Extract
Antimicrobial prophylaxis reduces the incidence of surgical site infection
for many
procedures1.
Nonetheless, these medications are not always given
appropriately2. This
failure may stem from a diffusion of
responsibility3:
both the surgeon and the anesthesiologist may assume that the other has
attended to this issue. The root cause, we believe, is the reliance on human
action, which invites avoidable error and demands fail-safe routines.To that end, we have incorporated a check on antimicrobial prophylaxis into
the "time-out" mandated by the United States Joint Commission on
Accreditation of Healthcare Organizations (JCAHO) in the "Universal
Protocol for Preventing Wrong Site, Wrong Procedure, Wrong Person
Surgery."4 Our
time-out comprises four questions, the first three of which are part of the
JCAHO standard:
Antimicrobial prophylaxis reduces the incidence of surgical site infection
for many
procedures1.
Nonetheless, these medications are not always given
appropriately2. This
failure may stem from a diffusion of
responsibility3:
both the surgeon and the anesthesiologist may assume that the other has
attended to this issue. The root cause, we believe, is the reliance on human
action, which invites avoidable error and demands fail-safe routines.
To that end, we have incorporated a check on antimicrobial prophylaxis into
the "time-out" mandated by the United States Joint Commission on
Accreditation of Healthcare Organizations (JCAHO) in the "Universal
Protocol for Preventing Wrong Site, Wrong Procedure, Wrong Person
Surgery."4 Our
time-out comprises four questions, the first three of which are part of the
JCAHO standard:
Who is the patient?What procedure is being done?Which is the correct operative side?Have antibiotics been administered?
Who is the patient?
What procedure is being done?
Which is the correct operative side?
Have antibiotics been administered?
These four questions are summarized by the mnemonic "The 4 Ps,"
which stand for patient, procedure, position (operative side), and
prophylactic antibiotics.
The "4P" method is a fail-safe, a compensatory process, to
prevent undetected error. It is not intended to be the primary means of
ensuring that antimicrobial prophylaxis is administered. This is a practical
distinction. Consider the following situation: a patient is allergic to
penicillin, and the drug of choice may be vancomycin, which must be infused
slowly. If the fact the patient did not receive prophylaxis is discovered only
at the last minute—when asking the fourth question—then there may
be long delay, during which time the patient remains under anesthesia. (What
is worse is that the patient may have received a spinal anesthetic with a
duration of action that is not long enough to account for both the delay and
procedure itself. Thus, the delay in starting the procedure may create a need
for general anesthesia as well.)
Although it is preferable to detect an error even at the last minute than
not to detect it at all ("better late than never"), it is
incumbent on the surgical team to ensure that antibiotics are given in a
timely manner. Written orders accompanying the documentation for admission
(even at an outpatient center) can be helpful in that regard. At the
suggestion of a process action team convened at the Philadelphia Veterans
Affairs Medical Center to optimize the administration of prophylactic
antibiotics, we have further refined our process of administering prophylactic
antibiotics by reiterating a verbal order at the time of signing the surgical
site—i.e., when verifying with the patient the correct limb and the
proposed procedure. This verification is done outside the operating room,
which usually gives ample time for infusing the antibiotic, even if vancomycin
is used.
The primary responsibility for determining whether an antibiotic is
indicated, and choosing which one to administer, rests with the surgeon, we
believe. Our training, knowledge, and experience make us the best arbiter of
this issue. Nonetheless, the anesthesiologist will be the doctor who actually
administers the drug, and therefore he or she should actively participate in
the process of ensuring that correct prophylaxis is applied. Just as surgeons
ask, "Does this patient have adequate anesthesia for surgery to
begin?" the anesthesiologist should be responsible for asking,
"Does this patient have adequate antibiotic prophylaxis for surgery to
begin?" The 4P method is an explicit means for all operating-room staff
to engage this question.
