LTP12 common mistakes
LTP12 · Understand the traps and how to avoid them.
When calculating a reference interval from a small sample, the candidate reports the central 95% interval as mean ± 1.96 SD without checking for normality or using a nonparametric method.
The candidate memorizes the normal-distribution shortcut and applies it to all data, ignoring skewness, outliers, or small sample size, which are common in reference interval studies.
What to do instead
Always inspect the distribution (histogram, skewness, outliers). For n ≥ 120 and non-normal data, use the nonparametric method: sort values and take the 2.5th and 97.5th percentiles (with rank-based confidence intervals). For smaller samples, use robust methods or transference, and document the approach.
When performing a method validation, the candidate calculates the correlation coefficient (r) between the new method and a comparative method and concludes that the methods agree if r > 0.95.
The candidate confuses correlation with agreement; r measures linear association, not bias or systematic difference. High r can coexist with large constant or proportional bias.
What to do instead
Use a difference plot (Bland-Altman) and regression analysis (Deming or Passing-Bablok) to assess bias. Calculate the mean difference and limits of agreement, and compare against clinically acceptable error limits (e.g., CLIA or biological variation).
When a quality control (QC) result exceeds the mean ± 2 SD but is within ± 3 SD, the candidate immediately rejects the run without applying multirule or trend analysis.
The candidate treats any 2 SD violation as a hard rejection, confusing the warning rule (1_2s) with rejection rules (1_3s, 2_2s, R_4s, 4_1s, 10x).
What to do instead
Apply the Westgard multirule algorithm: 1_2s is a warning only; reject only if a rejection rule is violated (e.g., 1_3s, 2_2s, R_4s, 4_1s, 10x). Document the rule violated and the corrective action.
When reviewing a proficiency testing (PT) result that is unacceptable, the candidate files a corrective action that only states 'recalibrated and retested' without investigating root cause or checking for clerical errors.
The candidate focuses on the immediate fix and overlooks the required root-cause analysis, which may reveal a clerical error, method bias, or improper handling.
What to do instead
Follow a structured investigation: verify unit/transcription errors, review calibration and QC history, check reagent and sample integrity, and perform a root-cause analysis (e.g., fishbone diagram). Document corrective actions and preventive measures, and monitor future PT performance.
When collecting a blood specimen for coagulation testing, the candidate fills a citrate tube to less than 90% of the draw volume and still processes the sample.
The candidate underestimates the effect of the blood-to-anticoagulant ratio on coagulation results, or runs out of blood and assumes partial fill is acceptable.
What to do instead
Discard underfilled citrate tubes and recollect. The tube must be filled to the indicated volume (typically 2.7 mL or 3.2% citrate) to maintain the 9:1 blood-to-anticoagulant ratio. Document the rejection and recollect if possible.
When a hemolyzed potassium result is critical, the candidate reports it as a critical value without noting the hemolysis or investigating a possible spurious result.
The candidate follows the critical value policy mechanically and does not consider preanalytical interference from hemolysis, which can falsely elevate potassium.
What to do instead
Check the hemolysis index; if significant, report the result with a comment about possible interference and recommend a repeat specimen. Notify the clinician of the possible spurious critical value and document the actions taken.
When troubleshooting an out-of-control QC result, the candidate changes reagent lot or recalibrates immediately without first checking for a simple error such as a wrong control level, expired control, or transcription mistake.
The candidate jumps to complex fixes and overlooks common preanalytical or clerical errors, wasting time and potentially introducing new variables.
What to do instead
Follow a systematic troubleshooting sequence: verify control lot, expiration, storage, and correct level; check for transcription errors; review the Levey-Jennings chart for shifts or trends; then check reagents, calibrators, and instrument. Document each step.
When selecting a new method, the candidate compares only the manufacturer's claimed precision and does not verify that the method meets the required analytical performance specifications for the intended clinical use.
The candidate relies on marketing claims and does not consider the total allowable error, clinical decision limits, or the laboratory's own quality requirements.
What to do instead
Define the analytical performance specifications based on clinical need (e.g., biological variation, CLIA). Verify precision, bias, and total error against these specifications during method verification. Use appropriate statistical tools and document the acceptance criteria.
When reporting a result for a test with a known interfering substance (e.g., biotin, drugs of abuse), the candidate reports the result without checking for interference or adding a comment.
The candidate is unaware of common interferences or assumes the analytical method is specific, leading to potentially misleading results.
What to do instead
Review the method's interference profile and the patient's history. If interference is suspected, use an alternative method, dilute and reanalyze, or add a comment. Document the interference and any corrective action.
When a specimen is mislabeled or unlabeled, the candidate relabels it using the patient's name from the requisition without verifying identity, and proceeds with testing.
The candidate prioritizes workflow and assumes the requisition is correct, overlooking the risk of misidentification and violating specimen labeling policies.
What to do instead
Reject and recollect mislabeled or unlabeled specimens. Do not relabel. Follow the laboratory's specimen acceptance and rejection criteria, and document the rejection. If recollection is impossible, follow the policy for discrepancy resolution with clinician approval.