It is a critical function of public health to respond to introductions of diseases and conditions of unknown etiology, with recent examples including HIV/AIDS, Zika virus disease, Ebola virus disease, e-cigarette vaping product use–associated lung injury (EVALI), and 2009 H1N1. In late December 2019, investigation of a cluster of pneumonia cases of unknown origin in Wuhan, China, resulted in the identification of a novel coronavirus. The virus, later named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was distinct from both the 2002-2004 severe acute respiratory syndrome coronavirus (SARS-CoV-1) and Middle East respiratory syndrome coronavirus (MERS-CoV), although closely related. Establishment of a national surveillance case definition for the disease caused by the virus, coronavirus disease 2019, or COVID-19, was one of multiple public health surveillance actions rapidly implemented to detect disease introduction and spread of the virus in the United States. The focus of this article is on developing surveillance case definitions and case-based morbidity surveillance utilizing the reportable disease model. Case-based surveillance is used to guide immediate public health decisions and apply containment and mitigation steps. This surveillance strategy can also play an important role in identifying cases and their contacts into the isolation/quarantine processes and more so when case counts are manageable. Challenges in developing the surveillance for an illness caused by a highly infectious, novel virus spreading through a population with no existing immunity and the necessary public health surveillance data infrastructure are discussed.
Jurisdictional response at the state, territorial, local, and Tribal (STLT) health departments to identify and report COVID-19 cases has been indispensable to the national response and is the primary source of COVID-19 morbidity data used to track the pandemic in the United States. When the novel coronavirus outbreak was first detected, the Centers for Disease Control and Prevention (CDC) began working with STLT agencies to collect, compile, and analyze information regarding the outbreak of unknown etiology. Since COVID-19 first emerged, infections have continued to grow exponentially. As of mid-October 2020, states reported more than 8.1 million confirmed and probable cases of COVID-19, and of these reported cases, more than 219 000 people have died.1 Despite what is likely to be many more months of response, it is important to document the challenges of initial surveillance implementation and identify unmet needs in order to better respond to this pandemic and prepare for future emerging diseases.
Early Efforts to Detect Introductions: Creation of a “Person Under Investigation” Status and Initial Identification of COVID-19 Community Transmission
To rapidly detect potential introductions of the novel coronavirus, a person under investigation (PUI) category2 was established as a predecisional status to track identified potential infections of respiratory illnesses of unknown etiology. As in the SARS report under investigation (RUI) category3 in 2003, PUI status was a sensitive, nonspecific designation based solely on clinical or epidemiological criteria and used prior to any definitive laboratory test results or information available to classify the infection as a “case” of COVID-19. The first PUI definition,2 established January 17, 2020, required both clinical and travel-associated risk factors. The PUI category was revised several times, and on February 28, CDC published a Health Alert Network Health Update4 introducing a new PUI testing criterion that allowed COVID-19 testing for patients in whom there was a high index of clinical suspicion without known travel history and a likely source of exposure. Identified PUIs were investigated by STLT agencies and were either (a) excluded or “ruled out” or (b) enumerated as SARS-CoV-2 infections following confirmation through CDC laboratory testing. Prior to availability of laboratory findings, local and state health departments monitored PUIs to ensure that public health measures for preventing disease transmission remained in place if SARS-CoV-2 was subsequently confirmed. To improve the likelihood of early detection of disease introduction, the PUI identification approach included enhanced airport health screening of passengers from areas with identified transmission along with funneling returning travelers to established screening airports and other repatriation efforts.5 While the PUI surveillance strategy was strenuous and aggressive, it was intended to contain the virus from spreading. Expanding the PUI testing criteria earlier in the outbreak to include persons with broader travel exposure (not limited to Wuhan) and with influenza-like illness (ILI) without an alternative diagnosis would have increased the number of individuals detected through public health surveillance efforts. The limited testing capability and strict PUI testing criteria, which required symptoms and travel history, meant COVID-19 infections were undetected and ultimately contributed to sustained community transmission.6 The PUI status was no longer maintained once broader community transmission was occurring, laboratory diagnostics became more widely available, and more established clinical and epidemiological criteria were defined by public health.
