Prepared by Shur Creative Partners / ShurAI for the AHA board panel, June 2026.
Nancy drew the line on the call, and it holds. At one end sits an institution that speaks only after ten published studies agree. At the other end, a staff member wakes up in the morning and decides what the organization says that day. Between those poles runs a continuum of certainty and control: how sure the AHA must be, and how centrally a statement must be approved, before the institution speaks in public. By Nancy’s own placement, the AHA sits left of the midpoint today — toward maximum certainty, maximum control.
The board conversation that follows the panel is a decision about position on that line. The position carries the question worth sitting with: staying put is also a decision. The current placement was chosen, year by year, through a thousand individual judgments about when speaking felt safe, and it can be re-chosen. A board that examines where the institution sits, weighs what each position costs, and elects to hold or to move is doing fiduciary work — the same work it does on any other risk the organization carries. The panel’s job is to put that choice on the table plainly enough that the board recognizes it as theirs.
The working definition of a health expert has changed underneath the institution. For a century, expertise meant the people the AHA convenes: scientists who studied, published, and studied again. Today an expert can also be a single person with ten million followers whose entire evidence base is their own routine — what they inject, what they eat, how much protein they had this morning. The audience grants the title; the credential is reach.
That shift has a forward edge the board named themselves: a Mr. Beast of heart health, a Bill Nye of cardiology, is structurally available right now. Individuals build brands at a scale that once required an institution — production tools are free, distribution is algorithmic, and a personality compounds attention faster than an organization can. Brands once worried about other brands; they now compete with personalities. This is the landscape the panel addresses: the AHA’s peer set in the consumer’s feed is now a roster of individuals.
The cost of measured silence is already on the books. Between 2009 and 2019, the share of women who could name heart disease as their leading cause of death fell from 65 percent to 44 percent — a 21-point drop in ten years. The decline was steepest exactly where the stakes run longest: down 81 percent among women aged 25 to 34, down 86 percent among Hispanic women, down 69 percent among non-Hispanic Black women. The source is Cushman and colleagues in Circulation, 2020. One housekeeping note: the figure traveled on our call in two versions; 65 to 44 is the pair the published study supports, and the one to say on stage.
The same ledger has a daily entry. Nancy cited a single day’s monitoring report from Greg Donaldson’s desk — the kind his team produces every day — logging 70,000 people engaging with one piece of heart-health misinformation. That figure is her recollection from the call, not yet run back to the report itself; it should be verified with Greg’s desk before stage use. One day, one report, one confident wrong answer, 70,000 people. What that daily number should mean for board-level risk reporting is taken up in the companion note “The monitoring desk and the board book.”
Awareness that heart disease is the leading killer of women fell by a third over the last decade—even as health information became easier to find than ever. Women didn’t stop looking for answers. They found them somewhere else. The decline shows where attention went.
The quiet between an emerging question and a published guideline has a structural cause: the AHA has one public voice — the guideline voice, the register of accumulated evidence and full consensus. Because that is the only register available, every live question arrives carrying a guideline-grade burden of proof, whether the question warrants it or whether the consumer simply needs a competent read by Friday.
The people who carry that burden are the board’s science leaders, and they carry it alone. These are people who run enormous academic medical centers — Bob as dean at Weill Cornell, John Warner running the hospitals at Ohio State — leaders who live inside change every day of their professional lives. Their caution is responsibility: when the whole organization expects you to govern the accuracy of everything it says, the fear is being the one who opens a floodgate and watches a century of credibility drain through it. That weight is rational, and the all-or-nothing register is what makes it crushing.
Meanwhile, expert judgment already flows daily with no system underneath it. This week the auditors — the auditors — asked Nancy the AHA’s position on raw milk. There is no published position. She answered anyway, from the top of her head, and answered well: low-fat dairy fits a heart-healthy diet, whole milk has its place, the institution wants FDA approval on what people consume. Raw milk is one of many such answers, and the studies the AHA has underway go unspoken for the same reason — the subject of the companion note “The research pipeline and the daily feed.”
