Claim

In a pragmatic cluster-randomized trial across 16 Kenyan primary-care facilities, LLM-assisted clinical officers did not significantly change 14-day treatment-failure rates versus control: 2.2% versus 2.0% (adjusted odds ratio 0.77, 95% CI 0.55–1.08, P=0.13).

Type QUANTITATIVE · truth-evaluable: yes · status at snapshot: ACTIVE · created seq 16255

33 words; atomic claims are usually under 25a comma-separated series may list several propositions

Referenced 145 times: checked 3, looked up by assistants 0, viewed 92, shared 50. A count of attention, not evidence.

Topics

Statistics · 1efd24fe Medicine · 1efd24fe AI · 1efd24fe

Assessment

Supported

Score: 0.9925 under ledger-default@0.3.0 at snapshot 27745

2 of 4 review checks done · 1 counted · 1 by the author only

3 independent evidence lineages · review checks 1 of 4 · stability HIGH

Main unresolved issue: 4 counted evidence items have no independence review; repeats of one origin may be counted as several coverage:07e9deff-2932-81e8-a0af-b0b8f9c94c53

Not yet independently audited.Evidence reviewed so far supports this claim, but only 1 of 4 review checks has been done.A quoted passage from birmingham.ac.uk and nature.com was not found on the page when Galedra fetched it.

Why?

Strongest counted support: The trial reports nearly identical 14-day treatment-failure rates and a non-significant primary comparison between LLM-assisted and control arms. weight 2.000000 evidence

Suppressed as dependent: The passage directly reports the intervention and control rates, adjusted odds ratio, confidence interval, P value, and non-significant comparison.

Review gaps: qualifiers reviewed, opposing search done, independence reviewed

What would most change this: One independent, direct measurement that contradicts this claim would move it from Supported to Supported.

Hide calculation

0.9925 under ledger-default@0.3.0 at snapshot 27745 · state Supported · Review checks: 1 of 4 · stability HIGH · 4 sources, 3 independent lineages

Model ledger-default@0.3.0 (the default model, not the answer) · snapshot 27745 · trace sha256:275c3d768697060da204ee84ef74bbbebda79d8782205d201b8a2b38c126ae03

Last changed at entry seq 25774; nothing logged since has borne on this claim.

How the number was reached

Each counted link is weighed by multiplying the factors the model declares; a factor it does not declare is not shown. Every factor name below links to what that number means.

Link	Working	Weight	Counted as
a6b938f4 support	relevance DIRECT 2.0 × observation MEASUREMENT 1.0 × interpretation 0 steps, 1 − 0.15 × 0 1.000000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance INDEPENDENT 1.0 × edition NAMED_OR_UNRELATED 1	2.000000	2.000000
9509f981 support	relevance STRONG 1.2 × observation DIRECT_TEXT 0.9 × interpretation 0 steps, 1 − 0.15 × 0 1.000000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance INDEPENDENT 1.0 × edition NAMED_OR_UNRELATED 1	1.080000	1.080000
30db13f8 support	relevance DIRECT 2.0 × observation DIRECT_TEXT 0.9 × interpretation 0 steps, 1 − 0.15 × 0 1.000000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance INDEPENDENT 1.0 × edition NAMED_OR_UNRELATED 1	1.800000	0.000000 (dependent strongest only, kept a6b938f4)
e3b2a82e support	relevance DIRECT 2.0 × observation DIRECT_TEXT 0.9 × interpretation 0 steps, 1 − 0.15 × 0 1.000000 × authenticity UNVERIFIED 1.0 × extraction UNVERIFIED 1.0 × provenance INDEPENDENT 1.0 × edition NAMED_OR_UNRELATED 1	1.800000	1.800000

prior 0.50 → log-odds 0.000000 + 4.880000 (the counted weights) = 4.880000 → probability 0.9925

Under neighbouring settings of the same model: 0.9682 to 0.9982, a spread of 0.0300.

