# Interpreting Proficiency Testing Results in Microbiology

Proficiency testing (PT) in microbiology is the periodic analysis of blinded specimens with known reference results, distributed by an external provider, with the laboratory's performance graded against those references or against peer-group consensus. Interpreting PT results means converting the graded report, its error flags, and its peer comparisons into specific, documented corrective actions in the preanalytic, analytic, or postanalytic steps of your own workflow.

PT is required for regulated analytes under the Clinical Laboratory Improvement Amendments (CLIA) and anchors any credible program of laboratory quality management [1]. Internal quality control (QC) tells you that your control materials behaved on the day you ran them. PT tells you that your entire process, performed by your own staff on specimens you did not prepare, lands on the same answer as the reference and as hundreds of peer laboratories (laboratories using similar methods) [1]. That outside view catches failures internal controls miss, and it predicts real performance. In a national network of tuberculosis laboratories, microscopists with low PT scores worked in laboratories with lower estimated sensitivity on patient smears, and targeted retraining improved both [2]. PT also functions as continuing education. The American Proficiency Institute (API) distributes ungraded Educational Sample challenges built around well-characterized strains with emerging resistance mechanisms, and the accompanying critiques reach hundreds of small and medium hospital laboratories across the United States [1]. This guide explains how to interpret proficiency testing results in microbiology, step by step, across identification, susceptibility, molecular, sequencing, typing, and microscopy challenges.

## What Proficiency Testing Measures in a Microbiology Laboratory

A PT event (providers also call it a survey, panel, or test event) is a set of specimens shipped on a fixed schedule, tested under routine conditions, and reported back by a deadline. Programs range from a single annual event to several cycles per year. Panels are engineered rather than random. Providers select well-characterized strains, spike defined matrices, dilute materials to target concentrations, and include negative members so that both false positives and false negatives can be detected.

The Canadian Blood Services program for bacterial screening of apheresis platelets (platelet concentrates collected from a single donor) shows how much engineering sits behind a panel. Developers spiked platelet preparations with Staphylococcus epidermidis, Pseudomonas aeruginosa, and Streptococcus pneumoniae at 10 and 100 colony-forming units (CFU) per mL, stored them at 2 to 8 degrees C, and inoculated culture bottles with 4 or 6 mL on days 7 through 10 to reproduce the real interval between collection and screening. S. epidermidis and P. aeruginosa grew reliably at both concentrations and volumes, but S. pneumoniae did not, so it was excluded from the implemented four-member panel of positives and negatives [3]. The lesson for interpretation is direct. A pass certifies performance only for the organisms, concentrations, and matrices the panel actually contains.

Providers grade against one of three bases. A reference (expected) result comes from reference characterization, often with molecular confirmation of the challenge strain. A consensus result comes from the participant group itself, usually the majority answer for qualitative analytes. An ungraded result carries no score and exists for education. Most reports give acceptability per sample and per analyte, place your result within your peer group, and summarize the group distribution. Quantitative analytes are often converted to a standard score (a z score) that expresses your deviation from the peer mean in standard deviation units, and many schemes treat values within plus or minus 2 as acceptable.

PT only measures what patients experience if you test it the way you test patient work. That means the same staff, the same instruments and reagent lots, the same interpretation rules, and the same reporting steps, with no extra repeats beyond your routine policy. Record lot numbers and instrument identifiers on the PT worksheet, keep the provider's critique with your report, and retain both for inspection. Regulators and accreditors expect these records to be retrievable, so file the worksheet, the critique, and any corrective action together.

| Term | Who runs it | Specimens | Question it answers | Microbiology example |
|--|--|--|--|--|
| Proficiency testing (PT) | External provider | Blinded panels of known content | Does our whole process match the reference and our peers? | Graded identification and susceptibility challenges [1] |
| Internal quality control (QC) | Your laboratory | Known controls run with each batch | Did the method perform as expected today? | QC organisms run with each susceptibility panel |
| External quality assessment (EQA) | External body | PT plus rechecking, inspection, and training | Is the laboratory's overall quality system sound? | National tuberculosis EQA combining PT with slide rechecking [2] |
| Rechecking (blind review) | Reference laboratory | Your own routine specimens | Are routine results accurate in real time? | Rechecking of AFB smears read by network microscopists [2] |
| Educational (ungraded) challenge | External provider | Characterized strains with no score | What emerging problem should we learn about? | API Educational Samples for susceptibility testing [1] |

External quality assessment (EQA) is the umbrella term for all of this, and PT is the blinded, scored component. Some EQA programs add rechecking of routine work, on-site visits, or phased assessments, and those components answer different questions than a PT score does.

