Read on Twitter
1/n
Important validation update on #Abbott and #Roche #COVIDー19 #Antibody tests. Validation carried out on convalescent sera from hospital (severe) and community (mild) confirmed #SARSCoV2 cases: https://www.gov.uk/government/publications/covid-19-laboratory-evaluations-of-serological-assays
Important validation update on #Abbott and #Roche #COVIDー19 #Antibody tests. Validation carried out on convalescent sera from hospital (severe) and community (mild) confirmed #SARSCoV2 cases: https://www.gov.uk/government/publications/covid-19-laboratory-evaluations-of-serological-assays
2/n
Abbott test showed 100% specificity and just over 93% sensitivity for sera >14 day after symptom onset. Manufacturer claimed 100% sensitivity for sera >14 day post-onset.
Abbott test showed 100% specificity and just over 93% sensitivity for sera >14 day after symptom onset. Manufacturer claimed 100% sensitivity for sera >14 day post-onset.
3/n
In the #PHE validation, the Roche assay also showed 100% specificity and 87% sensitivity for sera obtained >14 day post symptom start. Meaning the assay would miss a previous infection in 1 out of 10 cases. Manufacturer claimed a sensitivity of 100% in their validation...
In the #PHE validation, the Roche assay also showed 100% specificity and 87% sensitivity for sera obtained >14 day post symptom start. Meaning the assay would miss a previous infection in 1 out of 10 cases. Manufacturer claimed a sensitivity of 100% in their validation...
4/4
Take home message. Positive test means you have had a past #SARSCoV2 infection. Negative test doesn’t tell you anything - most likely not infected in the past, but you wont know for sure. And as the number of cases increases the chances if it being a true negative falls...
Take home message. Positive test means you have had a past #SARSCoV2 infection. Negative test doesn’t tell you anything - most likely not infected in the past, but you wont know for sure. And as the number of cases increases the chances if it being a true negative falls...
5/4
Correction - re-reading the test procedure it looks like all #Covid-19 samples were obtained in hospital setting. Will seek clarification...
But data still holds true.
Correction - re-reading the test procedure it looks like all #Covid-19 samples were obtained in hospital setting. Will seek clarification...
But data still holds true.
6/4
‘...none of the patients with previously positive PCR who tested negative by this assay had been hospitalised for COVID-19 disease and most likely had a mild disease outcome.’ So, yes they included non-hospital admissions and importantly mild cases sometimes tested negative
‘...none of the patients with previously positive PCR who tested negative by this assay had been hospitalised for COVID-19 disease and most likely had a mild disease outcome.’ So, yes they included non-hospital admissions and importantly mild cases sometimes tested negative
7/4
This statement was included in the reports for both assays. Therefore, antibody testing to confirm community (mild) disease might lack sensitivity. This fits with the healthcare worker study @bealelab flagged to me yesterday. I will dig further...
This statement was included in the reports for both assays. Therefore, antibody testing to confirm community (mild) disease might lack sensitivity. This fits with the healthcare worker study @bealelab flagged to me yesterday. I will dig further...
