Members of the General Assembly’s Joint COVID-19 Legislative Workgroup raised questions Wednesday about the Maryland Department of Health’s data reporting methodology, citing inconsistencies with other tracking organizations.
House Health and Government Operations Chairwoman Shane E. Pendergrass (D-Howard) grilled Dr. Jinelene Chan, the acting deputy secretary for public health services, about the numbers the Maryland Department of Health has been reporting, saying that they are in constant conflict with numbers reported by other COVID-19 tracking organizations, like Johns Hopkins.
Chan has stepped in for Fran Philips, who retired from her position mid-health crisis late last month.
“I’ve been struggling with the numbers — the positivity numbers — for a fairly long time, and it’s not clear to me that the [Johns] Hopkins’ numbers show a higher positivity rate than the state numbers that you get that you post,” said Pendergrass, adding that her husband, a mathematician, can’t seem to make the rates match, either.
Maryland remains in phase two of its reopening plan. Chan noted fluctuating case and hospitalization rates in July prompted Gov. Lawrence J. Hogan Jr. (R) to hold off on initiating the next phase of the plan, but explained that the “historically” low positivity rates officials have seen in August thus far are encouraging.
As of Wednesday, the Maryland Department of Health was reporting a positivity rate of 3.61%. Johns Hopkins’ online COVID-19 tracker registered Maryland’s positivity rate at 5.2%, and COVID Act Now, a consortium of health experts, reported the state at 5.3% with an at-risk outbreak level warning.
Senate President Bill Ferguson (D-Baltimore City) said that rather than looking at the Department of Health’s daily dashboard, he has switched to COVID Act Now, noting the positivity rate posted there today is “above the threshold of stability.”
Chan told the workgroup that at state and national levels, data is being defined and reported in multiple ways, leading to inconsistencies and difficulty in “compar[ing] apples to apples.”
“What I would say is that, in Maryland, we have been consistent from the very beginning, and so in terms of the numbers that we’re reporting out, we have not changed our methodology from the beginning of the pandemic to now,” Chan said to Pendergrass.
Pendergrass interrupted Chan’s answer to acknowledge that the state has been consistently inconsistent.
“There is consistency,” the chairwoman asserted. “You consistently don’t add up on a daily basis to the number of tests.”
Chan said that the data used in the Hopkins dashboard comes from the Maryland Department of Health.
During the meeting, Pendergrass said that doing simple math to match the results published by the state doesn’t compute, either. And she’s correct, the state numbers don’t quite fit.
Wednesday’s 541 positive results reported by the Maryland Department of Health added to the 10,838 negative results reported do not match the day’s testing volume, which is 18,015.
Chan said that there are “so many potential nuances” when it comes to dealing with data, and that the Health Department has tried to consistently “pick numbers that we thought could be as consistent as we can make them.”
Chan explained that the Department of Health publishes the number of test results reported back to the agency in a single day, not based on a test-collection date because the published numbers would be constantly shifting.
“I think our feeling is that … it would be, perhaps, more confusing to explain the constant shifting in the daily numbers, and so we selected what we thought would be the most consistent reporting number for the positivity which is [the] number of reported test results in a given day,” Chan told Pendergrass.
The Johns Hopkins Coronavirus Resource Center website’s FAQ page states that the dashboard’s daily positivity rate is found by dividing the number of people who test positive for COVID-19 by the total number of people who have been tested.
Sen. Clarence K. Lam (D-Howard) also suggested that Hopkins, COVID Act Now and other tracking organizations may be trying to “weed out” individuals who have been tested more than once, whereas the Department of Health is counting the overall number of completed tests leading to different results.
He explained to the workgroup that the state’s positivity rate should be looked at in the “spectrum of other metrics,” because it can be raised or lowered based on the number of tests administered. Lam said that the “laser focus” on a positivity rate that constantly fluctuates based on testing capacity could more easily be replaced by ICU capacity because the standards of admitting a COVID-positive patient to the hospital are unchanging.
Lam did say, however, that it’s important to note both the Johns Hopkins’ positivity rate for Maryland, as well as COVID Act Now’s, has consistently hovered around 5% since early July.
“While there is a difference in how these are calculated,” he said, “there’s certainly you know, a consistency amongst these other data sources that show that we are consistently above 5%.”
Chan told the workgroup that she would ensure that lawmakers would get more information regarding the Department of Health’s methodology.