This article was originally posted by the Cornell Sun on Oct. 20.
ITHACA, N.Y. -- A little over a month into the fall semester, Cornell and the surrounding Ithaca community has kept COVID-19 cases low, with test positivity rates consistently remaining at a fraction of 1 percent.
Critical to Cornell’s ability to reopen safely is its surveillance testing program, which subjects everyone on campus to regular COVID-19 tests. What does the data from this testing say about where Cornell stands?
Faculty and Staff
Throughout the semester students have made up the vast majority of positive COVID-19 tests, totaling 129 of the 139 cases on campus. However, the week of Oct. 11 the faculty and staff saw a spike in cases — an uptick that mirrors that of nearby counties.
The four cases in a span of two days among faculty and staff members prompted an?email?from Provost Michael Kotlikoff and Vice President Mary Opperman, urging them to avoid any unnecessary travel outside of the Ithaca area.
“Staff and faculty are strongly encouraged to avoid unnecessary travel outside the greater Ithaca region, as any such trips present additional opportunities for COVID-19 to enter our community,” the two wrote. “If you must travel outside the area, avoid crowds and public transportation.”
Kotlikoff and Operman wrote that the four cases were not connected to any form of an on-campus cluster, encouraging all staff members to be diligent in adhering to public health guidelines like mask wearing and complying with surveillance testing schedules.
“Earlier in the semester, following a few positive clusters, our students rose to the challenge of strictly adhering to?public health precautions?as they protected themselves and others,” Kotlikoff and Opperman wrote. “Now we must all do our part and abide by all New York state and Cornell public health guidance.”
Tompkins County has kept COVID-19 cases low since the start of the pandemic, peaking at 48 active cases before students arrived for the fall semester and only seeing three deaths —?two were transfers from New York City.
The county, with a?population?of around 102,180, tested anywhere from 500 to 2,000 individuals a day in the middle to end of August. Testing ramped up significantly on Sept. 2 when Cornell began its surveillance testing program. After Cornell began surveillance testing, the county conducted 2,500 to 6,000 tests daily. Cornell’s testing program is designed to process up to?50,000 tests a week?— most recently running 34,983 tests on the week of Oct. 4.
Tompkins County saw its largest spike in cases following the arrival of students for the fall semester — on the week of Sept. 6. This spike followed the identification of the first cluster of COVID-19 cases Aug. 29, and the uptick to the?yellow alert level?five days later.
Since Sept. 29, cases in Tompkins County have moderately risen despite Cornell maintaining a low positivity rate. The county saw 54 new cases on the week of Oct. 11, and?three possible public exposures. This follows the?trend of other counties in the Southern Tier, including Broome, Chemung and Steuben counties, which have seen upticks in COVID-19 cases. Early October also saw the first instance of multiple hospitalizations in Tompkins County since the end of July and the?first death of a local resident.
Cornell’s testing data was obtained from its?COVID-19 dashboard, and Tompkins County’s data was pulled from?its health department’s website. Although Tompkins County’s data includes tests and positive cases from Cornell, there are several days where Cornell recorded a higher number of tests than Tompkins County. Tompkins County Health Department and Cornell did not explicitly explain this discrepancy at the time of publication and referred to their respective dashboards instead.
The low number of new daily cases — despite a high frequency of testing — indicates that the University, and the county at large, have effectively adhered to social distancing and largely mitigated the spread of COVID-19.
The overall trend is also indicative that one of the largest fears of reopening —?uncontrolled community spread?— has been abated. Community spread is when the virus is so common in a community that it becomes difficult to effectively trace where individuals were infected, making it difficult to control the spread of the virus in the long term.
New York State
Despite its initial status as the global epicenter of the pandemic, New York State has been lauded for its response to COVID-19 by public health officials like?Dr. Anthony Fauci M.D. ’66, as the state has consistently reported low positivity rates.