## How to Interpret Proficiency Testing Results: A Step-by-Step Method

![Flow of six steps: check the preanalytic record, identify the grading basis, classify errors, find patterns, split causes, then correct and document.](/article-images/illustrations/how-to-interpret-proficiency-testing-results-microbiology-six-step-method.webp)

Treat the PT report the way you treat a culture workup, as a structured investigation rather than a grade. The six steps below turn a score sheet into actions.

### Step 1: Verify the preanalytic record

Before you accept any failure as real, reconstruct what happened to the sample. Check receipt condition, shipping delays, storage temperature, reagent and lot numbers, instrument used, and the staff who performed the work. Panel members can degrade in transit or storage, and providers occasionally withdraw a member after the fact because it proved unstable, as happened with one organism in the Canadian platelet screening panel [3]. A documented preanalytic problem changes the interpretation from "our laboratory failed" to "this sample was compromised", and both conclusions require paperwork.

### Step 2: Identify the grading basis

Find out whether each sample was graded against a reference result, a peer consensus, or not at all, because this determines what your score can mean. Reference-graded challenges measure accuracy against truth. Consensus-graded challenges measure agreement with your peers, which is a different thing. If most peers share a method limitation, a consensus grade can mark a defensible result wrong, and if an error propagates through the group, consensus can hide it. Grading design also has to cope with unstable materials. In an EQA of HIV tropism testing across 14 laboratories, triplicate testing of some clinical samples showed proviral DNA variability so large that the geno2pheno false-positive rate (FPR) spanned 0 to 96.7 percent within a single sample, so the organizers graded against a consensus median FPR and defined adequate performance in advance as no more than 2 R5 and no more than 1 X4 specimen miscalled [4]. In that assay, a low FPR predicts a CXCR4-using (X4) virus and a high FPR predicts a CCR5-tropic (R5) virus. Read the grading rules before you read your scores.

### Step 3: Classify every error precisely

For antimicrobial susceptibility testing (AST), use the standard CLSI error vocabulary. A very major error (VME) is a susceptible result reported for an isolate that is truly resistant. A major error (ME) is a resistant result reported for a truly susceptible isolate. A minor error (MinE) is a discrepancy between intermediate and susceptible or resistant at a breakpoint (the concentration or zone diameter that separates the interpretive categories). Categorical agreement (CA) is the percentage of results landing in the correct interpretive category. The distinctions are clinical, not bureaucratic. A VME can lead to ineffective therapy, an ME can withhold a useful drug, and a MinE at a breakpoint boundary is often a dilution or criteria artifact. For molecular and antigen PT, classify each result as false positive, false negative, or invalid. An invalid result is a distinct outcome from a false result, and providers grade it separately, so read whether invalids count against you before you decide how to respond. For identification PT, note whether the error is at genus level or only species level. For sequencing PT, distinguish failure to generate data from failure to interpret it.

### Step 4: Search for patterns

![Decision path that routes a failed PT challenge by pattern: one sample, one analyte, one method, or many samples, ending in a root cause review.](/article-images/illustrations/how-to-interpret-proficiency-testing-results-microbiology-error-cause-triage.webp)

Four patterns carry four meanings.

1. One sample missed while everything else is correct points to a sample-specific cause, such as a low copy number, a difficult matrix, or handling damage.
2. One analyte missed across samples points to a method or reagent problem for that drug, gene target, or antigen.
3. One method or platform wrong across analytes points to the instrument, software, or database.
4. Many samples wrong across the board points to a workflow breakdown, contamination, or a systematic process failure.

Then compare with the peer group. If many laboratories missed the same sample, the challenge itself or a shared reagent or database limitation is the more likely cause. If only your laboratory missed it, look inward. Provider critiques publish these distributions, and reading them is part of interpreting your own result.

