Clostridium difficile Infection in Ohio Hospitals and Nursing Homes During 2006

Article (PDF Available)inInfection Control and Hospital Epidemiology 30(6):526-33 · June 2009with59 Reads
DOI: 10.1086/597507 · Source: PubMed
Abstract
Healthcare data suggest that the incidence and severity of Clostridium difficile infection (CDI) in hospitals are increasing. However, the overall burden of disease and the mortality rate associated with CDI, including the contribution from cases of infection that occur in nursing homes, are poorly understood. To describe the epidemiology, disease burden, and mortality rate of healthcare-onset CDI. In 2006, active public reporting of healthcare-onset CDI, using standardized case definitions, was mandated for all Ohio hospitals and nursing homes. Incidence rates were determined and stratified according to healthcare facility characteristics. Death certificates that listed CDI were analyzed for trends. There were 14,329 CDI cases reported, including 6,376 cases at 210 hospitals (5,217 initial cases [ie, cases identified more than 48 hours after admission to a healthcare facility in patients who had not had CDI during the previous 6 months] and 1,159 recurrent cases [ie, cases involving patients who had had CDI during the previous 6 months]) and 7,953 cases at 955 nursing homes (4,880 initial and 3,073 recurrent cases) . After adjusting for missing data, the estimated total was 18,200 cases of CDI, which included 7,000 hospital cases (5,700 initial and 1,300 recurrent cases) and 11,200 nursing homes cases (6,900 initial and 4,300 recurrent cases). The rate for initial cases was 6.4-7.9 cases/10,000 patient-days for hospitals and 1.7-2.9 cases/10,000 patient-days for nursing homes. The rate for initial cases in nursing homes decreased during the study (P < .001). Nonpediatric hospital status (P = .011), a smaller number of beds (P = .003), and location in the eastern or northeastern region of the state (P = .011) were each independently associated with a higher rate of initial cases in hospitals. Death certificates for 2006 listed CDI among the causes of death for 893 Ohio residents; between 2000 and 2006, this number increased more than 4-fold. Healthcare-onset CDI represents a major public health threat that, when considered in the context of an increasing mortality rate, should justify a major focus on prevention efforts.
infection control and hospital epidemiology june 2009, vol. 30, no. 6
original article
Clostridium difficile Infection in Ohio Hospitals
and Nursing Homes During 2006
Robert J. Campbell, PhD; Lynn Giljahn, MPH; Kim Machesky, MPH; Katie Cibulskas-White, BS; Lisa M. Lane, MS;
Kyle Porter, MAS; John O. Paulson, MS; Forrest W. Smith, MD; L. Clifford McDonald, MD
context. Healthcare data suggest that the incidence and severity of Clostridium difficile infection (CDI) in hospitals are increasing.
However, the overall burden of disease and the mortality rate associated with CDI, including the contribution from cases of infection that
occur in nursing homes, are poorly understood.
objective. To describe the epidemiology, disease burden, and mortality rate of healthcare-onset CDI.
methods. In 2006, active public reporting of healthcare-onset CDI, using standardized case definitions, was mandated for all Ohio
hospitals and nursing homes. Incidence rates were determined and stratified according to healthcare facility characteristics. Death certificates
that listed CDI were analyzed for trends.
results. There were 14,329 CDI cases reported, including 6,376 cases at 210 hospitals (5,217 initial cases [ie, cases identified more than
48 hours after admission to a healthcare facility in patients who had not had CDI during the previous 6 months] and 1,159 recurrent
cases [ie, cases involving patients who had had CDI during the previous 6 months]) and 7,953 cases at 955 nursing homes (4,880 initial
and 3,073 recurrent cases) . After adjusting for missing data, the estimated total was 18,200 cases of CDI, which included 7,000 hospital
cases (5,700 initial and 1,300 recurrent cases) and 11,200 nursing homes cases (6,900 initial and 4,300 recurrent cases). The rate for initial
cases was 6.4–7.9 cases/10,000 patient-days for hospitals and 1.7–2.9 cases/10,000 patient-days for nursing homes. The rate for initial cases
in nursing homes decreased during the study ( P
! .001 ). Nonpediatric hospital status ( P p .011 ), a smaller number of beds ( P p .003),
and location in the eastern or northeastern region of the state ( P p .011 ) were each independently associated with a higher rate of initial
cases in hospitals. Death certificates for 2006 listed CDI among the causes of death for 893 Ohio residents; between 2000 and 2006, this
number increased more than 4-fold.
conclusion. Healthcare-onset CDI represents a major public health threat that, when considered in the context of an increasing
mortality rate, should justify a major focus on prevention efforts.
Infect Control Hosp Epidemiol 2009; 30:526-533
Clostridium difficile is an anaerobic, spore-forming, toxigenic
bacteria that is the most commonly recognized cause of in-
fectious nosocomial diarrhea and one of the most common
healthcare-associated pathogens.
1,2
C. difficile infection (CDI)
presents with a spectrum of severity ranging from uncom-
plicated diarrhea to pseudomembranous colitis that can lead
to toxic megacolon and death; CDI is especially problematic
among elderly individuals, the group in which infection is
most common. Admission to inpatient healthcare facilities
and antibiotic use are associated with an increased risk of
CDI.
