Showing posts with label corona virus. Show all posts
Showing posts with label corona virus. Show all posts

Friday, June 24, 2022

A look at Race/Ethnicity and South Central PA COVID Vaccination and Case Mortality Rates

I have been using the stat package R to create graphics. Above is a graph from R that uses the same variables in a graph from the last post: case mortality and the average number of mentally unhealthy days in the last month. I was able to add another layer to the graph by color coding each county by the full vaccination rate. Counties with higher vaccination rates are lighter colored and those with lower rates are darker colored. Graphics like these are useful for showing the effects of more than two variables. 

The main thrust of this post looks at correlations with COVID vaccination and Case Mortality rates in different ethnicities. If the rates by race/ethnicity are higher than overall rates, that suggests that there is more of an issue with that group in the 10 county region. If the rates are lower, then there is a potential confounding for this variable. Some counties do not have values for different ethnicities as they may have small populations of that group.

% Screened for a Mammogram

There was no overall correlation for vaccination or case mortality.  There was a significant negative correlation for white percent screened and case mortality (-0.92).  Only 6 out of the 10 counties had rates for whites and African Americans and 5 for Hispanics and Asians. For these counties, 84% of the variability in case mortality is accounted for by the % of whites screened. For the other groups the correlations were negative but not as strong. Those rates decreased the overall rate.

% Flu Vaccination

There was a significant positive correlation (0.704) for the flu and COVID vaccination rates and a negative correlation for flu vaccination and case mortality (-0.895) (pictured above with color codes for COVID vaccination rates). The white vaccination rates were virtually identical to the overall rates. All 10 counties had white rates while 6 had rates for Asians, 8 for African Americans, and 9 for Hispanics. Correlations were weakest for African Americans.

% Driving Alone to Work

There was a significant positive correlation between COVID case mortality rates and the % driving alone to work (0.73). Even though only five counties had rates for whites (pictured above), there was a significant positive correlation for them at 0.946. Centre county has a lower rate for driving alone to work for the overall and white rate due to better public transportation with Penn State University present. There were 4 counties with rates for Hispanics and African Americans and 2 for Asians. The rates were positive for the other groups but much weaker for Hispanics and African Americans.

These correlations for other groups highlight disparities within the counties. I have written before on racial disparities in Cambria County and presented at the Juneteenth celebrations. My slides from the presentations are presented below. This is just for one county. Other disparities are present in the other counties.

**Related Posts**

Sunday, June 19, 2022

County Health Rankings and COVID Case Mortality

I finished being at the Juneteenth festivities in Johnstown and am ready to get back to county health rankings (CHR) and local COVID numbers. Case mortality (the number of deceased divided by the number of cases) will be the focus of this post. Two weeks ago, I focused on vaccination rates in the 10 county region. Case mortality and vaccination rates were negatively correlated (meaning that as one variable increases the other decreases) accounting for 54% of the variability. There were eight other CHR statistics correlated with vaccination rates. Eighteen other CHR statistics were correlated with case mortality. 

The correlation with the average number of mentally unhealthy days in the last month is summarized in the graph above. The graph shows a strong positive relationship with case mortality accounting for 70.1% of the variability. The regression equation states that for every one day increase in the average number of mentally unhealthy days there is a predicted 1.5% increase in the case mortality rate. There is also a significant but weaker negative correlation that I summarized two weeks ago between mentally unhealthy days and COVID vaccination rates accounting for 42.2% of the variability.  The % of the variability accounted for is simply the correlation coefficient squared.

The strongest negative correlation for case mortality is with the % in the county who are vaccinated for the flu. This correlation accounts for 79.9% of the variability in COVID case mortality. The regression equation says that for every 1% increase there is a predicted 0.06% decrease in case mortality rate. If 100% of the variability were accounted for, all of the counties would fall on the regression lines. Surprisingly this relationship is even stronger than the one with case mortality and COVID vaccination rates which were also negative and only accounted for 54% of the variability.

The 18 significant correlations are summarized in the table below. The positive correlations were with years of potential life lost, both the average number of physically and mentally unhealthy days, % smokers, % physically inactive, the teen birth rate, % unemployed, the social association rate, the injury death rate, and the % who drive alone to work. The negative correlations are as follows: the % with access to exercise opportunities, the % with an annual mammogram, % with flu vaccinations, % completed high school, % with at least some college, the higher the income level in the 80th percentile in the county, the % with severe housing problems and those with a high housing cost burden.

Variable Correlated with Case Mortality

Correlation Coefficient

% Variability Explained

Years of Potential Life Lost Rate (YPLL)



Average Number of Physically Unhealthy Days



Average Number of Mentally Unhealthy Days



% Smokers



% Physically Inactive



% With Access to Exercise Opportunities



Teen Birth Rate



% With Annual Mammogram



% Vaccinated for the flu



% Completed High School



% Some College



% Unemployed



80th Percentile Income



Social Association Rate



Injury Death Rate



% Severe Housing Problems



Severe Housing Cost Burden



% Drive Alone to Work



One should always be careful about inferring cause and effect relationships between  correlated variables. Variable A could cause variable B or vice versa. There is always a potential 3rd variable that could explain the correlation such as poverty. Many of these variables are also correlated with each other. This method does allow one to see how they could be interrelated. Next I will look at how different ethnicities correlate with case mortality.