### Step 5: Separate technical from interpretive causes

Many PT failures are interpretation failures, and they need different fixes than instrument problems. During the SARS-CoV-2 point-of-care testing (POCT) PT program in British Columbia, a root cause investigation of poor rapid antigen performance found gaps both in how tests were executed and in how results were read, and the issues clustered among new testers without diagnostic experience [5]. In the Zambian national HIV PT exercises, lay counselors and nurses produced more false-negative, false-positive, and indeterminate readings than laboratory personnel, a pattern pointing to line interpretation rather than test chemistry [6]. In sequencing PT, the same split appears as pipeline versus curation, where tool and database choices drive identification failures [7]. Your corrective action differs for each cause.

### Step 6: Correct, document, and recheck

Close the loop. Perform a root cause analysis, implement a corrective action, document both, and verify improvement on the next event. The British Columbia program reinforced training and mentoring, retrained personnel, rewrote its sample processing instructions, and produced a training video [5]. In the Mexican tuberculosis network, intensive retraining after poor PT scores produced gains that were still measurable two years later [2]. An undocumented fix protects no one during an inspection and teaches nothing at the next cycle.

### A worked example

Suppose your laboratory receives a graded AST report with five Gram-negative challenges. Four are fully acceptable. On the fifth, you reported meropenem susceptible and the reference result is resistant, while most of your peer group called it correctly. Work the steps. Step 1 confirms the vial arrived intact and was tested by the usual staff on the usual platform. Step 2 confirms the challenge was reference graded, so accuracy against truth is the standard. Step 3 classifies the result as a VME, the highest-risk error category. Step 4 finds a single-sample, single-drug pattern with no method-wide problem. Step 5 separates causes: the MIC value itself may have been in range and the error occurred at transcription, or your reading of the panel may have been off by one dilution. Step 6 assigns the action, which might be double reading of unusual phenotypes, a software alert for improbable susceptibility combinations, or a review of how your instrument flags them. Documented that way, one failed sample becomes a permanent improvement in the reporting chain.

## Interpreting Identification and Susceptibility Testing Results

### Organism identification scores

Across five years of API Educational Sample challenges, seven Gram-positive and eight Gram-negative well-characterized strains distributed from 2007 through 2011, graded identification accuracy was 95.3 percent for Gram-negative challenges (range 91.0 to 99.2 percent) and 97.0 percent for Gram-positive challenges (range 94.2 to 100.0 percent) among participating laboratories [1]. Identification errors in a PT setting usually trace to three causes: an organism outside your identification system's database, an outdated database version, or a transcription mistake. A single miss on an unusual organism suggests a database limitation. Repeated misses across samples suggest a workflow problem. Check whether the provider requires the exact species or accepts a close relative, because grading tolerance varies.

### Categorical susceptibility results

For susceptibility results, read the categorical accuracy for each drug and each method you used. In the same API dataset, categorical accuracy was generally greatest for disk diffusion (91.0 percent for Gram-negative and 97.0 percent for Gram-positive challenges) compared with commercial automated and manual minimum inhibitory concentration (MIC) methods combined (89.9 and 96.1 percent) [1]. A multicenter Spanish study of 60 laboratories testing 12 Enterobacteriaceae strains with defined beta-lactam resistance mechanisms found overall CA of 85.5 percent, with VME at 5 percent, MinE at 9.7 percent, and ME at 5.5 percent [8]. The error mix tells you where to look. Excess minor errors usually mean borderline isolates or interpretive criteria differences. Excess very major errors demand an immediate review of how you read and transcribe MICs and zone diameters, because those errors reach patients as inappropriate therapy.

### Resistance mechanism challenges

Mechanism-level challenges are harder than routine susceptibility panels. In the Spanish study, CA was lowest for carbapenems (56.7 to 78.1 percent). Beta-lactam and beta-lactamase inhibitor combinations were also difficult, with VME of 8 percent for amoxicillin/clavulanic acid under European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria and MinE of 45.7 percent for piperacillin/tazobactam under Clinical and Laboratory Standards Institute (CLSI) criteria. At the mechanism level, the OXA-48 carbapenemase was correctly identified by 75 to 87 percent of centers, the VIM-1 metallo-beta-lactamase by 65 percent, the KPC-3 carbapenemase by 71 percent, strains carrying both an extended-spectrum beta-lactamase (ESBL) and a carbapenemase by 32 to 68 percent, and a strain whose only beta-lactamase was the chromosomal K1 enzyme by 7 percent [8]. Dual mechanisms and chromosomal or inducible mechanisms defeat simple phenotypic rules. If your laboratory missed such a challenge, the corrective action is usually an algorithm change, such as adding inhibitor-based tests, molecular confirmation, or referral criteria, rather than instrument calibration. Also check which criteria your peer group used, because 53.3 percent of the Spanish centers used CLSI criteria and 46.7 percent used EUCAST, and that choice changed their error profile.