3
Because CDI is not a reportable condition nationwide,
there are few national or regional data on its incidence in
the United States (US). Nonetheless, data based on discharges
from acute care hospitals suggest that both the incidence and
severity of CDI are increasing, both in the Midwest and
nationally.
4,5
There are no similar data to indicate the inci-
dence or disease burden of CDI in nursing homes. In recent
years, there has been an increasing trend for US states to
make healthcare-acquired infections publicly reportable. Cur-
rently, approximately half of all states have mandated public
reporting for some healthcare-acquired infections and at least
another dozen are either currently considering legislation or
have study bills in place.
6
Ohio initiated public reporting for
all initial and recurrent cases of healthcare-onset CDI that
occurred in acute care hospital and nursing home patients,
effective January 1 through December 31, 2006. The purpose
of this public reporting system was to better determine the
burden of healthcare-onset CDI infection among Ohio res-
idents and to establish facility-level baseline CDI rates to assist
in identifying unusual disease activity. In the present article,
we summarize the major findings drawn from Ohio’s 2006
From the Ohio Department of Public Health (R.J.C., L.G., K.M., K.C.-W., L.M.L., J.O.P., F.W.S.) and the Center for Biostatistics, Ohio State University
(K.P.), Columbus, Ohio; and the Centers for Disease Control and Prevention, Atlanta, Georgia (C.M.).
Received November 13, 2008; accepted January 16, 2009; electronically published May 5, 2009.
2009 by The Society for Healthcare Epidemiology of America. All rights reserved. 0899-823X/2009/3006-0004$15.00. DOI: 10.1086/597507
c. difficile in ohio during 2006 527
statewide reporting to provide better understanding of the
regional epidemiology and disease burden of CDI. In addi-
tion, we assessed the changes that occurred between 2000 and
2006 in the mortality rate associated with CDI.
methods
All 210 Ohio acute care hospitals and 955 Ohio nursing
homes were required to report the number of CDI cases that
occurred in their facilities between January 1 and December
31, 2006. The long-term care facilities under surveillance were
limited to nursing homes (ie, facilities that housed low acuity
patients who required long-term skilled care) by legislative
mandate. Hospitals and nursing homes reported cases by fax
or telephone to the local health jurisdiction for their area.
Local health departments initially made a paper report to the
Ohio Department of Health (ODH), followed by an electronic
report to the ODH that was made through a secure public
health system. Hospitals or nursing homes that opened or
closed during the study period were not required to report
and are not included in these analyses.
Patient-days became reportable for hospitals on April 1,
2006, retroactive to January 1, 2006. Effective July 1, 2006,
nursing homes were also required to report patient-day in-
formation for this same time period. The reporting period
for cases identified more 48 hours after admission to a health-
care facility that occurred in patients who had not had CDI
during the previous 6 months (hereafter, “initial cases”) and
cases involving patients who had had CDI during the previous
6 months (hereafter, “recurrent cases”) changed from weekly
to monthly beginning July 1, 2006. Only those hospitals and
nursing homes that reported the number of initial and/or
recurrent CDI infections and the number of patient-days for
a given month were included in the analyses. Case and pa-
tient-day reports that were received at ODH by January 30,
2007, were included.
All rates were calculated as the number of cases per 10,000
patient-days. Missing months of data (numerators and/or
denominators) were estimated by imputing the value from
the preceding or subsequent reporting period, if available;
otherwise imputed values were randomly selected from the
distribution of values for similar facilities. Values were only
imputed if facilities reported at least 1 month of data (ie, no
facilities had all 12 months of data imputed). All statistical
analyses used only data that were not imputed. Sensitivity
analyses were performed using only data from facilities with
complete data reported for the entire year. Facilities were
classified as urban, suburban, or rural using 1990 census data,
in accordance with the following classifications: urban coun-
ties contained at least 1 metropolitan area with population
50,000 or greater, suburban counties had urban areas and
were contiguous with urban counties, and rural counties were
all other nonurban, nonsuburban counties. The hospital’s
number of beds, as well as its pediatric and teaching status,
were obtained from the Ohio Hospital Association. The list
of Ohio nursing homes was obtained from the ODH Division
of Quality Assurance. Hospitals and nursing homes were con-
tacted to obtain the facility’s number of beds if this infor-
mation was missing,
Statistical analyses were performed with SAS, version 9.13
(SAS Institute), and Microsoft Excel (Microsoft). Repeated-
measures Poisson regression models were used to evaluate
variables associated with CDI rates and to conduct trend
analyses (for rates of initial cases only). All variables with a
P value less than .20 on univariate analysis were included in
the initial multivariate model. Variables with a P value greater
than .05 were then removed sequentially, starting with the
highest P value. Statistical significance was defined as P less
than .1.
Death data for the years from 2000 through 2006 were
obtained from Vital Statistics at the ODH, and deaths were
classified as caused by CDI if the primary or any other re-
ported cause of death on the death certificate was Interna-
tional Classification of Diseases and Related Health Problems,
10th Revision, code A04.7 (“enterocolitis due to Clostridium
difficile”).
