key: cord-0686572-4jhbup2h
authors: Edwards, Jessie K; Lessler, Justin
title: What Now? Epidemiology in the Wake of a Pandemic
date: 2020-07-22
journal: Am J Epidemiol
DOI: 10.1093/aje/kwaa159
sha: 2a3d367797b5f9ac80f273424d6da12ad35a71e3
doc_id: 686572
cord_uid: 4jhbup2h

The COVID-19 pandemic and the coming transition to a post-pandemic world where COVID-19 will likely remain as an incident disease present a host of challenges and opportunities in epidemiologic research. The scale and universality of this disruption to life and health provide unique opportunities to study phenomena, and health challenges, in all branches of epidemiology, from the obvious infectious disease and social consequences, to less clear impacts on chronic disease and cancer. If we are to both take advantage of the largest natural experiment of our lifetimes, and provide evidence to inform the numerous public health and clinical decisions being made every day, we must act quickly to ask critical questions and develop new methods to answer them. In doing so we should build on each of our strengths and expertise, and try to provide new insights rather than becoming yet another voice commenting on the same set of questions with limited evidence.

As the world confronts the threat of COVID-19, epidemiology is having a moment in the spotlight. This focused attention brings both praise and criticism, and has highlighted the role of epidemiology, particularly infectious disease epidemiology. However, while epidemiology's role in combating an emerging infectious disease is well defined, as this disease transitions to endemicity and as we enter the post pandemic period, epidemiology will be no less important, but have a fundamentally different role.

Despite vast differences in scale between the COVID-19 pandemic and other infectious disease outbreaks, infectious disease epidemiologists and virologists have adapted a standard playbook (1) to learn rapidly about the characteristics and transmission dynamics of SARS-CoV-2, the virus that causes COVID-19. For example, after the report of a novel coronavirus cluster in

China's Hubei province on December 31, 2019 and subsequent cases in China, Thailand, and beyond, researchers used early data to estimate the number that each infected person was likely to infect (2) , the time from becoming infected to transmitting disease or developing symptoms (3), the proportion of those infected who end up dying due to COVID-19 (4, 5) , and the probability of transmission from asymptomatic infected individuals (6) . These estimates continue to be refined as more data become available, and, as the outbreak has progressed, epidemiologists have turned attention to strategies to prevent infection and study the clinical course of disease.

This work on the disease itself is critical, but the scale of both the COVID-19 pandemic and the global response mean its broad effects will extend far into the future. We know from the global

response to the emergence of HIV in the 1980s that epidemiologic research into a novel pathogen can go in unexpected directions as we respond to unforeseen challenges, that often require new methods and data sources. Here, we explore the opportunities and challenges for epidemiology in the wake of the COVID-19 pandemic as deal with the aftermath of mass infection, death and social disruption.

We are in the midst of a massive public health crisis. As of the end of June, 2020, over 10 million cases of COVID-19 have been reported worldwide, resulting in over 500,000 deaths, and cases and deaths continue to rise in many areas (7) . Because testing has been uneven and incomplete, the actual scope of the pandemic is known to be much larger (8) . Moreover, in the United States, early research into the distribution of infections and mortality has revealed a woefully unprepared public health system and deep inequalities, particularly by race (9) . These 

diseases. We will also need to understand the effects of multiple exposures to SARS-CoV-2, especially at later ages, and on immunity and the clinical response to later infections.

Moreover, improving the long-term response to COVID-19 will require a robust understanding the sequelae of disease among children because, though not thought to be a major driver of transmission or a major source of morbidity or mortality at the moment, as the disease moves towards endemicity we expect the average age of infection to shift towards younger individuals.

In addition to learning how to manage patients who have already had COVID-19, 

The pandemic and its aftermath will give rise to many fundamental scientific questions about both the impact of social disruption and the spread of novel pathogens. These questions will take time to answer, and their immediate public health implications may be unclear. Likewise, there are many decisions in public health, clinical medicine, and policy that must be made immediately as we emerge from the pandemic, and these decisions need to be grounded in epidemiologic evidence whenever possible.

The timeline to answer both types of questions may be short. Some decisions will be made with or without scientific knowledge as a guide, while others may wait for evidence but be less effective if delayed. Likewise, the window to answer some deep scientific questions might close if studies are not begun, or at least planned, before we enter the post pandemic period.

