Thread by @DrEricDing

📍HUGE—a new study in New Orleans, Louisiana estimates there was a local infection fatality ratio (IFR) of 1.63%—much higher than thought for #COVID19. the IFR is among all infected, unlike CFR mortality in just those diagnosed. Antibody in May only 6.9%.🧵wwwnc.cdc.gov/eid/article/26…

2) seroprevalence was highest (9.8%) in Blacks, followed by multiracial (7.1%), Asians (5.5%), & Whites (4.5%). Hispanics had 5.3% seroprevalence.

➡️ The IFR was statistically similar for Whites (1.58%), Blacks (1.72%), and multiracials (1.40%) but lower for Asians (0.61%).

3) “No COVID-19–related data on Hispanic persons were collected by the Louisiana Department of Public Health during the study period.”

➡️ oh cmon @LADeptHealth!

4) How huge is this 1.63% IFR finding? CDC suggests the IFR is 0.65%. This is 2.5x worse than CDC estimate. cdc.gov/coronavirus/20…

5) this 1.63% IFR is close to the reanalyzed Princess Diamond cruise ship IFR of 2.0%. Again this is IFR not CFR (which is just among diagnosed) IFR uses a denominator estimate all infected endcoronavirus.org/papers/diamond…

6) The earlier 0.6% IFR is already shocking enough - 0.6% is already 50x to 500x higher IFR than 2009 flu pandemic according to @DrMikeRyan.

0.6% IFR: ~1 in 200 infected die of #COVID19

2009 H1N1 influenza: 1 in 10,000 to 1 in 100,000 infected died

7) Thus, if New Orleans truly has IFR of 1.63%—it’d be 125x—1250x worse than 2009 pandemic based on Dr Mike Ryan’s above math. My god. I hope this CDC-journal published paper about New Orleans is wrong. But Princess 💎 🚢 new analysis also being 2% is worrisome. (HT @VanGennepD)

8) Good to point out that IFR can have lots of fluctuation between populations depending on comorbid risk factors (NOLA has high obesity prevalence for example) and hospital system capacity.