Current issues of ACP Journal Club are published in Annals of Internal Medicine


Increasing age, severe illness, and some treatments were associated with hospital acquired Clostridium difficile carriage and diarrhea

ACP J Club. 1991 Jan-Feb;114:24. doi:10.7326/ACPJC-1991-114-1-024

Source Citation

McFarland LV, Surawicz CM, Stamm WE. Risk factors for Clostridium difficile carriage and C. difficile-associated diarrhea in a cohort of hospitalized patients. J Infect Dis. 1990;162:678-84.



To identify risk factors associated with asymptomatic Clostridium difficile carriage or C. difficile-associated diarrhea. The frequency of stool cytotoxin in patients who developed C. difficile-associated diarrhea was also studied.


A prospective cohort of all consenting patients admitted to a general medicine ward during an 11-month period in 1985 to 1986.


University medical center in Seattle, Washington, USA.


Of 728 patients admitted, 487 (67%) were eligible for the study. Others were excluded because they were expected to stay < 48 hours (n = 172), were too ill to give consent (n = 57), or did not speak English (n = 12). 59 of 428 eligible patients refused, and 29 with an initial positive culture were excluded.

Assessment of prognostic factors

Rectal swab or stool specimens were taken from each patient within 48 hours of admission and every 3 to 4 days thereafter. Stool samples for cytotoxin assay were collected only from culture-positive patients. Patients were interviewed and their medical records reviewed to obtain information on other diagnoses, diet, other medications, and severity of illness.

Main outcome measures

C. difficile carriers were defined as those with at least 1 positive culture but no diarrheal symptoms thereafter. C. difficile-associated diarrhea was diagnosed if the patient had diarrhea that was temporally associated with a positive culture and had no other apparent cause.

Main results

Of 399 patients with negative admission cultures, 83 patients (21%) acquired C. difficile. Of these, 52 (63%) remained asymptomatic, 26 (31%) developed antibiotic-associated diarrhea, and 5 (6%) developed diarrhea with no associated antibiotic use. Stool cytotoxin was recovered from 17 of 31 (55%) patients with C. difficile-associated diarrhea and 8 of 52 (15%) asymptomatic carriers (P < 0.001). In a multivariate analysis, asymptomatic carriage was associated with antacid use (relative risk [RR] 2.04, CI 1.14 to 3.68); and extreme severity of disease (RR 12.53, CI 1.70 to 92.26). Similarly, multivariate analysis showed that C. difficile diarrhea was associated with use of stool softeners (RR 1.74, CI 1.02 to 3.00); cephalosporin use for > 1 week (RR 2.07, CI 1.12 to 3.84); use of gastrointestinal stimulants (RR 3.06, CI 1.67 to 5.60); use of enemas (RR 3.26, CI 1.51 to 7.02); use of penicillin for 8 to 14 days (RR 3.41, CI 1.48 to 7.86); extreme severity of underlying disease (RR 5.18, CI 1.20 to 22.3); age 41 to 60 years (RR 5.76, CI 1.23 to 27.0); age 61 to 75 years (RR 9.56, CI 2.17 to 42.1); and age 76 to 105 years (RR 7.86, CI 1.80 to 34.4).


C. difficile was commonly associated with asymptomatic carriage and nosocomial diarrhea. Intrinsic host factors and agents altering the bowel's normal flora were associated with increased risk for acquisition of C. difficile.

Sources of funding: Not stated.

Address for article reprint: Dr L. V. McFarland, Department of Medicinal Chemistry, BG-20, 311 Bagley Hall, University of Washington, Seattle, WA 98195, USA.


An earlier report describing the same cohort of patients followed from admission to 1 medical ward (1) provided strong evidence that C. difficile is usually a nosocomially acquired organism. Cases clustered in space and time. Patients who acquired C. difficile were more likely to have a roommate with the same strain. The organism was frequently recovered from the hands of personnel caring for infected patients and from environmental surfaces. This report focused on other risk factors for acquisition of asymptomatic or symptomatic C. difficile infection, but did not control for having an infected roommate. Because they are based on a prospective study in a nonepidemic period, these studies avoid some of the potential problems in earlier studies. Care must be used in extrapolating to other settings because the frequency of recognized C. difficile infections varies widely with hospital, ward, and time. A recent report from the Minneapolis Veterans Affairs Hospital (2) emphasized the importance of treating C. difficile as a nosocomial infection. Education, focused on glove use when handling stool, dramatically reduced the incidence of C. difficile diarrhea, showing that simple infection control measures can help prevent this illness.

Andrew T. Pavia, MD
University of Utah School of MedicineSalt Lake City, Utah, USA


1. McFarland LV, Mulligan ME, Kwok RY, Stamm WE. Nosocomial acquisition of Clostridium difficile infections. N Engl J Med. 1989;320:204-10.

2. Johnson S, Gerding DN, Olson MM, et al. Prospective, controlled study of vinyl glove use to interrupt Clostridium difficile nosocomial infection. Am J Med. 1990;88:137-40.