A case patient was defined as a patient who had a laboratory
diagnostic test result positive for C. difficile (enzyme im-
munoassay, cytotoxin, antigen, and/or culture), pseudomem-
branes seen on endoscopy, or a positive histologic test result
from a tissue specimen obtained during surgery or autopsy;
if the case patient was identified by laboratory test results,
the result had to be a first positive result—that is, it had to
have been more than 2 weeks since the patient’s last result
positive for C. difficile. The presence or absence of diarrhea
was not included in the definition for logistical reasons; we
relied instead on recommended laboratory practices that in-
dicate C. difficile testing should be performed only on un-
formed stool. Only healthcare-onset cases of CDI were cap-
tured in this reporting scheme.
A healthcare-onset initial case was defined as occurring in
a patient who had a laboratory diagnostic test, endoscopy, or
biopsy result positive for C. difficile more than 48 hours after
admission to a healthcare facility. A subsequent episode of
CDI that occurred more than 6 months after an initial in-
fection was also classified as an initial infection. A healthcare-
onset recurrent case was defined as occurring in a patient
who had had an initial case of CDI and then, within 6 months
after the initial case, had a subsequent laboratory diagnostic
test, endoscopy, or biopsy result positive for C. difficile.
Monthly patient-days were defined as the sum of the daily
facility census for that month. All patient-specific data used
to determine case assignment (ie, initial or recurrent) were
captured and saved at the facility level, and only monthly
summary data of the number of initial and recurrent cases
(ie, numerators) and the number of patient-days (ie, denom-
inators) were sent to the health department. There was no
formal data sharing agreement between facilities to ensure
that patients were correctly identified as having recurrent or
528 infection control and hospital epidemiology june 2009, vol. 30, no. 6
initial cases of CDI, other than the medical records that ac-
companied patient transfers.
results
During the 12-month reporting period, a total of 14,329 cases
of CDI were reported. Hospitals ( n p 210 ) reported 6,376
cases, of which 5,217 were initial and 1,159 were recurrent.
Nursing homes ( n p 955) reported 7,953 cases, of which
4,880 were initial and 3,073 were recurrent. After adjusting
for missing data, the estimated total number of cases was
18,200; hospitals accounted for 7,000 cases (5,700 initial and
1,300 recurrent cases), and nursing homes accounted for
11,200 cases (6,900 initial and 4,300 recurrent cases). The
mean percentage of acute care hospitals that reported both
the number of initial cases and the number of patient-days
each month was 90% (range, 77%–95%), and the mean per-
centage of nursing homes that reported both the number of
initial cases and the number of patient-days each month was
75% (range, 57%–87%).
The mean number of initial cases for all Ohio hospitals
was 435 cases/month (475 cases/month after adjusting for
missing data). The rate of initial cases (hereafter, “initial case
rate”) ranged from 6.4 to 7.9 cases/10,000 patient-days during
the reporting period (Figure 1). The mean number of re-
current cases for Ohio hospitals was 97 cases/month (108
cases/month after adjusting for missing data). The rate of
recurrent cases ranged from 1.1 to 2.0 cases/10,000 patient-
days for the reporting period (Figure 1). There was a down-
ward trend in the rates of both initial and recurrent cases
during the year. There was concern that, in the early months
of the reporting system, facilities may not have reported 0 as
a number of cases, thereby potentially biasing the case esti-
mates upward, so analyses were conducted using only the 130
hospitals (62%) that reported complete data during the 12-
month period. The decrease in the initial case rate for these
hospitals was not significant (
P p .104
).
The mean number of initial cases for all Ohio nursing
homes was 407 cases/month (575 cases/month after adjusting
for missing data). The reported initial case rate ranged from
1.7 to 2.9 cases/10,000 patient-days for the reporting period
(Figure 1). The mean number of recurrent cases reported for
Ohio nursing homes was 256 cases/month (358 cases/month
after adjusting for missing data) per month. The reported
rate of recurrent cases ranged from 0.8 to 2.4 cases/10,000
patient-days for the reporting period (Figure 1). There was
also a downward trend for the rate of both initial and re-
current cases in nursing homes during the year; this decrease
was statistically significant for initial cases (
P ! .001
). An anal-
ysis restricted to the 401 nursing homes (42%) that reported
complete data yielded similar results.
The mean initial case rate per 10,000 patient-days in hos-
pitals had a strong inverse relationship with the number of
beds (
P p .014
). Rates were highest in hospitals with 24 or
fewer beds (20.6 cases/10,000 patient-days) and lowest in
hospitals with 400–499 beds (5.9 cases/10,000 patient-days)
(Table and Figure 2). The mean initial case rate was slightly
higher in hospitals in rural counties (8.0 cases/10,000 patient-
days) than in hospitals in suburban counties (7.4 cases/10,000
patient-days) or urban counties (7.1 cases/10,000 patient-
days), but this difference was not statistically significant (Ta-
ble). The initial case rate in teaching hospitals was 6.3 cases/
10,000 patient-days, compared with 7.6 cases/10,000 patient-
days in nonteaching hospitals (
P p .063
); the rate was 2.6
cases/10,000 patient-days in pediatric hospitals, compared
with 7.4 cases/10,000 patient-days for nonpediatric hospitals
(P p .027). The rate for initial cases in hospitals was generally
higher for hospitals located in the northeast and east central
regions of Ohio than for hospitals located in other parts of
figure 1. Rates of Clostridium difficile infection in Ohio acute care hospitals ( n p 210) and nursing homes ( n p 955) from January 1
to December 31, 2006, according to month. See Methods for definitions of initial and recurrent cases.
