Georgia COVID-19 dashboard updated after five-day absence

In a news release Monday evening, DPH sought to reassure the public its data is accurate, though its description of the problem differed from what DPH said occurred last week.

“Issues with electronic laboratory reporting (ELR) that were identified last week have been thoroughly investigated and the problem fixed,” the agency said in a statement. “It is important to note that data for case counts, hospitalizations and deaths come from sources other than ELR, and as such were not impacted by the problem.”

DPH’s electronic lab reporting system is how labs, hospitals and other providers submit electronic testing data.

The Atlanta Journal-Constitution requested an interview with Dr. Kathleen Toomey, the state’s commissioner of public health, but the agency did not make her available on Monday.

Data published Monday by DPH showed the seven-day rolling average of new confirmed and probable infections declined Sunday and Monday, the first consecutive days that has happened since Georgia’s fifth surge started after Thanksgiving.

Still, case rates remain extraordinarily high, more than double the peak of last summer’s delta surge. On Monday, the seven-day rolling average of new and probable infections was 20,409.

Dr. Harry J. Heiman, a clinical associate professor at the Georgia State University School of Public Health, said it is “hard to know what to conclude given the data interruptions and numerous factors that threaten to worsen the current surge.”

“To be clear, the persistent high case numbers together with high positivity rates demonstrate that we are only capturing a portion of the positive cases which have likely not peaked,” he said.

The state’s overall case counts are likely an underestimation of spread. High test positivity rates mean the state isn’t testing enough to capture all the infections, and positive at-home rapid tests are likely underreported.

Metro Atlanta counties, including Cobb, Cherokee, Clayton, DeKalb, Fayette, Fulton and Gwinnett all reported in recent days either a decline or leveling off of their rolling average of new infections.

Hospitalizations, however, continue to climb.

On Monday at 4:30 p.m., Georgia reported 4,992 people were hospitalized statewide for COVID-19, most since mid-September.

About 85% of inpatient hospital beds statewide are in use, including about 87% of intensive care beds.

While omicron spreads much faster than prior strains of the virus, initial research suggests a smaller percentage of infections result in hospitalization or death.

But that still results in a bad math equation for Georgia’s hospitals which could be swamped by a large number of patients needing hospitalization.

A new DPH explanation

Since the beginning of the pandemic, numerous independent experts have questioned Georgia’s accounting of coronavirus cases, tests, deaths and vaccinations. The flaws stem from years of the state underfunding public health, critics say, and have been exposed by a rapid spread of a deadly virus and an onslaught of data.

The agency typically updates its COVID-19 dashboard on its website each weekday, with Monday reports including data from over the weekend. But DPH was unable to publish its Jan. 3 report, which included the holiday weekend, citing an “overwhelming” volume of data.

The next day, Jan. 4, DPH reported 2.6 million PCR tests since its update four days earlier on New Year’s Eve, a figure that appeared to be an error. But the AJC reported last week that DPH acknowledged the testing figures reported Jan. 4 actually revealed a data processing problem that stretched back nearly two years.

A DPH press release on Jan. 5 labeled the testing data issue a “coding error.”

In emails last week, DPH spokeswoman Nancy Nydam said a processing issue caused DPH to under-report daily COVID-19 tests and positive results in at least one prominent location on its website since the beginning of the pandemic.

Nydam told the AJC last week the problem did not reflect an overall undercount of tests administered or infections and that vital data in other locations on its website were accurate.

Each day, providers and labs report testing data to DPH and that data goes into its charts and cumulative totals. But the tally of tests reported each day involves a more complicated computer query, Nydam said last week.

Last week, DPH blamed the issue on a data hiccup involving time stamps on incoming tests, which resulted in a portion of tests falling through the cracks in its column of daily report tabulations.

The issue came to light on New Year’s Eve when DPH’s systems were sluggish. To correct the issue, Nydam said, DPH ran the numbers a different way and through that process the 2.6 million tests appeared on Jan. 4 as the total number of tests reported for the previous four days.

On Monday, DPH offered another explanation.

In its news release, DPH said tens of thousands of test results arrive electronically each day with a classification of positive, negative or inconclusive. Some tests do not include complete information from the provider, and if so, those results are not assigned a classification and must be manually reviewed and classified.

These manually reviewed tests total 2.6 million since the start of the pandemic, the release said.

After encountering a processing issue on New Year’s Eve, DPH said it changed its coding process, which caused all 2.6 million tests that were previously accounted for manually to be reclassified with a Jan. 4 date.

DPH delayed its report Wednesday and scuttled its updates on Thursday and Friday because of electronic reporting issues.

“Once the coding issue was identified as the root cause of the problem, it was fixed, and extensive testing and verification of the system was completed over the past four days,” the release said.

Nydam did not immediately respond to follow up questions about the discrepancies in the agency’s explanations last week and those offered on Monday.