## Interpreting Molecular, Antigen, and Sequencing PT Results

### Molecular detection and antigen panels

Molecular PT grades qualitative calling, and the report should be read against sample strength. In the Canadian SARS-CoV-2 POCT program, six shipments per year each carried four samples, either positive or negative, and enrollment grew from 23 sites in March 2021 to more than 100 by December 2021. Performance was consistently acceptable for sites using nucleic acid technology (NAT, amplification-based testing), while sites using rapid antigen detection (RAD) performed poorly, especially on weakly positive samples [5]. A weak positive near your assay's limit of detection (LoD, the lowest concentration your assay detects reliably) is the discriminating challenge, and a miss there means something different from a false positive on a negative sample.

PT panels have ceilings as well as floors. In an evaluation of a commercial SARS-CoV-2 reverse transcriptase PCR (RT-PCR) kit against two in-house assays, all three assays returned correct results for all eight PT samples, yet clinical sensitivity on 289 patient specimens was about 50 percent for each assay (51.6 percent for the commercial kit, 49.8 and 50.5 percent for the in-house assays), because patient samples carried viral loads near the assays' limits. The commercial kit had an LoD of 1.8 × 10^-1 TCID50/mL (TCID50 is the 50 percent tissue culture infectious dose, a standard infectivity unit), one log10 lower than the in-house assays [9]. A clean PT panel verifies that your assay calls positives and negatives correctly at moderate concentrations. It does not certify performance at the detection limit, where many clinical samples sit.

### Serology and antibody PT

Antibody PT adds an isotype dimension and a manufacturer dimension. In a pilot EQA of anti-SARS-CoV-2 antibody testing, 16 laboratories reported 294 interpreted results and 796 raw data results across eight serum samples. The overall error rate was 2.7 percent for IgG, 6.9 percent for IgA, and 16.7 percent for IgM. Diagnostic specificity was very high overall, but sensitivity ranged from 67 to 98 percent across manufacturers [10]. Read your result within your method's peer group, request the raw replicate data when your provider offers it, and treat a pass on a high-sensitivity platform differently from a pass on a marginal one. A high IgM error rate across the whole group is an assay-generation problem. A high error rate for your method alone is your problem.

### Sequencing and bioinformatics PT

Sequencing PT grades a pipeline plus a judgment. Whole-genome sequencing (WGS) has reached routine diagnostic microbiology to the point that proficiency testing is treated as a core quality control element for implementation and accreditation [11]. Three lessons recur across published schemes.

First, the pipeline matters as much as the bench. In a COMPARE in silico virus PT (in silico means generated by computer simulation rather than from real sequencing runs), 13 European institutes analyzed the same simulated Illumina HiSeq dataset, and the choice of tools, programs, and databases changed whether viral sequences were identified correctly, with slightly mutated and highly divergent genomes the hardest to recover. Participants also wrote an interpretation for a fictitious case report, and that virological evaluation mattered beyond the raw pipeline output [7].

Second, reference databases and [preprocessing](/blog/data-analysis/what-is-data-preprocessing) are failure points of their own. In a COMPARE metagenomics PT, 12 participants analyzed an artificial shotgun dataset resembling sequencing of a contaminated trout sample, submitting per-read taxonomic assignments at species level plus a report classifying findings as pathogen, background, or contamination. The software classified reads reliably overall, but incomplete reference databases and inappropriate preprocessing caused difficulties [12].

Third, read the metrics every cycle. The SARS-CoV-2 WGS proficiency testing program covering one New Zealand and 14 Australian public health laboratories assessed performance on genome coverage, Pango lineage assignment (the standardized SARS-CoV-2 lineage nomenclature), and sequence quality. When the 2021 results were regraded under the more stringent 2022 metrics, more participants would have failed, and the fail rate per sample was higher in 2021 than in 2022 [13]. A metric change can move your pass or fail without any change in your laboratory. Compare performance across years only after confirming the metrics stayed the same.

