Tuesday, on a ConnectLocal post of the Georgia Department of Public Health COVID-19 update, a reader asked why Stephens County’s infection rate appears to be higher than that of surrounding counties. ConnectLocal examines a range of demographic factors to see what role, if any, they may play in COVID-19's prevalence.
In an attempt to discover a cause behind that statistic, ConnectLocal examined a number of variables to determine if there was any causal effect.
Charting demographic factors in correlation to April 15 COVID-19 testing statistics provided by GDPH showed significant variances in a number of those factors, but none that tracked with prevalence of confirmed tests of different counties.
In Stephens County, population of 25,925, a cumulative total of 24 people have tested positive for SARS-CoV-2, the virus that causes COVID-19. Using this raw data, Stephens County has a base level infection rate of 0.093% - or approximately one out of every 10,000 people. While this rate is incomplete, and does not account for the percentage of the population that has been tested to result in those 24 positive tests, nor does it account for any differences in testing criteria between counties, it currently provides the most widely-used figure used to compare populations in terms of the transmission of COVID-19, including comparisons of countries, states and counties.
Among immediately adjacent counties, Stephens County’s base level infection rate is highest, with Habersham county, population 45,328, falling second, reporting a cumulative total of 29 confirmed cases, for a base level infection rate of 0.064%, or 6 out of every 10,000 people.
Among the geographic units selected by ConnectLocal for this review, following is a ranking of base level infection rates.
Upson County – 51 per 10,000 residents (0.513%)
United States – 19 per 10,000 residents (0.186%)
Hall County – 18 per 10,000 residents (0.183%)
Decatur County – 17 per 10,000 residents (0.178%)
Georgia – 14 per 10,000 residents (0.144%)
Stephens County – 9 per 10,000 residents (0.093%)
Habersham County – 6 per 10,000 residents (0.064%)
Anderson County (SC) – 6 per 10,000 residents (0.055%)
White County – 5 per 10,000 residents (0.045%)
Banks County – 4 per 10,000 residents (0.042%)
Rabun County – 4 per 10,000 residents(0.041%)
Franklin County – 3 per 10,000 residents (0.030%)
Hart County – 2 per 10,000 residents (0.019%)
Evans County – 2 per 10,000 residents (0.019%)
Oconee County (SC) – 2 per 10,000 residents (0.018%)
Taliaferro County – 0 per 10,000 residents (0%)
Glascock County – 0 per 10,000 residents (0%)
Although fatalities from COVID-19 are concentrated among those older than 65 years of age, according to state, national and international figures, the majority of confirmed cases occur between the ages of 20 and 60. To adjust for that figure, ConnectLocal recalculated the base infection rate for each geographical unit after removing those under the age of 18 from the population figure used in the calculation.
Following is a ranking of age-adjusted infection rates among the included geographical units.
Upson County – 66 per 10,000 residents over 18 (0.659%)
Hall County – 25 per 10,000 residents over 18 (0.245%)
United States – 24 per 10,000 residents over 18 (0.240%)
Decatur County – 24 per 10,000 residents over 18 (0.236%)
Georgia – 19 per 10,000 residents over 18 (0.189%)
Stephens County – 12 per 10,000 residents over 18 (0.119%)
Habersham County – 8 per 10,000 residents over 18 (0.083%)
Anderson County (SC) – 7 per 10,000 residents over 18 (0.067%)
White County – 6 per 10,000 residents over 18 (0.057%)
Banks County – 5 per 10,000 residents over 18 (0.054%)
Rabun County – 5 per 10,000 residents over 18 (0.049%)
Franklin County – 4 per 10,000 residents over 18 (0.038%)
Evans County – 3 per 10,000 residents over 18 (0.026%)
Hart County – 2 per 10,000 residents over 18 (0.024%)
Oconee County (SC) – 2 per 10,000 residents over 18 (0.022%)
Taliaferro County – 0 per 10,000 residents over 18 (0%)
Glascock County – 0 per 10,000 residents over 18 (0%)
Although there are a few changes in the ranking as a result of removing those under the age of 18 from the calculation, the changes are neither uniform, not significant enough to indicate that the age of a county’s population plays a role in the prevalence of COVID-19 among their residents.
Similarly, examination of each county’s demographics in terms of gender, race, household size, education, health insurance, employment, income, poverty levels and population per square mile show that these factors may vary, sometimes even significantly, between counties, but that those variances fall randomly along the continuum of counties listed in ranked order by number of COVID-19 confirmed cases, indicating that there is no definitive causal effect of any of those demographic factors.
“After looking at county-breakout date from a number of states, including Georgia’s, I have found no correlation between any standard demographic division, such as age, gender, race or income, that accounts for differences in how widespread COVID is in each of the counties,” Oregon statistics professor and author Chris Alsen told ConnectLocal. “That, to me, leaves four factors that may play a role in determining why some areas appear to have more cases than others. One is how many tests a county is conducting, compared to another county, the second is how stringent the screening is before tests are conducted, the third is the random nature of this, or any virus' severity of symptoms on an infected individual, and the fourth is adherence to social distancing measures within each county."
Geographical Unit selection:
Upson, Decatur and Hart counties - counties closest to Stephens County in population
Habersham, Banks, Franklin and Oconee (SC) counties - adjoining counties
Hall County, Anderson (SC), White, Banks, and Rabun counties - NE/Upstate region
Evans County - One of last three counties to remain virus free, April 14, when Evans had it's first positive test.
Taliaferro and Glascock County – last two Georgia counties with zero confirmed cases