Trends in Clostridium Difficile Infection and Associated Risk Factors in an Urban County, Denver, Colorado, 2011-2013

Wednesday, June 22, 2016: 11:10 AM
Tikahtnu A, Dena'ina Convention Center
Anna Oberste , Denver Public Health Department, Denver, CO
Kathryn Henderson DeYoung , Denver Public Health, Denver, CO
Helen Johnston , Colorado Department of Public Health and Environment, Denver, CO
Wendy M. Bamberg , Colorado Department of Public Health and Environment, Denver, CO
Arthur Davidson , Denver Public Health, Denver, CO
BACKGROUND: Approximately 75% of patients with Clostridium difficile infections (CDI), which cause mild to severe diarrhea and can be fatal, have symptom onset outside of healthcare settings. Known risk factors for CDI include recent broad-spectrum antibiotic exposure, advanced age, severe underlying morbidities, immunocompromised status, long-term hospital stays, and long-term-care facility residence. We sought to understand specific trends and risk factors for CDI in Denver County, a large metropolitan county with a population of 645,000.

METHODS:  Active population- and laboratory-based surveillance of CDI occurs in Denver County as part of the Colorado Emerging Infections Program; CDI cases between 2011 and 2013 with completed medical record reviews (22%) were analyzed. Cases were stratified by epidemiologic class. Cases whose stool samples were collected in an outpatient setting or within the first 3 days of a hospital admission were defined as community onset (CO). CO cases were further defined as healthcare facility associated (CO-HCFA) if the patient had an overnight stay in a hospital or long-term care facility in the 12 weeks prior to their positive test, or community associated (CO-CA) if not. Cases whose stool samples were collected in a long-term care facility or after the first 3 days of a hospital admission were defined as healthcare-facility onset (HCFO). Multivariate regression was used to determine associations between age, underlying conditions, medication exposures, and health care exposures for each epidemiologic class.

RESULTS: Compared to CO-CA cases, CO-HCFA cases were more likely to have had surgery (OR 24, 95% confidence interval [CI]: 11-52) in the previous 12 weeks, have a hematological malignancy (OR 4.8, CI: 1.4-16.5), or have been exposed to immunosuppressive medications (OR 2.4, CI: 1.0-3.0).  Compared to CO-CA cases, HCFO cases were more likely to have dementia (OR 18.2, CI: 6.9-48), have had surgery within the previous 12 weeks (OR 7.9, CI: 2.6-23.9), or have congestive heart failure (OR 1.7, CI: 0.8-3.7; p<0.05). More CO-HCFA and HCFO cases had exposure to antibiotics, proton pump inhibitors, and histamine-2 blockers compared to CO-CA cases and had at least one chronic condition.

CONCLUSIONS:  

Within Denver County, CDI in this analysis reflect national trends in age, chronic disease burden, and antibiotic exposure by epidemiologic classification. Although there were significant differences between the risk profiles of facility associated and community acquired CDI cases, no risk factors emerged as a primary driver for CO-CA cases when compared to other epidemiologic classes.