The COVID-19 pandemic is one of the defining moments in history for this generation’s health care leaders. In 2019, most of us wrongly assumed that this virus would be similar to the past viral epidemics and pandemics such as 2002 severe acute respiratory syndrome–CoV in Asia, 2009 H1N1 influenza in the United States, 2012 Middle East respiratory syndrome–CoV in Saudi Arabia, and 2014-2016 Ebola in West Africa. Moreover, we understood that the 50% fatality rate of Ebola, a single-stranded RNA virus, was deadly on the continent of Africa, but its transmission was through direct contact with blood or other bodily fluids. Hence, the infectivity of Ebola to the general public was lower than SARS-CoV-2, which is spread by respiratory droplets and contact routes in addition to being the virus that causes COVID-19.1 Many of us did not expect that SARS-CoV-2, a single-stranded RNA virus consisting of 32 kilobytes, would reach the shores of the United States from the Hubei province of China, the northern Lombardy region of Italy, or other initial hotspots. We could not imagine its effects would be so devastating from an economic and medical perspective. Until it did.
The first reported case of SARS-CoV-2 was on Jan. 20, 2020 in Snohomish County, Wash., and the first known death from COVID-19 occurred on Feb. 6, 2020 in Santa Clara County, Calif.2,3 Since then, the United States has lost over 135,000 people from COVID-19 with death(s) reported in every state and the highest number of overall deaths of any country in the world.4 At the beginning of 2020, at our institution, Wake Forest Baptist Health System in Winston-Salem, N.C., we began preparing for the wave, surge, or tsunami of inpatients that was coming. Plans were afoot to increase our staff, even perhaps by hiring out-of-state physicians and nurses if needed, and every possible bed was considered within the system. It was not an if, but rather a when, as to the arrival of COVID-19.
Epidemiologists and biostatisticians developed predictive COVID-19 models so that health care leaders could plan accordingly, especially those patients that required critical care or inpatient medical care. These predictive models have been used across the globe and can be categorized into three groups: Susceptible-Exposed-Infectious-Recovered, Agent-Based, and Curve Fitting Extrapolation.5 Our original predictions were based on the Institute for Health Metrics and Evaluation model from Washington state (Curve Fitting Extrapolation). It creates projections from COVID-19 mortality data and assumes a 3% infection rate. Other health systems in our region used the COVID-19 Hospital Impact Model for Epidemics–University of Pennsylvania model. It pins its suppositions on hospitalized COVID-19 patients, regional infection rates, and hospital market shares. Lastly, the agent-based mode, such as the Global Epidemic and Mobility Project, takes simulated populations and forecasts the spread of SARS-CoV-2 anchoring on the interplay of individuals and groups. The assumptions are created secondary to the interactions of people, time, health care interventions, and public health policies.
Based on these predictive simulations, health systems have spent countless hours of planning and have utilized resources for the anticipated needs related to beds, ventilators, supplies, and staffing. Frontline staff were retrained how to don and doff personal protective equipment. Our teams were ready if we saw a wave of 250, a surge of 500, or a tsunami of 750 COVID-19 inpatients. We were prepared to run into the fire fully knowing the personal risks and consequences.
But, as yet, the tsunami in North Carolina has never come. On April 21, 2020, the COVID-19 mortality data in North Carolina peaked at 34 deaths, with the total number of deaths standing at 1,510 as of July 13, 2020.6 A surge did not hit our institutional shores at Wake Forest Baptist Health. As we looked through the proverbial back window and hear about the tsunami in Houston, Texas, we are very thankful that the tsunami turned out to be a small wave so far in North Carolina. We are grateful that there were fewer deaths than expected. The dust is settling now and the question, spoken or unspoken, is: “How could we be so wrong with our predictions?”
Models have strengths and weaknesses and none are perfect.7 There is an old aphorism in statistics that is often attributed to George Box that says: “All models are wrong but some are useful.”8 Predictions and projections are good, but not perfect. Our measurements and tests should not only be accurate, but also be as precise as possible.9 Moreover, the assumptions we make should be on solid ground. Since the beginning of the pandemic, there may have been undercounts and delays in reporting. The assumptions of the effects of social distancing may have been inaccurate. Just as important, the lack of early testing in our pandemic and the relatively limited testing currently available provide challenges not only in attributing past deaths to COVID-19, but also with planning and public health measures. To be fair, the tsunami that turned out to be a small wave in North Carolina may be caused by the strong leadership from politicians, public health officials, and health system leaders for their stay-at-home decree and vigorous public health measures in our state.
