalization received transitional services.

The standard post-hospital care for those patients includes sending them home with some advice and scheduling follow-up visits for them with cardiologists within 14 days of discharge. The researchers found that patients who received this standard post-hospitalization care with an average age of 75 had an average life expectancy of 2.9 years and 2.9 hospitalizations during their remaining lifetime. In comparison, nurse home visits decreased the number of hospitalizations by 10 readmissions per 100 patients and increased life expectancy by approximately four months, the study found.

“If these interventions were successfully implemented at scale, they could provide important substantial benefits with very good value,” said co-author Douglas Owens, MD, the Henry J. Kaiser Jr. Professor and professor of medicine at Stanford.

Reduced hospitalizations for congestive heart failure, according to the research, produces substantial cost savings that partially offset the costs of delivering the interventions. Though nurse home visits increase lifetime health care costs by $4,622, the substantial health benefits that they deliver justify their costs: $19,570 quality adjusted life years gained, which is considered highly cost-effective.

Hospital and insurance administrators take note

“Our results have important implications for decision-makers in hospital administration as well as in insurance and policy settings,” the authors wrote. They concluded:

• Transitional care services should become the standard of care for post-discharge management of patients with heart failure;
• The increasing reimbursement restrictions and regulations affecting HF hospital readmissions, through such programs as the Centers for Medicare & Medicaid Services Hospital Readmission Reduction Program, makes this research particularly informative to decision-makers;
• Hospital administrators could use the research to determine which transitional services are most cost-effective for its rural population, overall patient base and hospital system.

The other Stanford researcher on the study was Paul Heidenreich, MD, a professor of medicine and health research and policy at the Stanford University School of Medicine and, by courtesy, professor of health research and policy at the Palo Alto Veterans Affairs Health Care System.

Stanford Health Policy researchers, led by Josh Salomon, have been awarded a five-year grant from the Centers for Disease Control and Prevention (CDC) to conduct health and economic modeling to guide national and local policies and programs focusing on some of the most important infectious diseases in the United States.

The CDC grant establishes the Prevention Policy Modeling Lab at Stanford, continuing a multi-institution collaboration that began when Salomon was a professor at Harvard prior to joining Stanford in 2017.

“The overall mission of the Prevention Policy Modeling Lab is to leverage the best available evidence to inform strategic decision-making about major public health problems,” Salomon said. “We do this by combining techniques from decision science, simulation modeling and health economics to estimate and project major patterns and trends in these diseases and to evaluate different clinical and public health strategies to address them.”

The initiative will focus on policy and practice in the areas of tuberculosis, HIV, hepatitis, sexually transmitted infections and adolescent health. The grant from the Centers for Disease Control and Prevention supports a wide range of modeling activities, including those that assess: 

• Projections of future morbidity and mortality
• Burden and costs of diseases
• Costs and cost-effectiveness of interventions
• Population-level program impact
• Optimized resource allocation

Stanford researchers who are involved in the Modeling Lab include Douglas K. Owens, Margaret Brandeau, Eran Bendavid, Jeremy Goldhaber-Fiebert, Jason Andrews, Samuel So and Mehlika Toy. The consortium also includes partners at Harvard, Yale, Michigan, Boston University, Boston Medical Center and the MA Department of Public Health.

“As a multi-institution consortium, on any given problem we’re able to assemble a team that includes both subject matter experts and collaborators who specialize in statistics, epidemiology, data science, economics and decision analysis,” Salomon said. “The policy models that we develop allow us to synthesize a wide array of different types and sources of evidence to shed light on the essence of the problem and to weigh the likely benefits and costs of responding in different ways.”

Prior work from the consortium on the potential impact and cost-effectiveness of expanding testing for hepatitis C virus was cited in the recent decision by the U.S. Preventive Services Task Force to revise their screening recommendations to cover all adults. The Modeling Lab has also examined prospects and strategies for eliminaitng tuberculosis in the United States and policies relevant to the rising threat of antimicrobial-resistant gonococcal infection among other topics.

Understanding the value of chronic disease care is critical to confronting the challenges of aging societies. In a new ebook published by the Centre for Economic Policy Research, APARC Deputy Director and Asia Health Policy Program Director Karen Eggleston provides a framework for assessing the social value of health spending.

The world population is aging faster than ever before and governments must confront the increasing burden of healthcare spending on their economies. At a time when the economics of aging is inseparable from the economics of healthcare, successful adaptations to older population age structures necessitate better understanding of the value of medical care. Policymakers, in particular, must incorporate value into considerations of healthcare cost growth, so they can determine the extent to which average health improvements offset added cost, reduce cases in which health spending rises without sufficient corresponding health outcomes, and reward those in which “we are getting what we pay for.”

A new book chapter, authored by APARC Deputy Director and Asia Health Policy Program Director Karen Eggleston, provides a framework for assessing the social value of health spending. Titled “Understanding ‘value for money’ in healthy aging,” the chapter is part of an ebook, Live Long and Prosper? The Economics of Ageing, published by the Centre for Economic Policy Research (CEPR).

[To receive more stories like this directly to your inbox subscribe to our newsletters.]