Cornell saw positivity rates peak several days after its initial move-in date for students living on campus, recording positivity rates of 3.57 percent on Aug. 30 and 3.44 percent on Sept. 2. The Aug. 30 spike followed the initial identification of the first cluster associated with Cornell student athletes by two days. Sept. 2 was when surveillance testing began on campus and was the first instance in which some individuals had been tested since their initial arrival to Ithaca weeks earlier.
Since the first cluster of cases on campus, positivity rates on campus have remained low, consistently plateauing at 0.02 percent, as Cornell fluctuates between zero and two new COVID-19 cases daily.
Positivity rate is the percentage of tests that come back positive —?a figure that is often used to assess the levels of virus in a population. The World Health Organization?recommended?in May that positivity rates remain below 5 percent for at least two weeks before places could consider reopening. Several days later, the Centers for Disease Control and Prevention set a lower bar for reopening — stating that its?goal?was to achieve no more than a 10 percent positivity rate among the entire population, including those that are asymptomatic and pre-symptomatic.
Cornell’s positivity rate would likely be higher than other areas because it is using asymptomatic screening — testing everyone regardless of symptoms —?whereas most areas are only testing individuals for cause. In the latter scenario, it is possible that individuals infected with COVID-19, but do not show symptoms, will not be tested and will never be accounted for in the positivity rate.
Cornell maintaining positivity rates below New York State’s is also indicative of the efficacy of the University’s reopening procedures. However New York State’s size makes it difficult to generalize its success through a statewide positivity rate. The course of the pandemic varies between New York City and the more rural areas in upstate New York.
The data from Cornell is also more susceptible to sudden spikes or dips because its sample size is much smaller than the entirety of New York State. The difference in population size explains the two large spikes that Cornell saw in late August and early September.
Duke University is another school that has managed to successfully reopen for the fall semester, and the North Carolina-based school is following a very?similar testing protocol as Cornell. Students are tested weekly, with the potential to be tested twice a week, and faculty members who report to campus frequently may also be tested weekly. Like Cornell, Duke is also using the same anterior nares tests to conduct surveillance testing.
Both schools saw an initial spike in cases, before cases decreased and subsequently plateaued.
Duke’s success in managing COVID-19 on its campus represents a stark contrast to its neighbor, the University of Northern Carolina at Chapel Hill, which was forced to quickly?reverse course?after reopening because of an uncontrolled campus outbreak.
The success of the two schools speaks to the efficacy of their testing schedules and methodologies. Both schools briefly experienced an initial spike as the testing program identified cases upon entry and initial spread, but in the long term, Cornell and Duke have largely contained the spread of COVID-19.
In addition to comprehensive testing, both schools have also employed strategies to de-densify campus and slow the spread of COVID-19. At Cornell, two-thirds of classes are?taught virtually, while at Duke,?most upperclassmen were barred from living on campus at the last minute?and encouraged to study at home.
Duke University’s data was taken from its?COVID-19 dashboard, which reports data weekly. The positivity rate was calculated by dividing the number of total tests by the number of positives over the course of the week.
Yale was the only other Ivy League university to at least partially reopen in September, with the others opting for fully remote semesters or later starts.
However, unlike Cornell, Yale did?restrict which students could return?to campus, welcoming back first-years, juniors and seniors in the fall and sophomores, juniors and seniors in the spring. Yale is using surveillance testing to monitor cases of COVID-19 on its campus, testing students biweekly in a similar fashion to Cornell.
Yale’s?COVID-19 dashboard?reports the last seven days of data, so comparisons were limited only to the past week. Both schools have kept cases of COVID-19 low, with Yale’s positivity rate peaking at 0.49 percent during the last week.
While Cornell has maintained a low positivity rate since early October, Yale has seen more dramatic fluctuation, likely owing to the much lower number of students on Yale’s campus. Since the beginning of October, Yale has not tested more than 2,000 individuals in a single day, meaning a single positive case will have a great impact on the school’s positivity rate than it would at Cornell.