This thing uses "socioeconomic factors driving risk" to make the map. There are a whole bunch of these listed when you click on an area and most of them make absolutely zero sense with relation to COVID. I pulled many of their socioeconomic factors from maybe 50 different areas I clicked on in several states and grouped them into these categories (this is probably not comprehensive):
- Income level
- Poverty level
- Transportation availability
- Transportation cost
- Unemployment rate
- Access to employment
- Labor market engagement
- Industrial & Commercial job density
- Environmental health hazard
- Housing instability
- Retail job density
- Commercial retail availability
- Frequency of prolonged commute (60+ min)
- School performance
Just off the top of my head I threw together the following list of the most important and biggest factors I would use in such a model. There's almost zero overlap with what Jvion chose. But what do I know? I'm not a pro. AND I'm not cowed by PC and Wokeness:
- International airport with lots of traffic from China and Italy.
- High concentration of Chinese and Italian communities with lots of family members traveling to and from China and Italy.
- Democrat Governor and Mayor.
- Politicians urging "Hug a Chinese" a month ago.
- Local population density (at the zip code level).
- Fraction of population using public transportation.
- Prevalence of known health factors by zip code (BMI, hypertension, diabetes).
- Per capita general hospital and ICU beds.
- High density of older people and nursing homes.
- Known existing hotspot.
- Demographics and age distribution.
- Fraction of population with health insurance
This thing is ridiculously and hopelessly useless, IMHO. Just zoom into Seattle and look at Marysville, one of the hottest of hot spots because of the introduction of the disease by ONE Chinaman and the prevalence of nursing homes in the area. Their map shows low susceptibility in Marysville.
About Jvion: Jvion enables healthcare organizations to prevent avoidable patient harm and lower costs through its AI-enabled prescriptive analytics solution. An industry first, the Jvion Machine goes beyond simple predictive analytics and machine learning to identify patients on a trajectory to becoming high risk and for whom intervention will likely be successful. Jvion determines the interventions that will more effectively reduce risk and enable clinical action. And it accelerates time to value by leveraging established patient-level intelligence to drive engagement across hospitals, populations, and patients. To date,
the Jvion Machine has been deployed across about 50 hospital systems and 300 hospitals, who report average reductions of 30% for preventable harm incidents and annual cost savings of $6.3 million.
Data Sources: Jvion analyzed de-identified data on 30 million Americans. Data analyzed includes de-identified claims, USDA, EPA, Transportation and other third-party data sources like food and retail access, length of job commute, and transportation. [POF comment - would ANY of us amateur FR sleuths choose ANY of these factors to model COVID susceptibility?]