c. difficile in ohio during 2006 529
table. Rate of Initial Cases of Clostridium difficile Infection in Ohio Acute Care Hospitals and Nursing Homes During 2006,
According to Facility Characteristics
Hospitals Nursing homes
(n p 210) (n p 955)
Unadjusted mean rate,
P
Unadjusted mean rate,
P
Characteristic cases/10,000 patient-days Univariate Multivariate cases/10,000 patient-days Univariate Multivariate
County type NS NS .064 NS
Urban 7.1 2.5
Suburban 7.4 1.8
Rural 8.0 2.0
No. of beds .014 .003 NS NS
6–24 20.6 3.1
25–49 12.4 2.5
50–99 9.8 2.2
100–199 7.7 2.2
200–299 6.8 2.3
300–399 7.3 2.5
400–499 5.9 0
x500 6.2 NA
Region .175 .011
!.001 !.001
Northwest 5.7 1.3
West central 5.7 0.9
Southwest 7.2 2.5
Central 6.0 1.8
Southeast 8.0 2.3
East Central 7.9 2.8
Northeast 8.4 2.9
Teaching status .063 NS NA NA
Teaching 6.3 NA
Nonteaching 7.6 NA
Pediatric status .027 .011 NA NA
Pediatric 2.6 NA
Nonpediatric 7.4 NA
note.
An initial case was defined as occurring in a patient who had a laboratory diagnostic test, endoscopy, or biopsy result positive for C.
difficile more than 48 hours after admission to a healthcare facility. A subsequent C. difficile infection that occurred more than 6 months after an
initial infection was also classified as an initial infection. NA, not applicable; NS, nonsignificant ( P
1 .1).
the state. Multivariate analysis showed that nonpediatric hos-
pital status (P p .011 ), a smaller number of beds ( P p
.003), and location in the east central or northeastern region
(P p .011 ) were each independently associated with a higher
initial case rate (Table).
The mean initial case rate in nursing homes was higher in
urban counties (2.5 cases/10,000 patient-days) than in rural
counties (2.0 cases/10,000 patient-days) or suburban counties
(1.8 cases/10,000 patient-days) ( P p .064) (Table). In con-
trast to hospitals, in nursing homes there was no correlation
between the initial case rate and the number of beds (Table
and Figure 2). There were, however, higher rates of initial
cases in nursing homes in the east central and northeastern
regions of Ohio ( P
! .001 ). County-level mean rates for hos-
pitals and nursing homes were positively correlated with one
another for both initial cases (Pearson r p 0.22; P p .054)
and recurrent cases (Pearson r p 0.29; P p .011 ) (data not
shown).
The number of Ohio resident deaths for which CDI was
the primary or any other reported cause of death increased
during the period from 2000 to 2006 (Figure 3). In 2006,
there were 893 deaths for which the death certificate reported
CDI as any cause of death, 528 (59%) of which reported CDI
as the primary underlying cause of death. Age-adjusted mor-
tality rates were consistently higher and increased fastest in
the oldest age groups (data not shown), and they were similar
for male and female subjects. However, rates for white sub-
jects were higher than those for black subjects. Mortality rates
were highest in northeast Ohio.
discussion
Our results indicate a substantial CDI burden for the state
of Ohio, with an estimated total of 12,600 initial and 5,600
recurrent cases. If the data for Ohio (which has a population
of approximately 11.5 million) are extrapolated to the entire
US population (300 million), then there may have been
333,000 initial and 145,000 recurrent cases of CDI nationwide
530 infection control and hospital epidemiology june 2009, vol. 30, no. 6
figure 2. Mean rates of Clostridium difficile infection per month in Ohio hospitals ( n p 210) and nursing homes ( n p 955) from
January 1 to December 31, 2006, according to the number of beds in a facility. See Methods for definitions of initial and recurrent cases.
in 2006. Nonetheless, the true overall burden of CDI in Ohio
and elsewhere may be much greater, because community-
onset disease, which was not included in this report, may
represent up to half of all cases of CDI.
7
This report highlights the significant disease burden of CDI
in nursing homes, where over half of all healthcare-onset cases
of CDI occurred in this study. Furthermore, long-term care
facilities other than nursing homes, such as rehabilitation or
mental health facilities or group homes, were not included
in this study, so the total number of healthcare facility–onset
cases of CDI is likely higher than this estimate. Similar to the
results of other recent reports, we found evidence of a rapidly
increasing mortality rate for CDI during the period from 2000
to 2006; in 2006, there were 893 death certificates that in-
dicated CDI as a cause of death. No other infectious disease
listed as a cause of death showed any increase during the
same time period, and the number of mortality codes in 2006
for CDI was substantially higher than that for any other in-
fectious disease except unspecified septicemia (data not
shown). It is possible that many patients who have CDI re-
corded on their death certificate die while they have the dis-
ease rather than from the disease, whereas other deaths that
truly are attributable to CDI go unrecognized. Nonetheless,
recent reports indicate that the attributable mortality rate for
CDI ranges from 5.7% to 15% for initial, hospital-onset
cases.