In addition to the PUI strategy, several jurisdictions developed sentinel surveillance7–9 to determine the extent, if any, of community transmission in the United States. Sentinel surveillance was used to identify COVID-19 infections in the health care system among persons with ILI who tested negative for influenza and did not have another diagnosis or relevant travel history. For example, the New York City Department of Health and Mental Hygiene collected and analyzed de-identified remnant nasopharyngeal swab specimens from patients with ILI and no known virologic diagnosis evaluated at 6 sentinel emergency departments,6 and in Santa Clara County, California, specimens with identifiers were collected and analyzed from 4 urgent care centers from patients presenting with respiratory symptoms (eg, fever, cough, or shortness of breath) who had no known risk factors for COVID-19.7 While jurisdictional sentinel surveillance implementations differed, each proved effective in establishing the occurrence of community transmission and documenting SARS-CoV-2 infections previously undetected by the active case finding via traveler screening, repatriation efforts, and PUI surveillance.
Establishing COVID-19 as a Nationally Notifiable Condition and Creating a National Surveillance Case Definition
Early in the COVID-19 pandemic, CDC and the Council of State and Territorial Epidemiologists (CSTE) developed surveillance criteria to standardize the identification and quantification of COVID-19 cases. A surveillance case definition is a set of uniform criteria used to define a disease or condition for public health surveillance. Surveillance case definitions enable public health officials to classify and count cases consistently across reporting jurisdictions and are not intended to be used by health care providers to make a clinical diagnosis or treatment decisions. CSTE is the national member-based organization representing STLT epidemiologists across disciplines and disease categories. Convened for the first time in the 1950s, CSTE, in collaboration with CDC, holds the responsibility for defining and recommending which diseases and conditions are nationally notifiable and then voluntarily reported by states to CDC. CSTE position statements represent membership's consensus opinions on policy and surveillance topics or issues affecting public health.
Developed through the position statement process, surveillance case definitions include 3 major components: clinical, laboratory, and epidemiologic criteria. These 3 criteria are used to construct a tiered case classification scheme that can include confirmed, probable, or suspected case status depending on the disease or condition under surveillance. For nonemergent issues, an annual process is used to establish or revise national case definitions, which allows broad, membership vetting, approval by the full CSTE Council (with 1 voting member per state and territory), and preplanning time for CDC, jurisdictions, and health care to ensure appropriate reporting laws, processes, and polices are in place. Because of the urgent and time-sensitive needs of the COVID-19 response, use of the interim position statement process (voting approval by the Executive Board rather than full Council) was employed to ensure effective surveillance case-counting methods were in place. A collaborative group led by CSTE members representing multiple jurisdictions, the Association of Public Health Laboratories (APHL), and subject matter experts at CDC developed the COVID-19 standardized surveillance position statements.
On April 5, 2020, CSTE adopted the first interim position statement (Interim-20-ID-01)10 to establish the national surveillance case definition for COVID-19, add COVID-19 to the National Notifiable Condition List, and designate the disease as “immediately notifiable, urgent (within 24 hours).” While all states and territories have laws with provisions that require reporting of unusual cases, outbreaks, or newly emerging conditions, Interim 20-ID-01 recommended all states and territories enact laws to make COVID-19 explicitly reportable in their jurisdictions. On August 5, 2020, the CSTE Executive Board approved the adoption of the second interim position statement for COVID-19, Interim-20-ID-02,11 superseding Interim-20-ID-01. The surveillance case definition was revised within 4 months to reflect the rapid advancement in the science and public health understanding of the clinical, laboratory, and transmission characteristics of SARS-CoV-2 infection and COVID-19 disease. During this time, CSTE received feedback from its members regarding the implementation and application of the original surveillance case definition, and antigen detection tests and serologic tests were developed and authorized for use by the US Food and Drug Administration. Interim 20-ID-02 clarifies classification of cases identified through use of antigen detection and serologic testing. Interim 20-ID-02 was broadly vetted among the CDC/CSTE COVID-19 Core Group, State Epidemiologists, and CDC Response Incident Management and Office of Infectious Diseases leadership prior to approval by the 2019-20 CSTE Executive Board. CSTE anticipates that additional revisions to the surveillance case definition will likely be needed before the close of the pandemic as further scientific advancements, such as new laboratory diagnostics, are discovered.
