American Journal of Infection Control 41 (2013) 1178-81

Contents lists available at ScienceDirect

American Journal of Infection Control American Journal of

Infection Control

journal homepage: www.ajicjournal.org

Major article

Successful reduction in catheter-associated urinary tract infections: Focus

on nurse-directed catheter removal
Michael F. Parry MD a, b, *, Brenda Grant RN b, Merima Sestovic RN b
a
Columbia University College of Physicians and Surgeons, New York, NY
b
Department of Infectious Diseases and Infection Prevention, Stamford Hospital, Stamford, CT

Key Words:
Catheter-associated urinary tract infections
CAUTI
Nurse-directed catheter removal protocol
Hospital-acquired urinary tract infections
Background: Despite using sterile technique for catheter insertion, closed drainage systems, and
structured daily care plans, catheter-associated urinary tract infections (CAUTIs) regularly occur in acute
care hospitals. We believe that meaningful reduction in CAUTI rates can only be achieved by reducing
urinary catheter use.
Methods: We used an interventional study of a hospital-wide, multidisciplinary program to reduce
urinary catheter use and CAUTIs on all patient care units in a 300-bed, community teaching hospital in
Connecticut. Our primary focus was the implementation of a nurse-directed urinary catheter removal
protocol. This protocol was linked to the physician’s catheter insertion order. Three additional elements
included physician documentation of catheter insertion criteria, a device-specific charting module added
to physician electronic progress notes, and biweekly unit-specific feedback on catheter use rates and
CAUTI rates in a multidisciplinary forum.
Results: We achieved a 50% hospital-wide reduction in catheter use and a 70% reduction in CAUTIs over
a 36-month period, although there was wide variation from unit to unit in catheter reduction efforts,
ranging from 4% (maternity) to 74% (telemetry).
Conclusion: Urinary catheter use, and ultimately CAUTI rates, can be effectively reduced by the diligent
application of relatively few evidence-based interventions. Aggressive implementation of the nurse-
directed catheter removal protocol was associated with lower catheter use rates and reduced infection rates.
Copyright Ó 2013 by the Association for Professionals in Infection Control and Epidemiology, Inc.
Published by Elsevier Inc. All rights reserved.

Catheter-associated urinary tract infections (CAUTIs) have
become a target for systematic reduction efforts because of the fact
they are the most common hospital-acquired infection (HAI), and
they lead to excess cost, length of stay, and patient morbidity.
Evidence-based interventions exist to reduce these infections,1-4
and the Centers for Medicare and Medicaid Services will no
longer pay for the costs incurred by the rising case complexity
because of hospital-acquired CAUTIs.5 Although public reporting of
such infections is not yet common, and the National Healthcare
Safety Network (NHSN) has limited its reporting to only symp-
tomatic infections, the total impact of catheter-associated bacteri-
uria (symptomatic and asymptomatic) is high. At Stamford
Hospital, from 2000 to 2008, catheter-associated bacteriuria was
the most common HAI, representing 28.2% of all HAI cases.
* Address correspondence to Michael F. Parry, MD, Stamford Hospital, 30
Shelburne Road, Stamford, CT 06902.
E-mail address: mparry@stamhealth.org (M.F. Parry).
Conflicts of interest: M.F.P. is a consultant to Sanofi-Pasteur and Ethicon and
receives research support from Glaxo-SmithKline. The rest of the authors have
none to report.
Although some episodes of bacteriuria will resolve spontaneously
when the catheter is removed, 40% to 60% of asymptomatic
catheter-associated bacteriuria will persist6,7 and require medical
follow-up and/or specific intervention for symptoms.
At our hospital, surveillance activities over the past decade
and expanded use of the electronic medical record (EMR) have
established baseline facility-wide and unit-specific rates of infection
for multiple conditions and devices. However, despite extensive
hospital-wide educational programs and hand hygiene improve-
ment initiatives, which had resulted in lowering of central line-
associated bacteremia and ventilator-associated pneumonia rates,
CAUTI rates and urinary catheter (Foley) use rates had not fallen
over a 5-year period. These, therefore, became a target for rate
reduction. Our goal was to both reduce indwelling urinary catheter
use and CAUTIs on all patient care units at Stamford Hospital.
METHODS
Stamford Hospital is a 300-bed, community, major teaching
hospital in southwestern Connecticut. Clinical care is provided on