**Related Posts**

Sunday, June 5, 2022

County Health Rankings Statistics and COVID Vaccination Rates

I have talked in the last few posts about COVID rates and County Health Rankings (CHR). I flagged significant correlations between the rates and the rankings (greater than 0.632 or less than -0.632 for 10 counties). While the rankings are easy to explain and are often quoted in the media, the individual statistics (more than 60 or them) used to create the rankings provides more specific information about the variables that predict local COVID rates.

This post focuses on COVID vaccination rates and individual CHR statistics.  The graph above shows the correlation between Vaccination rates (for those who have received the first 2 shots) and COVID case mortality rates. It shows a negative relationship between the two where every 1 % increase in the vaccination rate yields a predicted 0.046% decrease in the case mortality rate. This relationship accounts for 54% of the variability in the Case mortality rate. Next I will summarize the CHR statistics that are significant with the vaccination rates.

There were 8 CHR variables that were significantly associated with the vaccination rate. The first that was significant was the average number of mentally unhealthy days (a quality of life statistic). This model states that for every one day increase in mentally unhealthy days, there is a predicted 19.3% decrease in the vaccination rate. This relationship accounts for 42.18% of the variability in the vaccination rate. 

To save space, I will summarize verbally the remaining significant correlations. The next significant correlation is the % of the county who are physically inactive which is also negative. This relationship accounts for 42.3% of the variability. Paradoxically, the % with access to exercise opportunities was positively correlated with the vaccination rate accounting for 49% of the variability.

The next significant correlation with the vaccination rate was with the primary care physician rate in the county. These rates were positively associated accounting for 42.3% of the variability. Likewise the flu vaccination rate was positively associated accounting for 49% of the variability. The high school completion rate and the % completing at least some college were both positively associated with the vaccination rate accounting for 44.9% and 39.7% of the variability respectively. Finally the income amount in the 80th percentile in the county was positively associated accounting for 39.7% of the variability. 

Next week I will summarize case mortality correlations.

**Related Posts**

2022 County Health Rankings are Out: Cambria Still Ranks Low

Friday, May 27, 2022

County Health Rankings predict COVID Mortality in the 10 County Area


Last year I looked at COVID mortality and County Health Rankings (CHR) numbers for the whole state of PA. This year, the state does not make cumulative mortality numbers readily available. This makes updating the numbers subject to copying error. For this year, I thought I would focus on ten county region surrounding Cambria County.

The 10 county area is listed in the table at the top. The univariate correlation coefficients are presented in the table below for the COVID measures above and the CHR rankings for health outcomes and health factors. Health outcomes is a composite of length of life and quality of life measures. Health factors is a composite of health behaviors, clinical care, social and economic, and physical environment factors. The correlation matrix is presented below. Because of the small sample size, correlation coefficients of 0.632 or higher or -0.632 or lower were flagged as significant and presented in bold below.


% COVID Fully Vaccinated 

Case Mortality %

COVID Case Rate /100,000

COVID Mortality /100,000

Hosp  /100,000

County Health Outcomes Rank

Health Factors Rank

% COVID Fully Vaccinated 


Case Mortality



COVID Case Rate




COVID Mortality /100,000





Hosp /100,000






County Health Outcomes Rank







Health Factors Rank








Length of life rank








Quality of Life Rank








Health Behaviors rank








Clinical Care rank








Soc & Econ Factors rank








Phys env rank








The COVID case and population adjusted mortality rates were most strongly associated with the CHR rankings of length of life, health behaviors, and clinical care. A positive correlation with the rankings suggest that the lower the ranking is, the higher the COVID mortality. Case mortality is simply the number of COVID deaths divided by the number of COVID cases.

The above graph shows the scatter plot for the length of life rank showing a linear association with COVID case mortality. The R squared statistic of 0.485 means that 48.5% of the variability in case mortality is accounted for by the length of life ranking. The regression equation says that for every unit a county is ranked lower, there is a predicted increase of 0.02% in the case mortality. There were no strong outliers in in this plot. 

The strongest correlation for the health factors rankings was clinical care at 0.808. The scatter plot above shows a stronger correlation between that and case mortality accounting for 64.5% of its variability. If it were 100% of the variability, all of the counties would fall on the regression line. Like length of life, the regression equation predicts that for every unit lower that a county is ranked, a 0.02% increase in case mortality should happen.

These sub rankings are themselves composites of dozens of statistics. The individual statistics should shed more light on what variables may be driving COVID vaccination, case, mortality, and hospitalization rates. The devil is always in the details. This will be the next step.

**Related Posts**