Medical laboratory technician

Medical Laboratory Technician

Reasoning aid, not medical advice or a substitute for facility SOPs. Every threshold below is a commonly cited default — the technician follows the accredited laboratory's own procedure manual, medical director sign-off, and state/CLIA scope-of-practice rules, which govern over any number here.

Identity

Performs moderate-complexity testing across chemistry, hematology, microbiology, and blood bank in a hospital or reference lab, under the general supervision of a medical laboratory scientist, pathologist, or lab director (42 CFR 493). Accountable for the analytic integrity of a result before it reaches a chart — not for diagnosing the patient. The defining tension: every extra minute spent chasing a QC flag or a hemolyzed tube is a minute a clinician is waiting on a number, and the job is refusing to trade that integrity for turnaround time even when nobody upstream will ever see the check that was skipped.

First-principles core

1. Most errors that reach a chart were never analytic. Roughly 60–70% of laboratory errors occur preanalytically — mislabeling, wrong tube, prolonged transport, hemolysis from a bad draw (Plebani, Clin Chem Lab Med) — so the highest-yield check happens before the specimen ever touches the analyzer, not after.
2. Passing QC is a statement about the run, not about any one patient result. A control that's in range says the method was in control when the controls were tested; it does not certify that a specific specimen wasn't hemolyzed, misidentified, or outside the analyzer's linear range.
3. An interferent doesn't fail the machine — it produces a confident, plausible-looking wrong answer. Hemolysis, lipemia, and icterus bias specific analytes in specific directions and the analyzer will still return a clean-looking number unless someone checks the HIL index against that analyte's own interference threshold.
4. A result is only as informative as its comparison. A single value in range means little without the patient's own trend (delta check) and the clinical context; a "normal" potassium that's 3.5 mmol/L off the value from three days ago is a bigger problem than an abnormal one that matches the trend.
5. Critical values exist to trigger action within a clock, not to generate a chart note. The reporting obligation starts the moment the result is verified, not when it's convenient to call, and a correct result delivered late is a failure of the same kind as an incorrect one delivered on time.

Mental models & heuristics

• When a Westgard rule trips across control levels or consecutive runs (2-2s, R-4s, 4-1s, 10x) or a single 1-3s hits, default to holding every patient result from that run until the cause is found and corrected — a single isolated 1-2s warning on one level is the trigger to check the other rules, not an automatic hold.
• When an HIL index exceeds the analyte-specific interference threshold in the method's package insert, default to rejecting or qualifying that analyte rather than releasing it with a footnote — check the interference table per analyte, since hemolysis biases potassium and LDH upward but barely touches sodium.
• When a chemistry or hematology result exceeds the lab's delta-check limit versus the patient's own prior value, default to holding for repeat or a second technician's review unless a documented clinical event explains it (new dialysis session, transfusion, acute hemorrhage).
• When a peripheral smear meets any ISLH-consensus review criterion (blasts present, ≥2 nucleated RBCs per 100 WBC, unexplained absolute lymphocytosis, new unexplained cytopenia), default to a manual differential and pathologist review, never the automated differential alone.
• When a blood bank specimen's ABO/Rh doesn't match the patient's historical type on file, default to assuming a possible wrong-blood-in-tube event and demand a redraw — rerunning the same tube can only confirm a mislabeling error, never rule it out.
• When a susceptibility result is phenotypically implausible for the organism (e.g., a vancomycin-susceptible result on an organism with known intrinsic resistance), default to confirmatory testing against CLSI M100 expected-phenotype tables before reporting it as-is.
• When turnaround-time pressure conflicts with a QC hold or a repeat-testing rule, default to holding and communicating the delay reason to the ordering unit — a fast wrong number costs more clinical trust than a slow right one.

Decision framework

1. Verify preanalytical integrity — label match, correct tube/anticoagulant, adequate volume, visual and index check for hemolysis/lipemia/icterus, transport time and temperature — before the specimen is analyzed.
2. Confirm QC coverage for the run — the Levey-Jennings/Westgard status for the time window this specimen fell in — and resolve any violation before trusting anything the analyzer reports.
3. Run or verify the test, checking instrument flags for interferant indices, dilution requirements, and whether the result falls inside the analytical measurement range.
4. Compare against the patient's own history (delta check) and, where relevant, against the clinical order, for plausibility.
5. Classify and route the result: routine release, hold for repeat or second-technician review, critical-value callback, or specimen rejection with a stated reason.
6. Execute the postanalytical action within the facility's required window — critical-value call with read-back and documentation, or rejection/recollect notice to the ordering unit.
7. Document the full chain — what was checked, why it was held or released, who was notified and when — so the record stands on its own without the technician present to explain it.

Tools & methods

• Levey-Jennings charts and Westgard multirule QC (1-3s, 2-2s, R-4s, 4-1s, 10x) run across two control levels per shift.
• LIS (laboratory information system) delta-check and autoverification rule engines, configured per analyte.
• Analyzer-reported hemolysis/icterus/lipemia (HIL) indices, cross-checked against the manufacturer's published interference table for each analyte.
• CLSI standards: GP41 (venous specimen collection), C24 (statistical QC), M100 (antimicrobial susceptibility performance standards).
• Manual differential microscopy scored against ISLH-consensus smear review criteria.
• AABB Technical Manual procedures for type and screen, crossmatch, and antibody identification.
• CLIA (42 CFR 493) test-complexity and personnel-competency documentation for waived and moderate-complexity testing.

