Autoverification Getting Started and Beyond

1
Autoverification Getting Started and Beyond

• Karen A. Gallion
• B.S., M.T. (ASCP)

2
Overview

• Accreditation requirements
• System capabilities
• Laboratory policy
• Getting started
• Continuing the process performance and approval

3
Accreditation

• Review your accrediting organizations policy
  regarding autoverification.
• College of American Pathologists (CAP)
• CLIA
• COLA
• JCAHO

4
CAP
Autoverification is the process by which patient
results are generated from interfaced instruments
and sent to the LIS, where they are compared
against laboratory-defined acceptance parameters.
If the results fall within these defined
parameters, the results are automatically
released to patient reporting formats without any
additional laboratory staff intervention. Any
data that fall outside the defined parameters is
reviewed by laboratory staff prior to
reporting. http//www.cap.org/apps/docs/laborator
y_accreditation/checklists/laboratory_general_sep0
7.pdf (pg 75)
5
CAP

• Is there a policy signed by the laboratory
  director approving the use of autoverification
  procedures?
• Is there documentation that the autoverification
  process was validated initially, and is tested at
  least annually and whenever there is a change to
  the system that could affect the autoverification
  logic?
• For all test results subject to autoverification,
  does the laboratory ensure that applicable
  quality control samples have been run within an
  appropriate time period, with acceptable results?

http//www.cap.org/apps/docs/laboratory_accreditat
ion/checklists/laboratory_general_sep07.pdf (pg
75-77)
6
CAP

• Are results compared with an appropriate range of
  acceptable values prior to autoverification?
• Are results checked for flags or warnings prior
  to autoverification?
• Does the audit trail in the computer system
  identify all test results that were autoverified,
  and the date/time of autoverification?

http//www.cap.org/apps/docs/laboratory_accreditat
ion/checklists/laboratory_general_sep07.pdf (pg
75-77)
7
CAP

• Does the autoverification process include all
  delta checks that the laboratory performs prior
  to manual release of test results?
• Does the laboratory have a procedure for rapid
  suspension of autoverification?

http//www.cap.org/apps/docs/laboratory_accreditat
ion/checklists/laboratory_general_sep07.pdf (pg
75-77)
8
Lab Information System Capability

• What type of LIS do you have?
• Vendor
• Home grown
• What are its limitations in regards to
  accreditation?
• Does it meet all of the requirements?
• If not, can programming be requested easily?
• If not, how cost affective will changes be?
• Do you have a contact for questions regarding the
  system?
• What support does the lab have for database
  maintenance?
• Who is going to be responsible for maintaining
  the files that allow AV to occur or fail?

9
Lab Information System Capability

• What support does the lab have for testing time
  and staff?
• Startup testing vs annual review
• What support does the lab have in regards to
  programming changes?
• On-site vs Vendor
• Software version validation
• Does the system have a mechanism for determining
  effectiveness of AV (TAT report or other metric)

10
Laboratory Autoverification Policy

• Who should be involved in developing the policy?
• Laboratory Director
• Quality Assurance Officer
• Lab IS representative
• Primary Section Heads (areas where AV is going to
  be implemented and used)

11
Marshfield Autoverification Policy

• Excerpt from Proc. 0235-0019-5060-0002 of the
  Marshfield Laboratories Policy and Procedures
  Manual
• SCOPE
• This policy is applicable to all laboratories and
  laboratory personnel within the Marshfield Clinic
  Division of Laboratory Medicine.
• POLICY STATEMENT
• The objective of this policy is to specify the
  design, implementation, and validation of the
  computer based logic mechanism that will allow
  previously defined test results to be reported
  without further intervention from laboratory
  technical personnel. This system will help ensure
  the reliability and timeliness of results since
  every result is passed through the same
  previously verified algorithmic process.

12
Marshfield (continued)

• In addition to test method specific criteria, all
  of the following conditions must be met before
  results are eligible for autoverification
• Quality control has been performed as documented
  in start-up for autoverification.
• Specimen ID is not a temporary ID.
• The specimen ID has a status of "arrived in
  Lab".
• No data alarms for instrument of specific test
  are present.
• Result is not included as part of a
  multiple-run situation.
• Reflex result, change result, and calculation
  actions pass without error or user intervention
  is not required.
• All result fields are filled.
• All results fall within the previously defined
  range for autoverification of that test, as
  determined by the laboratory director or
  designee.
• No delta-check errors are present.