National Notification of COVID-19 Case Data to CDC
In addition to providing standardized classifications for counting cases, the CSTE position statement Interim-20-ID-01 recommended COVID-19 become nationally notifiable and states transmit case data to CDC for national aggregation. The CDC's National Notifiable Diseases Surveillance System (NNDSS) data submission process supports the collection of a core set of variables but lacks flexibility and rapid modification capabilities. The NNDSS COVID-19 case surveillance database includes case-level data voluntarily sent to CDC from US states and autonomous reporting entities, including New York City and the District of Columbia, as well as US territories and freely associated states. Jurisdictions send de-identified data about COVID-19 cases to CDC including demographic characteristics (eg, age, race, and ethnicity), disease severity indicators and outcomes, clinical data, laboratory diagnostic test results, and presence or absence of any comorbidity. The individual case-level NNDSS data serve as the foundation to support assessment of identifying those at highest risk for severe complications (elderly and those with certain underlying health conditions12) and support analyses that identified racial and socioeconomic disparities.13 States and territories worked in coordination with CDC to submit core data elements through the standard NNDSS process; however, depending on jurisdiction processes and capacity, others completed direct data entry or upload alternate file formats to the CDC Data Collation and Integration for Public Health Event Response (DCIPHER) platform. While the NNDSS processes support individual case-level details to be collected, enhanced data elements, such as those related to exposures, are entered or submitted to CDC via DCIPHER where the data were subsequently combined with the core data submitted via NNDSS. In addition, NNDSS did not support rapid data collection of simple daily case count totals, and alternate processes to collect these data had to be established between STLT agencies and CDC.
Undercounting and Lack of Identification of Acute Infections
Individual case counts have been used throughout the response to track the spread of the virus, identify clusters and outbreaks, measure disease burden, and guide public health containment and mitigation measures. However, as SARS-CoV-2 became more widespread, the utility of individual case counts for measuring the full spectrum of disease burden has diminished. Several reasons contribute to this, including the limited availability of testing and the recognition of the large role that asymptomatic and mildly symptomatic infections play in transmission. These important disease characteristics have resulted in many people who likely became mildly ill with SARS-CoV-2 and never sought treatment. In addition, as the outbreak intensified, due to limited testing supplies and personal protective equipment, testing was Restricted—at times available only to hospitalized patients—leaving many people who may have sought care but were never officially tested or diagnosed. Thus, the official case counts continue to represent only a fraction of the true burden of COVID-19 in the United States. As described later, another contributing factor to undercounting was due to the public health data supply chain and constraints reporting the data needed for case ascertainment to public health. However, despite exponential growth of the pandemic and the lack of identification of acute infections (as further evidenced through seroprevalence studies14,15), individual case-based surveillance (counting confirmed and probable cases) continues as the primary surveillance strategy.
Challenges in the Public Health Data Supply Chain and Its Impact on Public Health Action
The COVID-19 pandemic has exposed inadequacies of the US health system16 and strains on the public health system.17 One primary challenge to effective case-based surveillance efforts is the data reporting process itself. The COVID-19 pandemic has put unprecedented stress on the public health data supply chain. The ability of public health authorities to respond to the COVID-19 pandemic has been hampered by 2 drivers: manual processes for data transmission and incomplete information when reports to public health are made. Delays or lack in data reporting from providers and the sheer volume of cases have impeded public health's ability to implement case-based control measures. While public health has been scrutinized for missing information (eg, race and ethnicity) and the inability to fill in those data gaps despite authority and supporting laws, it is critical we look further upstream to the initial reporting and provision of data to public health by health care. Even in the best scenarios, receiving vast amounts of missing information will delay a public health response; however, if the data already stored electronically in the electronic health record (EHR) (eg, patient address, phone number, race, ethnicity, and comorbidities) were provided at the time of initial report, inefficiencies would be eliminated and days and weeks spent tracking down missing information could be saved.