## Interpreting PT for Typing, Microscopy, and Rapid Testing

### Typing and profile interpretation

Typing PT grades two separate skills. In a 2012 trial organized by the European Union Reference Laboratory for Listeria monocytogenes, 28 national reference and public health laboratories performed [pulsed-field gel electrophoresis](/knowledge/diagnostics/molecular/pulsed-field-gel-electrophoresis) (PFGE, a method that separates large DNA fragments after restriction digestion to produce a strain fingerprint) and interpreted the profiles, and 16 achieved classifications of good or satisfactory. The trial carried separate components for running the method and for interpreting profiles, and successful participation was a prerequisite for submitting profiles to the shared European database [14]. If your typing PT fails, determine which component failed. Dirty gels, incomplete digests, and missing bands are bench problems. Wrong band matching or wrong cluster assignment are interpretation problems, and those matter most for surveillance because a mislabeled profile can merge or split an outbreak investigation.

### Microscopy PT

Microscopy PT is graded per slide, and low-positive specimens do the discriminating. In the Mexican national program for acid-fast bacilli (AFB) smear microscopy, a network of 637 laboratories was assessed, with 586 laboratories (92 percent) inspected and 430 technicians each reading 10 slides with known AFB numbers. Of those technicians, 196 (46 percent) scored below 80 percent and received intensive training. Mean scores rose from 65 to 90 percent (P < 0.0001) and remained at 83 percent when the group was retested in 2001. The main factors affecting scores were the type of laboratory and the number of low-positive slides (1 to 9 AFB per 100 fields) in the panel. Laboratories whose routine work was also rechecked scored better (P = 0.002), and PT scores were associated with the estimated sensitivity of the microscopist's laboratory (P = 0.01) [2]. A failed microscopy PT usually reflects scanning discipline, such as how many fields you read and for how long, rather than staining chemistry, and the low-positive slides are where laboratories separate.

### Rapid serologic testing

Rapid test PT exposes interpretation. In the Zambian national HIV PT system, sites received panels of five dried tube specimens (DTS), with 282 sites participating in 2009 and 488 in 2010. Overall accuracy was 93.1 percent (95 percent CI 91.2 to 94.9) in 2009 and 96.9 percent (95 percent CI 96.1 to 97.8) in 2010. Laboratory personnel were more accurate than non-laboratory personnel in 2009, but that difference had disappeared by 2010. Across both exercises, lay counselors and nurses had more difficulty interpreting results, with more false-negative, false-positive, and indeterminate readings [6]. When your PT errors cluster in weak bands or invalid results rather than in strong positives and negatives, the corrective action is reading training, job aids, and a second-reader rule, not instrument service.

## Reading Trends Across Cycles

A single PT event is a snapshot. Longitudinal interpretation asks whether your accuracy and precision are improving, stable, or drifting. The Immunology Quality Assessment (IQA) program, which has assessed CD3(+)CD4(+) and CD3(+)CD8(+) lymphocyte subset proficiency for laboratories supporting HIV clinical trial networks since 1999, reviewed ten years of results (January 2003 through July 2012) with longitudinal mixed effects models. That design separates intralaboratory variability (precision, measured through repeated measures variance) from interlaboratory agreement (accuracy), and its stated goals were to increase interlaboratory consistency and decrease intralaboratory variability [15]. You can apply the same logic at bench level. Plot your acceptability rate per analyte across cycles. A laboratory that passes every event but swings between the top and bottom of its peer group has a precision problem that pass or fail hides. A laboratory drifting slowly toward the peer mean may be correcting itself. Both patterns are invisible on any single report. When you change methods or instruments, annotate the timeline, because a step change in your trend usually dates to the switch rather than to slow drift.

