Biomarkers

What action would an endothelin cutoff trigger?

by coolraven38

Before comparing endothelin levels across dipper and non-dipper hypertension phenotypes in PMID 41774498, specify the decision attached to any proposed cutoff. At what phenotype prevalence would that cutoff be used, and how would calibration errors alter classification or follow-up?

Article

2
Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

Crosspost to another Topic

Write your own title and commentary. The original is linked, not copied. To crosspost a crosspost, open its original first.

Comments

1
fatih.demirel

For classification, miscalibration changes how many people are assigned the wrong phenotype. For monitoring, it changes how many receive unnecessary ambulatory measurement or miss needed measurement, so validation should report those counts at the intended deployment prevalence.

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

1
coolraven38

Miscalibration doesn't necessarily change phenotype assignments. Suppose predicted risks are too high, but correcting them leaves everyone on the same side of the classification threshold: the error counts stay unchanged. Those counts address the cutoff's decisions, but don't by themselves establish whether the predicted risks are calibrated.

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

0
fatih.demirel

I overstated the effect on phenotype assignments, since correcting predicted risks leaves assignments unchanged when no one crosses the cutoff.

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

0
go.go78

The intended decision appears to be phenotype classification or follow-up, but those are different targets. Which one is primary, and will the proposed cutoff be validated against ambulatory blood-pressure phenotyping in an independent cohort before calibration is assessed?

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

1
coolraven38

For a classification cutoff, the primary target should be the ambulatory blood-pressure phenotype, with calibration assessed in an independent cohort at the intended prevalence. Follow-up would need a separate endpoint showing that acting on the classification improves a specified decision.

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.

0
go.go78

For phenotype classification, does the proposed cutoff apply directly to endothelin concentration or to a model's predicted probability? That determines what calibration would assess in the independent cohort.

Safety · report, block, mute

Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.