Read on Twitter
Preprint 
Using viral sequencing + some epi data, we show that a majority #SARSCoV2 infections in healthcare workers at our institution occur in the community and can only rarely be traced to a #COVID19 patient source. @GageKMoreno @dho @tcfriedrich 1/12 https://www.medrxiv.org/content/10.1101/2021.01.28.21250421v1
Using viral sequencing + some epi data, we show that a majority #SARSCoV2 infections in healthcare workers at our institution occur in the community and can only rarely be traced to a #COVID19 patient source. @GageKMoreno @dho @tcfriedrich 1/12 https://www.medrxiv.org/content/10.1101/2021.01.28.21250421v1
This was a team effort. Many thanks to the infection control specialists, epidemiologists, and ID docs directly involved in this study. Thanks also to all healthcare workers and infection control teams for their ongoing dedication to patient and community health. 2/12
We hypothesized that infection control guidelines at a major academic medical center (described here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7591740/) are generally successful in protecting healthcare workers from #SARSCoV2 infection in a healthcare setting. 3/12
We tested this hypothesis by sequencing viruses from 96 healthcare personnel (HCP). We compared these viruses against ~140 possible patient contacts and our rich community surveillance sequence dataset: https://nextstrain.org/community/gagekmoreno/Wisconsin-SARS-CoV-2/ncov/wisconsin/2021-1-8. 4/12
60.4% of infections could *not* be linked to a patient or co-worker so they likely occurred in the community. Here’s a schematic of one such case. This healthcare worker cared for a single #COVID19 patient 2 wks prior to testing
. Divergent viruses=not a transmission pair. 5/12
. Divergent viruses=not a transmission pair. 5/12
10.4% of infections appeared to be linked to a co-worker. These HCP worked in the same department. HCP 16-2 and HCP 16-1 had unmasked contact during a lunch break. Additional messaging and guidelines to reduce HCP-to-HCP spread in and out of the workplace would be 
. 6/12
. 6/12
Only 4.2% of HCP infections were linked to a patient! In one of these cases, a HCP provided care to 15 #COVID19-patients prior to testing . This HCP’s virus was identical to the virus found in one of the patient contacts (patient 10-G). 7/12
. This HCP’s virus was identical to the virus found in one of the patient contacts (patient 10-G). 7/12
A recent study from the Netherlands similarly found no evidence for widespread nosocomial transmission of #SARSCoV2. This suggests our conclusions may hold across institutions and healthcare systems. 8/12 https://www.sciencedirect.com/science/article/pii/S1473309920305272
If infection control measures + PPE protect against symptomatic infection, even in #COVID19-dense settings → similar measures used consistently & correctly might be able to provide adequate protection in other congregate settings like correctional facilities and schools! 9/12
We used rapid whole-genome sequencing of current #SARSCoV2 outbreaks in hospitals to retrospectively reconstruct the likely source of HCP infection and to prospectively adjust and improve infection control practices and guidelines. Other hospitals can do the same! 10/12
As the majority of #SARSCoV2 infections in healthcare workers appear to be tied to the community, our results emphasize the importance of measures to reduce *community spread* = masks, physical distancing, robust testing programs, and rapid distribution of vaccines. 11/12
Relatedly, check out this video from our clinical partners @UWHealth showing how the infection control team has been instrumental in protecting HCP, patients, and our community since the beginning of the pandemic. 
12/12 https://www.uwhealth.org/news/healthcare-heroes-infection-control-and-prevention-at-uw-health/53637
12/12 https://www.uwhealth.org/news/healthcare-heroes-infection-control-and-prevention-at-uw-health/53637