Error prevention increases with redundancy. That is, it is far better that
both the surgeon and the anesthesiologist consider themselves individually
responsible than for neither to do so. The choice of antibiotic and the need
for it for a specific procedure can be defined by surgeons in guidelines
shared with the anesthesia staff, obviating the need for the anesthesiologist
to wrestle with questions that he or she may not feel trained to consider.
We introduce the 4P approach without any explicit studies proving its
validity. Nonetheless, we assert that this method has inherent face validity
and content validity. The face validity refers to the (perhaps obvious) fact
that asking about the administration of antibiotics is an apt means to detect
instances of a failure to administer them. Content validity describes the
sense that the experts—authors, peer reviewers, readers,
etc.—agree that this is reasonable.
Harder to assess is the construct validity—that is, is the 4P
approach effective at preventing errors of failure to administer antibiotics?
In our brief experience using this method over a period of two months, there
were at least three instances in which the failure to give preoperative
prophylactic antibiotics was detected and corrected. This anecdotal
experience, of course, does not prove the effectiveness of the method, as (in
the extreme example) a universal ban on the performance of surgery would also
prevent the occurrence of surgery without appropriate antibiotics.
Effectiveness, then, refers to something more subtle—how many instances
are prevented, and at what cost? This is yet unknown.
The precise number of instances in which doctors would fail to give
necessary antibiotics without the 4P method is unknown. (From our experience,
this would be far more common than operating on the wrong patient or the wrong
limb; although, admittedly, the cost of this antibiotic error is far smaller
too.) Still, if it is assumed that 98% of patients who should receive
antibiotics currently do so and that the 4P method would increase this rate to
99%, that the risk of surgical site infection is 1% with prophylaxis and 2%
without, and that there are five million orthopaedic surgical procedures
performed annually in which prophylaxis is indicated, a case can be made that
the adoption of the 4P method would prevent 500 surgical site infections
annually.
The prevention of 500 infections is not the only benefit. We assert that
adding a question regarding antibiotics to the time-out can make the entire
time-out more effective. The reason for this is that screening tests that are
truly positive only rarely (if ever) may lose their power. It is well known
from the experience of baggage examiners at the airport that it is difficult
to remain attentive when a scant minority of test results (if that) are truly
positive. Likewise, one could imagine that, after a few years of following the
JCAHO Universal Protocol without a single instance of wrong-site surgery
explicitly prevented, the operating-room staff may be lulled into complacency.
Operating by rote, the participants over time may not even listen to the
answers. By incorporating a question regarding prophylactic
antibiotics—a question that is not uniformly answered in the
affirmative, even in the absence of error, as prophylaxis is not always
indicated—we can invigorate the entire JCAHO Universal Protocol and
promote vigilance.
The costs of the method are small, but they are not zero. It is certainly
possible to overwhelm the time-out with so many extraneous but seemingly
reasonable questions, including those that are important but just not timely
(such as questions regarding allergies), that it ceases to be a time-out but
rather a procedure unto itself. It is also possible that an effective
fail-safe method may detract from efforts to ensure a priori that antibiotics
were ordered, in the same way that automobile air-bags are thought to decrease
the use of seat belts.
Our personal cost-benefit estimation strongly favors including a question
regarding antibiotics into the JCAHO time-out, and we have done so. The
Philadelphia Veterans Affairs Medical Center has officially included this
question in its policies. The JCAHO may wish to conduct its own assessment of
validity before making this a national standard. Because the issue of
antibiotic prophylaxis is particularly germane to orthopaedic surgery, our
orthopaedic organizations may wish to consider this issue independently.
Individual surgeons, of course, are free to adopt this approach on their own
without an official mandate, and we recommend that they do.
