Claim ledger

The Average Is Nobody's Result

Of 255 studies on AI-assisted colonoscopy, 21 split the result by who held the scope. They disagree.

Read the essay On Substack Published Tests Law IV: Instruments Over Theory

Claims
35
Verified at the primary source
23
Removed before publication
0
Last checked

35 claims: 23 verified, 11 executed, 1 reported.

Download this ledger: CSV · JSON · Markdown · CC BY 4.0 · Query it from Claude or ChatGPT

What the essay claims

Walking in: "The trial says the assistant improves detection by eight points. That is the effect. My people will get roughly that."

Walking out: "That is an average over operators the tool may be affecting in opposite directions, and almost nobody checks. When they do check they disagree about who benefits — in randomised trials, on the same endpoint. The average is a property of their operator mix, not of the tool, so it was never going to transfer to mine. I can check my own in an afternoon."

The load-bearing move: The average is not a weak estimate of a single effect. It is a mixture of effects that can have opposite signs, weighted by a staffing decision. Povyakalo 2013 is the existence proof; the count is the evidence that the field mostly does not look; the eight that print numbers are the evidence that even looking is not enough unless the numbers are published.

The claim ladder

Strongest first. Each rung names whose behaviour it measures and what it does not reach.

#ClaimWhose behaviour it measuresWhat it does not reach
1Of 255 abstracts on AI-assisted colonoscopy, 21 report the effect split by an operator property and 234 report an average and nothing elseThe reporting choices of one literature, at the abstract layer, as of 2026-08-02These are deposited abstracts. A study can split in its full text and never say so here. The count measures what is visible at the layer the field summarises itself in. It is also a floor: three patterns each found what the others missed
2The 21 contradict each other about who benefits. 7 find the larger effect in weaker operators, 4 in stronger, 3 no interaction, 3 nothing significant after stratifying, 1 divergence over time. Both directions appear in randomised trials on the same endpointPublished subgroup results in one procedureDoes not adjudicate. Several are underpowered after stratification, several are post-hoc, and the piece must not pick a winner. The spread is the finding
3Only 8 of the 21 print per-stratum estimates. None reports a test of the interaction. The rest report which subgroup reached significance, which is not a test of a differenceThe reporting and inference choices of the studies that did lookDoes not establish the interaction is real in any of them. It establishes that the published evidence cannot tell you
4The standard has no field for it. CONSORT-AI 5(iv) asks what expertise users needed; the guidance asks for differences across population subgroups; no item asks for the result across operator subgroupsThe reporting guideline, verified item by item at PMC7598943Scoped to CONSORT-AI. DECIDE-AI's items were not read and no claim is made about guidelines in general
5Povyakalo 2013 is the existence proof. Reanalysing a CAD mammography study whose average effect was null: +0.016 sensitivity (95% CI 0.003–0.028) for the 44 least-discriminating readers on 45 easier cancers, −0.145 (95% CI 0.034–0.257) for the 6 most-discriminating on 15 harder ones50 readers, 180 mammograms, one reanalysisExploratory and post-hoc by the authors' own description, with thresholds derived from the same regression and six readers in the key cell. It proves an average can hide opposite signs. It is not a law, and colonoscopy does not replicate it
6The average is a property of your operator mix, not of the tool. Change who is on shift and the measured effect changes with no change to the toolNothing. An arithmetic consequence of rung 2Does not quantify how much any published estimate would move. A reason the number does not transfer, not a correction to it

What would make it wrong

Any of these lands and the piece is wrong, not adjustable.

Full-text checking of a random sample of the 234 finds that most do split in the full text. Highest-probability failure by a distance.

A fourth search pattern finds a large new tranche (>8) all three of mine missed. The floor is then too low to support "rarely".

An independent adjudication of the same 255 abstracts returns materially more than 21 under the stated definition. The count is then an artefact of my reading, not the literature.

A reporting-guideline item for operator-stratified results turns out to exist in CONSORT-AI. Rung 4 fails outright. (Verified against the item list; the residual risk is a later update.)

Someone has already published this count. The prior-art sweep found the ecological version (39216648, 38272274) and the mechanism (Povyakalo) but not the count. If the count exists, the object is not an object.

The evidence, row by row

Each row is a claim the essay makes, then the evidence recorded for it and the source it was checked against. For verified rows that is usually the exact words, where they sit and the day they were read. The label says how far it was checked.

What the labels mean
Verified
Checked against the primary source itself by a checker who did not draft the piece, usually with the exact words, where they sit and the date read.
Executed
A run or observation done for the piece: code, a query, a count, or a check made live in an app. The claim is what it returned. Run records are not published.
Checked
Checked against its source by a separate checker in the older ledger format, which recorded no quote, locator or date.
Reported
Not verified word for word against a primary source: carried from a secondary source, from one that could not be opened in full, or supported only in part. The essay words it accordingly.
Struck
Drafted, checked and removed before publication. Kept here, crossed out, with the reason.
Excluded
Considered and deliberately left out of the piece. Kept here with the reason.
Sources (12)
  1. First-party runs 11 rows
  2. Povyakalo 6 rows
  3. Endoscopy 2026 2 rows
  4. Clin Gastroenterol Hepatol 2023 1 row
  5. Dis Colon Rectum 2026 3 rows
  6. Gut 2022 1 row
  7. Gelman & Stern 2 rows
  8. CONSORT-AI extension 2 rows
  9. COLO-DETECT 4 rows
  10. Gastrointest Endosc 2025 systematic review 1 row
  11. Gastrointest Endosc 2024 1 row
  12. Bainbridge 1 row
  1. 1 Executed Read

    Of 255 abstracts with text in the corpus, 21 report the assistant's effect split by a property of the operator and 234 report an average and nothing else

    Result: "HEADLINE: 21/255 studies report the assistant's effect split by a property of the OPERATOR. = 8.2% 234 report an average over operators and nothing else."

    first-party run by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party run by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 1, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-1 (source read 2026-08-02).

    Report an error
  2. 2 Executed Read

    Among primary records by PubMed's own publication type the figure is 21 of 157

    Result: "Among primary records only: 21/157 = 13.4%"

    first-party run by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party run by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 2, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-2 (source read 2026-08-02).

    Report an error
  3. 3 Executed Read

    Of the 21, only 8 print per-stratum effect estimates for every stratum; 3 print one stratum and a bare null; 10 report only a direction or a significance verdict

    Result: "SPLIT_FULL 8 per-stratum estimates for every stratum · SPLIT_PARTIAL 3 one stratum estimated, the other a bare null · SPLIT_DIRECTION 10 a direction or a p-verdict, nothing poolable"

    first-party run by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party run by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 3, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-3 (source read 2026-08-02).

    Report an error
  4. 4 Executed Read

    The splits disagree: 7 find the larger effect in weaker operators, 4 in stronger, 3 no interaction, 3 nothing significant after stratifying, 1 divergence

    Result: "LOWER 7 ... HIGHER 4 ... UNIFORM 3 ... NONE 3 ... DIVERGES 1"

    first-party run by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party run by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 4, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-4 (source read 2026-08-02).

    Report an error
  5. 5 Executed Read

    Three deliberately dissimilar detectors were run; the obvious one reached 15 records of which 7 are real splits, missing 14

    Result: "pattern A: reached 15 of which real splits 7 precision 0.47 coverage 0.33 of the 21 found" · "Pattern A alone, unread, would have reported 15 records of which 7 are real, missing 14."

    first-party run by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party run by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 5, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-5 (source read 2026-08-02).

    Report an error
  6. 6 Executed Read

    Pattern B reached 39 records at precision 0.36 and coverage 0.67 of the found set; pattern C reached 13 at precision 0.69 and coverage 0.43

    Result: "pattern B: reached 39 of which real splits 14 precision 0.36 coverage 0.67 of the 21 found" · "pattern C: reached 13 of which real splits 9 precision 0.69 coverage 0.43 of the 21 found" · "'coverage' is the share of the found positives a pattern reached, NOT recall"

    first-party run by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party run by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 6, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-6 (source read 2026-08-02).

    Report an error
  7. 7 Executed Read

    Every record any pattern reached was hand-read, and the script refuses to print the headline while one is unread

    Result: "HEADLINE: UNAVAILABLE — 20 record(s) reached by a pattern are unread." (the refusal branch, observed firing before adjudication was complete)

    first-party run by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party run by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 7, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-7 (source read 2026-08-02).