Half-even rounding, six places for weights and four for the probability. The same inputs under the same model reproduce this exactly; that is what the trace hash above is for. How a score is worked out explains every factor, why it is there, and what the model sets it to.

Your view

Registered views

Nobody has registered a view yet.

Registered views are personal belief, not evidence: they never change Galedra's assessment. Affiliations are self-declared and shown only for groups of 5 or more.

Checked as part of

Statements this claim was broken out of. The claim is scored on its own; a statement's reading only counts its claims.

“In a pragmatic cluster-randomized trial across 16 Kenyan primary-care facilities, LLM-assisted clinical officers did not significantly change 14-day...” · 1 claim · 29 September 2026
“In a pragmatic cluster-randomized trial across 16 Kenyan primary-care facilities, LLM-assisted clinical officers did not significantly change 14-day...” · 1 claim · 29 September 2026
“In a pragmatic cluster-randomized trial across 16 Kenyan primary-care facilities, LLM-assisted clinical officers did not significantly change 14-day...” · 1 claim · 29 September 2026

Evidence for

The trial reports nearly identical 14-day treatment-failure rates and a non-significant primary comparison between LLM-assisted and control arms.

From Generative AI-enabled clinical decision support system in primary care: a pragmatic, cluster-randomized trial Read 2026-09-29 Relevance DIRECT Steps 0 Counted
The passage directly reports the intervention and control rates, adjusted odds ratio, confidence interval, P value, and non-significant comparison.

From Generative AI-enabled clinical decision support system in primary care: a pragmatic, cluster-randomized trial Read 2026-09-29 Relevance DIRECT Steps 0 Counted
Independent institutional reporting repeats the trial's null 14-day treatment-failure comparison and notes improved documentation and treatment planning.

From AI clinical support tool improved clinician decisions in real-world primary care trial Read 2026-09-29 Relevance STRONG Steps 0 Counted

Secondary report from a collaborating institution; it corroborates the primary result but derives from the same trial.
The Nature Medicine trial reports no significant 14-day treatment-failure difference between LLM-assisted and control clinical officers: 2.2% versus 2.0%, adjusted OR 0.77, 95% CI 0.55–1.08, P=0.13.

From Generative AI-enabled clinical decision support system in primary care: a pragmatic, cluster-randomized trial Read 2026-09-29 Relevance DIRECT Steps 0 Counted

Primary article directly matches the Kenyan setting, cluster-randomized design, event rates, effect estimate, and significance result.

Evidence against

None.

Qualifications

None.

Suppressed as dependent

Evidence	Group	Kept instead
3ec52420		a6b938f4

Inferences

Concluded from

None.

Used as a premise in

None.

An inference is interpretation, not evidence: it never changes the assessment of its conclusion. Each premise is a claim with its own evidence.

Threads

None. Threads are for disagreements about how a determination was made — a statement whose figures are not in the passage it rests on, two sources that may share an origin. Something about the world belongs in the evidence above.

Claim edges

None. Edges are stored and displayed; v0.1 does not propagate probability across them.

Review checklist

Review checks: 1 of 4

2 checks by this claim's own author, alongside the independent review above. 1 QUALIFIER CHECK · 1 EVIDENCE VERIFICATION. A check by the person who recorded a claim does not raise its review coverage.

Check	Done	By
qualifiers reviewed	no
opposing search done	no
independence reviewed	no
primary source reviewed	yes	a6b938f4, 30db13f8, e3b2a82e

Summary

Supported. The main support is: The trial reports nearly identical 14-day treatment-failure rates and a non-significant primary comparison between LLM-assisted and control arms. ed3955ac

1 further item repeats the same origin and is not independent evidence. 3ec52420

Review checks: 1 of 4. coverage:07e9deff-2932-81e8-a0af-b0b8f9c94c53

Generated by stub-v0.1; every sentence cites graph ids.

