§ 493.941 - Hematology (including routine hematology and coagulation).  


Latest version.
  • § 493.941 Hematology (including routine hematology and coagulation).

    (a) Program content and frequency of challenge. To be approved for proficiency testing for hematology, a program must provide a minimum of five samples per testing event. There must be at least three testing events at approximately equal intervals per year. The annual program must provide samples that cover the full range of values that would be expected in patient specimens. The samples may be provided through mailed shipments or, at HHS' option, may be provided to HHS and or its designee for on-site testing.

    (b) Challenges per testing event. The minimum number of challenges per testing event a program must provide for each analyte or test procedure is five.

    Analyte

    Table 1 to Paragraph (b)—Analyte or Test Procedure

    Cell identification
    or white
    .
    White blood cell differential.
    Erythrocyte count.
    Hematocrit (excluding spun microhematocrit).
    Hemoglobin.
    Leukocyte count.
    Platelet count.
    Fibrinogen.
    Partial thromboplastin time.
    Prothrombin time (
    1) An approved program for cell identification may vary over time. The types of cells that might be included in an approved program over time are—

    Neutrophilic granulocytes

    Eosinophilic granulocytes

    Basophilic granulocytes

    Lymphocytes

    Monocytes

    Major red and white blood cell abnormalities

    Immature red and white blood cells

    (2) White blood cell differentials should be limited to the percentage distribution of cellular elements listed above.

    (
    seconds or INR).

    (c) Evaluation of a laboratory's analyte or test performance. HHS approves only those programs that assess the accuracy of a laboratory's responses in accordance with paragraphs (c) (1) through (5) of this section.

    (1) To determine the accuracy of a laboratory's responses for qualitative and quantitative hematology tests or analytes, the program must compare the laboratory's response for each analyte with the response that reflects agreement of either 80 percent or more of ten 10 or more referee laboratories or 80 percent or more of all participating laboratories. The score for a sample in hematology is either the score determined under paragraph (c) (2) or (3) of this sectionBoth methods must be attempted before the program can choose to not grade a PT sample.

    (2) For quantitative hematology tests or analytes, the program must determine the correct response for each analyte by the distance of the response from the target value. After the target value has been established for each response, the appropriateness of the response is determined using either fixed criteria based on the percentage difference from the target value or the number of standard deviations (SDsSD) the response differs from the target value.

    Criteria

    Table 2 to Paragraph (c)(2)—Criteria for Acceptable Performance

    The criteria for acceptable performance are:
    Analyte or test
    Criteria for acceptable performance
    Cell identification90% 80% or greater consensus on identification.
    White blood cell differentialTarget ±3SD based on the percentage of different types of white blood cells in the samples.
    Erythrocyte countTarget ±6%±4%.
    Hematocrit (Excluding spun hematocritshematocrit)Target ±6%±4%.
    HemoglobinTarget ±7%±4%.
    Leukocyte countTarget ±15%±10%.
    Platelet countTarget ±25%.
    FibrinogenTarget ±20%.
    Partial thromboplastin timeTarget ±15%.
    If a laboratory reports a prothrombin time in both INR and seconds, the INR should be reported to the PT provider program.
    Prothrombin time (seconds or INR)Target ±15%.

    (3) The criterion for acceptable performance for the qualitative hematology test is correct cell identification.

    (4) To determine the analyte testing event score, the number of acceptable analyte responses must be averaged using the following formula:

    (5) To determine the overall testing event score, the number of correct responses for all analytes must be averaged using the following formula:

    [57 FR 7151, Feb. 28, 1992, as amended at 58 FR 5229, Jan. 19, 1993; 68 FR 3702, Jan. 24, 2003; 87 FR 41241, July 11, 2022]