    Report an error
  8. 8 Executed Read

    The PubMed query returns 259 records, 255 with abstracts

    Result: "PubMed hits: 259 (retmax 400, fetched 259)" · "classified : 255 (no abstract: 4)"

    first-party run by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party run by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 8, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-8 (source read 2026-08-02).

    Report an error
  9. 9 Executed Read

    In a seeded 20-record sample of the primary average-only class, 9 had open full text and none reported an operator split in its results

    Result: Sample drawn with random.Random(20260802) from 136 eligible records; PMC full text retrieved for 9; operator-split patterns returned 0 result-section hits in 8, and in the ninth the only hit was future-tense text in a trial protocol

    first-party full-text check by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party full-text check by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 9, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-9 (source read 2026-08-02).

    Report an error
  10. 10 Executed Read

    Eleven of the twenty sampled records are not in PMC and their full texts were not read

    Result: NCBI ID converter returned "Identifier not found in PMC" for 11 of the 20 sampled PMIDs

    first-party full-text check by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party full-text check by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 10, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-10 (source read 2026-08-02).

    Report an error
  11. 11 Verified Read

    Reanalysing a CAD mammography study, use of CAD was associated with a 0.016 increase in sensitivity (95% CI 0.003-0.028) for the 44 least discriminating radiologists on 45 relatively easy cancers

    "Use of CAD was associated with a 0.016 increase in sensitivity (95% confidence interval [CI], 0.003-0.028) for the 44 least discriminating radiologists for 45 relatively easy, mostly CAD-detected cancers."

    Povyakalo, Alberdi, Strigini & Ayton, Med Decis Making 2013 · deposited abstract, Results · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Povyakalo, Alberdi, Strigini & Ayton, Med Decis Making 2013, deposited abstract, Results, https://pubmed.ncbi.nlm.nih.gov/23300205/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 11, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-11 (source read 2026-08-02).

    Report an error
  12. 12 Verified Read

    For the 6 most discriminating radiologists, sensitivity decreased by 0.145 (95% CI 0.034-0.257) on the 15 relatively difficult cancers

    "However, for the 6 most discriminating radiologists, with CAD, sensitivity decreased by 0.145 (95% CI, 0.034-0.257) for the 15 relatively difficult cancers."

    Povyakalo et al., Med Decis Making 2013 · deposited abstract, Results · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Povyakalo et al., Med Decis Making 2013, deposited abstract, Results, https://pubmed.ncbi.nlm.nih.gov/23300205/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 12, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-12 (source read 2026-08-02).

    Report an error
  13. 13 Verified Read

    The original study detected no significant average effect, and the reanalysis found CAD helped the less discriminating readers and hindered the more discriminating ones

    "It indicates that, despite the original study detecting no significant average effect, CAD helped the less discriminating readers but hindered the more discriminating readers."

    Povyakalo et al., Med Decis Making 2013 · deposited abstract, Conclusions · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Povyakalo et al., Med Decis Making 2013, deposited abstract, Conclusions, https://pubmed.ncbi.nlm.nih.gov/23300205/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 13, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-13 (source read 2026-08-02).

    Report an error
  14. 14 Verified Read

    That reanalysis covered 50 readers interpreting 180 mammograms both with and without computer support

    "reanalyzing data from a published study where 50 professionals (\"readers\") interpreted 180 mammograms, both with and without computer support"

    Povyakalo et al., Med Decis Making 2013 · deposited abstract, Background · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Povyakalo et al., Med Decis Making 2013, deposited abstract, Background, https://pubmed.ncbi.nlm.nih.gov/23300205/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 14, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-14 (source read 2026-08-02).

    Report an error
  15. 15 Verified Read

    The reader and case strata were derived after the fact from the same regression, and the authors call the method exploratory

    "Using regression estimates, we obtained thresholds for classifying a posteriori the cases (by difficulty) and the readers (by discriminating ability)." · "Our exploratory analysis method reveals unexpected effects."

    Povyakalo et al., Med Decis Making 2013 · deposited abstract, Method and Conclusions · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Povyakalo et al., Med Decis Making 2013, deposited abstract, Method and Conclusions, https://pubmed.ncbi.nlm.nih.gov/23300205/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 15, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-15 (source read 2026-08-02).

