![]() This usually involves demonstrating overlap of the confidence interval for the laboratory’s estimate with the manufacturer’s claim, or demonstrating that laboratory estimate falls within the confidence interval of the manufacturer’s claim. Then, the acceptability of performance is assessed by comparing the two estimates. Typically, laboratory verification of a manufacturer’s claim involves the laboratory’s performing a “smaller” experiment than that performed by the manufacturer to establish the claim. Other programs, such as MedCalc, Analyze-it, or Minitab also support probit analysis and may be more familiar and available to clinical laboratory scientists. We expect that the results here are within the uncertainty or confidence ranges of the “claimed” LoDs, but determining those confidence limits requires more complicated calculations with statistical software, such as the Maplesoft program described by Vaks. See even more stories about COVID-19 Laboratory Challenges. For the EP-17A2 worked example, our estimate was 0.064 CFU/mL compared to 0.077 CFU/mL.For the literature example, our estimate was 0.006 PFU/mL compared to 0.005 PFU/mL.For the EP-17A2 example, our estimate for LoD was 39.2 compared to 36.5.For purposes of exploring the use of probit analysis, part 1 of this series described how the method of Finney was implemented via an Excel spreadsheet to compare estimates of LoD with the results of 2 examples in the CLSI EP17-A2 document and 1 example from the literature. ![]() Probit analysis is a statistical technique that fits the “positivity rate” from replication experiments to a sigmoid or S-shaped curve that represents the cumulative probability function for a Gaussian distribution. Neill CareyĪn earlier post on this website discussed the use of probit analysis for determining the limit of detection (LoD) for Nucleic Acid Amplification Tests (NAAT). Probit Analysis 2: Issues with Verification of Limit of Detection James O. An earlier post on this website discussed the use of probit analysis for determining the limit of detection (LoD) for Nucleic Acid Amplification Tests (NAAT).
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |