On Wednesday, May 20, when the Colorado Department of Public Health and Environment published weekly coronavirus outbreak figures from around the state, one entry stood out: Van Cise-Simonet Detention Center, Denver’s largest jail, reported 398 new confirmed or probable COVID-19 cases, bringing the facility’s total to 581, the largest outbreak in the state.
The spike led to headlines in local newspapers and TV newscasts. The next morning, state and city officials sprang into action, scrutinizing the data, theorizing about what went wrong and spitballing solutions.
Their discussion about the surge in cases wasn't, however, focused on slowing the spread of the pandemic novel coronavirus in the jail.
Officials at the state and city health departments, Denver Health, the Denver County Sheriff’s Department and the Denver mayor’s office — including some agency heads, the sheriff and the jail’s chief officer, who has since been appointed sheriff — were all included in a hasty effort to lower the number of cases just publicly released.
“There is a lot of confusion in the media today regarding our numbers and we are hopeful that this can be clarified,” the jail’s chief officer, Elias Diggins, who was promoted from a top deputy to Denver County sheriff earlier this week, wrote to several other top officials, despite the media reports accurately conveying the numbers reported by Denver Health, the jail’s health care provider.
“What can be done to correct the misperception, based on the CDPHE website, that the Downtown Detention Center is ‘the biggest outbreak in Colorado?'” Dr. Bill Burman, the Denver Department of Public Health and Environment director, wrote in an email to top officials at the state health department.
In the following days, they settled on a new method to count and report their cases, which cut the number by more than half. The next Wednesday, the state health department’s outbreak file included only cases where the transmission of the virus was believed to have happened within the jail — excluding from the count inmates with confirmed or probable COVID-19 infections at the time of their intake.
The new lowered numbers came without a public explanation, and it’s not currently possible to find online the total number of inmates in the jail who are confirmed or probable cases.
Officials involved in the change have since said it was driven by epidemiology, but a spokesperson for the Colorado Department of Corrections, which runs the prison with the largest current outbreak in the state, said they were never made aware of the counting method changes the state health department worked with the Denver jail to achieve. The Corrections Department said it had not changed its reporting to mimic Denver's.
Rebecca Wallace, an attorney with the American Civil Liberties Union of Colorado, called the change in the jail’s counting and reporting method “completely misleading.”
She said that without indicating that the numbers exclude some cases, contradicting the same outbreak file's own data dictionary, the full picture is being obscured for the public.
“If, in reality, they’re using this data to limit the bad PR for the jails, that’s a real dereliction of duty,” Wallace said. “It unduly diminishes the public concern about what’s happening behind bars.”
Metadata embedded in one particular state record shows that after the jail changed how it was counting cases — and after The Gazette filed public records requests for documents related to the change — state health department staff changed their internal documents to allow the newly adopted counting method.
The agencies have provided incomplete and sometimes conflicting explanations about what numbers would have been reported using the original counting method and who exactly called for the change. They’ve also refused to allow some top officials involved in the decision-making process to be interviewed about it.
Using the Colorado Open Records Act, The Gazette obtained dozens of emails, sent by 12 staff members representing the Denver Sheriff’s Department, Denver Health, CDPHE and the city health department, and received by 26 other staff members from those same agencies and the Denver mayor’s office. The emails offer a partial picture of what led to the lower number of publicly reported cases at the Denver jail.
Some of the officials and agency employees agreed to discuss the effort on the record. Some would not. Others discussed the matter only on the condition that they not be identified or quoted.
Shortly after 9:30 a.m. on May 21, a data analyst at Denver Health emailed two Denver Health doctors and Laura Podewils, a Denver Health epidemiologist who had asked the analyst to pull up the numbers. The analyst wrote that the previous day’s numbers might have included some inmates who were released.
Podewils then emailed two other epidemiologists, Sarah Janelle from CDPHE and Abby Davidson from DDPHE. In the emails, the three epidemiologists agreed that the combination of widespread testing in the jail and the requirement to report “probable” cases, in addition to lab-confirmed cases, led to a spike.
Davidson emailed a detailed explanation to Diggins, copying Bob McDonald, DDPHE’s chief medical officer, and Danica Lee, DDPHE’s director.
Diggins responded, adding then-Sheriff Fran Gomez to the email conversation, expressing his displeasure with a report about the numbers on 9News, Denver’s NBC affiliate.
Though he called for the “confusions in the media” to “be clarified” in his email, others included in the emails wrote that the numbers were accurate and “how outbreak numbers are typically reported.”
Podewils sent a link to a CDPHE document that explained how outbreak numbers were expected to be counted. The document supported the claim that the previous day’s numbers were correctly reported, according to the state’s protocols.
Others joined the email conversation, including McDonald and Lee.
“We also would like to offer a few talking points for consideration…” Lee wrote to Diggins and others, including Gomez, before listing some positive aspects to focus on: The health department’s collaboration with the jail, control measures, testing of asymptomatic cases, adherence to U.S. Centers for Disease Control and Prevention guidance and an effort to issue isolation orders for inmates infected with COVID-19 when discharged from the jail.
Diggins, along with spokespeople for the Denver Sheriff’s Department and DDPHE, also offered ideas for shaping the response to the spike in numbers.
Later that day, Burman wrote to the CDPHE director, Dr. Rachel Herlihy, and CDPHE chief medical officer, Dr. Eric France, asking them, “what can be done to fix this?”