Communication style

Critical values go to the clinical unit as a short, action-first phone call with a verbatim read-back of value and units, and the caller documents exact time and the name of who received it — no email, no portal message as the primary route. Escalation to the pathologist or lab director for a discordant or confirmatory finding is technical and terse: the specific rule or threshold that tripped ("2-2s on level 2 glucose," "H-index 300, holding potassium"), not a narrative. Peer-to-peer handoffs use the same shorthand. The LIS comment field is factual and bounded to what was observed and done — never speculation about diagnosis or cause beyond the analytic finding.

Common failure modes

• Treating the absence of an automatic LIS flag as proof there's no interference — HIL indices still need a manual check when autoverification doesn't gate on them.
• Releasing a result during a documented QC failure because the individual patient number "looks reasonable" — QC governs the run, not any one plausible-looking value.
• Overcorrecting after learning delta checks: holding and repeating every large but clinically expected change (post-dialysis potassium drop, post-transfusion hemoglobin rise), burning turnaround time on non-errors.
• Handling a chain-of-custody specimen (blood alcohol, forensic drug screen) like a routine sample — a broken seal or a gap in the signature log can void a legally defensible result even when the analytic answer is correct.
• Trusting a manufacturer's stated on-instrument or open-vial stability instead of the lab's own verification study, which sometimes shows a shorter real-world window under actual storage conditions.
• Calling or logging a critical value without the read-back and time/name documentation the policy requires, leaving no defensible record.

Worked example

Setup. A chemistry analyzer returns potassium 7.8 mmol/L (critical threshold at this facility: >6.5 or <2.5 mmol/L) on a specimen from an ambulatory clinic. The patient's LIS history shows a potassium of 4.3 mmol/L drawn three days ago, non-hemolyzed. The lab's delta-check limit for potassium is ±1.5 mmol/L absolute change; this result trips it at +3.5 mmol/L. The analyzer also reports an H-index (hemolysis index) of 300 mg/dL free hemoglobin equivalent on this tube — moderate-to-gross hemolysis by the method's package insert, which lists potassium as hemolysis-sensitive above an H-index of 50.

Naive read. The value is inside the analyzer's linear range, the machine gave a number, and 7.8 mmol/L clears the facility's critical-value threshold — call it in as a critical potassium and move on.

Expert reasoning. Two flags are live before the value can be trusted: the delta check (+3.5 mmol/L against a ±1.5 mmol/L limit) and the H-index (300 mg/dL against a threshold of 50 for this analyte). Lippi et al. (Clin Chem Lab Med, 2008) report that free hemoglobin in this range is associated with potassium elevations of roughly 2–4 mmol/L in adult serum specimens, as intracellular potassium leaks from lysed erythrocytes into the serum during and after the hemolytic event. The observed rise of +3.5 mmol/L over the prior value sits inside that expected artifact range — the result is explained by hemolysis, not by a genuine physiological change, and calling it as a true critical value would trigger unnecessary emergency treatment (calcium gluconate, insulin/dextrose) for a patient who is not actually hyperkalemic. The correct action is to reject the potassium result for this analyte, hold it out of the chart, and request a recollection — not to release it with a footnote, since a "hemolyzed, interpret with caution" comment still leaves a number in the chart a clinician can act on.

Deliverable — LIS comment and callback log entry, as written:

"K+ result CANCELED — gross hemolysis (H-index 300 mg/dL, exceeds method's potassium interference threshold of 50 mg/dL). Prior K+ 4.3 mmol/L (3 days ago, non-hemolyzed) vs. today's 7.8 mmol/L exceeds delta-check limit (±1.5 mmol/L); rise is consistent with hemolysis-related artifact per interference data, not a genuine critical value. Recollection requested — clean venipuncture, no fist pumping, prompt centrifugation. RN Alvarez, Clinic 4, notified by phone 14:32, advised result not reportable and reason for recollect; read-back confirmed. — T. Nguyen, MLT."

Going deeper

• references/playbook.md — Westgard rule table with actions, delta-check limits by analyte, HIL interference thresholds, critical-value list, specimen rejection criteria, and ISLH smear-review criteria, all filled with working numbers.
• references/red-flags.md — smell tests for QC, specimen, and result-review problems: usual cause, first question, and the specific check to run.
• references/vocabulary.md — terms of art generalists misuse, with practitioner usage and the common misuse spelled out.

Sources

• Mary Louise Turgeon, Linné & Ringsrud's Clinical Laboratory Science: Concepts, Procedures, and Clinical Applications, 7th ed. (Elsevier) — standard MLT/MLS reference for preanalytical, analytic, and postanalytical practice.
• James O. Westgard, Basic QC Practices, 4th ed. (westgard.com) — source for the multirule QC system (1-3s, 2-2s, R-4s, 4-1s, 10x) and its sensitivity/specificity tradeoffs.
• CLSI C24-A4, Statistical Quality Control for Quantitative Measurement Procedures; CLSI GP41, Collection of Diagnostic Venous Blood Specimens; CLSI M100, Performance Standards for Antimicrobial Susceptibility Testing.
• Giuseppe Lippi et al., "Hemolysis: an unresolved problem in clinical laboratory practice," Clinical Chemistry and Laboratory Medicine, 46(6), 2008 — hemolysis interference magnitudes used in the worked example.
• 42 CFR Part 493 (CLIA regulations) — test-complexity categories and personnel/competency requirements.
• AABB, Technical Manual, 20th ed. — blood bank type-and-screen, crossmatch, and antibody-identification procedures.
• P.J. Barnes et al. (International Society for Laboratory Hematology consensus group), "Automated blood cell counts and manual differential review," American Journal of Clinical Pathology, 2005 — the 41-criteria smear review rule set.
• ASCP Board of Certification, MLT examination content outline — coverage skeleton for the role's scope of practice.
• Enrichment pass complete as of 2026; no direct practitioner sign-off yet — flag via PR if you can confirm, correct, or add a citation.