Some of the health systems in the United States have created “reemergence plans” to care for those patients who have stayed at home for the past several months. Elective surgeries and procedures have begun in different regions of the United States and will likely continue reopening into the late summer. Nevertheless, challenges and opportunities continue to abound during these difficult times of COVID-19. The tsunamis or surges will continue to occur in the United States and the premature reopening of some of the public places and businesses have not helped our collective efforts. In addition, the personal costs have been and will be immeasurable. Many of us have lost loved ones, been laid off, or face mental health crises because of the social isolation and false news.
COVID-19 is here to stay and will be with us for the foreseeable future. Health care providers have been literally risking their lives to serve the public and we will continue to do so. Hitting the target of needed inpatient beds and critical care beds is critically important and is tough without accurate data. We simply have inadequate and unreliable data of COVID-19 incidence and prevalence rates in the communities that we serve. More available testing would allow frontline health care providers and health care leaders to match hospital demand to supply, at individual hospitals and within the health care system. Moreover, contact tracing capabilities would give us the opportunity to isolate individuals and extinguish population-based hotspots.
We may have seen the first wave, but other waves of COVID-19 in North Carolina are sure to come. Since the partial reopening of North Carolina on May 8, 2020, coupled with pockets of nonadherence to social distancing and mask wearing, we expect a second wave sooner rather than later. Interestingly, daily new lab-confirmed COVID-19 cases in North Carolina have been on the rise, with the highest one-day total occurring on June 12, 2020 with 1,768 cases reported.6 As a result, North Carolina Gov. Roy Cooper and Secretary of the North Carolina Department of Health and Human Services, Dr. Mandy Cohen, placed a temporary pause on the Phase 2 reopening plan and mandated masks in public on June 24, 2020. It is unclear whether these intermittent daily spikes in lab-confirmed COVID-19 cases are a foreshadowing of our next wave, surge, or tsunami, or just an anomaly. Only time will tell, but as Jim Kim, MD, PhD, has stated so well, there is still time for social distancing, contact tracing, testing, isolation, and treatment.10 There is still time for us, for our loved ones, for our hospital systems, and for our public health system.
Dr. Huang is the executive medical director and service line director of general medicine and hospital medicine within the Wake Forest Baptist Health System and associate professor of internal medicine at Wake Forest School of Medicine. Dr. Lippert is assistant professor of internal medicine at Wake Forest School of Medicine. Mr. Payne is the associate vice president of Wake Forest Baptist Health. He is responsible for engineering, facilities planning & design as well as environmental health and safety departments. Dr. Pariyadath is comedical director of the Patient Flow Operations Center which facilitates patient placement throughout the Wake Forest Baptist Health system. He is also the associate medical director for the adult emergency department. Dr. Sunkara is assistant professor of internal medicine at Wake Forest School of Medicine. He is the medical director for hospital medicine units and the newly established PUI unit.
The authors would like to thank Julie Freischlag, MD; Kevin High, MD, MS; Gary Rosenthal, MD; Wayne Meredith, MD;Russ Howerton, MD; Mike Waid, Andrea Fernandez, MD; Brian Hiestand, MD; the Wake Forest Baptist Health System COVID-19 task force, the Operations Center, and the countless frontline staff at all five hospitals within the Wake Forest Baptist Health System.
1. World Health Organization. Modes of transmission of virus causing COVID-19: Implications for IPC precaution recommendations. 2020 June 30. https://www.who.int/news-room/commentaries/detail/modes-of-transmission-of-virus-causing-covid-19-implications-for-ipc-precaution-recommendations.
2. Holshue et al. First case of 2019 novel coronavirus in the United States. N Engl J Med. 2020;382: 929-36.
3. Fuller T, Baker M. Coronavirus death in California came weeks before first known U.S. death. New York Times. 2020 Apr 22. https://www.nytimes.com/2020/04/22/us/coronavirus-first-united-states-death.html.
4. Johns Hopkins Coronavirus Resource Center. https://coronavirus.jhu.edu/us-map. Accessed 2020 May 28.
5. Michaud J et al. COVID-19 models: Can they tell us what we want to know? 2020 April 16. https://www.kff.org/coronavirus-policy-watch/covid-19-models.
6. Centers for Disease Control and Prevention. https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/cases-in-us.html. Accessed 2020 June 30.
7. Jewell N et al. Caution warranted: Using the Institute for Health Metrics and Evaluation Model for predicting the course of the COVID-19 pandemic. Ann Intern Med. 2020;173:1-3.
8. Box G. Science and statistics. J Am Stat Assoc. 1972;71:791-9.
9. Shapiro DE. The interpretation of diagnostic tests. Stat Methods Med Res. 1999;8:113-34.
10. Kim J. It is not too late to go on the offense against the coronavirus. The New Yorker. 2020 Apr 20. https://www.newyorker.com/science/medical-dispatch/its-not-too-late-to-go-on-offense-against-the-coronavirus.