Hello, I'm Monica Kasting from Purdue
University, and I'm going to talk about our article that was recently published in the
American Journal of Public Health titled County Level Factors Associated with Influenza
and COVID -19 Vaccination in Indiana, 2020 through 2022. As a brief background,
the COVID -19 pandemic highlighted rural urban disparities, including access to
health care, differences in health beliefs, higher vaccination hesitancy and lower
vaccination confidence. While initial pandemic hotspots centered in densely populated
urban areas, infections in rural areas began to rise as the pandemic progressed.
Nationally, there are significant rural urban vaccination coverage disparities across multiple
vaccines. Addressing these disparities must begin with accounting for the triad of local culture,
geographic location, and economics of the region. Therefore, this study aimed to assess existing
differences among adults in both COVID -19 and influenza vaccination by county in Indiana
and determine what county -level factors are associated with those vaccination rates.
We chose to focus on these two vaccines because they both involve respiratory viruses,
both have generated greater vaccine hesitancy than other adult vaccines, and are both
routinely recommended for all adults. We use several sources of data for
our analyses. We examine COVID -19 and influenza vaccination rates across Indiana's
92 counties among adults aged 18 years and older from December 2020 through March 2022.
For COVID -19 data, we use the Indiana State Department of Health's Immunization
Registry, which is called the Children and Hoosier Immunization Registry Program,
or CHIR. CHRP had mandatory reporting for all COVID -19 vaccinations.
These data are publicly available by request. CHRP also includes information on
county -level vaccination rates by age group, race -ethnicity, and sex. All data were aggregate
and no individual -level data were available. However, there's not mandatory reporting in CHRP
for adult influenza vaccinations. Therefore, for flu data, we use the CDC's U .S. Influencers'
Surveillance Dashboard or FluVax View. FluVax View reports weekly national influenza
vaccination data by county. From the database, we extracted influenza vaccination rates using
the federal information processing codes, which are unique codes assigned to each county
in the U .S. and we filtered down to only Indiana counties. For our other measures, we derive
population -level estimates from the U .S. Census. We also included county -level socio -demographic
and health data from the County Health Rankings Report, which is supported by the Robert
Wood Johnson Foundation and the University of Wisconsin Population Health Institute.
From county health rankings, we extracted data on socio -demographic distribution of each
county, including median household income, percent of the county that's rural and percent
of the county that is Hispanic, among others. We extracted data on health indicators like
the number of primary care providers per capita and the percentage of the population who smoke.
vaccination rates using linear regression weighted by the total adult population in each county,
and then conducted model selection using forward, backward, and stepwise selection to determine
the best multiple linear regression model for each vaccination rate. We selected
the best model based on the adjusted R -squared value and examined tolerance values for
variables in each final model to ensure issues with multi -colonarity were negligible,
and we assessed residuals for normality. The selected reduced model for each vaccination
rate is reported, and the squared semi -partial correlation coefficient for each variable
is reported as a measure of effect size. You can see the mean percentage of counties that
were rural was 54 .5%. When looking at socio -demographic characteristics, the mean percentage
of residents who were African -American were 3 % with a wide range from 0 .2 to almost 30%.
There were high rates of smoking with a mean of 22 % and obesity with a mean of 36%. The mean COVID
-19 vaccination rate across the 92 counties in Indiana was 58 % and ranged from 31 .2 to 87 .6%.
The mean influenza vaccination rate was 42 .9 % and ranged from 33 .7 to 53 .1%. This figure
displays the variability in COVID -19 influenza vaccination rates across Indiana counties,
with darker color indicating the higher vaccination rate. COVID -19 vaccination
rates are shown in panel A on the left, and flu vaccination rates are shown in
panel B on the right. As you can see, there is more variability with COVID -19
vaccination than was flu vaccination, and overall flu rates were lower than COVID -19
vaccination rates. The final selected model for COVID -19 contained seven variables and
had an adjusted R -squared of 0 .867. Based on this model, an increase in primary care
providers per capita, median household income, percentage of Medicare enrollees
who had a mammography screening, and percentage of African -American residents
were all associated with increases in the percentage with COVID -19 vaccination.
