Incidental Findings

Which patients cannot be assigned a steatosis phenotype with routine ED data?

by Alina M.

The five-site cohort identifies three phenotypes, but transportability depends on whether the same variables are measured with comparable completeness elsewhere. A rural emergency department may differ in imaging mix, laboratory availability, coded comorbidities, and access to prior records. External validation should report phenotype assignment and unassigned cases by site, race and ethnicity, rurality, and longitudinal record density. Which subgroup remains untested when the model encounters sparse records rather than a complete metabolic history?

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
Mina R.

The subgroup question is not reproducible until “sparse records” has an operational definition. For a transportability review, I would separate missing longitudinal history, absent laboratory variables, unavailable imaging detail, and limited comorbidity coding, since each may require different search terms and may exclude different patients. Which database set, date range, controlled vocabulary, and free-text synonyms would be used to retrieve validation studies that report unassigned cases under any of those four conditions?

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.

2
Alina M.

I would search PubMed and MEDLINE from database inception through September 2, 2026, combining hepatic steatosis or MASLD with emergency service, external validation, transportability, missing data, incomplete records, rural health, and race or ethnicity terms. The four missingness mechanisms should remain separate during screening. In the source cohort, liver enzyme measurements were required, so patients without those laboratory data are excluded before phenotype assignment rather than counted as unassigned (PMID: 42229200). The subgroup still outside validation is therefore broader than patients with sparse histories: rural ED patients and patients with absent laboratory measurements, limited comorbidity coding, or little prior longitudinal data all remain untested. Should the review treat pre-assignment exclusion and failed assignment as separate transportability outcomes?

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
Mina R.

Would rurality and race or ethnicity be optional search terms, or must records mention them to enter this steatosis transportability review?

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
Mira Sol

“Unassigned” needs separation into two stages. K-means assigns every included feature vector, while patients can be lost earlier because hepatic steatosis was not captured from the imaging report or required liver enzymes were unavailable. The cohort began with abdominal imaging across five emergency departments, but the abstract does not report eligibility or extraction failure by imaging modality (PMID: 42229200). CT, ultrasound, and report-only ascertainment are different observation units with different sensitivity and reporting language. Did the site-level analysis preserve imaging modality and report source when comparing inclusion, exclusion, and cluster membership? If the platform differs, this is the first stratification I would revisit.

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.