The State of Iowa acknowledged on Wednesday that the system it uses to track COVID-19 cases has been assigning some portion of the positive and negative test results to incorrect dates.
The issue was first reported by the Times Citizen, Bleeding Heartland blog and the Associated Press on Monday, days after a lead epidemiologist at the Iowa Department of Public Health admitted in an email that there was a problem and that the state was working to correct it.
Through the tracking of daily test results, the Times Citizen identified inconsistencies and changes in the dates that cases were being reported. While the county’s and state’s cumulative case counts for Hardin County were almost always identical, a portion of the cases being reported by the state as recently as last weekend were being assigned to dates as far back as April and May.
The Times Citizen reached out to the Iowa Department of Public Health with questions about the inconsistencies on July 31. Department spokesperson Amy McCoy said the changes in case numbers were because “The website is constantly being updated.” But in an Aug. 14 email, IDPH Epidemiologist Rob Ramaekers responded to Iowa City nurse practitioner Dana Jones’ questions about the backdating of cases, writing “We recognize this is a problem and have been working on logic to handle it.” Jones shared that email with the Times Citizen and a number of other media outlets.
On Wednesday, Pat Garrett, a spokesman for Gov. Kim Reynolds, acknowledged the problem for the first time and reported that it was being corrected on Wednesday.
“In an effort to report the number of individuals tested without duplicating results from those who were tested multiple times, an individual’s most recent test result, whether positive or negative, was unintentionally attributed to the date of their first test result,” Garrett wrote in an email statement to Iowa media. In other words, if a person tested negative for COVID-19 in March, and then tested positive for the disease in July, the positive test result was being erroneously assigned to the test date in March.
In an interview with The Gazette on Tuesday, Dr. Caitlin Pedati, the state’s medical director and state epidemiologist, said IDPH became aware of the problem in late July. She said it was caused by the limitations of the Iowa Disease Surveillance System, a computer system that’s used to track communicable diseases. And while Pedati said the “data issue” didn’t affect the state’s contact tracing or any individual person, it did affect positivity rates, which are being used by the state to determine whether schools can shift to virtual learning.
The data published on the state’s coronavirus website was significantly altered on Wednesday afternoon. In Hardin County’s case, the alterations changed the number of cases per day for 98 out of the 161 days, stretching back to March. Despite those changes, the total number of cases remained the same at 194. As an example, though, before the data revision, the state website was reporting that there had been 14 new COVID-19 cases on June 22. After the changes, the state is reporting there were no new cases of COVID-19 that day.
The change also affected the number of people tested per day. The total number of tests, like the total number of cases, did not change. But the number of tests performed each day changed for all 161 days charted on the website. Garrett didn’t respond to follow-up questions about the revision of testing numbers.
In his email to media on Wednesday, Garrett said the data revision would do three things:
- The state’s all-time average positivity rate would stay the same at 9.4 percent
- Nearly 80 percent of counties would see a net decrease in their 14-day positivity rate
- The remaining counties would see an average of a less than 1 percent increase in their current 14-day positivity rate.
The data revision in Hardin County altered both the number of tests that have been performed in the last 14 days, and the number of positive results, which changed the 14-day positivity rate for some past weeks, but barely altered the current 14-day rate. Before the revision, the state was reporting that 247 Hardin County residents had been tested for COVID-19 in the previous 14 days, and 16 had tested positive, resulting in a positivity rate of 6.47 percent. After the revision, the state is reporting that 327 Hardin County residents have been tested for COVID-19 in the last 14 days, and 21 have tested positive, resulting in a positivity rate of 6.42 percent.
The data revisions did not significantly alter the number of Hardin County residents the state reports as having recovered from the disease. As of Wednesday afternoon, that number stood at 134. The county reported its first COVID-19 death this week.