In addition, an increase in uninsured residents, percentage of female residents,
and percentage of people who smoke were all associated with decreases in the percentage
with COVID -19 vaccination. The final model for flu contained five variables and
had an adjusted R -square to 0 .702. Based on this model, an increase in the
percentage of residents who were uninsured and those who completed high school were
associated with increases in the percentage with flu vaccination, and an increase in the
percentage of African -American residents, Hispanic residents, and percentages of adults
who smoke were all associated with decreases in the percentage with flu vaccination.
While both COVID -19 and flu vaccinations are routinely recommended for all adults,
county level factors associated with each very greatly between the two vaccines.
The reduced model for COVID explained a slightly higher percentage of the
variance, 86 .7 % than the model for flu, 70 .2%. This may reflect greater polarization
surrounding COVID -19 vaccination, leading to a greater influence in socio -demographic
factors with COVID -19 compared to flu. Of note were variables reflecting access to care,
like the number of primary care providers per capita, which were significant for COVID, but not
for flu. One possible explanation is the unique way in which the COVID -19 vaccine was rolled out,
which involved different dates which people were eligible based on age or occupation,
for instance. The need to sign up for a vaccine online and difficulties particularly
early on in finding available appointments, all of which may have increased logistical
barriers and resulted in the differences we found in our data. Another finding of interest
was the association between each vaccine and county -level racial ethnic distribution.
Their percentage of Hispanic residents in the county was associated with influence
of vaccination but not COVID -19, and the percentage of African -American residents in a
county was negatively associated with flu vaccine, but positively associated with COVID -19 vaccine.
This could be due to targeted outreach programs, at medical discrimination and
distrust in the medical community, or a broader reflection of access to health care.
It's possible differing findings by the Hispanic population may be better explained if
we controlled for individual patient characteristics. That was not possible with our
aggregate data. Future research should explore these racial ethnic disparities while taking
individual patient characteristics into account. Importantly, smoking was strongly
negatively associated with vaccination rates for both vaccines. This may seem
counterintuitive because smoking is a risk factor for severe respiratory diseases. However,
research shows preventive health behaviors cluster together and people who engage in one healthy
or unhealthy behavior are more likely to engage in another. Likewise, people who smoke are less
likely to receive routine preventive services. In addition, while the percentage of the county
that was rural was significantly associated with both COVID -19 and flu vaccinations in
univariable regression analyses. These associations were no longer significant when
other variables were included in the model. It's possible that the rural urban differences
were accounted for with other county -level factors, including income or smoking status. But
the reasons for the rural urban disparities across a multitude of health indicators are multifaceted
and are likely a complex elimination of access, infrastructure, attitudes, and beliefs.
Our findings should be interpreted in the light of some limitations. First, the
data are cross -sectional and causal relationships cannot be established.
Second, the data are aggregate and are subject to ecological fallacy.
In addition, our data focused only on Indiana and findings may not be applicable
to other states or jurisdictions. And lastly, because CHRP does not track adult influence of
vaccination, we use two different data sources to examine the two vaccination rates.
It's possible that the two data sources differ in their accuracy and reporting.
Therefore, the results of this study should be interpreted with caution and further
studies are needed to understand the complex associations between systems -level factors we
examined and vaccination rates. In conclusion, while both COVID -19 and flu vaccines
protect against respiratory viruses and are recommended for all adults. The
factors associated with uptake of each are varied. Variables reflecting access to care,
like the number of primary care providers per capita and median household income, were
significant for COVID -19, but not flu vaccination rates. The percentage of uninsured residents in
the county was significant for both vaccines, but in opposite directions, so that the rate was
negatively associated with COVID and positively associated with flu. The polarization surrounding
COVID -19 vaccination may have led to a greater influence of socio -demographic factors with COVID
-19 vaccination as compared with flu vaccination. Further research, including individual
-level data, is needed to better understand these associations and develop effective
interventions to address county -level factors and improve vaccine uptake. In closing,
I'd like to thank the rest of our research team and acknowledge our research support and
funding for this work. Thank you for your time. For more information, please view
our full article by scanning the QR code.