Reengineered Workflow in the Anatomic Pathology Laboratory: Costs and Benefits by ProQuest


More Info
									       Reengineered Workflow in the Anatomic Pathology
                                                        Costs and Benefits
                                              Erin E. Grimm, MD; Rodney A. Schmidt, MD

T   echnology and workflow in the anatomic pathology
     (AP) laboratory have been relatively static during the
last few decades, even though the volume of cases pro-
                                                                           clude mistakes in patient identification and mistakes in
                                                                           designation of tissue site. This study found that the prean-
                                                                           alytic error rate for surgical specimens was 4.3 per 1000
cessed and the complexity of cases have increased. In oth-                 at a major US academic hospital. The College of American
er industries, static processes in a changing environment                  Pathologists2 has also addressed the question of error rates
cause businesses to lose their competitive edge. Robust                    through a Q-probes survey that asked laboratories, most
business models adapt to changing environments and take                    of which were located in the United States, to report the
advantage of new tools. With tightening budgets through-                   number of corrected surgical pathology reports during a
out health care, the AP Department at University of Wash-                  designated period. Laboratories reported an average of 2
ington (UW) in Seattle has critically evaluated workflow                    amendments per 1000 surgical reports. Twenty percent of
in the AP laboratory and reengineered workflow using bar                    these amendments were due to patient identification er-
code technology and tailored software applications. The                    rors.
result has been improved efficiency that has created cost                      The 2 studies are snapshots of errors occurring at par-
savings while improving patient safety. This presentation                  ticular steps, but these studies do not accurately reflect
highlights common workflow problems experienced by                          total error. They do not capture errors corrected before
AP laboratories and describes how UW has used hard-                        signout. Neither do they retrospectively review randomly
ware and software to automate as many manual processes                     selected cases for accuracy. No study has addressed total
as possible. Lastly, we discuss questions for the future of                error rates.
laboratory automation in AP.                                                  Roger Resar,3 a leader in the field of quality improve-
   One of the flaws in current laboratory workflow is that                   ment methodologies, related average error rates to the
the basic processes of most laboratories were designed                     methodology used by the system. With systems based on
during a time when laboratories processed fewer cases per                  humans following a defined protocol, errors occur at rates
day. With increasing volume, small inefficiencies in the                    of a few per 100. An example is a protocol used in an
workflow are magnified. Most laboratories are not maxi-                      emergency department for treatment of community-ac-
mizing use of their resources, especially their most valu-                 quired pneumonia, in which the protocol is available for
able resource, labor.                                                      use but is not required and no checks are in place to see
   The second problem relates to patient safety. Medical                   that the protocol is used correctly. A more intensive meth-
errors occur in the pathology department, just as they do                  od is a protocol with error mitigation techniques. With this
in all areas of medicine. Most laboratories rely on the vig-               method, an error rate of a few per 1000 is achievable.
ilance of pathologists and staff to catch these errors. How                Transfusion medicine commonly uses such methods. Pro-
well does vigilance work? One study by Makary1 quanti-                     cedures require that the patient’s blood type, as deter-
fied preanalytic errors (errors occurring before the speci-                 mined by laboratory testing, is later checked against his-
men reaches the laboratory). Examples of such errors in-                   torical patient blood type records to make sure the blood
                                                                           types match. The methods we have discussed so far
   Accepted for publication October 1, 2008.                               achieve an error rate of a few per 1000.
   From the University of Washington Medical Center, Seattle (Dr              In the AP laboratory, what 
To top