Colorado has been hit hard by the novel coronavirus spreading across the world, with some of the highest known infection rates in the country.
But state officials now say the numbers understate the reality of the presence of the virus in Colorado. And the most current estimates predict as few as 900 COVID-19 deaths in Colorado if social distancing measures are highly successful, but as many as 33,200 deaths, or one out of every 172 Coloradans, without social isolation.
The number of positive cases has nearly doubled every two days, putting the spread a little below the national pace. The number of deaths has nearly doubled every three days, outpacing the rest of the country.
Colorado’s sparsely populated, but normally-tourist-saturated mountain resort counties have some of the highest rates of reported infections in the nation.
Gunnison, Eagle and Pitkin counties had the third, seventh and 13th highest rates of reported cases in the nation, respectively, according to a compilation of cases published Friday by The New York Times.
Eagle County has the highest rate among counties with more than 50,000 residents outside New York. The three high-country counties are home to world-renowned ski resorts, which annually attract international visitors.
An early outbreak in Pitkin County impacted a group of Australians who have a tradition of traveling together to Aspen each year to ski.
The number of cases-per-person in Denver County puts it in the top 7% of all U.S. counties. As of March 28, the state reported a total of 2,061 positive cases. The state ranks ninth highest for the number of tracked infections per person, according to the compilation by The New York Times as of March 27.
But the number of reported positive cases represents an unknown fraction of the actual cases in Colorado, state health officials acknowledged this week. The real number of cases right now, they said, is thousands more.
During a Friday press conference, when the official statewide count was 1,734, Colorado Gov. Jared Polis said, “We know it’s several times that number in Colorado.”
On top of the undercount due to a lack of complete testing, numbers released by the state have been inconsistent and incorrect at times.
Between March 17 and 18, for example, the number of cumulative positive cases listed for Boulder County went down. Colorado Department of Public Health and Environment officials said the earlier number was wrong, and that others had also been reported incorrectly in the preceding days.
In some parts of the state, the number of deaths reported by CDPHE has lagged what county officials have reported. Earlier this week, CDPHE’s website added a note that statewide totals also counted cases where individuals were not tested, but are presumed to be positive because of close contact with someone who did test positive.
Despite all the shortcomings of the reported cases, the numbers provided by the state match the spread pattern in other areas, as well as known virus spread patterns. The number of positive cases in Colorado since the first case was confirmed on March 5, as well as the number of deaths attributed to the virus since the first death on March 14, have followed an exponential growth curve, which is the typical shape of the beginning of a viral outbreak.
And while the necessarily undercounted number of positive cases grows, experts around the state have been developing a model that helps estimate the real number, as well as analyze the spread of the virus. They are predicting how the virus might continue to spread, as well as evaluating response tactics like social distancing.
Dr. Jonathan Samet, the dean of the University of Colorado School of Public Health, is leading the effort, which included faculty and volunteers from the Colorado School of Public Health and School of Medicine, and the University of Colorado at Boulder.
The group used data from Colorado, as well as models from other researchers around the world to develop a model specific to the state.
Polis first showed some of the results of the model at the Friday press conference, which laid out best-case and worst-case scenarios for the state. The different scenarios represent the outcomes based on the amount of social distancing Coloradans practice.
“This is a construct related to how much we are mixing going from complete mixing (0% reduction), to no mixing at all (100% reduction),” Dr. Samet explained. “We have perhaps reached the 50%-60% level,” given that some people are still working, and trips to essential stores have been reduced, but not eliminated.
David Bortz, a math biologist in the University of Colorado at Boulder’s applied mathematics department who worked on the model, said the high level of uncertainty about how things unfold is key to putting the predictive models in perspective.
“Worldwide, there’s a lot of uncertainty,” Bortz said. “It’s a very different disease than we’ve seen. The range of possibility is so big.”
Polis said the team of physicians and researchers believe the social distancing measures implemented over the past few weeks will slow the spread of the virus, and that their impact will become evident about twelve days after implementation, meaning they could begin to show up in the data in the coming days and weeks.
The first measure, canceling gatherings of 250 people or more, went into effect on March 13 and may become evident in the data beginning as early as a few days ago. The most recent “stay-in-place” order was implemented on Thursday, so may begin to become evident around April 7, Polis said.
“We would hope you start to see something kick in two weeks or three weeks from now," Samet said. "We’re in a critical time period now. My clinical colleagues are starting to see a lot of cases. We’re (still) on that steep uptick right now.”