    Report an error
  16. 16 Verified Read

    The authors said such differential effects should be assessed when evaluating CAD and similar warning systems

    "Such differential effects, although subtle, may be clinically significant and important for improving both computer algorithms and protocols for their use. They should be assessed when evaluating CAD and similar warning systems."

    Povyakalo et al., Med Decis Making 2013 · deposited abstract, Conclusions · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Povyakalo et al., Med Decis Making 2013, deposited abstract, Conclusions, https://pubmed.ncbi.nlm.nih.gov/23300205/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 16, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-16 (source read 2026-08-02).

    Report an error
  17. 17 Verified Read

    In a population-based randomised trial of 4,824 surveillance colonoscopies, CADe increased ADR among low-performing endoscopists (45.5% vs 52.1%; aRR 1.15, 95% CI 1.01-1.30) but not among high-performing ones (65.9% vs 63.5%; aRR 0.96, 95% CI 0.88-1.05)

    "Of 5,309 randomized surveillance colonoscopies, 4,824 surveillance colonoscopies were included in the final analysis" · "CADe increased ADR among low-performing (ADR<54.5%) endoscopists (45.5% vs 52.1%; aRR 1.15 [95% CI 1.01-1.30]), but not among high-performing endoscopists (65.9% vs 63.5%; aRR 0.96 [95% CI 0.88-1.05])."

    Endoscopy 2026, Galician screening programme, PMID 42365851 · deposited abstract, Results · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Endoscopy 2026, Galician screening programme, PMID 42365851, deposited abstract, Results, https://pubmed.ncbi.nlm.nih.gov/42365851/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 17, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-17 (source read 2026-08-02).

    Report an error
  18. 18 Verified Read

    That subgroup analysis was prespecified

    "Prespecified subgroup analyses evaluated endoscopist baseline performance (median ADR in the standard arm)."

    Endoscopy 2026, PMID 42365851 · deposited abstract, Methods · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Endoscopy 2026, PMID 42365851, deposited abstract, Methods, https://pubmed.ncbi.nlm.nih.gov/42365851/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 18, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-18 (source read 2026-08-02).

    Report an error
  19. 19 Verified Read

    In a 3,059-patient multicentre randomised trial the gain was larger in experts: ADR of experts 42.3% vs 32.8% (P < .001) and of nonexperts 37.5% vs 32.1% (P = .023)

    "From November 2019 to August 2021, 3059 subjects were randomized to AI-assisted colonoscopy (n = 1519) and conventional colonoscopy (n = 1540)." · "ADR of expert (42.3% vs 32.8%; P < .001) and nonexpert endoscopists (37.5% vs 32.1%; P = .023)... were all significantly higher in the AI-assisted colonoscopy."

    Clin Gastroenterol Hepatol 2023, PMID 35863686 · deposited abstract, Results · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Clin Gastroenterol Hepatol 2023, PMID 35863686, deposited abstract, Results, https://pubmed.ncbi.nlm.nih.gov/35863686/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 19, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-19 (source read 2026-08-02).

    Report an error
  20. 20 Verified Read

    A 2026 study expected the greatest increase among low-volume and junior endoscopists

    "The greatest increase was expected among low-volume and junior endoscopists."

    Dis Colon Rectum 2026, PMID 41919624 · deposited abstract, Objective · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Dis Colon Rectum 2026, PMID 41919624, deposited abstract, Objective, https://pubmed.ncbi.nlm.nih.gov/41919624/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 20, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-20 (source read 2026-08-02).

    Report an error
  21. 21 Verified Read

    It reported that seniors rose from 51.7% to 59.6% (p = 0.02) and juniors from 56.2% to 62.9% (p = 0.12), and concluded the tool correlated with increased detection for experienced endoscopists

    "Of 12 senior and 12 junior endoscopists, the seniors had a statistically significant increase ( p = 0.02) from 51.7% to 59.6%, whereas juniors did not (56.2% to 62.9%, p = 0.12)." · "Computer-aided detection in colonoscopy correlated with an increased adenoma detection rate for experienced endoscopists."

