Cognitive Research Intelligence of Seizure Pathology
Unrecognized seizures in dementia populations hide in the clinical record as falls, behavioral episodes, and unexplained changes in condition. CRISP translates the peer-reviewed evidence on this overlap into structured intelligence — scored against GRADE and Cochrane standards, delivered within one business day.
Direct Answer
Your question answered before any analysis
Scored Evidence
Quality · Bias · Relevance · 0–10 each
Gap Analysis
What the literature does not establish
DSS Framework Mapping
Findings across all four domains
Seagull Health Commentary
What it means for your context
What You Receive
Your clinical question — about seizure recognition, standard of care, or risk in dementia populations — answered in plain language before any scored analysis. Written to be usable immediately, not navigated.
Every peer-reviewed paper on the dementia-seizure overlap evaluated for evidence quality, bias risk, and clinical relevance. Each score is visible. Nothing is hidden behind a composite number.
What the dementia-seizure literature does not establish — where evidence is thin, contested, or absent. In a clinical space where seizure activity is systematically underdocumented, gaps are findings.
Every finding connected to the four domains of the Dementia Seizure Spectrum™ — the clinical taxonomy defining how seizures actually present in dementia patients.
What the dementia-seizure evidence means for your specific context — legal, investment, clinical, or operational. Written by someone who has been inside the records, the nursing notes, and the documentation gaps of long-term dementia care. This is not a summary of the literature. It is an interpretation of what the evidence requires of organizations responsible for this population.
Why It's Defensible
A literature search that retrieves papers about seizures in dementia is not clinical intelligence. The question that matters in a deposition, a due diligence memo, or a regulatory filing is not what the dementia-seizure literature says — it is how the evidence was evaluated, what weight it was given, and why.
CRISP uses GRADE and Cochrane methodology because these are the standards the peer-reviewed literature uses to evaluate itself. They are the standards expert witnesses are challenged on when seizure recognition and standard of care are at issue. They are the standards that define what "high-quality evidence" means when the question involves an understudied population.
Every CRISP brief shows the scoring on every paper retrieved about dementia-related seizure activity. Evidence quality, bias risk, and clinical relevance — scored individually, composite formula disclosed. Nothing hidden behind a proprietary algorithm. That transparency is not incidental — it is the point.
Transparent Scoring
Every paper in the dementia-seizure literature retrieved for your brief is scored individually — evidence quality, bias risk, clinical relevance. Every score visible. No composite black box.
Reproducible Methodology
GRADE and Cochrane standards are internationally documented. Any qualified reviewer — opposing counsel, investment committee, regulatory body — can replicate the evaluation using the same criteria.
Disclosed Formula
Evidence Quality 30% · Bias Risk 30% · Clinical Relevance 40%. Clinical Relevance carries the highest weight because a rigorous study in the wrong population is worthless intelligence.
Real-Time Retrieval
PubMed queried at the time of your request. The dementia-seizure evidence as it exists when you need it — not a fixed training dataset.
How CRISP Works
Every CRISP brief follows the same six-step process regardless of market. The output format adapts to the context in which it will be used.
Every CRISP analysis begins with a specific question submitted to Seagull Health. The question can be clinical, legal, investment-focused, or operational — but it must sit within the dementia-seizure clinical overlap. CRISP is not a general-purpose research tool.
CRISP translates your dementia-seizure question into structured search logic using MeSH terminology — the controlled vocabulary of the biomedical literature. MeSH translation is what separates a clinical literature search from a keyword search. It retrieves the specific peer-reviewed evidence on seizure activity in dementia populations, not every document containing matching words.
CRISP connects to PubMed in real time. Results reflect the dementia-seizure evidence as it exists at the time of your query — not a fixed dataset from a prior training period. Human studies only. Clinical publication types only.
Every retrieved paper is evaluated across three dimensions and scored 0–10 on each. The composite score is calculated using a weighted formula. Every individual score is visible in the delivered brief. Nothing is proprietary.
Evidence Quality
Methodological rigor via GRADE. Systematic reviews and RCTs score higher than narrative reviews and case reports.
Bias Risk
Cochrane Risk of Bias methodology. Lower bias scores higher. The standard expert witnesses are challenged on.
Clinical Relevance
Direct applicability to the dementia-seizure overlap. Highest weight — a rigorous study in the wrong population is not useful intelligence.
Every scored finding is mapped to the four domains of the Dementia Seizure Spectrum™ Framework — connecting published evidence to the observable signals that appear in nursing records, monitoring alerts, clinical assessments, and due diligence documentation. This is where the intelligence gets used.
