As many have noted, there has been an uptick in confirmed #COVID19 cases in multiple states starting roughly at the beginning of June. There is a major question surrounding how much "re-opening" efforts are contributing to increases in transmission. Figure from @nytimes. 1/10

I thought to look into a specific state and start with Arizona. Here, we see pretty dramatic increases in confirmed cases starting about May 26. Arizona's stay-at-home order expired on May 15. Figure from the @azdhs COVID-19 data dashboard. 2/10

This specific increase in cases at the end of May seems to be limited to Southern Arizona with Maricopa County contributing the most to this increase. 3/10

Connecting the increase in transmission in Arizona directly to specific reopening policies is difficult. Although the general timing makes sense as we've seen a reduction in social distancing, this is not definitive. Figure from @CATT_Lab ( https://data.covid.umd.edu/ ). 4/10

Generally, I've been disappointed with the lack of US public data on exposure setting obtained through case investigation. There have been a couple recent manuscripts examining this, one focused on transmission in Hong Kong and one focused on Japan. 5/10

Work lead by @dilloncadam and @bencowling88 examines case investigation and contact tracing data from Hong Kong to emphasize how important social settings were for propagating large clusters ( https://www.researchsquare.com/article/rs-29548/v1). 6/10

Work lead by Furuse and Oshitani concludes that clusters derived primarily from healthcare settings, but that social settings "with heavy breathing in close proximity, such as karaoke parties, clubs, bars, and gymnasiums" also contributed strongly ( https://wwwnc.cdc.gov/eid/article/26/9/20-2272_article). 7/10

We should have all we need to address the reopening question. The simple analysis would be: have we seen an increase in the proportion of cases with exposure deriving from social settings? 8/10

I imagine that household transmission will be a more continuous occurrence, while with reopening we would expect a new layer of infections derived from social settings. 9/10

However, I haven't been able to find public data on exposure setting in any of the state data dashboards. This data has to exist with state health departments (though could be spotty) and could be readily analyzed by public health epidemiologists. 10/10