Date of Award

January 2013

Document Type

Open Access Thesis

Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

James L. Hadler

Second Advisor

Linda M. Niccolai


Background: Clostridium difficile infection (CDI) is the most frequent cause of hospital-acquired infectious diarrhea in developed countries. Approximately 19-20% of affected patients will experience a symptomatic recurrence following their first episode of Clostridium difficile-associated diarrhea (CDAD). Risk factors for the initial CDAD episode have been well-documented, however, epidemiologic risk factors for recurrent CDAD have not been described in as much detail. It is hypothesized that initial treatment could be a risk factor for recurrent CDAD. The CT Emerging Infections Program (EIP) conducts laboratory-based surveillance for CDAD in New Haven County, however, routine surveillance does not collect treatment data. Therefore, we conducted a pilot case-control study on CDAD patients at Yale New Haven Hospital (YNHH) during the years 2010-2011 to examine initial CDAD treatment as a risk factor for recurrent CDAD.

Methods: Cases and control patients were identified from CT EIP CDI surveillance data. A patient with recurrent CDAD is defined as having had another positive C. difficile stool specimen between two to eight weeks after the last positive C. difficile stool specimen. Both cases and controls were persons hospitalized at YNHH in 2010-2011; cases had recurrent disease, controls had only a single (incident-only) episode of CDAD. Cases and controls were matched within +/- two years of age. Medical charts of cases and controls were reviewed to collect treatment information related to the incident CDAD episode and severity of the incident case of CDAD, as defined by the Society for Healthcare Epidemiologists of America. Cases and controls were compared on categorical variables with the Fisher's Exact Test or the chi-squared test. Differences in continuous variables were analyzed with the Student's t test. Stratified analyses were conducted by severity of incident infection and sex, using the Mantel-Haenszel chi-squared test.

Results: Eighty-one persons with recurrent CDAD were eligible cases, matched to 122 controls with incident-only CDAD. Persons starting on vancomycin as compared to metronidazole were found to have a lower risk of recurrence, regardless of initial disease severity. Persons treated with a course of vancomycin <10 days had a higher risk of recurrence than those treated longer. Initial treatment with vancomycin appeared to substantially reduce the risk of recurrence: recurrent cases were less likely to have received initial vancomycin therapy (odds ratio [OR]: 0.59; 95% confidence interval [CI]: 0.20, 1.76) and received shorter courses of vancomycin (OR: 2.11; 95% CI: 0.57, 7.82) than controls.

Conclusions: Persons started on vancomycin as compared to metronidazole have a lower risk of recurrence, regardless of initial CDI severity. Vancomycin is being underused at YNHH to initially treat severe CDAD. Our sample size was too small to demonstrate statistically significant differences between recurrent and control cases, but in light of these point estimates, YNHH may want to revisit their established clinical practices for treatment of CDAD. Ongoing surveillance for CDAD and recurrent CDAD may also want to include initial treatment information.


This is an Open Access Thesis.