That evening, around 9:30 p.m., Sarah Janelle, a CDPHE epidemiologist, wrote to Podewils, proposing that they could count and report only cases originating inside the jail, when someone with the virus in the jail transmitted it to someone else in the jail, and that they exclude from their counts inmates who were positive or probable coronavirus cases when they were booked into the jail: “We would like to try to exclude these people from the (Van Cise-Simonet Detention Center and Denver County Jail) line lists …”
Janelle’s email was the first time in the dozens of emails obtained by The Gazette that the idea that excluding the cases on booking, which ended up being implemented, was brought up. And the idea that inmates booked into the jail with COVID-19 could be excluded from the counts was different from the focus of others in the email exchange on the effect of including “probable” cases in the counts.
The next day, May 22, the emails continued and others welcomed Janelle’s idea.
“CDPHE may be changing how they count cases for the jail,” Lee wrote to Diggins.
May 27, the day the new lowered numbers were reported to the state and to the public, Diggins sent an email to several people at the Denver Department of Public Health and Environment, as well as an official in the Denver mayor’s office, asking them to join him on a conference call to discuss the matter.
Later that day, when the state health department published its weekly coronavirus outbreak figures from around the state, Van Cise-Simonet Detention Center reported a drop of 316 cases from the previous week.
Until then, no correctional facility had reported a lower number of cases than previous weeks. The numbers had only increased. Every other correctional facility has only counted increases in positive cases. In some facilities, “probable” cases have later been removed from the counts.
In response to questions about the decrease, a CDPHE spokesperson wrote in an emailed statement: “After discussions with DDPHE, we made a change to how we report outbreak cases at the Van Cise-Simonet Detention Center to more accurately depict the facility’s outbreak. … Previously our facility outbreak data included everyone who tested positive for COVID-19 at the detention center, regardless of when they were tested. Going forward, the facility outbreak report will not include people who tested positive when they were booked because that is not representative of a facility outbreak. … Our outbreak data is meant to detect and measure spread inside a facility that has an outbreak. These updated numbers reflect COVID-19 transmission within the facility.”
At the time, however, a spokesperson for the Colorado Department of Corrections said the agency hadn’t been informed about a change in counting and reporting methods.
With the new lowered number reported from the Denver jail, Sterling Correctional Facility, the state’s largest prison, had the largest reported outbreak numbers, but the DOC spokesperson said the facility was not excluding inmates from its counts based on whether they were being admitted at the time or not.
CDPHE’s staff was not able to say how many coronavirus cases the Denver jail actually had after the counting change.
It took the Denver health department nearly a month to provide a complete breakdown of the cases that explained the May 27 decrease and make clear what the number would have been using the original counting method. Where the city, county and state coronavirus cases information can be found online, the breakdown of positive and probable coronavirus cases included and excluded in the jail’s count is not available.
In one of the May 21 emails from Podewils, she referenced an internal document kept by CDPHE which details how outbreak cases were to be reported to the state. At the time she was pointing out that the large spike in cases was reported as the state protocol document specified.
But by the time the emails were provided to The Gazette, the document specifically included language excluding inmates who were booked with COVID-19. A Colorado Open Records Act request for the document’s metadata and previous versions, show that the language allowing the exclusion of inmates on booking was added to the document by Rachel Jervis, an enteric disease specialist at CDPHE, on May 30, after the jail already excluded them from the counts, and after The Gazette began asking questions about the jail’s counting method.
Dr. Lauren Brinkley-Rubinstein, a professor at the University of North Carolina and member of the university’s Center for Health Equity Research, whose research focuses on incarceration and health outcomes, said testing newly admitted inmates for COVID-19 can be a valuable tool for the jail and for the community’s public health efforts.
“For the facility to cut out that measurement from what they report doesn’t really make sense,” she said. “These are some of the most important numbers.”
Brinkley-Rubinstein, who also co-founded and serves as a co-lead investigator for covidprisonproject.com, a website that tracks coronavirus in incarceration settings, said reporting both the number of cases in inmates at intake, as well as cases where the transmission took place inside the jail, would increase transparency.
“When you’re really transparent around the data,” Brinkley-Rubinstein said, “that knowledge allows the resources to be focused in the places it needs to be: mitigation of more spread and devoting resources efficiently.”
Wallace, the ACLU attorney, said the change in the counting and reporting of cases could end up causing long-lasting problems.
She said her organization and other organizations that advocate on behalf of the rights of inmates rely on the data to be accurate. Earlier in the pandemic, when the ACLU initiated legal action to help inmates at Weld County jail, that was done based on the numbers being reported, she said, and if the data is being changed, legal and policy action can’t be properly weighed.
And if the total number of cases remains obscured, with artificially diminished figures in public reports and reports relied upon by policymakers, an outbreak could fester in the jail and cause an ongoing problem.
Sheriff Diggins’ spokeswoman initially said there was no change in the way the data is reported and said Diggins wasn’t involved in the issue.
“Chief Diggins is not involved with the reporting of COVID-19 data. I confirmed that the data reporting from Denver Sheriff Health Services (Denver Health) has not changed over the time period you pointed out. Denver Sheriff Health Services provides the COVID-19 data to Denver Public Health. Neither the Denver Sheriff Department or Denver Sheriff Health Services (Denver Health) report COVID-19 data directly to the Colorado Department of Public Health and Environment; therefore, Chief Diggins cannot speak to data that is reported to the state,” Daria Serna, the spokeswoman for the jail wrote.
But after Diggins’ emails about the matter, which had been obtained using CORA requests from DDPHE, were shared with Serna, she provided a statement that refers to approving expanded COVID-19 testing, but which doesn’t address the change in counting methods and pegs the responsibility to Denver Health.
“Although Chief Elias Diggins is not directly involved with reporting the Denver Sheriff Department COVID-19 cases to Denver Public Health or the Colorado Department of Public Health and Environment, he oversees operations for the department and was involved in the decision to approve the expansion of the COVID-19 testing program in Denver’s jails. Denver Health, the contracted medical provider for the Denver jails, is providing testing and reporting information to Denver Public Health,” Serna wrote in an email.