Contribution history

seq	event	detail
16255	APPENDED
16256	ACCEPT	AUTOMATIC_AFTER_VALIDATION
16261	LINK_EVIDENCE	by 1efd24fe (accepted)
16263	TASK_RESULT	by 1efd24fe (accepted)
16271	LINK_EVIDENCE	by 1efd24fe (accepted)
16333	LINK_EVIDENCE	by 1efd24fe (accepted)

Score trace

{
  "claim": "07e9deff-2932-81e8-a0af-b0b8f9c94c53",
  "links": [
    {
      "link": "34f740b6-6a7e-80b5-9b5a-2e31a6b938f4",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "54626ce9-f689-8bb7-a474-7efced3955ac",
      "direction": "SUPPORT",
      "magnitude": "2.000000",
      "relevance": "DIRECT",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "MEASUREMENT",
      "source_type": "PRIMARY_TEXT",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "2.000000",
      "interpretive_steps": 0
    },
    {
      "link": "b5fc9fc1-9cb8-846a-827a-00999509f981",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "629a8633-5f14-8a78-8715-ab07c0a2328f",
      "direction": "SUPPORT",
      "magnitude": "1.080000",
      "relevance": "STRONG",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "SECONDARY_TEXT",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "1.080000",
      "interpretive_steps": 0
    },
    {
      "kept": "34f740b6-6a7e-80b5-9b5a-2e31a6b938f4",
      "link": "b9d06368-75ef-88b0-a52d-91e830db13f8",
      "group": null,
      "reason": "dependent_strongest_only",
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "a2093ec5-82e0-87a0-a7cc-c0643ec52420",
      "direction": "SUPPORT",
      "magnitude": "1.800000",
      "relevance": "DIRECT",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "PRIMARY_TEXT",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "0.000000",
      "interpretive_steps": 0
    },
    {
      "link": "f891e311-46fb-8c5d-9dc4-7411e3b2a82e",
      "group": null,
      "edition": "NAMED_OR_UNRELATED",
      "evidence": "9b1c832b-b1d1-8f61-9c93-df80f2eef62c",
      "direction": "SUPPORT",
      "magnitude": "1.800000",
      "relevance": "DIRECT",
      "extraction": "UNVERIFIED",
      "provenance": "INDEPENDENT",
      "observation": "DIRECT_TEXT",
      "source_type": "PRIMARY_TEXT",
      "authenticity": "UNVERIFIED",
      "audit_confirmed": false,
      "effective_weight": "1.800000",
      "interpretive_steps": 0
    }
  ],
  "model": "ledger-default@0.3.0",
  "prior": "0.50",
  "variants": {
    "spread": "0.0300",
    "permissive": "0.9982",
    "conservative": "0.9682"
  },
  "code_hash": "sha256:c61a6fdf35a754a0d7e958df149036a083c0eaeb7d2d2e52704ec32b0c36def7",
  "contested": false,
  "stability": "HIGH",
  "claim_type": "QUANTITATIVE",
  "config_hash": "sha256:86768038080b8ab5233f1e4b7aba10af05b5bb44caa432095150b90e81209809",
  "probability": "0.9925",
  "provisional": true,
  "evidence_sum": "4.880000",
  "snapshot_seq": 25774,
  "prior_log_odds": "0.000000",
  "support_groups": 3,
  "model_dependent": false,
  "review_coverage": "0.25",
  "truth_evaluable": true,
  "assessment_state": "SUPPORTED",
  "review_checklist": {
    "qualifiers_reviewed": {
      "by": [],
      "ok": false
    },
    "opposing_search_done": {
      "by": [],
      "ok": false
    },
    "independence_reviewed": {
      "by": [],
      "ok": false
    },
    "primary_source_reviewed": {
      "by": [
        "34f740b6-6a7e-80b5-9b5a-2e31a6b938f4",
        "b9d06368-75ef-88b0-a52d-91e830db13f8",
        "f891e311-46fb-8c5d-9dc4-7411e3b2a82e"
      ],
      "ok": true
    }
  },
  "contradict_groups": 0,
  "rounding_boundary": false,
  "posterior_log_odds": "4.880000",
  "not_applicable_reason": null,
  "independence_unreviewed": 4
}