As Covid-19 surges again in the U.S., the high percentage of “recovered” cases might be cited as a sign that a vast majority of those infected quickly rid themselves of the virus.
But the “recovered” statistics are incomplete, inconsistent and call into question the accuracy of any total number of recovered cases, according to a review of 50 state public health sites by the Midwest Center for Investigative Reporting.
Indeed, the word “recovered” often is used for people still suffering serious after-effects.
When there is a definition of “recovered,” it varies widely from state to state. In Illinois, an infected person is considered “recovered” if they have not died within 42 days of when the person has been diagnosed with the virus. In Delaware, patients are considered recovered if they are symptom free within 7 days.
Half of the states in the U.S. either have no clear definition for “recovered” COVID-19 patients or do not track recoveries at all, according to the Midwest Center review.
While there is no standard definition for a “recovered” COVID patient, a more generalized indicator given by the CDC is: if the patient has not had a fever for at least 72 hours, symptoms have improved, and at least seven days have passed since their first symptom. If a patient is to be tested, they also need to receive two negative tests in a row at least 24 hours apart.
The Midwest Center’s review of state sites was done in the first week of July.
The review found some states had been changing their definitions of recovered over the duration of coronavirus, which had begun in January when the first cases were diagnosed.
Using the information posted on state public health websites, the survey found four primary categories on cases where the patient has “recovered.” The Midwest Center summarized those categories as:
- No Definition. 16 states (32 percent) have no definition for “recovered” and do not report “recovered” data or similar data.
- Day-Based. 19 states (38 percent) define “recovered” as being symptom-free for a certain number of days, which ranges anywhere between 3 and 42 days depending on the state.
- Alternative Definitions. 8 states (16 percent) do not record “recovered” data, but count cases using metrics such as hospital discharges or being symptom-free and no longer needing isolation.
- Tiered System. 7 states. (14 percent) These states have more complex requirements for a person to be considered “recovered.” Hawaii, for example, includes cases that meet isolation release criteria defined as: “Isolation should be maintained until at least 3 days (72 hours) after resolution of fever and myalgia without the use of antipyretics OR at least 10 days have passed since symptom onset, whichever is longer.”
“Recovered” can be misleading
Public health officials say the majority of people with COVID-19 only suffer mild symptoms, then get better usually after about two weeks, and some states report “recovered” numbers as a sign of progress in fighting the pandemic.
But medical studies and first person accounts have revealed that patients can struggle for weeks if not months to regain their health, and in some cases, patients are unsure they will ever be the same again. The lingering illnesses’ effects include fatigue, weakness of muscles, respiratory and cardiovascular issues, among other short- and long-term effects.
“[The term] ‘recovered’ on our FB [Facebook] page & reporting just means that the person is no longer infectious, and no longer in isolation, and if relevant no longer hospitalized," said Julie Pryde, public health administrator for the Champaign-Urbana Public Health Department, said in an email. "It does not mean, necessarily, that they are feeling better. There have been many reports of people having symptoms like shortness of breath, lung scarring, amputations, a young woman in Chicago had a double lung transplant, etc."
Pryde also said reports on the long-term effects of the virus will come out when there is more information: "Public Health is not following-up on the morbidity, just the mortality... It is likely that some people will have a lifelong disability from their infections."
Justine Kaplan, interim director of the master of public health program at the University of Illinois in Urbana, said a lack of knowledge about the virus leads to confusion over the effects.
“On the recovered definition, we don’t have that clarity, so it’s hard to ask people to be consistent when we’re not sure that’s the accurate message,” she said.
Kaplan said there are so many unknowns about the virus.
“That’s why when folks from public health or the clinical community are talking more specifically, they’ll keep referring to it as the novel virus, and ‘novel’ just means truly new and that’s an indicator of the lack of clarity we have about it,” she said.
Kaplan’s work at the College of Applied Health Sciences focuses on public health management and chronic disease prevention, with over 15 years of experience in the field. Much of her work also focuses on health in underserved communities. She said part of the vagueness in COVID-19 “recovered” definitions comes from a lack of testing in some communities.
Kaplan said some people who have been diagnosed with COVID-19 may not have had a test because some communities did not have testing early on in the pandemic — and some communities still don’t have access to reliable or available testing.
”So, sometimes in public health and medicine we say, ‘OK we’ll test you for something and then when you no longer test positive, aha that’s recovered.’ Well, if that was your plan, that doesn’t work if you didn’t test someone in the first place,” she said.
States, counties differ in data and definitions
As of July 7, the U.S. had recorded over 3.1 million COVID-19 cases. Of that number, there have been at least 133,000 deaths and over 960,000 recovered cases, according to Johns Hopkins University. According to its COVID-19 dashboard, the university team uses state-level “recovered” data from the COVID Tracking Project, including data for states that don’t have a “recovered” definition or don’t track recoveries on the state level, like Colorado. Some of the state’s county health districts report “recovered” cases, but not all counties do, and definitions may vary across counties.
COVID Tracking Project is run by The Atlantic magazine and publishing company. The project managers have not yet responded to a request to comment on their methodology.
States with alternative definitions, typically based on patient discharges from the hospital and needs to isolate, are mostly open or reopening except for Arizona, which is closing.
The state has no definition for “recovered” and instead uses a guide to “release from isolation.” If one were to test positive, then in order to release from isolation, they must have no fever for three days without use of medicine, other symptoms have improved, and at least 10 days have passed since symptoms appeared. Although similar to states that track “recovered” cases, the definitions and numbers differ greatly.
Two-thirds of states with day-based definitions are mostly reopened or reopening. Tiered system states, which have more complex requirements like Hawaii and Kentucky, are mostly reopened with Wyoming as the one state pausing.
In Wyoming, a lab-confirmed or probable case is defined as recovered when there is resolution of fever without the use of fever-reducing medications, there is improvement in respiratory symptoms (e.g. cough, shortness of breath) for 72 hours and at least 10 days have passed since symptoms first appeared. Cases with laboratory-confirmed COVID-19 who have not had any symptoms are considered recovered when at least 10 days have passed since the date of their first positive test and have had no subsequent illness provided they remain asymptomatic.
Tiered systems simply have more complex requirements in order for a person to be considered “recovered,” whereas alternative definitions do not use “recovered” at all.
Having reliable data is critical to better understanding the virus and finding possible solutions, said Timothy Wiemken, an associate professor at the Saint Louis University School of Medicine who focuses on infectious diseases, allergies and immunology.
“At this point in the pandemic, all data matters,” he said. “Anything that can be obtained in a valid and reliable manner helps us piece everything together in hopes of coming up with more specific and less difficult interventions.”
Wiemken said, in general, the United States has had difficulty obtaining reliable data because some people can’t access testing or there are delays in testing.
“Often, by the time people get their test results back they feel better and may not report back on recovery – so we lose those folks to follow-up unless they can be rigorously followed and their outcomes reported back to public health departments. Unfortunately, this isn’t happening everywhere,” he said.Half of the states with no definition for recoveries are reopening — and the other half are pausing reopening plans or closing the state. Texas, for example, began to reopen in phases, but has reimposed restrictions on restaurants and Governor Greg Abbott issued a mask mandate on July 2.
Pam Dempsey contributed to this report.
This story was updated June 13, 2020, to reflect new information on South Dakota.