The Association of Social Disparity with Invasive Group A Streptococcus Infections in Minnesota Using Census Tract Level Socioeconomic Status, 1996-2016
2018-10
Loading...
View/Download File
Persistent link to this item
Statistics
View StatisticsJournal Title
Journal ISSN
Volume Title
Title
The Association of Social Disparity with Invasive Group A Streptococcus Infections in Minnesota Using Census Tract Level Socioeconomic Status, 1996-2016
Authors
Published Date
2018-10
Publisher
Type
Thesis or Dissertation
Abstract
It has been difficult to determine the association, if any, between invasive Group A Streptococcus (iGAS) disease and social determinants of health (SDH) due to lack of data on an individual’s income, race, or living condition. While it is possible for a marginalized group to be genetically predisposed to specific infections, in most cases, socioeconomic factors such as poverty may be stronger determinants of disparities in infectious diseases. We used Census derived area-based information on SDH as a proxy for an individual’s data, and combined geocoding technology and proven statistical methods to investigate any association between iGAS cases with poverty, race, and living conditions. The Minnesota Department of Health Emerging Infections Program collected data on more than 3,600 iGAS cases from 1996 to 2016. The analysis was divided into three phases. For the first phase, iGAS cases from 1996 to 2016 were geocoded to their corresponding Census tract poverty levels and analyzed as outlined by the Harvard School of Public Health Disparities Geocoding Project method. Results demonstrated there would be 18% (population attributable fraction, or PAF, of 18%) reduction in iGAS if the exposure to poverty were reduced to zero. PAF was highest (29.4%) for 45-64 year-olds. The highest incidence rate ratio (IRR) of 2.26 (95% CI [2.02, 2.52]) was observed between the least impoverished (poverty level <=5%) and the most impoverished group (poverty level >=20%). RII (relative index of inequality) of 2.45 (95% CI [2.14, 2.77]) indicated that people in the most impoverished group were 2.45 times more likely to have iGAS than those in the least impoverished group. For the second phase, in addition to poverty, we appended area based racial diversity and overcrowded living condition from the U.S. Census and American Community Survey to each geocoded case. We divided the data into four groups of years (1996-2000), (2001-2005), (2006-2010) and (2011-2016) to observe any temporal pattern. A multilevel Poisson regression model was used to account for spatial similarity. While individual predictors -- poverty, racial diversity, and overcrowded living conditions -- were strongly associated with iGAS cases, multiple regression produced an inconsistent result due to moderate to high multicollinearity between these three measures. A principal component analysis used to reduce the correlated variables to a single latent variable and 73% of variance in the three SDH measures could be attributed to this variable. We created an index score by combining all three predictor variables using the first principal component of the correlation matrix. Every 10% increase in this combination of predictors is associated with an increase of the case count per 100,000 population in the first year-grouping (1996-2000) of 39% (p<.0001, 95% CI [31%-48%]), in the second grouping (2001-20015) of 21% (p<.0001, 95% CI [13%-29%]), in the third grouping (2006-2010) of 25% (p<.0001, 95% CI [17%-33%]), and in the fourth grouping (2011-2016) of 42% (p<.0001, 95% CI [33%-52%]). Finally, for the third phase of the analysis, over 2,000 cases were reported between 01/1/07 and 12/31/16. The residence of each case was geocoded to its corresponding Census tract allowing area-based measures of poverty from the American Community Survey to be appended to each case. We divided the data into two sets of 5 years each: (2007-2011) and (2012-2016) to observe any temporal pattern. A multilevel Poisson regression model was used to account for spatial similarity. For every 10% increase in CT poverty from 2007-2011, and from 2012-2016, the rate per 100,000 population for iGAS infection increased by 16% (p<.0001, 95% CI [8%-25%]), and 16% (p<.0001, 95% CI [8%-24%] respectively for whites. For every 10% increase in CT poverty from 2007-2011, and from 2012-2016, the rate for iGAS infection increased by 30% (p<.0001, 95% CI [16%-45%]), and 20% (p<.0001, 95% CI [8%-33%] respectively for non-whites. Since incidence rates are comparatively similar for both races, and racial diversity and poverty of a neighborhood are moderately correlated indicating more non-whites live in areas with higher poverty, the association of iGAS with race is mediated mostly by poverty. Invasive Group A Streptococcus infection disproportionately affects marginalized populations, likely due to factors associated with SDH and may be mitigated through proactive public health measures. Similar methods can be applied where an individual’s information on social determinants of health such as income, race, ethnicity, literacy, employment status, etc. is missing from the data. This may help accurately quantify any social disparity, and subsequently design targeted intervention for both chronic and infectious diseases. Since a long-term improvement of social infrastructure is warranted, vaccine development is an immediate and plausible solution.
Description
University of Minnesota Ph.D. dissertation. October 2018. Major: Environmental Health. Advisor: Michael Osterholm. 1 computer file (PDF); vi, 103 pages.
Related to
Replaces
License
Collections
Series/Report Number
Funding information
Isbn identifier
Doi identifier
Previously Published Citation
Suggested citation
Kamal-Ahmed, Ishrat. (2018). The Association of Social Disparity with Invasive Group A Streptococcus Infections in Minnesota Using Census Tract Level Socioeconomic Status, 1996-2016. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/201691.
Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.