Challenge 1: Case reporting
All states have laws that require physicians, hospitals, and laboratories to report COVID-19 and conditions of public health significance. While timely and accurate reporting is mandated through jurisdictional laws, required case reports are often delayed, incomplete, or never submitted. Despite many advancements in reporting of laboratory data through electronic laboratory reporting,18 case reporting continues to be largely an entirely paper-based process, with reports made via phone call or fax,19 consuming considerable human resources for both public health practitioners and health care providers. In addition, it is fraught with numerous inherent opportunities for error, including data entry mistakes and confusion about what to report and what information to include in a report, when to report, and how to report. Moreover, the sluggish nature of the manual process that is still in widespread use has important consequences. The most notable consequence is delayed detection and response to acute, chronic, and emerging public health threats. In many states, the large number of COVID-19 patients has severely strained the ability of hospitals, health care providers, and laboratories to submit reports to public health with complete demographic information, such as race and ethnicity. Because of the utter volume of reports, many health departments are not able to conduct thorough investigations of every case to obtain additional information. Consequently, our national picture and case records at the state and local level are largely missing data on patient demographics, symptoms, underlying health conditions, characteristics of hospitalizations (eg, ventilator use), and other factors (eg, recent travel history).
Challenge 2: Laboratory reports
Laboratory reports fail to include full patient address and phone number (missing as much as 50% of the time, oral personal communication), leading to delayed identification of hot spots, as well as the ability to rapidly contact individuals to conduct interviews and complete contact identification. Laboratory reports are usually missing race and ethnicity information (as much as 80%-85% of the time, oral personal communication). Providers commonly do not include patient demographic information on laboratory orders (paper), and electronic laboratory order interfaces have traditionally not been configured in a way to automatically pull the information from the EHR because demographic information usually is not necessary to perform a laboratory test. Although missing demographic information at the time of initial reporting has been a long-standing problem, its absence during the COVID-19 response is problematic, as there is a great need to obtain this information rapidly and at scale to impact policy changes.
While electronic laboratory result reporting from commercial or private laboratories to state public health agencies has been in place for many years18 and has had relatively good coverage and participation, current state laws often only support reporting of positive results indicative of conditions of public health significance. From the beginning of the COVID-19 pandemic, it was clear that reporting of all tests results would be necessary for managing various aspects of the pandemic. The negative findings were equally as important in some aspects as the positive findings, for example, to release those from quarantine (prior to the establishment of the known quarantine period), to assess the total persons tested, to assess if geographic, racial, or ethnic disparities or access to testing was present, to assess percent positivity, and to fully understand laboratory testing supply needs. Although these data have been collected to manage more localized outbreaks previously, the rapid scale-up by public health to receive this volume of data (ie, millions of records) has been a challenge. In addition, many new testing laboratories are becoming operational and are not familiar with public health reporting needs. Furthermore, “drive thru” or alternative testing locations are challenged to collect the full level of demographic information needed by public health.
Challenge 3: Point-of-care testing
The emergence of point-of-care testing20 in July 2020 is positioned to have profound effects on managing the pandemic. Point-of-care devices functioning in freestanding clinics and other nontraditional testing locations are largely not connected to EHRs or laboratory information management systems (LIMS). The challenges for public health surveillance for COVID-19 to incorporate point-of-care test results are not new and are faced when conducting surveillance for other conditions of public health importance, such as lead poisoning or influenza. Point-of-care tests are conducted at the “bedside,” and results are returned quickly to the patient and the provider with rapid information to treat and manage the patient. Unfortunately, without direct-device reporting or integration of results into EHRs or LIMS, the test results are often never sent to public health or, if submitted, they are handwritten via mail or fax as no electronic reporting system to send the data exists. This is another factor limiting case identification by public health.
Recommendations: Modernizing Public Health Surveillance
For years, public health professionals in the United States have lamented about decreasing budgets and lack of resources for prevention.21 The lopsided focus on spending more on treating preventable conditions while decreasing investments in public health systems that could be used to prevent such conditions has devastating effects. Ultimately, this funding approach has led to the health care delivery system and dedicated health professionals being quickly overwhelmed and threatened within days and weeks when responding to large-scale public health emergencies.
As a philosophical approach, getting data rapidly to public health has been left out of the current policy and funding mindset; instead, the framework is perpetuated that anemic funding is adequate and public health can find ways to fill in the data gaps. These slow, inefficient, and incomplete data processes from health care to public health occur across all diseases and conditions of public health significance. Even during this pandemic, there is still a perception that providing data to public health lacks urgency, despite the recognition that case and contact tracing is one of the primary means for controlling spread when few other measures exist.
