Technical Information


Recommendations for Testing a New Lot of Hematology Control and Establishing an Acceptable Range

Assay values on a new lot of control should be confirmed before it is put into routine use. Test the
new lot when the instrument is in good working order and quality control results on the old lot are
acceptable. The laboratory recovered mean should be within the assay range.

For greater control sensitivity each laboratory should establish its own mean and acceptable
range and periodically reevaluate the mean. The laboratory range may include values outside of
the assay range. Target values not listed on the Assay Sheet may be established by the user if
the control is suitable for the method.

TEst LInk
  1. Test the new lot in parallel with the old lot 2 to 3 times a day for a minimum of 5, preferably 10
    days.
    1. Verify that there are no trends and that the precision is acceptable.
    2. Delete obvious outliers (values > 3 SD from the mean). Repeat testing as needed.
  2. Calculate the mean for each measured parameter on the new lot. Calculate indicies from the
    measured parameters.
    1. A provisional mean may be calculated from 10 data points but recalculate the mean after 20 data points are accumulated.
    2. The laboratory mean should be within the assayed range.
  3. Calculate control limits from the laboratory mean and long-term standard deviation. The longterm
    SD is the average SD from previous lots. The SD for each parameter should be very
    similar from lot to lot at the same control level.
    1. Do not use the SD from parallel testing to establish a range. This SD is too narrow for control limits because it is obtained over a short time compared to the shelf life of the control.
    2. Typically a range is calculated from 2 to 3 times the SD value. For example if the WBC mean is 10.0 and 1 SD is 0.2, a range based on 2.5 SD is 10.0 + 0.5 (0.2 x 2.5). The range used depends on the clinical needs of the laboratory for error detection.
  4. Calculate and evaluate statistical parameters (mean, standard deviation, coefficient of variation) at least once per month to ensure continued acceptable performance. If the mean has changed significantly, usually more than 1 SD, check calibration and adjust the target mean as needed.

References:

  1. National Committee for Clinical Laboratory Standards. Internal Quality Control: Principles and Definitions; Tenative Guideline. NCCLS document C24-T. Villanova, PA; 1987.
  2. National Committee for Clinical Laboratory Standards. Performance Goals for the Internal Quality Control of Multichannel Hematology Analyzers; Proposed Standard. NCCLS document H26-P. Villanova, PA; 1989.
  3. Cembrowski GS, Carey RN. Laboratory Quality Managelemt QA and QC Chicago, IL: ASCP Press; 1989.