8-10
This suggests 297–782 deaths in 2006 in Ohio that
were attributable to initial cases of hospital-onset CDI, which
extrapolates to 7,752–20,000 such deaths nationally.
The overall initial case rate in acute care hospitals observed
in this study, which ranged from 6.4 to 7.9 cases/10,000 pa-
tient-days, is similar to previously reported rates of endemic
CDI in North American hospitals.
11-13
There are no similar
background data against which to compare the overall rate
for initial cases of CDI among nursing home patients, which
ranged from 1.7 to 2.9 cases/10,000 patient-days in this Ohio
study. We observed a lower initial case rate in nursing home
patients than in acute care patients, despite a similar rate of
recurrent cases in both groups, which may reflect the longer
mean length of stay and increased risk of colonization among
nursing home patients rather than a truly reduced risk of C.
difficile transmission. One recent study suggested that more
than 50% of asymptomatic nursing home patients may be
colonized with C. difficile and that such carriage is frequently
associated with a previous case of CDI or antimicrobial ex-
posure.
14
The correlation of county rates for acute care hos-
pitals and nursing homes in our data and the increasing pro-
portion of patients with CDI who are transferred between
acute care and nursing home facilities (as reflected in hospital
discharge data
2
) both suggest significant movement of re-
cently infected or colonized patients between acute and long-
term facilities. In turn, this suggests that the control of a
highly transmissible pathogen such as C. difficile may best be
addressed through communitywide prevention strategies that
align both healthcare quality and public health initiatives.
CDI rates in both acute care and nursing home facilities
appeared to decrease during the study period; however, this
trend was statistically significant only in nursing homes. Al-
though the effectiveness of public reporting of health process
and outcome data is still largely unproven, one goal of public
reporting initiatives is to motivate healthcare facilities and
providers to improve infection control measures
15
—in this
case, to implement CDI prevention measures. However, it is
unknown whether this significant decrease the rate of CDI
in nursing homes represents merely a trend toward baseline
c. difficile in ohio during 2006 531
figure 3. Deaths of Ohio residents during the period from 2000 through 2006 for which enterocolitis due to Clostridium difficile
(International Classification of Diseases and Related Health Problems, 10th Revision, code A04.7) was reported as the primary or any cause
of death. Data are from the Ohio Department of Health, Vital Statistics, death certificates for 2000–2006.
after an earlier statewide outbreak, seasonal variation, or ac-
tual improvement in infection control and antimicrobial use
practices in Ohio nursing homes.
In acute care hospitals, we found an inverse association
between the number of beds and the CDI rate. Typically,
larger hospitals are more likely to provide specialty care to
patients with greater severity of illness, who may in turn have
an increased risk of acquiring multidrug-resistant organisms,
including C. difficile. However larger hospitals may also be
more likely to have better-developed infection control and
antimicrobial stewardship programs and/or the expertise
needed for such programs.
16
Alternatively, smaller hospitals
may be more likely to provide care to patients with demo-
graphic factors other than increased severity of illness that
lead to an increased risk of CDI. For example, advanced age
is one of the most important host factors that increases patient
risk of CDI, and smaller hospitals may be more likely to
provide care to elderly patients. An analysis of national dis-
charge estimates from 2006 indicates that smaller hospitals—
variably defined in terms of urban versus rural, teaching
status, and regional location—had a larger percentage of pa-
tients aged 65 years or older (39.9% [2,140,559 of 5,358,106;
95% confidence interval {CI}, 36.3%–43.6%) than did me-
dium-sized hospitals (33.4% [3,351,891 of 10,024,172; 95%
CI, 31.1%–35.8%]) or large hospitals (33.2% [7,961,687 of
23,988,670; 95% CI, 30.9%–35.5%]; P
! .001 ) (unpublished
analysis of data from Healthcare Cost and Utilization Proj-
ect, Agency for Healthcare Research and Quality, Rockville,
Maryland).
One major limitation of this study was the fact that re-
porting was limited only to healthcare facility-onset cases.
There is increasing evidence that not only is the burden of
community-onset CDI large, but many cases appear in as-
sociation with inpatient healthcare exposures.
7,17,18
Unfortu-
nately, the Ohio public reporting efforts occurred before the
current interim CDI surveillance definitions became avail-
able.
19
The 6-month interval during which a subsequent ep-
isode of CDI was considered to be a recurrent case rather
than a new initial case was longer than the 8-week interval
used in the current definitions. Therefore, under these newer
definitions, the Ohio healthcare facility reporting for 2006
underestimates the number of initial cases. Another limitation
is the relatively large proportion of data that were missing
and therefore imputed for nursing homes (3,247 [40.8%] of
7,953 reported cases), relative to the amount of imputed data
for hospitals (624 [9.8%] of 6,376 reported cases), which
suggests that our calculation of the total burden and rates of
CDI in Ohio nursing homes may be less accurate than our
calculation for hospitals. There was no attempt to stratify
patient populations for risk with regard to age or underlying
illness. We did not include symptom criteria in our case def-
inition but relied instead on recommended laboratory prac-
tices, which indicate that C. difficile testing should be per-
formed only on unformed stool. In particular, the number
of recurrent cases may have been inflated if laboratories rou-
tinely accepted formed stool for testing, because patients
commonly shed toxin after resolution of symptoms. Con-
versely, the number of reported cases may be an underesti-
mate, owing to the insensitivity of commonly used toxin
enzyme immunoassays; in rare instances, severe cases of CDI
532 infection control and hospital epidemiology june 2009, vol. 30, no. 6
that occurred in patients with ileus may not have satisfied
the case definition.