A totally integrated, high-speed, networked health system—from laboratories to health care facilities to public health authorities—together with fast and reliable data, is required in order to protect Americans from health threats. Modernization is not just network upgrades; it is a commitment to build a world-class workforce with a new generation of skilled public health data scientists, laboratories, and electronic, interoperable enterprise data systems that are ready for the next public health emergency.
In the short term, nothing less than a national scale-up of surveillance system capability in every jurisdiction is needed. This scale-up should address critical gaps in the acceptance and management of electronic case reports for COVID-19 and routine surveillance activities, including public health investigation, case management, and contact tracing. The eCR Now initiative is a strategic initiative that enables rapid adoption and implementation of eCR for COVID-19 (https://www.cdc.gov/coronavirus/2019-ncov/hcp/electronic-case-reporting.html). The laboratory test order process must be improved to ensure it is interoperable with health care systems and fully supports the collection and sharing of critical data such as race and ethnicity information from health care to public health at the time of initial test order. Electronic direct-device reporting from point-of-care testing should be explored. In addition, we need to bolster public health surveillance systems now to handle the volume of laboratory reports, process them in real time, and improve our ability to rapidly analyze and summarize these data to provide information to the public and policy makers for decision making.
We must not continue with a piecemeal, fragmented approach to funding our public health data infrastructure. While some investment in surveillance system modernization has been made to support response efforts, Congress's further support and funding during emergencies are critical to support the COVID-19 pandemic response and to implement optimal data collection and data system management; it cannot be approached as “one and done.” We must invest now, and we must ensure we finally establish sustained, annual funding for public health surveillance systems to protect our national health security.
Our nation requires a robust, sustained commitment to build a 21st-century data superhighway to speed the seamless exchange of data for all diseases and conditions, to predict and prevent public health threats before they occur, and to ultimately improve and protect Americans' health.19 The need is now—to prioritize and support a public health surveillance enterprise that will speed the data collection and response for COVID-19 and future public health threats.
Implications for Policy & Practice
- Public health has been chronically underfunded and funded in a fragmented piecemeal approach. Congress should establish a sustained, long-term, predictable funding source for enhancing and maintaining a robust data infrastructure and public health surveillance systems to protect our national health security. For the current system to truly evolve, the federal government must commit to long-term funding to complete essential system upgrades both federally and at the state and local levels and annual optimal funding to support ongoing maintenance as technology improves and a strong data science workforce to run and update these systems.
- Provider incentives should be implemented to support rapid adoption of electronic case reporting eCR, and eCR should become a required base component for certified electronic health record (EHR) products.
- The laboratory test order process must be improved to ensure it is interoperable with health care systems and fully supports the collection and sharing of critical data such as race and ethnicity information from health care to public health at the time of initial test order.
- Explore and incentivize electronic direct-device reporting from point-of-care testing to help ensure testing that occurs outside of a traditional laboratory or health care setting can be easily reported to public health.
1. Centers for Disease Control and Prevention. Coronavirus disease 2019 (COVID-19)—situation summary. https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/summary.html
. Accessed October 20, 2020.
2. Centers for Disease Control and Prevention. Health Alert Network: Update and Interim Guidance on Outbreak of 2019 Novel Coron-avirus (2019-nCoV) in Wuhan, China. Atlanta, GA: US Department of Health and Human Services, CDC; 2020. CDCHAN-00426. https://emergency.cdc.gov/han/han00426.asp
. Accessed October 1, 2020.
3. Council of State and Territorial Epidemiologists. Position statement: (03-ID-12) inclusion of severe acute respiratory syndrome (SARS) reports under investigation and SARS-associated coronavirus (SARS-CoV) disease in the National Public Health Surveillance System (NPHSS), and revision of the interim US surveillance case definition for SARS. https://cdn.ymaws.com/www.cste.org/resource/resmgr/PS/03-ID-12revised.pdf
. Accessed October 1, 2020.
4. Centers for Disease Control and Prevention. Health Alert Network: Update and Interim Guidance on Outbreak of Coronavirus Disease 2019 (COVID-19). Atlanta, GA: US Department of Health and Human Services, CDC; 2020. CDCHAN-00428. https://emergency.cdc.gov/han/han00428.asp
. Accessed October 1, 2020.