## Common Mistakes and Limitations

1. **Reading PT as an exam rather than a diagnostic signal.** The score is the beginning of the analysis, not the end. A pass with wide peer scatter and a fail with a clear root cause carry different lessons.
2. **Running PT samples outside the routine pathway.** Extra repeats, your best technologist, or added controls make the grade meaningless, because PT exists to measure the process patients actually experience.
3. **Confusing error categories.** A minor error at a breakpoint boundary is not a very major error. Weight your response by clinical risk, and reserve urgent review for false-susceptible results.
4. **Skipping the ungraded challenges.** Educational samples carry no score but do carry emerging resistance mechanisms and method comparisons, and the critiques are written as teaching material [1]. Review them every cycle.
5. **Assuming a pass certifies clinical performance.** Panels use well-preserved materials at moderate concentrations, so a pass confirms correct calling at those levels and does not probe near-LoD sensitivity or performance on degraded clinical specimens [9].
6. **Ignoring the peer group.** The same reported result can be acceptable in one method group and an outlier in another, and manufacturer-level sensitivity differences are real [10]. Always read your result against the correct peer group.
7. **Forgetting which interpretive criteria you use.** CLSI and EUCAST criteria can disagree at breakpoints, and the choice changed error profiles across the Spanish beta-lactam study [8]. State your criteria in your PT records so grades stay traceable.
8. **Accepting a failure without checking sample integrity.** Panels degrade in transit, and providers occasionally withdraw unstable members. Document condition on receipt before you accept the grade.
9. **Comparing across cycles without checking the metrics.** Grading metrics change between cycles, and regrading under newer metrics has flipped prior results [13]. Confirm the metrics before you trend your history.
10. **Leaving corrective action undocumented.** Accreditors and inspectors ask for root cause, action, and verification. Write all three down and attach them to the PT report.

One limitation deserves plain language. PT is periodic, panels cover a fraction of your test menu, characterized challenge strains may not resemble local epidemiology, and consensus grading inherits the limitations of the peer group. A PT report never substitutes for case-specific judgment. Individual patient results and treatment decisions belong to the treating clinician working with the laboratory director.

## Quick Review

- PT grades your whole process, on blinded specimens you did not prepare, against reference results or peer consensus.
- Classify errors precisely: VME (false susceptible), ME (false resistant), MinE (intermediate discrepancy), false positive, false negative, invalid.
- Read patterns, not single data points. One sample, one analyte, one method, or the whole run each point to a different cause.
- Separate technical causes from interpretation causes, because their fixes differ.
- Review ungraded educational challenges, which carry emerging mechanisms and method lessons.
- A pass does not certify near-LoD sensitivity or performance on degraded clinical specimens.
- Document root cause and corrective action, verify improvement on the next cycle, and track your trend across years.

## Frequently Asked Questions

### What does proficiency testing mean in microbiology?

Proficiency testing is an external program that sends your laboratory blinded specimens of known content on a fixed schedule and grades your reported results against reference answers or peer consensus. It measures your whole testing process, from specimen handling to the final report.

### How is proficiency testing different from internal quality control?

Internal QC uses control materials you run with every batch to check method performance on that day. PT uses externally prepared, blinded panels to check your entire process and compares you with other laboratories. Both belong in a microbiology quality control program, and neither replaces the other.

### What is a very major error in susceptibility testing proficiency testing?

A very major error is a susceptible result reported for an isolate that is truly resistant. It is the most clinically dangerous susceptibility error because it can lead to ineffective therapy, and it warrants a more urgent investigation than a minor error at a breakpoint boundary.

### What should a laboratory do after a failed PT challenge?

Verify the sample's condition and your workflow records, classify the error, look for patterns across samples and peer groups, identify the root cause, implement and document a corrective action, and verify improvement on the next event.

### Why can a laboratory pass PT and still miss weak positive patient samples?

PT panels are usually built at moderate concentrations with well-preserved materials, so they confirm correct positive and negative calling without probing the limit of detection. Clinical specimens often sit near that limit, which is where misses happen.

### Are ungraded educational PT samples worth reviewing?

Yes. Ungraded challenges carry no score but are built around emerging resistance mechanisms and contemporary interpretation problems, and the published critiques are written specifically as teaching material for participating laboratories.

