DENVER (KDVR) — Outbreak data, it turns out, is not good for much beyond a general picture. Aside from that, it sheds little light on viral spread or the restrictions that go with it.

“I think the utility to the general public of outbreak data is really quite limited and has the potential to create more confusion than it has value,” said Dr. Glen Mays, chair of the Department of Health Systems, Management and Policy at the Colorado School of Public Health “I’m not saying they’re of no value, but I think they’re of primary value to inside baseball types like public health officials.”

The last week’s focus on schools illustrates Dr. Mays’ point.

Schools in spotlight

This week has seen several state-level discussions on when and how K-12 schools can open to in-person learning, including a meeting of superintendents and a press conference from Gov. Jared Polis concerning a plan to reopen schools.

Gov. Jared Polis echoed international and national guidance in a letter to parents when he said schools are safe places.

“What we know ten months into this pandemic is that schools are relatively low-risk
environments and can be a safe place for students and educators when they properly
implement a layered approach to risk reduction,” the letter reads.

This report refers to an earlier release from CDPHE, which says, “the relatively small number and size of these outbreaks indicates that school mitigation measures are limiting the introduction of COVID-19 from non-school community and household settings and preventing widespread transmission in schools.”

In a Dec. 15 press conference, Polis referred obliquely to international and U.S. Centers for Disease Control and Prevention guidance, and said Colorado schools are safe places despite being frequent outbreak locations.

The goal, he said, is to prevent community spread, and the small numbers coming from school outbreaks say district plans are doing just that.

“When you look at the school-based transmission, you see that the safety guidelines that have been in place have largely been very effective, meaning you see three (cases per outbreak), you see two, you see four. You don’t see 28, 86, 115,” he said.

State records back this claim. They also claim other more heavily-restricted locations do the same.

The data

If the public were using outbreak data as an indicator of what should be restricted, little would make sense. Numbers back much of the governor and studies say, but raise questions about why schools would open while locations with lower averages are restricted.

Schools are, by a large margin, the largest single source of outbreaks. Through Dec. 9, no other single location type, including hospitals, has seen as many individual outbreaks. This would seem especially dramatic considering schools have only been operational for a fraction of the pandemic’s timeline.

Polis is right in that school outbreaks have seen fewer infections than other locations. Hospitals and other closely-packed locales produce more.

The governor is also correct in that schools do not produce large scale outbreaks. Locations with massive, dense populations – prisons, colleges, hospitals – have huge infections per outbreak. State prisons top the list with 239 infected per each outbreak.

By contrast, schools have few infections per outbreak – on average, 5.7 infections.

This raises an important point, however: many of the locations with even lower average infections per outbreak are subject to higher levels of restrictions than schools are, notably the bars and restaurants smacked with Level Red restrictions on Nov. 20, along with other heavily restricted location types such as youth sports.

Even if these numbers weren’t what they are, state officials have insisted in the last month that school outbreaks are observed and counted at the school, but not necessarily contracted there.

This rationale, however, would apply equally to any other location who would claim they don’t drive viral spread.

According to Dr. Mays, this underscores that outbreak data alone simply isn’t a reliable metric for deciding what stays open and what shuts down. It requires more context, additional studies and expertise to interpret.

“Outbreak data are noisy, murky data, and the quality of that data depend on multiple factors,” he said. “I’m not part of the state government, but from my vantage point, our public health officials and policy makers are needing to triangulate across multiple sources of evidence.”

What those additional sources contain is unknown apart from CDC and World Health Organization guidance.