Validating a Reference Interval

1
Validating a Reference Interval

• Reference intervals (commonly referred to as
  normal ranges) for FDA approved reagents and
  kits are provided by manufacturers as a guide to
  the laboratory.
• A laboratory may transfer these intervals by a
  validation process using a similar population
  (normal, therapeutic) as that stated in the
  package insert. Transference can be for an
  analyte using the identical analytical system as
  the manufacturer or a system that differs.

Lab Pointer
Contributed by Donna D Castellone MS,
MT(ASCP)SH New York Presbyterian Hospital Weill
Cornell Medical Center New York City, New York
Revised mmiiii
www.CLOT-ED.com
2
Validating a Reference Interval

• Importance
• Reference intervals that are health associated
  are used for interpreting patient test results
  (normal vs abnormal)
• Reference intervals established for therapeutics
  are used to monitor patient treatment, for
  example, warfarin or heparin therapy
• Mean of reference interval for Prothrombin Time
  is used for calculating the INR

Lab Pointer
Revised mmiiii
www.CLOT-ED.com
3
Validating a Reference Interval

• NCCLS C28-A2 (2000)
• Laboratories may transfer (use) a reference
  interval prepared by a manufacturer if the
  laboratory performs validation studies. This is
  considerably easier to do than actually setting a
  reference interval for which NCCLS recommends a
  minimum of 120 reference subjects. The
  validation process must be as consistent as
  possible with pre-analytical and analytical
  conditions encountered by the manufacturer.
  Should demographics of laboratory and
  manufacturer populations differ, then
  transference can not be used.

Lab Pointer

• NCCLS H21-A4 (2003)
• Establishing a reference interval for coagulation
  assays should occur in the following situations
• For a new analytical method
• When a lot number of reagent is changed for an
  existing analytical method
• Change in instrumentation (old to new instrument)
• Change in blood collection system
• Annually

Revised mmiiii
www.CLOT-ED.com
4
Validating a Reference Interval

• Validation studies for transference require steps
  similar to those used for establishing a
  reference interval. Sections 4 5 of NCCLS
  C28-A2 outline these steps in detail. It is
  acceptable to use a minimum of 20 reference
  individuals, the selection of whom is critical.
  A questionnaire can be used in order to exclude
  or partition individuals (a priori sampling). In
  a posteriori sampling, exclusion and partitioning
  occur after sampling and testing have been
  performed. The reference interval is acceptable
  if no more than 2 of the 20 values (10) fall
  outside the manufacturers limits.

Lab Pointer
Questionnaire
Revised mmiiii
www.CLOT-ED.com
M Ledford-Kraemer
5
Validating a Reference Interval

• Be sure that your laboratorys reference interval
  reflects your patient population
• DO NOT use Emergency Room or Pre-operative
  patients
• These patients may have acute phase reactants
  (proteins that are elevated in times of stress,
  infection or inflammation such as FVIII, VWF, and
  Fibrinogen that will falsely shorten a PT APTT)
• Use an equal distribution of males and females
• Run over a period of time to introduce
  intra-laboratory variables and reduce bias

Lab Pointer
Remember that validating reference intervals is
one of the most important coagulation tasks you
will perform
Revised mmiiii
www.CLOT-ED.com