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "What does proficiency testing mean in microbiology?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Proficiency testing is an external program that sends your laboratory blinded specimens of known content on a fixed schedule and grades your reported results against reference answers or peer consensus. It measures your whole testing process, from specimen handling to the final report."
      }
    },
    {
      "@type": "Question",
      "name": "How is proficiency testing different from internal quality control?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Internal quality control uses control materials you run with every batch to check method performance on that day. Proficiency testing uses externally prepared, blinded panels to check your entire process and compares you with other laboratories. Both belong in a microbiology quality control program, and neither replaces the other."
      }
    },
    {
      "@type": "Question",
      "name": "What is a very major error in susceptibility testing proficiency testing?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "A very major error is a susceptible result reported for an isolate that is truly resistant. It is the most clinically dangerous susceptibility error because it can lead to ineffective therapy, and it warrants a more urgent investigation than a minor error at a breakpoint boundary."
      }
    },
    {
      "@type": "Question",
      "name": "What should a laboratory do after a failed PT challenge?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Verify the sample's condition and your workflow records, classify the error, look for patterns across samples and peer groups, identify the root cause, implement and document a corrective action, and verify improvement on the next event."
      }
    },
    {
      "@type": "Question",
      "name": "Why can a laboratory pass PT and still miss weak positive patient samples?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "PT panels are usually built at moderate concentrations with well-preserved materials, so they confirm correct positive and negative calling without probing the limit of detection. Clinical specimens often sit near that limit, which is where misses happen."
      }
    },
    {
      "@type": "Question",
      "name": "Are ungraded educational PT samples worth reviewing?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "Yes. Ungraded challenges carry no score but are built around emerging resistance mechanisms and contemporary interpretation problems, and the published critiques are written specifically as teaching material for participating laboratories."
      }
    }
  ]
}
</script>

## Sources

1. [Educational antimicrobial susceptibility testing as a critical component of microbiology laboratory proficiency programs: American Proficiency Institute results for 2007-2011.](https://pubmed.ncbi.nlm.nih.gov/23481025/)
2. [Implementation of proficiency testing in conjunction with a rechecking system for external quality assurance in tuberculosis laboratories in Mexico.](https://pubmed.ncbi.nlm.nih.gov/12797692/)
3. [Implementation of a proficiency testing programme for bacterial screening in platelet preparations in Canada.](https://pubmed.ncbi.nlm.nih.gov/17683356/)
4. [Results of external quality assessment for proviral DNA testing of HIV tropism in the Maraviroc Switch collaborative study.](https://pubmed.ncbi.nlm.nih.gov/23596247/)
5. [Ensuring diagnostic testing accuracy for patient care and public health- COVID-19 testing scale-up from an EQA provider's perspective.](https://pubmed.ncbi.nlm.nih.gov/38055697/)
6. [Accuracy in HIV Rapid Testing among Laboratory and Non-laboratory Personnel in Zambia: Observations from the National HIV Proficiency Testing System.](https://pubmed.ncbi.nlm.nih.gov/26745508/)
7. [Proficiency Testing of Virus Diagnostics Based on Bioinformatics Analysis of Simulated In Silico High-Throughput Sequencing Data Sets.](https://pubmed.ncbi.nlm.nih.gov/31167846/)
8. [Susceptibility testing and detection of β-lactam resistance mechanisms in Enterobacteriaceae: a multicentre national proficiency study.](https://pubmed.ncbi.nlm.nih.gov/29274802/)
9. [Evaluation of the commercially available LightMix® Modular E-gene kit using clinical and proficiency testing specimens for SARS-CoV-2 detection.](https://pubmed.ncbi.nlm.nih.gov/32516739/)
10. [Results of the first pilot external quality assessment (EQA) scheme for anti-SARS-CoV2-antibody testing.](https://pubmed.ncbi.nlm.nih.gov/32853163/)
11. [Practical issues in implementing whole-genome-sequencing in routine diagnostic microbiology.](https://pubmed.ncbi.nlm.nih.gov/29117578/)
12. [Proficiency Testing of Metagenomics-Based Detection of Food-Borne Pathogens Using a Complex Artificial Sequencing Dataset.](https://pubmed.ncbi.nlm.nih.gov/33250869/)
13. [Continued improvement in the development of the SARS-CoV-2 whole genome sequencing proficiency testing program.](https://pubmed.ncbi.nlm.nih.gov/38729860/)
14. [Pulsed-field gel electrophoresis proficiency testing trials: toward European harmonization of the typing of food and clinical strains of Listeria monocytogenes.](https://pubmed.ncbi.nlm.nih.gov/24066940/)
15. [The Immunology Quality Assessment Proficiency Testing Program for CD3⁺4⁺ and CD3⁺8⁺ lymphocyte subsets: a ten year review via longitudinal mixed effects modeling.](https://pubmed.ncbi.nlm.nih.gov/24911327/)