The public reporting described in this study resulted in a
significant resource burden borne by both healthcare facilities
and public health institutions. A cost survey conducted by
use of a questionnaire issued by the ODH estimated the state-
wide personnel costs associated with the reporting of CDI
from January through December of 2006. The estimated total
cost of reporting was $2,486,000 (95% CI, $2,159,000–
$2,813,000). This included $164,000 paid by the ODH,
$295,000 paid by local health departments, $560,000 paid by
acute care hospitals, and $1,467,000 paid by nursing homes.
The decision to make CDI publicly reportable was not ac-
companied by an increase in financial resources to cover these
costs for either healthcare facilities or public health institu-
tions. When considered in the context of the significant fi-
nancial impact of CDI on excess healthcare costs, supporting
surveillance with the ultimate goal of reducing the rate of
CDI may be a wise investment. Given recent estimates,
20,21
it
is likely that the excess healthcare costs for just initial, hos-
pital-onset cases of CDI were between $21 and $41 million
in Ohio and between $548 million and $1 billion when the
Ohio data were extrapolated to the nation.
Given the significant burden in terms of human suffering
and excess healthcare costs associated with CDI, resources
should be directed toward prevention. Evidence-based rec-
ommendations for infection control and antimicrobial ste-
wardship appear to reduce CDI rates in outbreak situations,
including outbreaks caused by the current epidemic strain.
22-24
However, additional data are needed to demonstrate the pre-
ventability of CDI in environments where it is endemic or
in a larger group of hospitals, much as has been demonstrated
with the reduction of other healthcare-acquired infections.
25,26
Although such data may not define the incremental impact
of each intervention, they could provide evidence of what
proportion of infections are preventable in a range of inpa-
tient healthcare settings and set the stage for national efforts
to reduce CDI rates and mortality.
acknowledgements
Potential conflicts of interest. All authors report no conflicts of interest rel-
evant to this work.
Address reprint requests to L. Clifford McDonald, Centers for Disease
Control and Prevention, 1600 Clifton Rd, MS A31, Atlanta, GA 30333
(CMcDonald1@cdc.gov).
The findings and conclusions in this report are those of the author(s) and
do not necessarily represent the views of either the Ohio Department of
Health or the Centers for Disease Control and Prevention.
references
1. Bartlett JG. Clinical practice. Antibiotic-associated diarrhea. NEngl J
Med 2002; 346:334–339.
2. McDonald LC, Owings M, Jernigan D. Clostridium difficile infection in
patients discharged from US short-stay hospitals, 1996–2003. Emerg Infect
Dis 2006; 12:409–415.
3. Noren T, Tang-Feldman YJ, Cohen SH, Silva J Jr, Olcen P. Clindamycin
resistant strains of Clostridium difficile isolated from cases of C. difficile
associated diarrhea (CDAD) in a hospital in Sweden. Diagn Microbiol
Infect Dis 2002; 42:149–151.
4. Ricciardi R, Rothenberger DA, Madoff RD, Baxter NN. Increasing prev-
alence and severity of Clostridium difficile colitis in hospitalized patients
in the United States. Arch Surg 2007; 142:624–631.
5. Redelings M, Sorvillo F, Mascola L. Increase in Clostridium difficile-
related mortality rates, United States, 1999–2004. Emerg Infect Dis
2007; 13:1417–1419.
6. Association for Professionals in Infection Control and Epidemiology.
Healthcare associated reporting laws and regulations. Available at: http://
www.apic.org/am/images/mandatory_reporting/mandrpt_map.gif. Ac-
cessed April 7, 2009.
7. Kutty PK, Benoit SR, Woods CW, et al. Assessment of Clostridium dif-
ficile–associated disease surveillance definitions, North Carolina, 2005.
Infect Control Hosp Epidemiol 2008; 29:197–202.
8. Loo VG, Poirier L, Miller MA, et al. A predominantly clonal multi-
institutional outbreak of Clostridium difficile–associated diarrhea with
high morbidity and mortality. NEnglJMed 2005; 353:2442–2449.
9. Pepin J, Valiquette L, Cossette B. Mortality attributable to nosocomial
Clostridium difficileassociated disease during an epidemic caused by a
hypervirulent strain in Quebec. CMAJ 2005; 173:1037–1042.
10. Dubberke ER, Butler AM, Reske KA, et al. Attributable outcomes of
endemic Clostridium difficile- associated disease in nonsurgical patients.
Emerg Infect Dis 2008; 14:1031–1038.