Bratzler DW, Houck PM; Surgical
Infection Prevention Guidelines Writers Workgroup; American Academy of
Orthopaedic Surgeons; American Association of Critical Care Nurses; American
Association of Nurse Anesthetists; American College of Surgeons; American
College of Osteopathic Surgeons; American Geriatrics Society; American Society
of Anesthesiologists; American Society of Colon and Rectal Surgeons; American
Society of Health-System Pharmacists; American Society of PeriAnesthesia
Nurses; Ascension Health; Association of periOperative Registered Nurses;
Association for Professionals in Infection Control and Epidemiology;
Infectious Diseases Society of America; Medical Letter; Premier; Society for
Healthcare Epidemiology of America; Society of Thoracic Surgeons; Surgical
Infection Society. Antimicrobial prophylaxis for surgery: an advisory
statement from the National Surgical Infection Prevention Project. Clin
Infect Dis. 2004;38:
1706-15.381706
2004
[PubMed][CrossRef]
Silver A, Eichorn A, Kral J, Pickett G,
Barie P, Pryor V, Dearie MB. Timeliness and use of antibiotic prophylaxis in
selected inpatient surgical procedures. The Antibiotic Prophylaxis Study
Group. Am J Surg. 1996;171:
548-52.171548
1996
[PubMed][CrossRef]
Warters RD, Szmuk P, Pivalizza EG,
Gebhard R, Ezri T. Perioperative antibiotic prophylaxis: the role of the
anesthesiologist. Anesthesiology.
2003;9 9:
515-6.9515
2003
[CrossRef]
Medicine is attempting to undergo a transformation in improving quality and
reducing error. The list of organizations (both government and private) vying
to lead the way grows each year. In particular, two organizations, the Joint
Commission on Accreditation of Healthcare Organizations (JCAHO) and the
Centers for Medicare and Medicaid Services (CMS), have shown remarkable
initiative by setting concrete standards and expectations.
The practice of surgery has been a focus of these efforts. In particular,
two parallel initiatives are having an impact on surgeons before their scalpel
touches the skin:
JCAHO: Surgical time-out for wrong-site surgery. (Effective July 1, 2004,
compliance with the "Universal Protocol for Preventing Wrong Site, Wrong
Procedure, Wrong Person Surgery" is required of all Joint
Commission-accredited
organizations1.)CMS: Measures for the prevention of surgical infection recommending that
prophylactic antibiotics be received within one hour prior to surgical
incision2.
JCAHO: Surgical time-out for wrong-site surgery. (Effective July 1, 2004,
compliance with the "Universal Protocol for Preventing Wrong Site, Wrong
Procedure, Wrong Person Surgery" is required of all Joint
Commission-accredited
organizations1.)
CMS: Measures for the prevention of surgical infection recommending that
prophylactic antibiotics be received within one hour prior to surgical
incision2.
Both organizations specifically outline the intended goal and/or outcome,
but both are careful to avoid specific recommendations regarding
implementation. JCAHO "recognizes the need for flexibility to
accommodate the logistical and procedural realities of the full range of
surgical
facilities."3
Dr. Bernstein and Dr. Meller present their method of addressing these two
initiatives. They should be commended for their proactive approach in
addressing two related issues. Their general recommendation to include
"a question regarding antibiotics into the JCAHO time-out" is
useful. The authors' specific recommendations are problematic in two regards:
timing and their use of a mnemonic.
The fourth P (for prophylactic antibiotic) illustrates a problem with
mnemonics. Often the available letter does not represent the best possible
description of the issue at hand. The phrase "prophylactic
antibiotic" does not readily translate into the authors' question,
"Have the antibiotics been administered?"
A more important problem with the fourth P is that it does not address the
issue of timing. To fully address this issue, any preoperative checklist
should include the familiar who (patient name), what (operation), where (which
extremity or body part), and when (time that the antibiotic was
infused). According to the CMS measures for the prevention of surgical
infection, "When was the antibiotic given?" is as
important as "Has the antibiotic been
given?"4 The
CMS recommendations stipulate that it is not enough to give the
antibiotic; it must be infused within one hour prior to surgery. Delivery of
the antibiotic one minute outside this window is considered a compliance
failure and is reported as such by the online database "Hospital
Compare."4
The use of a preoperative checklist has been called a "preoperative
verification process," "universal protocol," and, most
recently, "time-out." The medical use of the phrase
"time-out" is not consistent with the dictionary definition of
"time-out." Regardless of what this process is ultimately called,
some form of preincision checklist is now a mandated function.
