We’ve discovered some anomalies with the test results we received this week from one of our three providers. I’m going to explain in detail what we’ve discovered and how we discovered it, but the bottom line is this: Dozens of people whose initial tests showed them to be COVID-positive have been retested twice and all but one of those have turned out to be negative. . . .

When we examined the results a few days ago, we suspected something was wrong. Those results didn’t seem right for a number of reasons: Over 90% of the positive infections came from a single test provider; three-quarters of the positive tests were from people who reported no symptoms; the positive results were widely scattered across various groups in our population, with only one potential cluster that seemed more likely to be associated with their proximity to a particular testing location; and over 90% of the reported infections were for people who were fully vaccinated.

We then reviewed the detailed data and noted some very unusual patterns in the results that suggested there was a possible issue with a testing provider rather than a broader campus outbreak. When we consulted with that provider, we learned that they had begun using a different protocol than they had previously used at Rice, resulting in significant differences in how test results are decided. This change in testing protocol had not been disclosed to Rice. We asked that they immediately revert to their prior testing protocol and they have done so.

Then we retested about 50 people who initially tested positive. Each of them was tested two additional times, on two different days, by two different test providers, and all but one came back negative. Based on the anomalies and the two follow-up negative tests from other providers, we concluded that these people who were previously treated as positive were in fact negative, so they were released from isolation. The people whose positive tests were verified remain in isolation.

These testing data anomalies were part of the reason we decided to take most of our classes on-line for the first two weeks, until Sept. 3, as a precautionary measure. We’ll largely leave those plans in place, because many of our students and faculty have already made their own plans in accordance with that schedule. We will use this time to assess any other possible measures that we might put in place. We’re going to make some operational adjustments that will be announced shortly, but right now we anticipate returning to fully in-person classroom instruction in two weeks.

Students who we asked to delay moving on campus until after Sept. 3 are now welcome to move in as soon as they want.

Good catch, but how many other tests elsewhere are wrong?