116 Aspre Presention 1 Scratch Pad
116.1 Summary
- Brian Locke ASPIRE Presentation
- What I’m All About:
- Why Hypercapnia:
- Syndromes; Nosology
- Goal: identify everyone who fits the syndrome of Hypercapnic RF – so that we can begin this process.
- Critique of the Sufficient-Component Model
- Consider this as a missing data problem? What is the propensity of p(ABG)?
- Comp Sci. Dorks like systems.
- Initially: believer in the IHI / Evidence-Practice gap: If we could just get people to do the right (known) thing, we’d drastically improve care
- Overton window: range of policies acceptable to the population
- -> Exchange policies w process improvements: In healthcare, there is a narrow Overton Window of improvements one can make to improve care: ie. Sepsis 30cc/kg
- Drive to research: I’m interested in how we know what to do for each patient: how can we move the window? Start w “Clinical Epidemiology”?
116.2 Slide outline
116.2.1 Slide 1
- Brian Locke ASPIRE Presentation ### Slide 2
- What I’m All About:
- Comp Sci. Dorks like systems.
- Initially: believer in the IHI / Evidence-Practice gap: If we could just get people to do the right (known) thing, we’d drastically improve care
- Overton window: range of policies acceptable to the population
- -> Exchange policies w process improvements: In healthcare, there is a narrow Overton Window of improvements one can make to improve care: ie. Sepsis 30cc/kg
- Drive to research: I’m interested in how we know what to do for each patient: how can we move the window? Start w “Clinical Epidemiology”?
- How do we predict what will happen to a patient? Diagnose+Data
- How to do this? Think more sharply, collect and analyze data better. ### Slide 3
- Why Hypercapnia:
- Presenting my framing here not because the data is new, but I’m curious if y’all agree with the framework as I propose it:
- Why Hypercapnic Respiratory Failure?
- 80,000 hours argument ### Slide 4
- Syndromes; Nosology
- Is this the right approach? Studying Hypercap RF vs Each cause.
- Several underwhelming records: we still don’t really know what to do for sepsis; ARDs is slightly better but not clear benefits are specific to that group
- Not diseases
- Benefit of syndromes: easily clinically recognizable: think improvements in Sepsis quality of care.
- Bad for: identifying who will benefit from a given intervention – as patients presumably respond to diseases based on their physiologic state (ie. interventions have their effect by changing physiology: this is modern medicines ‘cosmology’). – key is to identify prognostic features; relative effect modifiers
- Promise is in phenotyping – breaking down the syndromes into subsets with predictable responses / implications, presumably based on physiology. Maybe it corresponds to current diagnosis, maybe other info (? For hypercap: maybe it has to do with cant vs wont breathe, or markers of control system stability vs respiratory frailty; the PIRO model (predisposition, insult, response, organ dysfunction) ### Slide 5
- Goal: identify everyone who fits the syndrome of Hypercapnic RF – so that we can begin this process.
- Why do this? I think our current definitions – E.g. OHS – miss a lot of people who share physiologic barriers. ### Slide 6
- Critique of the Sufficient-Component Model
- Consider PKU: Genetic defect causes disease when phenylalinine consumed.
- -> in practice, everyone eats phenylalinine: genetic attributable fraction 100%
- -> however, modification of diet avoids development. Thus, if phenylalanine consumption varies then environment contributes
- Consider, the hypothetical world where everyone has the PKU gene, but phenylalanine varies (lower diagram) – then, the disease is ALL environmental.
- what is the relationship to mendelian randomization?
- -> population attributable fraction (and percent variance explained) is not generalizable, and changes as covariates change. Applied to Hypercapnia? ### Slide 7
- Consider this as a missing data problem? What is the propensity of p(ABG)?
- Could P(ABG) – by provider ideall (or institution) work as an instrument? (to wiggle the likelihood of a diagnosis… could you test that? Are high p(ABG) environments or providers more likely to apply the diagnosis?
116.3 Learning objectives
- Brian Locke ASPIRE Presentation
- What I’m All About:
- Why Hypercapnia:
- Syndromes; Nosology
- Goal: identify everyone who fits the syndrome of Hypercapnic RF – so that we can begin this process.
116.4 Bottom line / summary
- Brian Locke ASPIRE Presentation
- What I’m All About:
- Why Hypercapnia:
- Syndromes; Nosology
- Goal: identify everyone who fits the syndrome of Hypercapnic RF – so that we can begin this process.
116.5 Approach
- TODO: Outline the initial assessment or decision point.
- TODO: Outline the next diagnostic or management step.
- TODO: Outline follow-up or escalation criteria.
116.6 Red flags / when to escalate
- TODO: List red flags that require urgent escalation.
116.7 Common pitfalls
- TODO: Capture common errors or missed steps.
116.8 References
TODO: Add landmark references or guideline citations.