Back

On the Effects of Misclassification in Estimating Efficacy With Application to Recent COVID-19 Vaccine Trials

Buonaccorsi, J. P.

2020-12-07 infectious diseases
10.1101/2020.12.04.20244244 medRxiv
Show abstract

The recent trials for proposed COVID-19 vaccines have garnered a considerable amount of attention and as of this writing extensive vaccination efforts are underway. The first two vaccines approved in the United States are the Moderna and Pfizer vaccines both with estimated efficacy near 95%. One question which has received limited attention, and which we address here, is what affect false positives or false negatives have on the estimated efficacy. Expressions for potential bias due to misclassification of COVID status are developed as are general formulas to adjust for misclassification, allowing for either differential or non-differential misclassification. These results are illustrated with numerical investigations pertinent to the Moderna and Pfizer trials. The general conclusion, fortunately, is that the potential misclassification of COVID status almost always would lead to underestimation of the efficacy and that correcting for false positives or negatives will typically lead to even higher estimated efficacy.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.