    Dis Colon Rectum 2026, PMID 41919624 · deposited abstract, Results and Conclusions · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Dis Colon Rectum 2026, PMID 41919624, deposited abstract, Results and Conclusions, https://pubmed.ncbi.nlm.nih.gov/41919624/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 21, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-21 (source read 2026-08-02).

    Report an error
  22. 22 Executed Read

    Those two reported increases are 7.9 points and 6.7 points, a gap of 1.2 points, and only the larger crossed p<0.05

    Result: 59.6 − 51.7 = 7.9 · 62.9 − 56.2 = 6.7 · 7.9 − 6.7 = 1.2. Subtraction only; no model, no reanalysis, and no claim that either estimate is wrong

    first-party arithmetic on row 21's quoted rates by The Durability Curve

    A run or observation done for this piece. The run record is not published.

    Link
    Cite

    A run or observation by The Durability Curve (first-party arithmetic on row 21's quoted rates by The Durability Curve). Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 22, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-22 (source read 2026-08-02).

    Report an error
  23. 23 Verified Read

    The same study reported the same pattern on two further operator axes: high-volume 51.3% to 59.4% (p = 0.01) against low-volume 56.5% to 63.1% (p = 0.13), and surgeons 49% to 62% (p = 0.02) against gastroenterologists 56.9% to 60.8% (p = 0.15)

    "Colorectal surgeons had a statistically significant increase in adenoma detection rate from 49% to 62% ( p = 0.02), but gastroenterologists did not (56.9% to 60.8%, p = 0.15)." · "When comparing 12 high-volume and 12 low-volume endoscopists, the high-volume group had a statistically significant increase (51.3% to 59.4%, p = 0.01), whereas the low-volume group did not (56.5% to 63.1%, p = 0.13)."

    Dis Colon Rectum 2026, PMID 41919624 · deposited abstract, Results · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Dis Colon Rectum 2026, PMID 41919624, deposited abstract, Results, https://pubmed.ncbi.nlm.nih.gov/41919624/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 23, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-23 (source read 2026-08-02).

    Report an error
  24. 24 Verified Read

    Pooling two randomised trials, CADe raised ADR (RR 1.29, 95% CI 1.16 to 1.42) while examiner experience did not (RR 1.02, 95% CI 0.89 to 1.16), and the authors concluded experience plays a minor role

    "use of CADe (RR 1.29; 95% CI: 1.16 to 1.42) and colonoscopy indication, but not the level of examiner experience (RR 1.02; 95% CI: 0.89 to 1.16) were associated with ADR differences in a multivariate analysis." · "Experience appears to play a minor role as determining factor for ADR."

    Gut 2022, AID-1 + AID-2, PMID 34187845 · deposited abstract, Results and Conclusion · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Gut 2022, AID-1 + AID-2, PMID 34187845, deposited abstract, Results and Conclusion, https://pubmed.ncbi.nlm.nih.gov/34187845/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 24, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-24 (source read 2026-08-02).

    Report an error
  25. 25 Verified Read

    The title of the paper naming the inference error is "The Difference Between 'Significant' and 'Not Significant' is not Itself Statistically Significant"

    "The Difference Between \"Significant\" and \"Not Significant\" is not Itself Statistically Significant"

    Gelman & Stern, The American Statistician 60(4):328-331, 2006 · author's deposited PDF, title, page 1 · stat.columbia.edu

    Link
    Cite

    Gelman & Stern, The American Statistician 60(4):328-331, 2006, author's deposited PDF, title, page 1, http://www.stat.columbia.edu/~gelman/research/published/signif4.pdf. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 25, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-25 (source read 2026-08-02).

    Report an error
  26. 26 Verified Read

    Their point is not that thresholds are arbitrary but that large changes in significance can correspond to small, non-significant changes in the underlying quantities

    "Rather, we are pointing out that even large changes in significance levels can correspond to small, nonsignificant changes in the underlying quantities."

    Gelman & Stern, 2006 · deposited PDF, abstract, page 1 · stat.columbia.edu

    Link
    Cite

    Gelman & Stern, 2006, deposited PDF, abstract, page 1, http://www.stat.columbia.edu/~gelman/research/published/signif4.pdf. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 26, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-26 (source read 2026-08-02).