0196-6553/$36.00 - Copyright Ó 2013 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ajic.2013.03.296
 M.F. Parry et al. / American Journal of Infection Control 41 (2013) 1178-81 1179

6 medical and surgical floors with 28 to 32 beds each, consisting of 20,000 0.25
two-thirds single rooms; an all-single room 30-bed maternity and 0.23

catheter use per patient day

0.20
0.22
10-bed pediatric unit; and a 16-bed, mixed medical-surgical 16,000
0.19 0.20 0.20
intensive care unit (ICU). An EMR system (Meditech Incorporated,
0.15 0.16

patient days
Westwood, MA) was installed in October 2005; sequentially 12,000
0.15
0.14 0.15
0.13 0.12 0.13
introducing nurse charting modules; physician order entry (CPOE);
and physician electronic progress note modules in October 2005, 8,000 0.10
May 2007, and June 2010, respectively.
The CAUTI reduction project started in January 2009. This 0.05
4,000
coincided with a definition change for CAUTI by the Centers for
Disease Control and Prevention through NHSN.8 Charting modules
0 0.00
were built by dedicated nurse and medical staff informatics teams. 2009 2 3 4 2010 2 3 4 2011 2 3 4
Both were overseen by a multidisciplinary committee, which 1 1 1
included leadership from the Information Systems Department, Calendar quarter
medical staff (including the Chief Medical Officer, the elected patient days catheter use