13
Marshfield (continued)

• EATS criteria are established for every test
  result that can be autoverified. Individual test
  statements can be written for single performable
  tests or EATS can be grouped under a larger
  performable code, i.e. HGMP, with smaller
  performable test codes such as HGB or PLT.
• Laboratory Information Systems (LIS) personnel
  are responsible for writing and testing the EATS
  statements prior to releasing them for use with
  patient tests. When test statements are written
  as components of a larger performable test code,
  the identical statements are valid for single
  performable tests and, as such, do not require
  revalidation by LIS personnel. In addition, LIS
  personnel can copy statements that they have
  tested to an equal performable code in use at a
  Marshfield Clinic regional center laboratory
  without the need to revalidate the statements.

14
Marshfield (continued)

• EATS statements must be revalidated whenever a
  change is made that could affect the
  autoverification logic. LIS personnel will make
  the requested changes, test the changes prior to
  implementation, and then send the updated EATS
  statements to the requesting manager/section
  manager. This manager is then responsible for
  updating the Autoverification Log as well as
  pulling some patient examples to verify the EATS
  changes.
• All EATS statements must be reviewed at least
  annually.

15
Marshfield (continued)

• PROCESS TO IMPLEMENT AUTO VERIFICATION
• The steps required for moving a test to auto
  verification are as follows
• 1. Identify result for auto verification.
• 2. Establish criteria for supernormal range (this
  can be something other than the normal range such
  as linear range, 120 of normal range, etc., as
  determined by responsible technical
  manager/clinical Ph.D.).
• 3. Complete auto verification documentation for
  analyte. Review with operations manager.
• 4. Work with LIS to install (EATS entries, coding
  expectations).
• 5. Verify performance in test mode. Look at
  decision points and all cases where results
  should not auto verify to ensure performance and
  reliability of resulting. File all validation
  documentation.

16
Marshfield (continued)

• 6. Initialize production performance through LIS.
  
• 7. Verify performance at least annually
  thereafter, or when there are changes to the auto
  verification system, and document with Patient
  Inquiry hardcopy. All records for auto
  verification should be documented by instrument.
• Each of your policies should reflect your
  laboratory practice.

17
Getting Started

• Start small
• Look for a test that has impact, but is
  controllable has limited variables
• Evaluate time saved vs time spent in set up and
  activation

18
Getting Started (continued)

• Create your protocol for request and set up
• Who is responsible for this?
• Have an easy request process
• Have realistic time tables for testing/activation

19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
Getting Started (continued)

• Set testing time
• Work with the lab section to set known
  evaluation/testing times
• Review testing promptly

23
Getting Started (continued)

• Evaluate test results
• Did samples pass if should
• Did samples fail if should
• Did appropriate tags occur
• Make sure all areas of the AV range have been
  evaluated

24
Review example
25
Getting Started (continued)

• Have a troubleshooting mechanism
• Can you find when AV was activated
• Can you find what failed/why
• Audit trail

26
(No Transcript)
27
(No Transcript)
28
(No Transcript)
29
Continuing the Process

• Activation of go live
• Set date/time for go live (this may be only
  during days at start or even a couple of hours)
• Evaluate in a timely fashion
• Have some type of sign off documentation

30
Documentation example
31
(No Transcript)
32
Continuing the Process (continued)

• Evaluate impact on system
• Has TAT been reduced
• Are staff able to focus on abnormal patient
  results
• Where do you go from here?

33
Marshfield Laboratories

• Coagulation (all interfaced instruments
  includes regional centers)
• PT
• APTT
• FIB
• 10 sites using AV for coagulation

34
Marshfield Laboratories

• Hematology (all large centers)
• Hemograms w/ Plt and Autodiffs
• Individual tests (WBC, RBC, PLT, HGB, HCT)
• 9 sites using AV for Hematology

35
Marshfield Laboratories

• Chemistry (main campus)
• Metabolic Basic Panels
• Individual tests (CREAT, BUN, etc.)
• Immunoassays (TSH, TNI, etc.)
• In development
• Metabolic Comprehensive Panel
• Regional Center chemistries

36

• Data from previous evaluation at Marshfield
• 80 Coag
• 95 CEA
• 86 PSA
• 97 TSH
• 50 Hospital Hematology
• 70 Clinic Hematology

37

• QUESTIONS