The danger in medicine is that we make the checklist more than it is. While
it is clearly here to stay, it is not clearly defined how such checklists
should be categorized from a "systems" standpoint. In other
industries, the use of a checklist is a means of monitoring the effectiveness
of the "system." We believe it is important to use a checklist to
monitor the system rather than to have the checklist serve as the system. Any
variances picked up by a checklist should be viewed as a failure of the
delivery system rather than as a success of the checklist.
Shigeo Shingo, one of the industrial engineers who revolutionized Toyota,
has said that a checklist wouldn't be necessary if people never forgot
things5. Shingo
introduced the concept of Poka-Yoke (Japanese slang for avoiding inadvertent
errors and translated as mistake-proofing). This has led to terms such as
"error management,"
"forcing-function,"6
and
"foolproofization."7
Error management is facilitated by forcing-function devices, which have three
attributes: inspection, setting function, and regulatory function. To
illustrate these concepts, consider the example of an airplane passenger who
enters a bathroom. Inspection is used to determine that there is no light, and
a switch (the setting function) is used to turn the light on, which
automatically locks the door and sets the outside signage to read
"occupied" (the regulatory function). As an added precaution, many
passengers might check to make sure the door is locked. If by chance
the door is not locked, most likely the "system" is broken. Rather
than view this as a success of the final door check, it should be viewed as a
failure of the locking mechanism.
Dr. Bernstein and Dr. Meller state that "there were at least three
instances in which the failure to give preoperative prophylactic antibiotics
was detected and corrected." In the airline industry, this would be
viewed as an identification of three near misses rather than as a success of
the mnemonic. While near misses are preferred to the actual occurrence of
harmful events, use of a checklist as a weak, human-attention-dependent
barrier to error is only a starting place. As the authors note, it is
"not intended to be the primary means of ensuring that antimicrobial
prophylaxis is administered."
Improved inspection techniques, setting functions, and regulatory functions
should be sought. Error detection should be moved as close to the actual time
and place of the error as possible, and should prevent the error in ways that
depend less on human attention and action. There are examples of forcing
functions in
medicine8, but
these responses to error are not routine. In this respect, health care lags
behind other safety-critical industries. Ideally, responses to error will be
improved as a "systems approach" to error reduction leads to the
development of forcing function devices. For example, incorporating "a
regulatory function" (an alarm) into an anesthesia software program has
a greater potential to reduce human error than does a human-dependent
"time-out."
Specific recommendations, once universally adopted, are difficult to amend.
The readership would best use this contribution to the Orthopaedic Forum as a
starting place to stimulate thought rather than as an endorsement of the
method proposed by Dr. Bernstein and Dr. Meller. Their recommendations fall
short of the timing guidelines recommended by the CMS. The introduction of the
recommendations may have a paradoxical effect by adding another
human-dependent element. That is why JCAHO correctly recognizes the danger of
making specific recommendations.
United States Department of Health and Human
Services. Hospital Compare—a quality tool for
adults, including people with Medicare.
.
Last updated September 1, 2005.
2005
www.hospitalcompare.hhs.gov/
Shingo S. The sayings of Shigeo
Shingo: key strategies for plant improvement. Dillon AP, translator.
Cambridge, MA: Productivity Press; 1987.
1987
Norman DA. The design of everyday
things. New York: Doubleday; 1990.
1990
Berwick DM, Godfrey AB, Roessner J.
Curing health care: new strategies for quality improvement. San
Francisco: Jossey-Bass; 2002.
2002
Grout JR. Preventing medical errors by
designing benign failures. Jt Comm J Qual Saf.
2003;29:
354-62.29354
2003
[PubMed]