President of the Medical Staff, and the Director of Infectious

Diseases/Hospital Epidemiologist), house staff, hospitalists, phar-
macy, nursing, and hospital administration. Order sets, physician
charting modules, and physician education strategy were pre-
sented to, approved, and promoted by the Medical Executive
Committee.
Based on Healthcare Infection Control Practices Advisory
Committee recommendations evolving in 2009 in draft form, and
published in final form in 2010,3 the CAUTI reduction initiative
started with the nurse-directed urinary (Foley) catheter removal
protocol. The nurse-directed protocol started as a pilot project after
agreement on the parameters for removal by the ICU Committee,
ICU physicians, and ICU nursing staff. It began as a paper checklist
for catheter removal (Appendix 1). Once successful, we obtained
Medical Executive Committee approval to roll out as a standing
house-wide protocol. This protocol is now part of the EMR daily
charting process for nursing. Three other elements were included
to reinforce appropriateness of catheter use. (1) A CPOE require-
ment for physician documentation of the criteria for catheter
insertion, mapped to the nursing check list. The physician order for
catheter insertion was linked to the order for a “Foley Maintenance
Protocol,” which included standard nursing care and the nurse-
directed removal protocol. (2) A device-specific charting module
was added to the physician electronic progress notes to serve as
a physician reminder of the catheter’s presence, indications for
its use, and the need to address a device removal plan.3 Biweekly
unit-specific reports on catheter use rates and CAUTI rates were
presented by each unit’s nurse manager in a multidisciplinary
forum (which we called The Quality Briefing). This Briefing was
attended by the Chief Operating Officer, Chief Medical Officer, and
all nurse managers, with pharmacy and infection prevention
representation.
Facility-wide electronic nursing documentation of patient
voiding method (eg, “indwelling catheter”) was required on every
shift. House-wide catheter-days were calculated by EMR extraction
from the nurses’ documentation. Silver hydrogel urinary catheters
(C. R. Bard, Medical Division, Covington, GA) were in use house
wide, and use of a catheter fixation device (StatLock; C. R. Bard) was
encouraged but not mandated. Catheter drainage apparatus con-
sisted of a gravity-dependent collection bag, antireflux valve, and
a sealed catheter-drainage tube connector. Bladder ultrasound
units were deployed on every patient care unit to assess patients for
urinary retention by December 2009.
CAUTIs were determined by the infection preventionists’ house-
wide surveillance activities using NHSN definitions.9 Aggregate
facility-wide and unit-specific CAUTI rates and catheter-use rates
were calculated biweekly for the Quality Briefing by the quality
improvement team and monthly by the hospital epidemiologist for
posting on the hospital intranet. These data were also presented at
Fig 1. Quarterly, hospital-wide urinary catheter use per patient-day (solid line) over
the 3-year study period. Bars represent total hospital patient days per quarter.
scheduled department head, nursing leadership, patient safety, and
medical staff meetings. Regular interdisciplinary meetings were
held to promote the program, determine further strategy, and
review outcomes.
Statistical trends for rate reduction in catheter use and CAUTI
were assessed using Poisson regression and linear regression
analysis. Analyses were done using SAS 9.2 (SAS Institute, Cary, NC)
and Microsoft Excel (Microsoft Corp, Redmond, WA). A 95% confi-
dence interval was used for significance testing.
RESULTS
The patient population evaluated over the 36-month study
period (January 2009 to December 2011) totaled 181,785 patient-
days (monthly average, 5,050  252 patient-days) and 30,747
catheter-days (monthly average, 854  207 catheter-days). Month-
to-month variation in patient-days was only 5%.
Hospital-wide use of indwelling urinary catheters (Fig 1), using
linear regression analysis, was reduced by 50.2%, from 0.223
catheters/patient-day to 0.112 catheters/patient-day over the
36-month period. This was equivalent to a 4.1% reduction in cath-
eter use per month (95% confidence limits: 0.8% to 7.3%).
Catheter use reduction on individual patient care units ranged from
4% (on maternity) to 75% (on telemetry) over the 3-year period.
Using Poisson Regression analysis, we found that CAUTI rates per
catheter-day fell by 3.3% per month over the 36-month period (95%
confidence limits: 1.29% to 5.37%) (Fig 2). The CAUTI rate per
patient-day fell by 5.29% per month (95% confidence limits: 3.24%
to 5.33%). Reduction in infections per patient-day was more
dramatic as a result of the combined reduction in infection rates
and catheter use.
DISCUSSION
We successfully reduced the facility-wide use of indwelling
urinary catheters by 50% and their associated infections by
a statistically significant 3.3% per month over a 36-month period
with our multifaceted program. The critical component of this
initiative was the nurse-directed catheter removal protocol. It not
only reduced urinary catheter utilization rates and ultimately
CAUTIs, but it also resulted in a culture change for the facility,
enhancing teamwork and ownership among the disciplines
involved in the process. The protocol was facilitated by embedding
1180 M.F. Parry et al. / American Journal of Infection Control 41 (2013) 1178-81

4000 6 witnessed accompanying this process at Stamford Hospital led to

3500 4.78
5.42 a celebration of “Foley-free days” in our ICU. Aggressive imple-
4.27
5 mentation of the nurse-directed catheter removal protocol is the

rate per 1000 cath-days

3000 3.96 most important intervention in reducing indwelling urinary cath-
4 eter utilization on the patient care unit.
catheter days

2500
3.79
2000 2.60 2.76 3
3.20
2.25 2.10 Acknowledgment
1500
2
2.12
1000 The authors thank the hospital administration, the Nursing
1.53
1 Service, and the Medical Staff at Stamford Hospital for their support
500
in our efforts to reduce hospital-acquired infections; Joan Olson for
0 0 her secretarial assistance; Brian Taylor, PhD, Director of Clinical
2009 2 3 4 2010 2 3 4 2011 2 3 4 Analytics at New York-Presbyterian Healthcare System for his
1 1 1
calendar quarter statistical advice; and Rohit Bhalla, MD, Vice-President for Quality
and Patient Safety at Stamford Hospital for his support.
catheter days CAUTI rate

Fig 2. Quarterly, hospital-wide CAUTI rates (solid line) per 1,000 catheter-days. Bars
represent total catheter-days per quarter.
References

