Diagnostic Logic Check¶
Criteria classification — instantiates Deductive Chain Validation
Checks whether a case actually meets the stated classification criteria and labels the evidential uncertainty that remains, so a criteria-based label is not mistaken for certainty.
A Diagnostic Logic Check asks whether a specific case genuinely satisfies the stated classification criteria — and then, crucially, marks how much evidential uncertainty survives even a clean match. Its defining move is holding two questions apart that everyday practice smears together: the deductive question "do the observed features meet the criteria for label L?" and the empirical question "how sure are we of those observations, and could this be something the criteria don't cover?" It verifies that the observations feeding the criteria are real and current, bounds the resulting label to exactly what the criteria license, and attaches a residual-uncertainty note so a criteria-match is never read as a confident, complete diagnosis. It is what keeps "meets the definition" from silently becoming "is certainly the thing."
Example¶
A rapid-response nurse is about to escalate a patient flagged as "meets criteria for systemic inflammatory response" — a label defined by a fixed set of named thresholds (temperature, heart rate, respiratory rate, white-cell count). A Diagnostic Logic Check runs the label back through its own criteria. Verify the observations: the elevated temperature was an axillary reading taken minutes after the patient came in from a cold corridor, and the white-cell count is still pending — so one criterion is on soft evidence and one is unmeasured. Bound the conclusion: "meets the inflammatory-response criteria" is not "has sepsis"; the label licenses a monitoring protocol, not a diagnosis. Label the residual uncertainty: the classification currently rests on three of four criteria with one measurement in question, so it could flip when the lab returns. The check doesn't overturn the flag — it stops the flag from being read as a settled sepsis call, and tells the team exactly which pending result would confirm or dissolve it.
How it works¶
The mechanism treats a criteria set as the rule and the case's observations as the premises, then does three things a generic argument scan would not. It re-checks that each observation is actually measured (not assumed, estimated, or stale) before letting it satisfy a criterion. It confines the output to the criteria's literal scope — a criteria-based classification is a category membership, never the broader clinical or causal story it is often mistaken for. And because diagnostic evidence is inherently probabilistic, it carries an explicit uncertainty label: which criteria rest on weak or pending evidence, and what would change the classification. That uncertainty note is the piece a purely deductive check omits and the reason this mechanism exists as its own thing.
Tuning parameters¶
- Observation-verification rigor — accept charted values as-is versus re-confirm how and when each was measured. Higher rigor catches the soft or stale reading but slows time-critical calls.
- Scope tightness — how strictly the label is held to its criteria versus allowed to imply the broader condition. Tighter scope prevents overreach but frustrates clinicians who want the classification to mean the diagnosis.
- Uncertainty granularity — a single "confidence: low/medium/high" tag versus a per-criterion note of what is pending and what would flip it. Finer labels guide the next test but add documentation load.
- Re-check trigger — whether the classification is revisited automatically when a pending result lands, or only on request. Automatic re-checks catch flips but generate alert traffic.
When it helps, and when it misleads¶
Its strength is refusing the two ways a criteria-based label goes wrong: being built on an observation nobody confirmed, and being read as more than the criteria say. By keeping validity (do the features meet the criteria?) separate from evidential confidence (are the features real, and is this the whole picture?), it prevents a tidy checkbox match from carrying unearned certainty into an escalation or a chart note.
Its failure mode is premature closure[1] — once the label attaches, the residual-uncertainty note gets ignored and the case is treated as solved, so the pending result that would have flipped it goes unwatched. The classic misuse is running the criteria and then quietly upgrading "meets criteria for X" into "has X," collapsing exactly the distinction the check exists to preserve. The guarding discipline is to make the uncertainty label actionable — naming the specific observation or pending result that would confirm or dissolve the classification — and to re-run the check when that evidence arrives.
How it implements the components¶
premise_verification— confirms each observation feeding a criterion is actually measured, current, and reliable before it is allowed to satisfy that criterion.conclusion_scope— bounds the output to the criteria's literal category ("meets criteria for X"), refusing the leap to the broader diagnosis or cause.residual_uncertainty_label— its signature: attaches an explicit note of how much evidential uncertainty remains after a match, and what would change the classification.
It applies fixed criteria to one verified case; it does not supply the generic prompt sweep — hunting hidden premises (premise_list), catching equivocal terms (definition_alignment), or ruling on abstract validity (validity_check). That portable, content-agnostic questionnaire is Logic Checklist.
Related¶
- Instantiates: Deductive Chain Validation — the Diagnostic Logic Check is the archetype's form for criteria-based classification, where evidential uncertainty must stay visible alongside the deductive match.
- Sibling mechanisms: Logic Checklist · Requirements Traceability Check · Legal Syllogism Review · Policy Eligibility Review · Syllogism Template · Rule-Engine Validation · Proof Checking
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: Diagnostic Logic Check operates as a bounded evaluation of existing evidence or work that produces a finding or disposition because it checks whether a case actually meets the stated classification criteria and labels the evidential uncertainty that remains, so a criteria-based label is not mistaken for certainty.
Independent corroboration: The frozen evidence defines Diagnostic Logic Check as 'Checks whether a case actually meets the stated classification criteria and labels the evidential uncertainty that remains, so a criteria-based label is not mistaken for certainty', so its operative form is Assessment, Review & Assurance.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Medicine & Healthcare
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Clinical reasoning cohered checking observed findings against explicit diagnostic criteria while preserving residual uncertainty.
Related originating lineages:
- Philosophy — Logic supplied the distinction between satisfying a definition and establishing an empirical identity with certainty.
Review resolution: Clinical reasoning cohered checking observed findings against explicit diagnostic criteria while preserving residual uncertainty. The retained alternate lineages materially shaped the mechanism's form.
Attribution caveat: The explicit criteria-versus-evidence split is a philosophical clarification of clinical classification practice.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; medium confidence.
Notes¶
The archetype warns that diagnostic settings are only partly deductive — much of the real work is probabilistic evidence and hypothesis testing. This mechanism deliberately handles only the deductive slice (does the case meet the stated criteria, and how sure are the inputs) and marks where the probabilistic work must take over. Its residual-uncertainty label is precisely the handoff point to that neighboring reasoning, not a substitute for it.
References¶
[1] Croskerry, P. "The Importance of Cognitive Errors in Diagnosis and Strategies to Minimize Them". Academic Medicine 78(8), 775–780 (2003). Defines premature closure as accepting a diagnosis and ending consideration of reasonable alternatives before the diagnosis is fully verified. registry ↩