Van Buren County had 24 confirmed COVID-19 cases as of July 29, according to Iowa's official coronavirus website. But the local health department had recorded 46 cases in that southeast Iowa county by that date.
The discrepancy in one small county's reported COVID-19 case numbers could point to big trouble for managing school closures this fall.
STATE COUNTS VS. "ACTUAL POSITIVE CASES"
The Van Buren County Hospital posted on its Facebook page July 29,
Van Buren County Hospital (VBCH) and Van Buren Public Health would like to notify the community of the discrepancy between the State Hygienic Labs COVID-19 positive counts and Van Buren County actual positive cases.
VBCH is providing patients with the COVID-19 antigen test. This is known as the Rapid test that gives patients results in 15-20 minutes. When we started using these test we were required to validate 5 positive and 5 negative test with the State Lab, which VBCH did.
We were informed yesterday by the Rob Ramaekers, Lead Epidemiologist at the Center for Acute Disease Epidemiology and Iowa Department of Public Health, that the antigen test is not a confirmatory test, according to CDC case definition. But he does go on to say, “However people tested with this test are considered a 'probable' case similar to serology antibodies. Currently the website only counts confirmed cases (PCR positive).”
The State website currently shows 24 as the positive count for Van Buren County, Van Buren County Public Health records show a positive count of 46 as of Wednesday, July 29th.
As of July 31, the state website shows 29 COVID-19 cases for the county, still well below the local count.
Staff at the Van Buren County Hospital told Bleeding Heartland in a July 31 telephone interview that the facility began using a Sofia rapid-testing machine for COVID-19 antigens in late June. The tests may be used for either symptomatic patients or those being screened for coronavirus ahead of a scheduled surgery or other inpatient procedure.
The hospital also collects samples for PCR (polymerase chain reaction) testing for COVID-19, depending on patient preference or what the health care provider orders. When given a choice, many patients select the antigen test because of the quick turnaround. Samples for PCR tests are sent to a private laboratory if the patient doesn't meet the State Hygienic Laboratory's current criteria for COVID-19 testing.
Van Buren's Public Health department treats positive results on PCR or antigen tests the same way, the July 29 Facebook post noted. After reported positives on either kind of test, staff proceed with case investigation and contact tracing.
"A NEW TYPE OF DIAGNOSTIC TEST"
PCR tests are considered the "gold standard" for diagnosing COVID-19. They can detect "even minuscule quantities of virus" in samples taken from a person's nose or throat. An antigen test works differently but can "quickly detect fragments of proteins found on or within the virus" from nasal swab samples.
The U.S. Food and Drug Administration approved the Sofia machines in May. A government news release described the antigen test as a "new type of diagnostic test designed for rapid detection of the virus that causes COVID-19." The big advantage: the "speed of the test, which can provide results in minutes." There are some limitations (emphasis added).
However, antigen tests may not detect all active infections, as they do not work the same way as a PCR test. Antigen tests are very specific for the virus, but are not as sensitive as molecular PCR tests. This means that positive results from antigen tests are highly accurate, but there is a higher chance of false negatives, so negative results do not rule out infection.
Guidance posted on the U.S. Centers for Disease Control website and an FDA fact sheet for health care providers, updated on July 17, list the same pros and cons about tests processed with these machines. From the fact sheet:
A positive test result for COVID-19 indicates that antigens from SARS-CoV-2 were detected, and the patient is infected with the virus and presumed to be contagious. [...]
The Sofia SARS Antigen FIA has been designed to minimize the likelihood of false positive test results. [...]
A negative test result for this test means that antigens from SARS-CoV-2 were not present in the specimen above the limit of detection. However, a negative test result does not rule out COVID-19 and should not be used as the sole basis for treatment or patient management decisions, including infection control decisions.
In other words, a person whose swab came back negative using the Van Buren hospital's rapid testing may or may not have coronavirus. But a person who received a positive result from the antigen test is highly likely to have COVID-19.
The CDC has said antigen tests can be used for diagnostic purposes, despite the possibility of false negatives. In contrast, federal guidance says antibody testing (also called serology testing) "should not be used as the sole basis to diagnose or exclude SARS-CoV-2 infections or to inform infection status."
"WE'D WANT TO MAKE SURE THAT THOSE ARE COUNTED AS CASES"
Dr. Eli Perencevich, an epidemiologist and infectious disease specialist at the University of Iowa, expressed concern about the state's counting method during a July 31 telephone interview. He observed, "CDC does consider an antigen test a diagnostic test. It's not an antibody test, so it's considered to be active disease."
Because the antigen tests are less sensitive than PCR tests, a positive result "is just as likely to be a true positive," but the process may miss some infections, "so there could be false negatives." Still, the antigen tests are "pretty good diagnostic tests" for people with symptoms suggestive of COVID-19, Perencevich added. Inaccurate results might be more of an issue when screening patients who are not apparently ill.