    Report an error
  27. 27 Verified Read

    The reporting standard for AI trials asks investigators to state what level of expertise was required of users

    "CONSORT-AI 5 (iv) Extension Specify whether there was human–AI interaction in the handling of the input data, and what level of expertise was required of users."

    CONSORT-AI extension, Nat Med 26:1364-1374, 2020 · checklist table, item 5 (iv), PMC full text · pmc.ncbi.nlm.nih.gov

    Link
    Cite

    CONSORT-AI extension, Nat Med 26:1364-1374, 2020, checklist table, item 5 (iv), PMC full text, https://pmc.ncbi.nlm.nih.gov/articles/PMC7598943/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 27, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-27 (source read 2026-08-02).

    Report an error
  28. 28 Verified Read

    Its guidance encourages exploring differences in performance across population subgroups, and no item asks for the result across operator subgroups

    "Beyond this, investigators should also be encouraged to explore differences in performance and error rates across population subgroups."

    CONSORT-AI extension, 2020 · discussion of item 19 extension, PMC full text; the full 14-item extension list was read and contains no operator-stratified reporting item · pmc.ncbi.nlm.nih.gov

    Link
    Cite

    CONSORT-AI extension, 2020, discussion of item 19 extension, PMC full text; the full 14-item extension list was read and contains no operator-stratified reporting item, https://pmc.ncbi.nlm.nih.gov/articles/PMC7598943/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 28, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-28 (source read 2026-08-02).

    Report an error
  29. 29 Verified Read

    The COLO-DETECT protocol pre-registered a subgroup analysis by colonoscopist type

    "Subgroup analyses will be conducted on colonoscopist type (i.e., non-BCSP accredited vs. BCSP accredited) and indication for colonoscopy (screening vs. symptomatic)."

    COLO-DETECT trial protocol, Colorectal Dis 2022, PMID 35680613 · protocol full text, statistical analysis section, PMC9796278 · pmc.ncbi.nlm.nih.gov

    Link
    Cite

    COLO-DETECT trial protocol, Colorectal Dis 2022, PMID 35680613, protocol full text, statistical analysis section, PMC9796278, https://pmc.ncbi.nlm.nih.gov/articles/PMC9796278/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 29, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-29 (source read 2026-08-02).

    Report an error
  30. 30 Verified Read

    The same protocol said a range of colonoscopist experience was anticipated and desirable, and specified how it would be measured

    "A range of experience amongst participating colonoscopists is anticipated and desirable; to facilitate subgroup analysis by colonoscopist experience, it will be assessed by BCSP accreditation status, lifetime procedure numbers and yearly procedure numbers for the last 3 years"

    COLO-DETECT protocol, 2022 · protocol full text, PMC9796278 · pmc.ncbi.nlm.nih.gov

    Link
    Cite

    COLO-DETECT protocol, 2022, protocol full text, PMC9796278, https://pmc.ncbi.nlm.nih.gov/articles/PMC9796278/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 30, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-30 (source read 2026-08-02).

    Report an error
  31. 31 Verified Read

    The COLO-DETECT results reported one average across 2032 participants at 12 NHS hospitals: adenomas detected in 56·6% versus 48·4%, adjusted odds ratio 1·47 (95% CI 1·21-1·78)

    "We did a multicentre, open-label, parallel-arm, pragmatic randomised controlled trial in 12 National Health Service (NHS) hospitals (ten NHS Trusts) in England" · "Between March 29, 2021, and April 6, 2023, 2032 participants (1132 [55·7%] male, 900 [44·3%] female; mean age 62·4 years [SD 10·8]) were recruited and randomly assigned" · "Adenomas were detected in 555 (56·6%) of 980 participants in the CADe-assisted colonoscopy group versus 477 (48·4%) of 986 in the standard colonoscopy group, representing a proportion difference of 8·3% (95% CI 3·9-12·7; adjusted odds ratio 1·47 [95% CI 1·21-1·78], p<0·0001)."

    COLO-DETECT, Lancet Gastroenterol Hepatol 2024, PMID 39153491 · deposited abstract, Methods and Findings · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    COLO-DETECT, Lancet Gastroenterol Hepatol 2024, PMID 39153491, deposited abstract, Methods and Findings, https://pubmed.ncbi.nlm.nih.gov/39153491/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 31, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-31 (source read 2026-08-02).

