Making sense of 'fresh' and 'late' COVID-19 cases in DOH testing results

Warren de Guzman and Edson Guido, ABS-CBN Data Analytics Team

Posted at Jun 11 2020 10:55 PM | Updated as of Jun 11 2020 11:31 PM

Health workers conduct their routine swab testing severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) on their fellow hospital employees at the Sta. Ana Hospital in Manila on April 14, 2020. George Calvelo, ABS-CBN News

MANILA -- The need to put the daily COVID-19 test result bulletins of the Department of Health (DOH) in the proper context is more dire than ever with the full impact of the shift from Modified Enhanced Community Quarantine (MECQ) to General Community Quarantine (GCQ) in the National Capital Region expected to be shown in the numbers out next week. To accomplish this, the DOH has introduced ‘fresh’ and ‘late’ test results. The ABS-CBN Data Analytics Team has broken down what this means.

Consider the DOH Bulletin format #88 from June 10.

 

There is geographical information, which is helpful to tell us where exactly the cases reported are located. There is information on recoveries and deaths. Then there is information on confirmed cases, reported as of June 10. There are a total of 740 confirmed cases reported. That is broken down into fresh cases, labelled as test results released to patient within 3 days, and late cases, labelled as test results released to patient after 4 days or more.

What is the significance of the 4-day threshold to differentiate fresh and late cases? The DOH is talking about the time between the completion and confirmation of positive results by laboratories and the confirmation or validation of those results by the DOH. RT-PCR or the Reverse Transcription Polymerase Chain Reaction COVID-19 test requires lab work to fully process samples collected from suspected COVID-19 patients.

As of this writing, the DOH has 56 accredited labs for COVID-19 testing. The test itself can take hours, even days, depending on the lab and other factors like transportation time for samples. However, that is excluded from the Fresh versus Late debate, which again only takes into consideration the time elapsed between a positive test result, and the confirmation or validation of that result by the DOH.

This chart breaks down the number of fresh and late cases reported each day. Red shows fresh cases, the lighter portion of the bar graph shows late cases. Ideally we want to see more fresh cases reported compared to late, as this would mean testing is being done rapidly, and results are being validated in a timely manner. This also means patients spend less time wondering if they have the killer virus or not.

The DOH, however, has started providing more specific data on exactly when COVID-19 test results are released. This chart shows us a granular breakdown of the days between the release of a lab test result to the patient and the actual report of DOH. The darkest red portion of the bar shows confirmed cases reported within a day of release of test results. The lightest portion in yellow shows results reported 11 days or more after their release from the labs.

On May 29, 30, 31 and June 1, a huge number of late results were reported. This reflects DOH efforts to clean up backlogs which had been piling up. But there was also a clear decline in fresh cases reported on those same days. As we move into June, we see a lot more red, reflecting a pick up in the reporting of fresh cases. This means the testing operation is moving at an optimum pace with little lag, and patients are hearing about their test results almost real time.

Throughout the chart, however, we have a lot of gray patches. These are test results with no data on when they were released from the lab. The DOH reports these as late, but without the proper data we cannot say for sure.

Those in orange are results released to patients within four to seven days. Any delay in validation will result in cases possibly being tagged as late. In the last three days (June 8-10), orange and gray bars accounted for most of the late cases.

The ideal is to have more red, and less of everything else. This would indicate more fresh cases are being reported, and less backlogs and late reports are coming out. The DOH is getting there as shown by the chart. But anything less would mean the data reported is not real time, and surges or dips in confirmed COVID-19 cases may be artificial, caused by delayed reporting and/or validation.

Card

Consider this last chart. This is a line graph showing even more granular data from COVID-19 patients. The red line shows us reported cases by DOH. The pink line shows us when patients reported feeling the onset of symptoms. The orange line shows us positive results released by labs. The yellow line shows us the number of specimens received by labs. All are 7-day moving averages.

The data released by the DOH is incomplete and imperfect, meaning there are a lot of gaps and not all test results have data input for three of the variables measured (excluding date of reporting). Until we get more complete data from the field, we should take these charts with a grain of salt. That said, these can still be informative.

The only chart that is angling toward flattening is the onset of symptoms, shown by the pink line. The medical definition of the onset of symptoms is the first appearance of symptoms, as recalled by the patient or by clinical examination by a doctor. The data, however, is problematic because case counts reported in recent days using this chart do not represent the final number after the inclusion of incoming reports. In fact, 14,097 of the 23,732 confirmed cases don’t have data on onset of symptoms.

Let’s consider the other three lines. All are rising, but they are not moving in unison. These lines tell us there is a disconnect between reported results, released results, and samples collected. Only the date of reporting has complete data but again, that trend is not real-time given the lag.

(Note: 3,132 cases don’t have data on when specimen was received by confirming laboratory while 5,205 cases don’t have data on date of release of result)

The DOH is trying to provide clear and specific information to decision-makers and the public. Its fresh and late cases have already helped us determine there is a lag between release and confirmation of test results. But the other data points show us there is a lag in testing and reporting, as well as holes in data collection. Things need to improve so the right decisions can be made in the fight against COVID-19.