Asking the right questions is the first, and most critical step in finding answers that improve public health. Epidemiologists should be central to shaping, asking, and answering both types of questions. Asking good questions cannot wait until we feel prepared to answer them.

Rather than asking questions that seem answerable with available data, we should be at the forefront of discussions on what we need to learn, while acknowledging that answering these questions may require new data sources and methods.

The state of the world during (and beyond) the COVID-19 pandemic is often described as The canonical approach to epidemiologic research is to design a study, collect all the data on any actions taken (e.g., which treatment was prescribed) and what happened (e.g., patient outcomes), analyze the data, combine results with other studies to meta-analyze the findings, and then make tepid policy recommendations. As we respond to and emerge from the pandemic, we do not have the luxury of this process. Decisions will be made rapidly, with or without our input. For example, because the first known patients with COVID-19 were infected only about 6 months ago, clinicians making decisions about how to treat survivors of COVID-19 must do so without knowledge of long-term outcomes under any possible treatment strategies.

Because the world has not seen global lockdown measures like those imposed since March in recent times, we cannot look to historical examples to determine the consequences of such actions. Moreover, while the pace of data collection for COVID-19 studies is accelerating, currently available data are limited in geographic scope and population coverage, meaning that data are not yet available for many geographic areas and population subgroups that would be affected by various types of public health decisions. To base these decisions on the best available evidence, we will need to find ways to synthesize what we know from various sources.

Answering timely questions in this new era will require thoughtful application of existing epidemiologic methods and innovative new approaches. As in all epidemiologic endeavors, a primary impediment to learning will be missing information (13) (14) (15) . New approaches to build off existing methods to handle various forms of incomplete or missing information will be critical. These may include borrowing approaches from infectious disease modeling (and other

fields) that allow researchers to combine knowledge on disease mechanism with limited available data to strengthen inferences and predictions; using quantitative approaches to generalize or transport results from a setting with available data to one without (16, 17) ;

developing new approaches to data fusion (18) to combine inferences from more than one data source; applying new two-stage study designs that leverage available, but imperfect, data sources while collecting key pieces of new information through small, targeted studies (19) (20) (21) ;

and developing new analytic approaches to learn from imperfect "natural experiments" like COVID-19 where many factors evolve simultaneously. Epidemiologists from all disciplines have the opportunity to shape the next phase of life as we emerge from the pandemic by identifying and answering questions related to their areas of expertise. While partnerships and collaborations with clinical colleagues and policy makers have long been a hallmark of epidemiologic research, this new era offers opportunities for new types of cross pollination of ideas that will strengthen our collective ability to respond to urgent questions as we emerge from the COVID pandemic and beyond.

Despite having a clear role during the emergence of a new infectious disease, the role of epidemiology in the next phases of the COVID-19 pandemic is still evolving. We contend that epidemiology has a role in shaping and prioritizing the questions that are asked and developing new methods, study designs, and data sources to answer these questions in light of the challenges posed by the massive upheaval caused by the pandemic. 

Opportunities and challenges in modeling emerging infectious diseases

The reproductive number of COVID-19 is higher compared to SARS coronavirus

The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application

Estimating the infection and case fatality ratio for coronavirus disease (COVID-19) using age-adjusted data from the outbreak on the Diamond Princess cruise ship

Estimates of the severity of coronavirus disease 2019: a model-based analysis. The Lancet Infectious Diseases

Presumed Asymptomatic Carrier Transmission of COVID-19

An interactive web-based dashboard to track COVID-19 in real time. The Lancet Infectious Diseases

Prevalence of Asymptomatic SARS-CoV-2 Infection

COVID-19 and Racial/Ethnic Disparities

COVID-19 US state policy database. (www.tinyurl.com/statepolicies)

Hypertension and COVID-19

Renin-Angiotensin-Aldosterone System Blockers and the Risk of Covid-19

Bayesian perspectives for epidemiologic research: III. Bias analysis via missing-data methods

Relaxation penalties and priors for plausible modeling of nonidentified bias sources

All your data are always missing: Incorporating bias due to measurement error into the potential outcomes framework

Generalizing Study Results: A Potential Outcomes Perspective

Transportability of Trial Results Using Inverse Odds of Sampling Weights

Causal inference and the data-fusion problem

A targeted maximum likelihood estimator for two-stage designs

Addressing an idiosyncrasy in estimating survival curves using double sampling in the presence of self-selected right censoring