having COVID-19 antibodies from 4.4 % to 11.9 % ( p < 0.001 ). Those individuals who reported that they had previously tested positive for COVID-19 with a PCR test were 18 times more likely to have a positive antibody test than those not ( 79.2 % vs . 4.2 %, p < 0.001 ).
Individuals living in Kanawha County had the highest rate of COVID-19 antibodies compared to Putnam and Cabell counties ( 9.25 %, 4.80 %, and 3.08 % respectively , p = 0.02 ). Healthcare workers overall in the study population had almost twice the positive antibody rate compared to CEW ( HCW 10.5 % vs . CEW 5.8 %, p = 0.004 ). When comparing at the county level , CEW were similar ( Kanawha 6.5 %, Putnam 5.3 %, Cabell 4.2 %, p = 0.71 ). When comparing HCW at the county level , positive antibody rates trended toward significance but were statistically similar as well ( Kanawha 12.1 %, Putnam 3.5 %, Cabell 0 %, p = 0.06 ).
Of the 84 positive antibody participants , only 42 ( 50 %) of participants had a previous positive COVID-19 PCR test . The other half of the patients either had not been previously tested for COVID-19 or had been tested with PCR testing and found to be negative ( 21 patients or 25 %, and 21 patients or 25 % respectively ). Of the 221 ( 20.5 %) participants that thought they had contracted COVID-19 at some point before enrolling in the study , only 57 ( 25.8 %) were found to have reactive antibodies ( Figure 3A ). Of the essential workers known to have a prior COVID-19 infection diagnosed with a positive PCR test ( 50 patients ), eight had non-reactive antibodies ( Figure 3B ).
DISCUSSION
In the fall and early winter of 2020 , the US was experiencing its first wave of the COVID-19 pandemic . At the time , WV had experienced 80,000 positive cases , 2,000 deaths , and a 4.9 % cumulative positive rate . Systems were strained with hospitalized patients . 2 Since there were no vaccines for protection at the time of the study , personal protective equipment ( PPE ), isolation precautions , quarantining , and good hygiene were the standard measures to avoid infection . For essential workers who continued to go to work with the public , the daily risk of exposure was much higher than those who were able to work from home . Al-
FIGURE 1 : Participant Reported Prior COVID-19 Exposure and Testing
59.3 %
Exposure
49.2 %
Previously PCR Tested
FIGURE 2 : Risks for Positive COVID-19 Antibody Test
Previous Positive PCR Test 79.2 %
No Previous Positive PCR Test 4.2 % ] - p < 0.001
Household COVID-19 Contact No Household COVID-19 Contact 4.9 %
Previously Tested with PCR
11.9 %
No Prior COVID-19 Test 4.4 % Exposure to COVID-19
11.3 %
No Exposure to COVID-19 2.8 %
Healthcare Worker |
10.5 % |
Community Essential Worker |
5.8 % |
Kanawha County |
9.3 % |
Putnam County |
4.8 % |
Cabell County |
3.1 % |
PCR : Polymerase Chain Reaction
Believed Previously Infected
] - p < 0.001
] - p < 0.001
] - p < 0.004
] - p < 0.02
though testing for acute infection was possible , supply was somewhat limited . The emergency use authorization of antibody testing by the FDA allowed a view into prior viral exposure and potential immunity for those not able to previously test whether due to lack of supply or symptoms .
20.5 %
4.6 %
Confirmed Positive PCR
46.7 % ] - p < 0.02
7.8 %
Positive Antibody Test
0 % 20 % 40 % 60 % 80 % 100 %
Our study demonstrated that participants poorly predicted previous infection with COVID-19 as only 26 % of patients positive for antibodies were suspicious that they had already experienced a COVID-19 infection . It is also noted that the timing of antibody testing could have occurred outside of the window of detection . Those individuals with known prior infection , known household contacts , known exposure to COVID-19 , and participants who had tested for COVID-19 regardless of results were much more likely to have positive antibody tests than those without these risks . Of those that had a positive antibody test , 75 % had previously been tested for COVID-19 with a PCR test , leaving 25 % of participants that did not know that they had previously been infected with COVID-19 , presumably due to an asymptomatic infection . Of the portion that did test , one-third of participants reported that their PCR was negative . This finding suggests that the actual prevalence of disease at the time was under detected with rates up to 50 % higher than reported at the time . Several antibody studies demonstrating that undiagnosed or asymptotic COVID-19 cases were much higher than detected when antibodies were used to define prior infections . 11-13
18 • www . wvsma . org