1. Hooton TM, Bradley SF, Cardenas DD, Colgan R, Geerlings SE, Rice JC, et al.
the catheter removal protocol in the physician order to insert or Diagnosis, prevention and treatment of catheter associated urinary tract
maintain an indwelling catheter (through CPOE). The protocol was infection in adult: 2009 Internal Clinical Practice Guidelines from the Infectious
Diseases Society of America. Clin Infect Dis 2010;50:625-63.
further reinforced by the biweekly quality briefing where nurse
2. Lo E, Nicolle L, Classen D, Arias KM, Podgorny K, Anderson DJ, et al. Strategies to
managers presented and discussed their unit rates, thereby prevent catheter-associated urinary tract infections in acute care hospitals.
increasing institutional awareness and unit competition. The Infect Control Hosp Epidemiol 2008;29:S41-50.
3. Gould CV, Umscheid CA, Agarwal RK, Kuntz G, Pegues DA. HICPAC Guideline for
success of similar catheter use reduction efforts has been shown by
prevention of catheter associated urinary tract infections, 2009. Available from:
others.10-13 http://www.cdc.gov/hicpac/pdf/CAUTI/CAUTIguideline2009final.pdf. Accessed
A total of 103 infections was documented over the 36-month October 25, 2012.
period, compared with 174 infections predicted (extrapolated 4. Saint S, Kowalski CP, Kaufman SR, Hofer TP, Kauffman CA, Olmsted RN, et al.
Preventing hospital-acquired urinary tract infection in the United States:
from the first quarter results) resulting in an estimated 71 CAUTIs a national study. Clin Infect Dis 2008;46:243-50.
prevented. With optimal performance, if current rates of catheter 5. Saint S, Meddings JA, Calfee D, Kowalski CP, Krein SL. Catheter-associated
utilization had been maintained for the entire 3-year period, a total urinary tract infection and the Medicare rule changes. Ann Intern Med 2009;
150:877-84.
of 126 infectious could have been prevented. Generally accepted 6. Harding GKM, Nicolle LE, Ronald AR, Preiksaitis JK, Forward KR, Low DE, et al.
direct costs and attributable mortality of hospital-acquired CAUTI How long should catheter-associated urinary tract infection in women be
are $750 per episode and 5% mortality.14 Savings from the program, treated? Ann Intern Med 1991;114:713-9.
7. Gordon DL, McDonald PJ, Bune A, Marshall VR, Grime B, Marsh J, et al. Diag-
therefore, are estimated at $100,000 and 6 lives saved. This does nostic criteria and natural history of catheter-associated urinary tract infec-
not include savings in indirect costs or the additional costs of tions after prostatectomy. Lancet 1983;2:1269-71.
patient morbidity. 8. Centers for Disease Control and Prevention: National Healthcare Safety
Network. CDC/NHSN surveillance definition of healthcare-associated infection
National and regional rates are not available for hospital-wide
and criteria for specific types of infections in the acute care setting. Available
urinary catheter use or CAUTI comparison at this time. However, from: http://www.cdc.gov/nhsn/PDFs/pscManual/17pscNosInfDef_current.pdf.
unit-specific CAUTI rates by year 3 of the program were lower than Accessed October 25, 2012.
9. Centers for Disease Control and Prevention: National Healthcare Safety
NHSN unit-specific means on 4 of 8 patient care units, and catheter
Network. Catheter-associated urinary tract infection (CAUTI) event. Guidelines
use rates were lower than NHSN means on 7 of 8 patient care and procedures for monitoring CAUTI. Available from: http://www.cdc.gov/
units.15 nhsn/PDFs/pscManual/7pscCAUTIcurrent.pdf. Accessed October 25, 2012.
Limitations of our study include the lack of statistical signifi- 10. Kenner LK, Wallace BR, Hagerty A, Urista C, Espinosa L, Shaikh Z. Reduction of
catheter-associated urinary tract infections on medical-surgical units in
cance in CAUTI reduction on individual units because of the small a teaching hospital. Abstract No. 593. Presented at the 21st Annual meeting of
number of symptomatic CAUTIs. We do not have information on the the Society for Healthcare Epidemiology of America, Dallas, Texas, April 1-
impact of this program on the occurrence of asymptomatic bacte- 4, 2011.
11. Hardenstine HH, Rivera K, Fowler KA, Wolfgang J, Rosa Ll. Reduction of
riuria. Over-estimation or under-estimation of infection rate could indwelling urinary catheter-associated urinary tract infections (CAUTI) through
also be attributed to the interpretation of “symptomatic” indicators system-wide interventions. Abstract No. 595. Presented at the 21st Annual
listed by the NHSN definitions,9 especially in complex ICU patients meeting of the Society for Healthcare Epidemiology of America, Dallas, Texas,
April 1-4, 2011.
who may be unresponsive and/or have multiple potential sources 12. Meddings J, Rogers MAM, Macy M, Saint S. Systematic review and meta-
of fever and abdominal pain. Measurement of catheter-days by analysis: reminder systems to reduce catheter-associated urinary tract infec-
abstracting from the EMR was dependent on accurate nurse tions and urinary catheter use in hospitalized patients. Clin Infect Dis 2010;51:
S550-60.
documentation. We actually found this to be more precise than
13. Trovillion EW, Skyles JM, Hopkins-Broyles D, Recktenwald A, Faulkner K,
manual methods, which had been decentralized to patient care Rogers AD, et al. Development of a nurse driven protocol to remove urinary
units, and we confirmed that this was reliable by demonstrating catheters. Abstract No. 592. Presented at the 21st Annual meeting of the
Society for Healthcare Epidemiology of America, Dallas, Texas, April 1-
a less than 5% difference between EMR abstraction and the infec-
4, 2011.
tion preventionists’ audits. 14. Scott RD. The direct medical costs of healthcare-associated infections in US
As clinical programs like ours successfully reduce urinary hospitals and the benefits of prevention. Available from: http://www.cdc.gov/
catheter use by multidisciplinary programs and empower nursing HAI/pdfs/hai/Scott_CostPaper.pdf. Accessed October 25, 2012.
15. Dudeck MA, Horan TC, Peterson KD, Allen-Bridson K, Morrell G, Pollock DA,
staff to remove urinary catheters that are not indicated, CAUTI rates et al. National Healthcare Safety Network (NHSN) report, data summary for
should fall dramatically in most hospitals. The culture change we 2010, device-associated module. Am J Infect Control 2011;39:798-815.
 M.F. Parry et al. / American Journal of Infection Control 41 (2013) 1178-81 1181

