When the state’s top health official announced this week that Washington would revamp how it counts negative COVID-19 test results, he said the change would give officials a more up-to-date snapshot of the disease.
But it also added even more delays to the state’s daily reporting of the positivity rate, a key metric that determines how many of those tested are infected with the virus.
Ongoing data troubles had already slowed reporting of the statewide positivity rate for 10 straight days — the latest in a series of technical problems hampering the state Department of Health’s (DOH) COVID-19 case reporting since the pandemic erupted here in late February.
State health officials have variously blamed an overwhelmed disease reporting system, software challenges, methodology changes and other fits and starts for periodically stalling timely, comprehensive and accurate data reporting.
“This is work that has never been done at this scale before, our data systems have never handled this quantity of data and the epidemiologists doing this work have been part of a tireless effort since the beginning of this response,” Lisa Stromme Warren, a DOH spokesperson, said in an email on Thursday.
“As we learn more about this virus, our plans, public guidance and even methodology will evolve.”
DOH officials largely have downplayed the impacts of data issues over recent months, saying they haven’t hurt Washington’s public health response. With the most recent problems, DOH stressed that the most important COVID-19 metric — the count of people who have tested positive for the new coronavirus — is unaffected and the state’s efforts to curb the virus aren’t hampered.
“Fortunately, we are confident that we know the general nature of the positivity rate, which has been relatively stable, and the delta between getting the exact number is not large enough to really upset decision making,” Gov. Jay Inslee said at a news conference Thursday.
But the positivity rate used by some across the country to help assess whether to open schools or businesses won’t be available in Washington for about another week as, six months into the pandemic, the state shifts its reporting, said Secretary of Health John Wiesman.
Going forward, the figure will be based on the results of the tests overall, versus the number of people who have tested positive or negative for the virus. That shift, public health officials said, will make it possible to count the recent tests for people who may have tested negative for the virus months ago.
Some researchers and scientists say that accuracy, integrity and timeliness of coronavirus infection data is important not only for crafting an effective public health response but building the public’s trust.
“The more transparent public health authorities are with the data that’s being used to drive decisions, the better,” said Dr. Amesh Adalja, a researcher focusing on pandemic preparedness and emerging infectious diseases at Johns Hopkins University Center for Health Security.
“We’re in an environment now where there’s a deep mistrust of government and how public health decisions are made. You want the public to have the same data you have so they understand why a restaurant has to remain closed or why other decisions are being made, so that they can have complete trust in what’s driving those decisions.”
Six months of problems
Timely, accurate data will become increasingly important as the pandemic moves into the late fall and winter months, when other traditional respiratory infections typically take off.
Without good data, policy makers, scientists and the public are in the dark about the true state of the pandemic’s spread in Washington, some have said.
But public health officials have not been tasked with tracking testing for a widespread pandemic in the past, and they are creating systems for tracking the virus in the absence of federal data standards, said Jennifer Nuzzo, an epidemiologist and professor at the Johns Hopkins center.
“Having interruptions, glitches and setbacks makes it harder, for sure. But that’s a hallmark of every public health response,” she said. “Recalibrating and figuring out what to do and changing course — I think that’s the sign of a healthy response rather than just staying the course and denying that there are problems.”
The positivity rate, she said, should be used to gauge how well a state is doing with providing testing, rather than determining the prevalence of the disease in a community.
For example, if a state has a high positivity rate, that means officials are probably missing a lot of infections that are out there.
Some of the state’s data problems stem from an electronic disease-surveillance system that wasn’t set up to handle the flood of data that’s been produced by the coronavirus, state health officials have said.
Before the pandemic, the statewide disease-tracking system — called the Washington Disease Reporting System, or WDRS — typically received electronic entries from local health departments and labs when a positive case of measles, E.coli or other public health hazard cropped up, Wiesman has said.
But the system wasn’t set up to take negative results, he said, and WDRS hadn’t experienced a sustained, high volume of daily reports like those created in the past few months.
Technical problems began to surface by late March, a few weeks after the outbreak exploded in Washington and thousands of test results started streaming into the system.
More than 90% were negative, and labs sometimes submitted the same results multiple times per day, requiring health officials to manually weed out duplicates. The state’s daily reporting ground to a halt.
Because a software vendor upgraded the system, Washington didn’t report new case data between March 31 and April 15.
Some of the same problems, largely stemming from negative case data, have persisted since then, ultimately leading to the change in methodology that Wiesman announced this week.
But the flood of data and duplicate negative reports haven’t been the extent of Washington’s reporting troubles. In June and July, the state reduced its death counts after determining that COVID-19 wasn’t a contributing factor in some instances.
This week, officials said they learned 135 tests had been logged as positive for coronavirus between Aug. 7 and 10, when actually they were negative. Those cases had been manually entered into the system as positive.
Officials also released preliminary data this week showing the percentage of people testing positive for COVID-19 gradually increased in Washington as the state tries to address its data issues.
Higher positivity rate
The gap in reporting the state’s positivity rate — or the percentage of total tests that labs report to health officials as positive — comes as it jumped to 6.7%, according to DOH data released Tuesday. That is a 1% increase from the time the state halted reporting on July 31.
The positivity rate nationwide based on a seven-day average is 7.8%, according to most recent figures from Johns Hopkins University & Medicine.
Nuzzo cautioned against comparing positivity rates among states, given health departments’ varied approaches to determining the metric.
Final figures on Washington’s positivity rates are expected within the next week. DOH officials said they are not certain how much the data will shift, although Inslee said that he did not expect a big adjustment when final figures are available.
Seattle Times reporters Mike Reicher and Ryan Blethen contributed to this report.