5. Myers JF, Snyder RE, Porse CC, et al. Identification and monitoring of international travelers during the initial phase of an outbreak of COVID-19—California, February 3-March 17, 2020. MMWR Morb Mortal Wkly Rep. 2020;69:599–602.
6. Daskalakis D. Health Alert #4: COVID-19 Updates for New York City. New York, NY: New York City Health; 2020. https://www1.nyc.gov/assets/doh/downloads/pdf/han/alert/2020/covid-19-update-030920
. Accessed August 20, 2020.
7. Spellberg B, Haddix M, Lee R, et al. Community prevalence of SARS-CoV-2 among patients with influenza like illnesses presenting to a Los Angeles medical center in March 2020. JAMA. 2020;323:1966–1967.
8. Bushman D, Alroy KA, Greene SK, et al. Detection and genetic characterization of community-based SARS-CoV-2 infections—New York City, March 2020. MMWR Morb Mortal Wkly Rep. 2020;69:918–922.
9. Zwald ML, Lin W, Sondermeyer Cooksey GL, et al. Rapid Sentinel Surveillance for COVID-19—Santa Clara County, California, March 2020. MMWR Morb Mortal Wkly Rep. 2020;69:419–421.
10. Council of State and Territorial Epidemiologists. Position statement: 20-ID-01: standardized surveillance case definition and national notification for 2019 novel coronavirus disease (COVID-19). https://cdn.ymaws.com/www.cste.org/resource/resmgr/ps/positionstatement2020/Interim-20-ID-01_COVID-19_NO.pdf
. Accessed August 31, 2020.
11. Council of State and Territorial Epidemiologists. Position statement: 20-ID-02: update to the standardized surveillance case definition and national notification for 2019 novel coronavirus disease (COVID-19). https://cdn.ymaws.com/www.cste.org/resource/resmgr/ps/positionstatement2020/Interim-20-ID-02_COVID-19.pdf
. Accessed August 31, 2020.
12. Preliminary estimates of the prevalence of selected underlying health conditions among patients with coronavirus disease 2019—United States, February 12-March 28, 2020. MMWR Morb Mortal Wkly Rep. 2020;69:382–386.
13. Moore JT, Ricaldi JN, Rose CE, et al. Disparities in incidence of COVID-19 among underrepresented racial/ethnic groups in counties identified as hotspots during June 5-18, 2020—22 states, February-June 2020. MMWR Morb Mortal Wkly Rep. 2020;69:1122–1126.
14. Menachemi N, Yiannoutsos CT, Dixon BE, et al. Population point prevalence of SARS-CoV-2 infection based on a statewide random sample—Indiana, April 25-29, 2020. MMWR Morb Mortal Wkly Rep. 2020;69:960–964.
15. Biggs HM, Harris JB, Breakwell L, et al. Estimated community seroprevalence of SARS-CoV-2 antibodies—two Georgia counties, April 28-May 3, 2020. MMWR Morb Mortal Wkly Rep. 2020;69:965–970.
16. Blumenthal D, Seervai S. Coronavirus is exposing deficiencies in U.S. health care. Harv Bus Rev. https://hbr.org/2020/03/coronavirus-is-exposing-deficiencies-in-u-s-health-care
. Published March 10, 2020. Accessed April 11, 2020.
17. Smith N, Fraser M. Straining the system: novel coronavirus (COVID-19) and preparedness for concomitant disasters. Am J Public Health. 2020;110(5):648–649.
18. Lamb E, Satre J, Ponte S, et al. Update on progress in electronic reporting of laboratory results to public health agencies—United States, 2014. MMWR Morb Mortal Wkly Rep. 2020;64(12);328–330.
19. Council of State and Territorial Epidemiologists. Driving public health in the fast lane: the urgent need for a 21st century data super highway. http://resources.cste.org/data-superhighway/mobile/index.html
. Accessed September 20, 2020.
21. Shah GH, Ye J, Leep CJ, Leider JP. Local health departmentsʼ approaches to deal with recession: what strategies are used to minimize the negative impact on public health services to community? J Public Health Manag Pract. 2016;22(6):537–541. https://digitalcommons.georgiasouthern.edu/health-policy-facpubs/145
. Accessed October 1, 2020.