Joom!Fish config error: Default language is inactive!
Please check configuration, try to use first active language

Case Study A3 - Summary of Panel - Answer (Question 3)

TraQ Program of the BC PBCO

Home Case Studies A-Level Case A3: Multiple Antibodies in a Mother Delivering Twins Case Study A3 - Summary of Panel - Answer (Question 3)
Thursday, 23 March 2017

Case Study A3 - Summary of Panel - Answer (Question 3)

Anti-Jka, anti-E, and anti-c have been confirmed but not all with an acceptable degree of probability (see below). The other possible antibodies have been excluded.

p Values

An easy way to calculate p values is to use the 3 pos/3 neg guideline, since this will produce a p value of 0.05:

  • The patient did not react with 5 cells that are Jk(a-) E- c- (1 in the initial panel and 4 in the exclusion panel). The criterion of 3 negatives has been met by all 3 likely antibodies.
  • Anti-Jka: The patient reacted with 4 cells that are Jk(a+) E- c- (3 in the initial panel and 1 in the exclusion panel).
  • Anti-E: The patient reacted with only 1 cell that is E+ Jk(a-) c- (1 in the exclusion panel).
  • Anti-c: The patient reacted with 3 cells that are c+ Jk(a-) E- (2 in the initial panel and 1 in the exclusion panel).

Anti-Jka and anti-c have met the criterion of a p value of at least 0.05, indicating that there is only a 5% chance of getting the results due to another antibody (chance).

However, anti-E has not because the patient reacted with only 1 cell that is E+ but negative for the other 2 presumptive antibodies, anti-Jka and anti-c.

  • To obtain the minimum acceptable p value for anti-E, 2 more cells that are E+ Jk(a-) c- would need to be tested and be positive.

Online p value calculator

See this online tool for calculating p values  - Enter values as follows:

  1. Change Outcome 1 to negative and Outcome 2 to positive
  2. Change Group 1 to antigen-neg (e.g., E-) and Group 2 to antigen-pos (e.g., E+)
  3. The 2 x 2 table would look like this:
      Negative Positive
    E- 5 0
    E+ 0 1
  4. Choose Fisher's exact test
  5. Use a one tailed test  - A one-tail p value is appropriate only when previous data or common sense tell you that a difference, if any, can only go in one direction. Obviously, with antibody identification, we expect cells that are antigen-positive to react (and vice versa).
  6. Hit the calculate button

Caution #1: Statistical calculations do not substitute for carefully examining serologic test results. It is critical to consider p values as just another tool. If acceptable, they provide additional converging evidence to support the conclusion regarding which antibody has been identified.

Caution #2: With complex antibody combinations, some blood banks prefer a minimum p value of 0.01 as an added measure of probability.

Last modified on Wednesday, 29 December 2010 12:57