Combined Symptom Screening and At-Home Tests for COVID-19

At-home rapid antigen tests are widely used for SARS-CoV-2 detection, but their limited sensitivity reduces reliability. Prior research has shown that these tests can miss infections, especially in asymptomatic or early-stage cases. No prior work had resolved how combining these tests with symptom screening might improve diagnostic accuracy. This gap motivated the current study to evaluate the value of integrating symptom data with at-home testing. Established knowledge includes the limitations of antigen tests and the role of symptoms in identifying infections. This paper's contribution is to quantify how symptom screening influences test accuracy. The study's focus is on improving community-level detection of SARS-CoV-2. It builds on existing research on test limitations but introduces a novel approach to enhance detection rates.

The goal was to assess how combining at-home antigen tests with computerized symptom screening affects the accuracy of diagnosing SARS-CoV-2. The specific problem addressed is the low sensitivity of rapid antigen tests alone. The motivation comes from the need for better community-level detection strategies. The study aimed to compare four diagnostic methods: symptom screening, antigen tests, combined methods, and vaccination status. Researchers wanted to determine if integrating symptom data could improve diagnostic accuracy. The study also sought to evaluate if a second antigen test improved results. The design allowed for comparison of different screening approaches. The ultimate aim was to identify the most effective method for detecting active infections.

The study used data from two phases with different SARS-CoV-2 variant periods. Phase 1 data were used to develop a symptom screening tool using logistic regression with interaction terms. Phase 2 data validated the accuracy of four diagnostic methods. The McFadden pseudo-R2 measured how much variation in RT-PCR results each method explained. Researchers evaluated the first and second at-home tests separately. They also tested the combination of symptom screening with antigen tests. A fourth method combined symptom screening with vaccination status. The study compared the pseudo-R2 scores of each approach. The data sources included primary data collected from 461 and 374 participants in each phase.

The first at-home test had low sensitivity (0.587) and high specificity (0.989). The second at-home test did not improve sensitivity. Computerized symptom screening had a pseudo-R2 of 0.158. The first at-home test had a pseudo-R2 of 0.274. Adding a second test lowered the pseudo-R2 to 0.140. Symptom screening improved the first test's sensitivity by 0.131 points. Symptom screening and vaccination status had the highest pseudo-R2 of 0.476. This combination explained 47.6% of RT-PCR variation. These results suggest that symptom screening can enhance or replace antigen tests in some cases.

The authors found that combining symptom screening with antigen tests improved diagnostic accuracy. Symptom screening alone had moderate accuracy but was less effective than combined methods. Adding a second antigen test did not enhance sensitivity. The most accurate method combined symptom screening with vaccination status. This approach explained nearly half of RT-PCR variation. The study suggests that symptom screening can either improve or replace antigen tests for symptomatic individuals. These findings are specific to the study's data and cannot be generalized beyond the tested population. The authors propose that integrating symptom data into diagnostic strategies may enhance community-level detection.