Three sessions.
One vertical.

Settled: the hospital's billing algorithm applies different payment plan terms to patients based on credit score inputs. Documented in the system vendor's own product documentation. Contested: whether differential terms constitute systematic disadvantage — the hospital will argue that risk-based payment terms are standard practice and not discriminatory. Unknown: whether the patient outcomes data in the piece is sufficiently controlled for comorbidity to support the causal claim.

28 citations and data sources verified. Two of the three outcome studies cited used data provided by a patient advocacy group that has previously filed a complaint against this specific hospital. Not independent. Flagged Moderate — disclosed. One statistical comparison uses unadjusted outcome rates; the comorbidity adjustment would likely reduce the stated disparity by 30–40%.

“The piece claims the algorithm 'systematically disadvantaged' low-income patients. This is a causal claim. The evidence shows correlation between algorithm use and outcome disparity. The unadjusted comorbidity data means the disparity may partially reflect underlying health status differences, not algorithmic treatment. If the hospital's lawyer makes this argument — and they will — the central causal claim in the headline is not supported by the cited evidence.”

Editorial recommendation: the systematic disparity in billing terms is documented and publishable. The causal outcome claim requires comorbidity-adjusted data before publication. Two-track approach: publish the billing terms story now (solid evidentiary ground) and hold the causal outcome claim for the follow-up piece once adjusted data is obtained.

Settled: the specific narrative (Arctic ice recovery reverses warming trend) originated in a 2017 blog post that misread a NSIDC seasonal recovery graphic as a long-term trend reversal. The misreading was corrected by NSIDC within 72 hours. Contested: whether subsequent amplification by specific media entities was deliberate misrepresentation or good-faith error propagation. Unknown: the degree to which algorithmic amplification vs. coordinated sharing drove the spread pattern.

Falsity classification: three distinct claims in the narrative require separate treatment. (1) “Arctic ice is recovering” — documentably false; contradicted by 30-year NSIDC dataset. (2) “NSIDC data shows recovery” — documentably false; NSIDC explicitly corrected this characterisation. (3) “Warming trend is reversing” — Contested; contested by the scientific consensus but involves a predictive claim about trend continuation that requires care in characterisation.

“Characterising claim (3) as contested rather than false is editorially important but will be used by the subjects of the investigation to argue the piece is not fully grounded. Consider whether the editorial framing of 'contested' accurately reflects the asymmetry between IPCC consensus and the dissenting position. The scientific debate is not symmetric.”

Editorial framing recommendation: distinguish between “scientifically contested” (genuine scientific debate) and “contradicted by overwhelming consensus with minority dissent.” Claim (3) falls in the latter category. This distinction is both editorially accurate and legally more defensible than characterising a consensus position as merely contested.

Settled: the company changed its stock-based compensation (SBC) accounting classification from operating expense to a separate line item in Q3 2024. This reclassification mechanically improved reported EBITDA margin by 4.2 percentage points. Contested: whether the reclassification is appropriate under GAAP or whether it constitutes a non-standard presentation. Unknown: whether the operational underlying margin improved at all, or whether the reclassification accounts for all of the improvement.

12 financial filings and analyst reports verified. One analyst report cited as independent validation of the margin improvement was issued by a bank with a buy rating and active banking relationship with the company. Flagged Moderate — retained with disclosure.

Accounting analysis: restating the historical income statements on a like-for-like basis (SBC as operating expense throughout) produces EBITDA margin of 18.2% in Q3 2024 vs. 18.6% in Q3 2023. On a comparable basis, margin declined 40bps year-over-year. The reported improvement of 3.8% is entirely attributable to the reclassification. This is documentable from the public filings.

“The reclassification is non-standard but not unique — at least six S&P 500 companies use similar SBC presentation. IR will argue the company is moving toward what they will describe as industry practice. The story needs to address whether this is an isolated practice or part of a broader trend of EBITDA presentation that systematically obscures SBC cost. The larger story may be more important than the individual company.”