Census Tract-Level Poverty and Rurality and Shiga Toxin-Producing Escherichia coli Incidence: Connecticut, 2000-2011

BACKGROUND:  Shiga toxin-producing Escherichia coli (STEC) O157 and other STEC strains are a well-known cause of enteric illness. National estimates are that STEC O157 causes approximately 96,534 illnesses every year in the US, with another 168,698 illnesses caused by non-O157 STEC serotypes. Determining economic and sociodemographic factors associated with disease incidence may provide new understandings of the transmission of these pathogens and their potential prevention.

METHODS:  Overall, 764 incident STEC cases were reported in Connecticut from 2000–2011, 471 O157 and 273 non-O157. Cases were geocoded based on home address and linked to census tract poverty and rurality levels using the 2000 census (2000-2005) and 2006-2010 American Community Survey (2006-2011). Census tract-level poverty was categorized into four levels: 0 – 4.9%, 5 – 9.9%, 10 – 19.9%, and >20% of the population living below the federal poverty line. Census tract-level rurality was categorized into quartiles: 0–24.9%, 25–49.9%, 50–74.9%, and >75% of housing units considered rural. Twelve-year age-adjusted overall STEC, O157 and non-O157 incidence rates were calculated for each poverty and rurality category and by race/ethnicity.

RESULTS:  Of the 764 cases, 744 (97.4%) were geocoded. Both census tract poverty and rurality were significantly associated with STEC incidence. Age-adjusted rates of all STEC infections revealed a trend of increasing census tract poverty level and decreasing STEC incidence (p<0.001); residents of the wealthiest census tracts were four times as likely to be diagnosed with STEC compared to residents of the highest poverty census tracts. Age-adjusted rates of all STEC infections showed a trend of increasing neighborhood rurality and increasing incidence (p<0.001); residents of the most rural census tracts were 1.7 times as likely to be diagnosed with STEC compared to residents of the most urban census tracts. The same significant associations were seen among O157 STEC and non-O157 STEC cases separately and were consistent across time, age, and race/ethnicity groups.

CONCLUSIONS:  STEC incidence increased as neighborhood poverty decreased, showing a dose-response relationship with socioeconomic status, and increased as neighborhood rurality increased. These findings can be used to more effectively target education and interventions, especially in high-income neighborhoods, which include more rural neighborhoods in Connecticut. Area-based socioeconomic measures provide additional insights into the epidemiology of infectious diseases and can be used further to elucidate possible control and prevention measures. Future study implications include the need to better understand what risk exposures are driving the differences between different SES groups.