11. Olson MM, Shanholtzer CJ, Lee JT Jr, Gerding DN. Ten years of pro-
spective Clostridium difficile–associated disease surveillance and treat-
ment at the Minneapolis VA Medical Center, 1982–1991. Infect Control
Hosp Epidemiol 1994; 15:371–381.
12. Miller MA, Hyland M, Ofner-Agostini M, Gourdeau M, Ishak M. Morbidity,
mortality, and healthcare burden of nosocomial Clostridium difficile-associated
diarrhea in Canadian hospitals. Infect Control Hosp Epidemiol 2002; 23:137–140.
13. Miller MA, Gravel D, Mulvey M, et al. Surveillance for nosocomial Clostridium
difficile–associated diarrhea (N-CDI) within acute-care hospitals in Canada:
results of the 2005 nosocomial infections surveillance program (CNISP) study
shows escalating mortality. In: Program and Abstracts of the 16th Annual
Scientific Meeting of the Society for Healthcare Epidemiology of America
(SHEA). Arlington, VA: SHEA, 2006:108. Abstract 151.
14. Riggs MM, Sethi AK, Zabarsky TF, Eckstein EC, Jump RL, Donskey CJ.
Asymptomatic carriers are a potential source for transmission of epi-
demic and nonepidemic Clostridium difficile strains among long-term
care facility residents. Clin Infect Dis 2007; 45:992–998.
15. Fung CH, Lim YW, Mattke S, Damberg C, Shekelle PG. Systematic
review: the evidence that publishing patient care performance data im-
proves quality of care. Ann Intern Med 2008; 148:111–123.
16. Tan ET, Robertson CA, Brynildsen S, Bresnitz E, Tan C, McDonald C.
Clostridium difficile–associated disease in New Jersey hospitals, 2000–
2004. Emerg Infect Dis 2007; 13:498–500.
17. Dubberke E, Yokoe DS, Mayer J, et al. Multicenter study of Clostridium
difficile–associated disease rates from 2000–2006. In: Program and Abstracts
of the 16th Annual Scientific Meeting of the Society for Healthcare Epide-
miology of America (SHEA). Arlington, VA: SHEA, 2008:62. Abstract 52.
18. Chang HT, Krezolek D, Johnson S, Parada JP, Evans CT, Gerding DN.
Onset of symptoms and time to diagnosis of Clostridium difficile–asso-
ciated disease following discharge from an acute care hospital. Infect
Control Hosp Epidemiol 2007; 28:926–931.
19. McDonald LC, Coignard B, Dubberke E, et al. Recommendations for
surveillance of Clostridium difficile–associated disease. Infect Control Hosp
Epidemiol 2007; 28:140–145.
20. Kyne L, Hamel MB, Polavaram R, Kelly CP. Health care costs and mor-
tality associated with nosocomial diarrhea due to Clostridium difficile.
Clin Infect Dis 2002; 34:346–353.
21. Dubberke ER, Reske KA, Olsen MA, McDonald LC, Fraser VJ. Short-
and long-term attributable costs of Clostridium difficile–associated disease
in nonsurgical inpatients. Clin Infect Dis 2008; 46:497–504.
22. Johnson S, Samore MH, Farrow KA, et al. Epidemics of diarrhea caused
by a clindamycin-resistant strain of Clostridium difficile in four hospitals.
NEnglJMed 1999; 341:1645–1651.
23. Muto CA, Blank MK, Marsh JW, et al. Control of an outbreak of infection
c. difficile in ohio during 2006 533
with the Clostridium difficile BI strain in a university hospital using a
comprehensive “bundle” approach. Clin Infect Dis 2007; 45:1266–1273.
24. Valiquette L, Cossette B, Garant MP, Diab H, Pepin J. Impact of a re-
duction in the use of high-risk antibiotics on the course of an epidemic
of Clostridium difficile–associated disease caused by the hypervirulent
NAP1/027 strain. Clin Infect Dis 2007; 45(suppl 2):S112–S121.
25. Pronovost P, Needham D, Berenholtz S, et al. An intervention to decrease
catheter-related bloodstream infections in the ICU. N Engl J Med 2006; 355:
2725–2732.
26. Reduction in central line-associated bloodstream infections among pa-
tients in intensive care units—Pennsylvania, April 2001–March 2005.
MMWR Morb Mortal Wkly Rep 2005; 54:1013–1016.
