Over the past few weeks, we have been sending samples of our wastewater to a lab in Massachusetts for testing. Each time, the level of Covid-19 virus has been “Not Detected.” (You can read the details of the process here.) Based on that result, can we feel assured that there is no Covid-19 on our campuses?
It’s not that simple.
Why we’re different. Our expectations of the sewage-testing approach appear to be different that most (or all) other users. We have been using Biobot, which is the most widely-used testing company in this business. (At the end of June, we switched to a different testing company.) Most of Biobot’s customers are municipalities or counties, handling sewage for tens of thousands of customers. For them, some level of Coivd-19 infection is probably a given. What they want to learn from the tests is whether the number of cases is growing or falling—in particular, they want advance warning of a major outbreak.
We are unusual in having a waste treatment facility that handles fewer than 1,000 residents. The question we are most anxious to answer is not “Is the case count growing or shrinking?” but rather “Do we have any cases at all?” In other words, we would ideally want the ability to detect a single case. That is not something a county or a city would require.
Can this approach detect just one case? The question remains: would Biobot, or our new testing lab, be able to detect an isolated case? What if it took multiple simultaneous cases of Covid-19 before the virus reached a level that lab tests could detect? Would it still be worthwhile having this analysis done? I reached out to Biobot for an answer to this question.
Eventually, an analyst from Biobot got back to me. She wrote that they are using a testing method with a “current assay detection limit of 3,560 copies/L.” That means if there are fewer than 3,560 copies of the virus per liter of sample, the virus will not be detected.
What does that mean for us? Well, the sampling machine here at Kendal collects samples at intervals throughout the day, and these are mixed together. A sample of 50ml (a 20th of a liter, or a little less than 1/4 cup) of the mixture is send to Biobot for analysis. If that sample contains at least 178 copies (1/20 of 3,560) of the virus—an amazingly small amount of material—Biobot’s assay would detect it. (An infectious person would “shed” many thousands, or perhaps millions, of copies of the virus.)
More than one case is required. Unfortunately, according to the Biobot who analyst wrote to me, a single, isolated case of Covid-19 would be expected to produce slightly fewer copies than their test could detect. They can detect one case in a population of about 830. However, our population using the sewage plant is about 900. The analyst concludes, “two cases in 900 could be detected through wastewater.” The situation with our new testing lab is presumably similar.
What this means is that a single case could go undetected by this process, but two or more overlapping cases would be caught. That’s not ideal, perhaps, but it may be all we need. It is still far better than any other option we currently have, short of repeatedly testing every resident in each community. If there is a single case, and no other resident is infected, the case might be missed by the wastewater analysis. If that person is asymptomatic, we might never know that the infection had occurred. But if the infection does not spread, the failure to detect it is not important. If it does spread, the wastewater test will detect it as soon as the second person gets it concurrently.
In any case, it is not clear what action Kendal/Crosslands would take if a wastewater test were sensitive enough to indicate that one of our campuses had a single case of Covid-19. But in the case of an outbreak, it would be detected, and our administrators would surely take action–perhaps the resumption of the kind of isolation procedures we’ve had in place during the past few months.
In my opinion, the wastewater analysis is a powerful and valuable approach to keeping us safe from Covid-19 and well worth the modest amount of time, effort, and funding it requires.