Seizure Events
Tonic-clonic, focal jerking, post-ictal confusion
Movement Changes
Lip smacking, hand automatisms, repetitive gesturing
Awareness Changes
Blank staring, sudden unresponsiveness, transient confusion
Behavioral Changes
Sudden mood shifts, acute agitation, post-episode fatigue
When fewer than 15 peer-reviewed papers directly address the submitted question, CRISP automatically activates Frontier Mode. This is disclosed before work begins — you are told what the evidence can and cannot support before the brief is produced.
Thin evidence is itself a significant finding. In the dementia-seizure space, the absence of direct literature on a question is often more strategically important than a shelf of moderate evidence. Frontier Mode expands retrieval across six concentric rings, labeling every finding by proximity to your original question.
Seagull Health reviews the CRISP output for accuracy and contextual relevance, writes the Seagull Health Commentary section, and delivers a structured PDF brief within one business day of scope confirmation. The brief contains all five components — Direct Answer, Scored Evidence, Gap Analysis, DSS Framework Mapping, and Commentary — in a format purpose-built for the context in which it will be used.
Delivered personally
Every brief is reviewed and delivered by Russ Barker, DHSc(c) — 25 years as a nursing home administrator, doctoral candidate in clinical nutrition and global health, dissertation focused on seizure risk assessment in long-term care. The output reflects that context. This is not automated delivery.
Triage Status
Based on the volume and quality of peer-reviewed literature directly addressing your dementia-seizure question. The triage state appears at the top of every brief — before any analysis begins.
The dementia-seizure literature directly addresses your question with high-quality studies. Evidence is sufficient to support clear conclusions and arguments.
Direct dementia-seizure literature exists but is limited in volume, quality, or recency. Evidence supports qualified conclusions with identified limitations.
Your dementia-seizure question is addressed only by lower-quality or indirect literature. Findings must be carefully qualified. Gap analysis takes on heightened importance.
Fewer than 15 papers directly address your dementia-seizure question. This is not uncommon — seizure recognition in dementia is a genuinely understudied clinical space. CRISP expands retrieval across six concentric rings. Thin evidence here is a finding, and it is strategically significant.
The Evidentiary Standard
Grading of Recommendations Assessment, Development and Evaluation. The international standard for evaluating the quality of clinical evidence and the strength of clinical recommendations. Used by the WHO, the Cochrane Collaboration, the American College of Physicians, and leading guideline bodies worldwide.
The internationally recognized methodology for assessing bias in clinical research. Developed by the Cochrane Collaboration — the global authority on evidence-based healthcare. Applied systematically to every paper CRISP retrieves.
These are not standards Seagull Health invented. They are the standards the peer-reviewed literature uses to evaluate itself — which is precisely why they hold up when the methodology is challenged.
Why this matters for due diligence
Whether the scrutiny comes from opposing counsel, an investment committee, or a regulatory body — here is what holds up:
Evidence quality scores are based on internationally published GRADE criteria — not Seagull Health's judgment
Bias assessment follows Cochrane methodology — reproducible by any qualified reviewer
Every score is visible. No composite black box. What isn't hidden cannot be challenged as opaque
Gap analysis explicitly states what the literature does not support — making the brief more defensible, not less
PubMed retrieval is real-time and documented — not a fixed dataset that can be challenged as outdated
What Transparency Means in Practice
Every paper retrieved on the dementia-seizure question shows its Evidence Quality, Bias Risk, and Clinical Relevance score individually — before the composite is calculated.
Clinical Relevance carries the highest weight (40%) because a rigorous study in the wrong population is worthless for dementia-seizure questions. The 30/30/40 formula and its rationale appear in every brief.
The MeSH terms and Boolean search logic used to retrieve the dementia-seizure literature are included in the brief. Retrieval is reproducible and can be challenged by any qualified reviewer.
What the dementia-seizure literature does not establish is documented as explicitly as what it does. In a field where seizure activity is systematically underdocumented, the gaps are often the most important findings.
Every finding is connected to its DSS domain — showing how published evidence maps to the observable signals that appear in nursing records, clinical assessments, and due diligence documentation.
The date and time of PubMed retrieval is documented in every brief. The dementia-seizure evidence reflects the literature as it existed at that moment — verifiable, timestamped, not a static training dataset.
By Market
A standard-of-care evidentiary brief mapping behavioral signals in the clinical record to published seizure presentations in dementia patients. Delivered within one business day. Flat fee.
Litigation intelligence →Pre-acquisition and post-close clinical intelligence identifying seizure-related liability exposure in memory care and skilled nursing assets.
Investment intelligence →Structured clinical intelligence on the dementia-seizure overlap for trial design, endpoint development, and safety signal interpretation.
Life sciences intelligence →DSS Framework licensing and CRISP intelligence layer integration for remote monitoring and care platforms built to detect seizure-related behavioral signals in dementia populations.
Technology partnerships →Request an Assessment
Clinical intelligence on the dementia-seizure overlap, commissioned for your specific question and delivered within one business day.