Quality-Adjusted Cost of Care

How do health economists incorporate value into measurements of health spending and how do they measure the social value of medical care? First they assume each additional year of life brings a given monetary value. Then they measure the growth in value to patients as monetized gains in “quality-adjusted life-years,” a metric that includes increases in life expectancy and quality of life. The difference between the change in health spending and the change in monetized gains on improved survival is the net change in quality-adjusted health spending or the net value of medical care.

Understanding the value of chronic disease care is especially critical in aging societies, as governments must transform their health systems to support patients who will live with chronic diseases for decades. Health benefits of medical care, however, are difficult to aggregate across disparate services and diseases, and hence focusing on management of a single important chronic disease allows researchers to develop metrics of quality improvement and value that are linked to rigorous clinical studies. Eggleston describes a recent international research collaboration, which she was part of, that did just that. The researchers studied quality adjustment for one disease of growing global prevalence, type 2 diabetes, in four different health systems: one in Europe (the Netherlands) and three in East Asia (Japan, Hong Kong, and Taiwan).

Results of the study suggest that, in each health system, the value of improved survival outweighs the increase in health spending. For example, in the case of Japan, Eggleston and her colleagues found a positive value net of $2,595 for $100,000 value of a life-year. They also compared net value across the four health systems and different patient samples, finding mean net value that ranged between $600 and $10,000 for a $100,000 value of a life-year. Moreover, net value was positive for all age groups and remains positive and significant for individuals well beyond traditional retirement ages. These results, says Eggleston, indicate “the importance of continuing investments in medical treatments and services that deliver health outcomes of commensurate or higher value.”

Policy Implications

Confronting the challenges of aging societies requires careful thinking about the value of investments in new technologies for managing chronic conditions. To promote healthy aging governments must be “resiliently persistent in measuring the value of innovations for healthy aging and rewarding those that deliver high net value,” argues Eggleston. The goal should be improving the “value for money” of medical care rather than applying largescale cost controls that might stifle important breakthroughs.

The four-system study by Eggleston and her colleagues provides a framework for developing methods for assessing quality improvement and the net value of chronic disease spending and, more broadly, for measuring the value of healthy aging.

Download Eggleston’s chapter as part of the entire ebook >>

Learn more about Dr. Karen Eggleston’s research agenda seeking to assess net value in diabetes management and to identify and analyze innovation for healthy aging.

In a new article published by the Milken Institute Review, APARC Deputy Director and Asia Health Policy Program Director Karen Eggleston offers an overview of the successes of China’s health reforms and the challenges ahead if the country is to increase the efficiency of health care delivery as it expands quality and usage.

Creating a high-quality universal health care system is an immense challenge anywhere, let alone in a country as large and diverse as China. But equal access to care will become ever more important as China converges on higher incomes, slower economic growth, population aging, and dependence on a skilled workforce to approach OECD living standards. With its health reforms over the past two decades, its growing technological prowess, and its application of innovative business models, China has made significant progress towards supporting higher-quality and more convenient health care for its 1.4 billion people. However, it is yet to deal with a host of challenges.

In a new article, Healing One-Fifth of Humanity, published in fall quarter 2019 of the Milken Institute Review, APARC Deputy Director and Asia Health Policy Program Director Karen Eggleston offers a progress report on China’s efforts to provide decent health care to all of its citizens, detailing how the reforms are working and what is left to do.

[To get more stories like this delivered directly to your inbox sign up to receive our newsletters.]

A Tale of Two Chinas

One complex challenge China faces is addressing the gap in living standards between its rising middle class and its poorest citizens. Eggleston’s research shows stark gaps between urban and rural China in multiple measures, including life expectancy (almost 10 years differences as of 2013), infant mortality, and under-age-5 mortality. Health outcomes differ along other dimensions, too — between urban regions with higher and lower per capita income and among individuals with more or fewer years of schooling.

There are also striking inequalities in the burden of chronic disease and in health care and risk protection. For example, diabetes is associated with greater excess mortality in rural China, although prevalence is higher in urban areas. Moreover, although China has attained universal health coverage and put in place policies to enhance access while decreasing households’ out-of-pocket spending burden, the coverage of rural insurance is less generous than coverage for urbanites. Therefore, the prospect of catastrophic medical spending on delayed care remains substantially higher for rural than urban residents.

Additional challenges abound in multiple other areas, from addressing patient-provider tensions and trust, to changing provider incentives to promote value rather than volume, to deciding which new medical therapies qualify as basic. In broad terms, argues Eggleston, “China must build an infrastructure that increases the efficiency of health care delivery as it expands quality and usage.”

An important issue is the extent to which leveling public policy will ameliorate disparities, even as an array of social and economic forces push to widen disparities in health, health care use and burden of medical spending over a lifetime. The good news, says Eggleston, is that the process of catching up on one critical component of social justice, that is, universal health care, has begun, and there is evidence suggesting “that health investments can narrow the gaps in outcomes by compensating for health disadvantages.” There is also good reason to believe that policies that go beyond direct medical intervention – notably, investments in the quantity and quality of schooling – can have even greater influence on health and survival than access to medical care.

“The challenge now,” concludes Eggleston, “is to persist in an endeavor that requires flexibility, sensitivity to competing interests — and lots and lots of money.”