Demographic and socioeconomic characteristics associated with hospital acquired Clostridium difficile infections at Hahnemann University Hospital: identifying risk factors for use of indications to receive a Clostridium difficile vaccine and for the development of a generalizable predictive prevention tool

Chelsea May Weldie

doi:10.17918/8emb-eg31

Clostridium difficile is a gut colonizing, gram-positive, sporulating, anaerobic bacteria that causes disease when toxins are released, leading to gastrointestinal inflammation and other associated side effects. C. difficile infections (CDIs) have been highly associated with age and length of inpatient hospital stay. Hahnemann University Hospital (HUH) serves a unique population of low income, minority, and young patients in a large, urban, tertiary acute care medical center providing this study a unique population of subjects to further test known risk factors of CDI and determine other potential risk factors that may not have been detected by the Centers for Disease Control and Prevention (CDC). We hypothesized that analysis of this cohort would reveal unique risk factors associated with patient characteristics and socioeconomic factors. Using a cross-sectional analysis of medical records from HUH, collected between 2014 and 2016, we analyzed 682 subjects in a 1:2 case-control ratio. Every lab confirmed Hospital-Acquired CDIs (228 subjects) were collected from medical records and considered as cases. Non-CDI controls (454 subjects) were randomly selected from the general hospital census and frequency matched for age and duration of inpatient visits. In the first aim, I conducted a test of proportions (chi-squared) to find if there are any significant differences between cases and controls by sex, race, ethnicity, where people were referred from, or insurance type-all potential confounders. This analysis demonstrated stark differences in insurance types (private versus nonprivate) between HA-CDI cases; 62% were Non-Private in comparison only 38% controls were Non-Private. CDI cases were more likely to be African American or Black, making up 53% of all HA-CDIs versus 48% of Controls. Just over half of all cases were males, the majority being between the ages of 46 and 64 years of age. Most patients with HA-CDI came from their homes before being admitted to HUH (66 %). These results do not reflect the CDC's findings that people over the age of 65 have a higher risk of developing CDI. Secondly, we analyzed potential environmental risk factors for HA-CDI in the Philadelphia and Southern New Jersey zip codes. In this analysis we collected zip code data on mean population, median household income, median age, household size, race, and education. These analyses showed no significant differences of any specific ecological factors, including neighborhood levels of poverty and education status when compared with Controls. This analysis showed that within Hahnemann's catchment area, HA-CDI risk factors do not reflect national trends but do reflect the local demographics. It also found that whether someone is on private versus non-private health insurance could be an indicator of risk for HA-CDI. In the discussion, I review why HA-CDI surveillance may be most effective at the local level focusing on the demographics of the patient population of that hospital. Risk factors considered to be generalized may not be when a hospital's population is not representative of the U.S. general population.

Files and links (1)

Abstract

Metrics

Details

Files and links (1)

Abstract

Metrics

Details

Drexel University Social media