    Report an error
  32. 32 Reported Read

    Its randomisation was stratified by age group, sex, indication and NHS Trust, and its abstract names no colonoscopist subgroup

    Stratification verbatim: "with stratification by age group, sex, colonoscopy indication (screening or symptomatic), and NHS Trust". The absence of a colonoscopist subgroup is an absence in the deposited abstract only — the paper is not in PMC and the full text was not read.

    COLO-DETECT, Lancet Gastro Hep 2024, PMID 39153491 · deposited abstract, Methods · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    COLO-DETECT, Lancet Gastro Hep 2024, PMID 39153491, deposited abstract, Methods, https://pubmed.ncbi.nlm.nih.gov/39153491/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 32, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-32 (source read 2026-08-02).

    Report an error
  33. 33 Verified Read

    The field's current answer is a study-level one: pooling twenty-eight randomised trials and 23,861 participants, an expert-only subgroup showed a similar effect size (RR 1.19, 95% CI 1.11-1.27) and the review concluded the benefit holds irrespective of endoscopist experience

    "Twenty-eight RCTs were selected for inclusion involving 23,861 participants." · "Subgroup analyses involving only expert endoscopists demonstrated a similar effect size (RR, 1.19; 95% CI, 1.11-1.27; P < .001)" · "AI-assisted colonoscopy significantly improved adenoma detection but not sessile serrated lesion detection irrespective of endoscopist experience, system type, or healthcare setting."

    Gastrointest Endosc 2025 systematic review, PMID 39216648 · deposited abstract, Results and Conclusions · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Gastrointest Endosc 2025 systematic review, PMID 39216648, deposited abstract, Results and Conclusions, https://pubmed.ncbi.nlm.nih.gov/39216648/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 33, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-33 (source read 2026-09-27).

    Report an error
  34. 34 Verified Read

    A second meta-analysis reached the same study-level conclusion across twenty-four randomised trials and 17,413 colonoscopies

    "Twenty-four RCTs involving 17,413 colonoscopies (AI assisted: 8680; non-AI assisted: 8733) were included." · "Type of AI system used or endoscopist experience did not affect overall improvement in ADR."

    Gastrointest Endosc 2024, PMID 38272274 · deposited abstract, Results and Conclusions · pubmed.ncbi.nlm.nih.gov

    Link
    Cite

    Gastrointest Endosc 2024, PMID 38272274, deposited abstract, Results and Conclusions, https://pubmed.ncbi.nlm.nih.gov/38272274/. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 34, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-34 (source read 2026-08-02).

    Report an error
  35. 35 Verified Read

    Bainbridge stated in 1983 that unused physical skills decay and a monitoring operator becomes an inexperienced one

    "Unfortunately, physical skills deteriorate when they are not used, particularly the refinements of gain and timing. This means that a formerly experienced operator who has been monitoring an automated process may now be an inexperienced one."

    Bainbridge, Automatica 19(6):775-779, 1983 · §1.1.1 Manual control skills · ckrybus.com

    Link
    Cite

    Bainbridge, Automatica 19(6):775-779, 1983, §1.1.1 Manual control skills, https://ckrybus.com/static/papers/Bainbridge_1983_Automatica.pdf. Checked in: Claim Ledger, "The Average Is Nobody's Result", The Durability Curve, row 35, https://durabilitycurve.com/claims/the-average-is-nobodys-result/#c-35 (source read 2026-08-01).

    Report an error

Cite and reuse

A single claim
Use the row's Cite button. It names the original source first, then this row, which carries its own link.
This ledger
Floyd, Harry (2026). Claim Ledger: The Average Is Nobody's Result. The Durability Curve. https://durabilitycurve.com/claims/the-average-is-nobodys-result/
The essay
Floyd, Harry (2026). The Average Is Nobody's Result. The Durability Curve. https://durabilitycurve.com/blog/the-average-is-nobodys-result/

The ledger data is licensed CC BY 4.0: reuse it, quote it or train on it, with credit to The Durability Curve and a link. Read the licence. Machine copies: JSON, CSV, Markdown.