    • "Clostridium difficile infection (CDI) is a leading cause of hospital-acquired infection (HAI)123. In many areas of the United States, CDI has surpassed methicillin-resistant Staphylococcus aureus as the most common type of HAI [1] with approximately 333,000 initial and 145,000 recurrent hospital-onset cases in the nation [4]. Certain patient populations have a disproportionately higher risk for CDI due to either host factors, frequent antibiotic use or both. "
    [Show abstract] [Hide abstract] ABSTRACT: Clostridium difficile infection (CDI) remains one of the major hospital acquired infections in the nation, often attributable to increased antibiotic use. Little research, however, exists on the prevalence and impact of CDI on patient and hospital outcomes among populations requiring such treatment. As such, the goal of this study was to examine the prevalence, risk factors, and impact of CDI among pneumonia and urinary tract infection (UTI) hospitalizations. The Nationwide Inpatient Sample (2009-2011), reflecting a 20% stratified sample of community hospitals in the United States, was used. A total of 593,038 pneumonia and 255,770 UTI discharges were included. Survey-weighted multivariable regression analyses were conducted to assess the predictors and impact of CDI among pneumonia and UTI discharges. A significantly higher prevalence of CDI was present among men with UTI (13.3 per 1,000) as compared to women (11.3 per 1,000). CDI was associated with higher in-hospital mortality among discharges for pneumonia (adjusted odds ratio [aOR] for men = 3.2, women aOR = 2.8) and UTI (aOR for men = 4.1, women aOR = 3.4). Length of stay among pneumonia and UTI discharges were also double upon presence of CDI. In addition, CDI increased the total charges by at least 75% and 55% among pneumonia and UTI discharges, respectively. Patient and hospital characteristics associated with CDI included being 65 years or older, Charlson Deyo index for comorbidity of 2 or more, Medicare as the primary payer, and discharge from urban hospitals, among both pneumonia and UTI discharges. CDI occurs frequently in hospitalizations among those discharged from hospital for pneumonia and UTI, and is associated with increased in-hospital mortality and health resource utilization. Interventions to mitigate the burden of CDI in these high-risk populations are urgently needed.
    Full-text · Article · Dec 2015
    • "AAD and CDAD have become and remain a serious problem for health care providers, leading to concerns around patient safety, and increased medical treatment costs. The considerable morbidity associated with CDAD results in a high economic burden, with extended length of hospital stays being the main cost driver as patients with CDAD spend on average an extra 7–21 days in hospital, compared with non-infected controls (Campbell et al., 2009; Dubberke and Wertheimer, 2009). Isolation of the patients who develop CDAD can also represent a substantial cost, as well as the closing and cleaning of wards. "
    [Show abstract] [Hide abstract] ABSTRACT: Introduction: Antibiotic-associated diarrhea (AAD) is common and frequently more severe in hospitalized elderly adults. It can lead to increased use of healthcare resources. We estimated the cost-effectiveness of a fermented milk (FM) with probiotic in preventing AAD and in particular Clostridium difficile-associated diarrhea (CDAD). Methods: Clinical effectiveness data and cost information were incorporated in a model to estimate the cost impact of administering a FM containing the probiotic Lactobacillus paracasei ssp paracasei CNCM I-1518 in a hospital setting. Preventing AAD by the consumption of the probiotic was compared to no preventive strategy. Results: The probiotic intervention to prevent AAD generated estimated mean cost savings of £339 per hospitalized patient over the age of 65 years and treated with antibiotics, compared to no preventive probiotic. Estimated cost savings were sensitive to variation in the incidence of AAD, and to the proportion of patients who develop non-severe/severe AAD. However, probiotics remained cost saving in all sensitivity analyses. Conclusion: Use of the fermented dairy drink containing the probiotic L. paracasei CNCM I-1518 to prevent AAD in older hospitalized patients treated with antibiotics could lead to substantial cost savings.
    Full-text · Article · Feb 2014
    • "Similarly, the proportion of Acinetobacter baumannii that was multidrug-resistant was comparable or greater in smaller facilities compared to larger, tertiary facilities. Furthermore, smaller bed size was independently associated with a higher rate of incident Clostridium difficile infection (CDI) cases reported among 210 Ohio acute care hospitals during 2006 [5]. Antimicrobial stewardship has been promoted for all hospitals to help cope with the challenges of CDI and emerging resistance to antibiotics678 . "
    [Show abstract] [Hide abstract] ABSTRACT: Background Antimicrobial stewardship has been promoted as a key strategy for coping with the problems of antimicrobial resistance and Clostridium difficile. Despite the current call for stewardship in community hospitals, including smaller community hospitals, practical examples of stewardship programs are scarce in the reported literature. The purpose of the current report is to describe the implementation of an antimicrobial stewardship program on the medical-surgical service of a 100-bed community hospital employing a core strategy of post-prescriptive audit with intervention and feedback. Methods For one hour twice weekly, an infectious diseases physician and a clinical pharmacist audited medical records of inpatients receiving systemic antimicrobial therapy and made non-binding, written recommendations that were subsequently scored for implementation. Defined daily doses (DDDs; World Health Organization Center for Drug Statistics Methodology) and acquisition costs per admission and per patient-day were calculated monthly for all administered antimicrobial agents. Results The antimicrobial stewardship team (AST) made one or more recommendations for 313 of 367 audits during a 16-month intervention period (September 2009 – December 2010). Physicians implemented recommendation(s) from each of 234 (75%) audits, including from 85 of 115 for which discontinuation of all antimicrobial therapy was recommended. In comparison to an 8-month baseline period (January 2009 – August 2009), there was a 22% decrease in defined daily doses per 100 admissions (P = .006) and a 16% reduction per 1000 patient-days (P = .013). There was a 32% reduction in antimicrobial acquisition cost per admission (P = .013) and a 25% acquisition cost reduction per patient-day (P = .022). Conclusions An effective antimicrobial stewardship program was implemented with limited resources on the medical-surgical service of a 100-bed community hospital.
    Full-text · Article · Oct 2012
Show more

  • undefined · undefined
  • undefined · undefined
  • undefined · undefined