Yes, that was my understanding. In order to test positive for the antibodies, you had to actually have the virus at some point. Maybe Doc can clarify.
If that's true, it means many, many more people have had the virus than the denominator currently being used. And that means many people with the virus have few or no symptoms. All that makes sense to me because up until now, the only people being tested were sick people with symptoms. So the reported mortality rate was really, "Of all the people sick with serious symptoms, X% die". Now we're learning, "Of all the people with the virus, Y% die". There's a difference, and it seems like a significant difference to me.
This just came from JAMA within the hour. It is from an integrated health system in California. This is part of the paper that is relevant to your question.
[Of 16 201 tests in adults, results from 1299 patients (8.0%) were positive for SARS-CoV-2. Of these patients, 377 (29.0%) were treated as inpatients and 113 (8.7%) were treated in the ICU.
The median age was 61.0 years (interquartile range, 50.0-73.0); 56.2% were men (
Table). The most common comorbidity was hypertension (n = 164, 43.5%). Of 166 patients who underwent testing for influenza A/B or respiratory syncytial virus (44.0% of the cohort), none tested positive. Bilateral infiltrates on chest film were seen in 63.4% (n = 239). Overall, 34 patients (9.0%) received a prednisone-equivalent dosage of 20 mg/d or more.
Most patients were treated on the general ward or intermediate care unit (n = 264, 70.0%); of whom 54.9% received supplemental oxygen through nasal cannula/face mask. A total of 113 inpatients (30.0%) required ICU admission and 110 (29.2%) received invasive mechanical ventilation.
Patients aged 60 to 69 years represented the most common age group both hospitalized (n = 93, 24.6%) and admitted to the ICU (n = 31, 27.4%) (
Figure). However, adults of all ages were admitted, and the proportion of younger and middle-aged adults (≤59 years) who were hospitalized (n = 172, 45.6%) was similar to the proportion of older adults (≥60 years) who were hospitalized (n = 205, 54.4%).
Of 321 patients with discharge dispositions, 50 (15.6%) died in the hospital. Of 253 patients treated on the ward with discharge dispositions, 16 (6.3%) died. Of 68 patients treated in the ICU with discharge dispositions, 34 (50.0%) died.]
Now those were all PCR tests. For now pending more experience we have to take ELISA antibody tests with somewhat of a grain of salt pending reliable validation and further experience.
The problem is test accuracy and specificity. The problem being that if antibodies detected were induced by exposure to Covid 19, yes, they had exposure. The question becomes is are we actually detecting a past infection with a different Corona virus. We know for certain that some tests have, especially from the UK, which have very strict criteria for these test, and has frustrated the public and politicians causing missed targets. However Professor Chris Whitty is correct, in that an inaccurate test is worse than no test.
We really are just at the beginning of understanding the epidemiology of this Covid 19 pandemic and I'm sure our understanding will undergo a lot of iterations over time.
That is the reality of medicine, data changes over time and becomes more precise. So that results in changing advice and recommendations to the public, which is confusing to individuals. That can not be helped.
That is why I'm very excited about the project getting under way here in Minnesota. I hope it will provide a lot of answers.
