1314. Evaluation of Differences in Population-Based Incidence of Clostridium difficile Infection across Diverse U.S. Geographic Locations, 2010
Session: Oral Abstract Session: Challenges in C. difficile Infection Surveillance
Saturday, October 20, 2012: 11:00 AM
Room: SDCC 29 ABCD

Background:  C. difficile infection (CDI) diagnosis and treatment are no longer restricted to hospital settings.  Accurate estimates of CDI nationally will require accounting for differences in population- and diagnostic-specific factors (e.g., age, testing practices) that influence incidence measures. We analyzed population-based data to identify these factors.

 Methods:   Population-based surveillance for persons ≥ 1 year of age was conducted in 21 counties in 7 U.S. states (8.5 million pop) in 2010.  A CDI case was defined as a positive C. difficile toxin or molecular assay on a stool specimen from a person without a prior positive assay in the past 8 weeks. Cases were classified as community-associated (CA) if stool was collected as an outpatient or ≤3 days of admission in a person with no overnight stay in a healthcare facility in the past 12 weeks; otherwise they were classified as healthcare-associated (HA).  We queried participating laboratories about molecular diagnostics (e.g., PCR) utilization. The U.S .Census and Area Resource File provided county-level demographics and healthcare utilization data.  Two regression models (CA- and HA-CDI) were built to evaluate factors associated with higher CDI incidence.  Site-specific incidence was calculated using 2010 U.S. Census and adjusted based on the regression models.

Results:  Of 10,062 cases identified, 32% were CA.  Overall CDI incidence per 100,000 was higher among persons who were female (137 vs. 99; P=0.01), white (140 vs. 75; P<.001), or > 64 years (632 vs. 59; P<.001).  Unadjusted incidence varied by site; CA-CDI ranged from 28–79/100,000 and HA-CDI ranged from 70–155/100,000.  By multivariate analysis independent predictors of higher CA-CDI incidence were age, race, sex, and PCR usage; for HA-CDI only age was a statistically significant predictor. After adjusting for relevant factors, the range of incidence narrowed greatly; CA-CDI ranged from 29–42/100,000 and HA-CDI ranged from 59–111/100,000 (Figure).

Conclusion:  Differences in CDI incidence across sites can be partially explained by differences in PCR usage, age, race and gender, especially for CA-CDI cases.  Variation in antimicrobial use and infection control practices, not captured in this analysis, may contribute to the remaining differences in CDI incidence.

 

Fernanda Lessa, MD1, Yi Mu, PhD2, Jessica Cohen, MPH1,3, Ghinwa Dumyati, MD, FSHEA4, Monica M. Farley, MD5, Lisa Winston, MD6, Kelly Kast, MPH7, Stacy Holzbauer, DVM8, James Meek, MPH9, Zintars G. Beldavs, MS10, L. Clifford Mcdonald, MD1, Scott Fridkin, MD, FSHEA1 and EIP CDI Surveillance Investigators, (1)Division of Healthcare Quality Promotion, Centers for Disease Control and Prevention, Atlanta, GA, (2)Centers for Disease Control and Prevention, Division of Healthcare Quality Promotion, Atlanta, GA, (3)Atlanta Research and Education Foundation, Atlanta, GA, (4)Infectious Diseases, University of Rochester, Rochester, NY, (5)Division of Infectious Diseases, Emory University School of Medicine, Atlanta, GA, (6)University of California, San Francisco/San Francisco General Hospital, San Francisco, CA, (7)Colorado Department of Public Health and Environment, Denver, CO, (8)CDC CEFO assigned to the MN Dept. of Hlth, St. Paul, MN, (9)CT EIP, New Haven, CT, (10)Oregon Health Authority, Portland, OR

Disclosures:

F. Lessa, None

Y. Mu, None

J. Cohen, None

G. Dumyati, Pfizer: Collaborator, Research support

M. M. Farley, None

L. Winston, None

K. Kast, None

S. Holzbauer, None

J. Meek, None

Z. G. Beldavs, None

L. C. Mcdonald, None

S. Fridkin, None

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 17th with the exception of research findings presented at the IDWeek press conferences.