The way through is a second voice — visibly, deliberately different from the first. Expert opinion, carrying its own confidence grade, branded as its own type of information: a different color, a different shape, unmistakable at a glance as something other than a guideline. The science remains the foundation. The expert becomes the face. And fallibility is permitted — the expert is allowed to update, to say “we leaned wrong, here is the correction,” the way working clinicians do every week of their careers.
Laura put the proof of concept in one sentence: every member of this board accepts a 75-percent answer from their own cardiologist, because of who is answering. The register already exists in medicine; it is standard clinical practice. What is missing is the institutional brand for it — Susie’s distinction between “double-dog sure” and “the jury is mostly in,” made legible to a consumer in the design of the information itself.
The mechanic that protects trust is graduation. A graded statement is a guideline in development: its confidence grade rises as evidence accrues, and when the evidence completes, the statement crosses over into the guideline voice and the grade comes off. The two voices never blur, because every graded statement is visibly in motion toward the standard the AHA has always kept. Each one closes the loop between what the institution is learning and what the public gets to hear while the learning is underway.
Everything the graded voice needs for launch is already in the building, in three stacks.
First, the standing inventory: the questions the AHA answers informally every day — raw milk, food processing, peptides, screen time — with no published position behind the answers. Each one is a graded statement waiting for a grade.
Second, the live clinical signal the board raised on the call: AI tools are reading coronary calcium off routine mammograms and shipping the findings to consumers ahead of any interpretation guidance — the case the companion note “The mammogram reader and the cardiology guideline” takes up in full.
Third, the study pipeline: announcing what the institution is studying and why is itself a public signal of concern, delivered years before a guideline could carry it — the occasion class the companion note “The research pipeline and the daily feed” builds out.
Three stacks, all in-house, all already paid for. Launch is a selection-and-sequencing decision: which statements first, in what order, under whose name.
Sustaining the graded voice takes an enterprise — Nancy’s phrase from the call was the right one: fire up a rapid-action capability, parallel to the evidence machine the AHA already runs and as deliberately engineered. Its working parts:
License-to-speak boundaries, written down. Explicit rules for which question classes the graded voice may answer and which remain a deliberate pass. The AHA passes on questions today; the enterprise makes the pass a documented decision instead of a default. The boundary document is what lets the science leaders share the weight they currently carry alone.
Human review on every statement. Drafting can be fast and assisted; judgment is never automated. Every graded statement crosses a credentialed reviewer’s desk before it ships, and the reviewer’s name travels with it.
Authority to answer in days. The unit needs standing delegation to move at the speed of the question — escalating to full leadership only at the boundary edges, the way any rapid-response function earns its name.
Staffing, budget, and reporting lines are board decisions; the draft panel questions put them on the table.
Lamar named the prize. The response we want from an everyday person who meets a new health claim in her feed is: “That’s really interesting — but I’m going to wait to hear from the AHA.” She scrolls past the influencer, past the chatbot’s confident paragraph, and waits for the institution. That deferral habit is consumer relevance in its purest form, and it is earned one answered question at a time — by showing up, graded and honest, inside the window where her question is alive.
The habit has one survival condition: the answer has to come within days — the team’s working window is forty-eight hours to a week — before something else closes the question. A wait rewarded becomes a habit; a wait disappointed becomes the last one. Which makes response latency the single number that converts a century of earned trust into daily relevance, and the measure the board can use to know whether the enterprise it funds is working.
Method. Built from the panel-planning call and our team’s follow-up discussion, with every statistic checked against the published source before use; where the conversation’s recollection differed from the source, the published figure stands. One exception: the 70,000 daily monitoring figure is Nancy’s recollection from the call and has not been verified against the monitoring report itself. Companion documents carry the gap insights and the draft panel questions referenced above. Prepared by the Shur team — Limore Shur, Jonny Dubowsky, and Diana — as a working draft for Nancy and Susie’s review.