The discrepancy in Van Buren's case numbers struck Perencevich as important because the PCR testing supply "is really being challenged" around the country. As COVID-19 cases "are exploding" in many states, materials needed to process PCR tests are not always available. Perencevich anticipates more use of antigen tests this fall, as school openings lead to more illnesses and greater need for testing. "I think we'd want to make sure that those are counted as cases."
The FDA stated in May,
Antigen tests are also important in the overall response against COVID-19 as they can generally be produced at a lower cost than PCR tests and once multiple manufacturers enter the market, can potentially scale to test millions of Americans per day due to their simpler design, helping our country better identify infection rates closer to real time.
Perencevich noted that the few dozen cases recorded in Van Buren County may be "a drop in the bucket" compared to the state's total (44,581 at this writing). "But for their schools, it's a big deal" if the state is significantly undercounting positives.
Under guidance the Iowa Department of Public Health and Iowa Department of Education released on July 30, school districts may request permission to close schools temporarily only under certain conditions, including at least a 15 percent positivity rate for COVID-19 tests conducted in a county over the previous fourteen days.
If sick people are getting antigen tests for a quick diagnosis, but the state doesn't count those numbers when determining community spread, schools may remain open even where a COVID-19 outbreak is spiraling out of control. The Van Buren County Community School District extends into portions of three other southeast Iowa counties.
Perencevich found it "shocking" that the Iowa Department of Public Health wouldn't count antigen tests as confirmed positives and said he didn't understand the thinking behind that policy.
Communications staff for the IDPH didn't respond to an inquiry about how many Iowa hospitals use rapid antigen testing machines and how many positive results from such tests have been reported to the state. The Iowa Hospital Association referred questions about hospital usage of antigen tests to the state.
This inconsistent counting method may not have the same impact on school districts located near Iowa's large medical centers. A spokesperson for UnityPoint confirmed on July 31 that their hospitals use PCR rather than antigen tests to screen patients. Staff for Linn County Public Health said via email, "we are only aware of PCR testing and not antigen testing at our local hospitals."
Polk County's Health Department referred questions to the IDPH or area hospitals, but staff indicated that when counting cases, health officials in Iowa's largest county "use the data provided" on the state's website.
Perencevich highlighted one other red flag about the state's plan to use the positivity rate as a metric for school closure decisions. The governor's public health emergency proclamations require a negative COVID-19 test within 72 hours before elective procedures. As hospitals screen before scheduled surgeries or other inpatient procedures, Perencevich said, they are "testing a ton of asymptomatic people," which is "really inflating the denominator" for the state's tests.
For that reason, he argued, Iowa should be gauging community spread by looking at active COVID-19 cases per 100,000 population in each county, rather than using the positivity rate for tests conducted in the past two weeks.
UPDATE: The Van Buren County Public Health Department has begun posting regular updates on local COVID-19 statistics on its Facebook page. As of July 31, the county had recorded 55 total cases: 30 active positive cases and 25 recovered from coronavirus.
SECOND UPDATE: IDPH spokesperson Amy McCoy explained the rationale for the policy in an August 5 email.
When we think about testing, we can think about three kinds of testing: PCR, antibody and antigen. The CDC describes them here in a recent article. Antigen tests have not yet been widely used at this time in Iowa, but are now commercially available for point-of-care testing.
There are special considerations with this type of testing, including that false negatives from symptomatic patients will also need to be confirmed through standard polymerase chain reaction (PCR) viral testing, the standard testing currently done to detect active infections.
Currently, according to the CSTE case definition, antigen tests that have positive results are considered probable positive results, not confirmed positive results. Only confirmed positive results are reported at coronavirus.iowa.gov at this time.
The Council of State and Territorial Epidemiologists is working on the case definition for this testing, and IDPH is looking forward to any updates and will adjust as appropriate. We hope to have that information soon.
As with many parts of the COVID-19 response, FDA-approved tests are developing rapidly and the state will continue to evaluate how antigen testing can be a meaningful tool as part of our COVID-19 response and reporting.
The fact remains that an unknown number of positive cases identified through antigen tests are not disclosed on the state's website and not used when calculating county-level positivity rates.
AUGUST 19 UPDATE: The state has changed much of the case data on the COVID-19 website after a member of the epidemiologist team admitted some cases had been backdated for months.
One thing that didn't change was the undercount of cases in Van Buren County, where local public health officials announced on August 19 that the county has identified 85 total cases, 47 active cases, and two deaths. Meanwhile, the supposedly corrected state website lists 42 total cases in Van Buren County and one death.
Top image: Van Buren County Hospital in Keosauqua, Iowa. Photograph originally published on the facility's Facebook page.