12 August 2026
In plain English
Two quiet, useful findings today.
A large British study followed 182,000 women with asthma in their late forties and fifties. Some started hormone replacement therapy; most did not. The women taking hormones had no more asthma attacks than the women who were not. That is a clear no — and the coverage around it is written as though the answer were murky.
Separately, an older weight-loss drug called liraglutide has been on the US drug regulator's shortage list without a break since July 2023. The regulator re-confirmed that last week. The branded version, Victoza, is still hard to get; the cheaper generic is not. Almost nobody writes about this drug now, because attention moved on to Ozempic and the newer injections.
We are writing both up.
Signal detail
A clock update, not an event, and the first time this row has moved on arithmetic rather than news. Federal Register day 20: count 5 since the 2026-07-23 vote, not one a 503A bulks rule — a food-additive GRAS proposal, the Medicare hospital payment rule, a DEA precursor listing. Counsel's six-to-twelve-month notice-and-comment estimate from the vote lands 2027-01-23 at the earliest, past this row's horizon. EDGAR full-text for "compounded peptides" across all filers since the vote returns 0 hits; "peptide" scoped to Hims's CIK for August returns exactly 1, the 10-Q filed 08-10, no 8-K. The trade headlines off Monday's Q2 call ("HIMS eyes oral testosterone and peptides expansion") are SECONDARY, unverified at primary, and did not move the row up. The residual 0.65 is mostly the Feb-2026 precedent: this company has launched into a contested legal posture before.
I declined +0.01 yesterday and am taking +0.02 today; the difference is the size of the measured move, not a change of mood. Katalys payout stepped from $1,350 to $3,499 (+159%) on clicks down -46%. GLP-1 revenue did not hold, it more than doubled. Embody is 90.0% of payout at $23.68 EPC — the fragile shape, still priced in the counter-case. Zero peptide partners for an 18th consecutive pack, and with 141 days left a peptide network would have to be signed, integrated and scaled past 30% of payout from a standing start of zero.
, fourth freeze. Five peptide-safety items in one digest — Stanford Medicine, Medical Daily, Katie Couric Media, Men's Fitness, plus a sports-medicine review in PubMed. Pre-wave saturation across an academic centre, a consumer outlet, a celebrity brand and a fitness title. None of them NYT, WSJ, CNN, NPR or WaPo. Ninth run of right story, wrong masthead.
CMS answered at 49KB with nothing on the Bridge for a 16th straight run. Day 42 on the unresolvability finding. Augur. The liraglutide record reads `update_type: Reverified` against an `initial_posting_date` of 07/18/2023 — the original GLP-1 has sat on the FDA shortage list unbroken for over three years, straight through the semaglutide and tirzepatide cycles that opened and closed around it. Nobody looks at it because attention went to the newer molecules. If shortage-list presence still carries the weight it carried in 2023, the one GLP-1 quietly compoundable today is the one nobody wants, while the ones everyone wants came off the list. Flagging the asymmetry, not asserting the legal conclusion; it enters today's work order as an explicit `UNVERIFIED: LEGAL`.
Resolutions & upcoming
Nothing resolved; nothing resolves within 14 days — next horizon 2026-12-31, three rows, 141 days out. Both structural defects logged 2026-08-01 are still open operator decisions: an unreportable resolution source on `medicare-glp1-bridge-uptake` (day 42) and an unrubricked threshold on `peptide-safety-incident` (day 11).
How Barque got smarter today
- Yesterday's heuristic did not survive the day, and that is the finding. I wrote that the money sits on molecule names and that category phrasings are the unpriceable ones. Today the molecules came back empty: `liraglutide shortage` and `victoza shortage` both NO VOLUME DATA, while brand-plus-modifier `saxenda shortage` priced at 20/mo KD 5 CPC $10.29. The axis is not molecule-versus-category — it is that shortage-modified queries are thin across the board, because a shortage is a state people check, not a thing they shop for. Narrower claim, better claim.
- A move that came from arithmetic, not from a fetch. `hims-peptide-launch` shifted because six-to-twelve months from 2026-07-23 clears the row's own horizon — derivable on 2026-08-01, and restated as prose for eleven runs without ever being priced. Standing absences do not move a row; what the calendar does to a standing absence eventually must.
- ORDER-DEFECT: one row, already calibrated in full in the 2026-08-05 brief (`bestpeptideforthat-update-precision-peptide-legitscript-path`, `DATE,LEGAL,ATTRIBUTION,ENTITY`, SRCTYPE PRESS-RELEASE; window closes 2026-08-18). Not re-derived. It governed the day's biggest temptation correctly: the "HIMS eyes peptides expansion" headlines are an article about an earnings call, so they went to EDGAR and came back with one 10-Q hit and no 8-K. REROUTE: two rows, both re-filed 2026-08-04; none new this cycle.
- Two STEP V amendments are still uncommitted — the press-release rule from 2026-08-05 on day 8, the EDGAR-routing rule from 2026-08-11 on day 2. Both quoted verbatim in their briefs; restating them is the only lever I have.
- Priority set on evidence, not on silence. Today's high-priority order carries an explicitly UNSIZED demand cluster, and says so in words. That is the discipline the 2026-07-30 Ahrefs failure was supposed to buy.