APPENDIX 1.

PAPER CHECKLIST FOR CATHETER REMOVAL

Nursing protocol for removal of Foley catheters

Foley catheter checklist
Date: ___________________________ MR#: __________________
Room#: ____________ V#: __________________
Does your patient have a Foley catheter? Y N
Is there an order for the catheter? Y N
Date Foley catheter inserted: ______________________________________
What service is your patient on? (circle) Medical Surgical Other
Which criteria for appropriate use of a Foley catheter does your patient meet?
____ 24 hour urine collection (if unable to obtain by voiding)
____ Epidural Catheter
____ Head injury, acute
____ Skin breakdown (in males not manageable with condom catheter)
____ On “spine precautions”
____ Acute neurogenic bladder
____ Clinical need, ie, chemically paralyzed and sedated
____ Crush injury or pelvic fracture
____ Hemodynamically unstable needing strict I/O (Q 1 hour urine output)
____ Renal/Urology Surgery
____ Colorectal Surgery (collaborate with MD for earliest removal)
____ Abdominal/Pelvic Surgery (collaborate with MD for earliest removal)
____ Comfort for End of life care
If none of the above criteria are met, remove the Foley and alert physician.
Was Foley removed? Y N
Name of physician notified:__________________________________________
If Foley not removed, explain why:___________________________________

MR#, medical room; V#, patient account number; I/O, input/output.