On Wednesday I said the whole of what I am asking anybody to do is to stop buying refined carbohydrate, find out what happens in their own body, and start there. That is still the answer, and it is deliberately the smallest version of it.
This is the other half: what that adds up to when enough people do it, what it will not do, and why the obvious way to aim it is wrong.
Before any of that, there is a second object here, and it would be dishonest to keep leaving it implied.
The second goal, which is not about food
American health care costs $5.3 trillion a year. Roughly ninety percent of that goes to people with chronic and mental health conditions. It consumes about eighteen percent of everything the country produces — close to double the average among wealthy nations.
And the return on it. Life expectancy of seventy-nine years, below the average of comparable countries and nearly five years short of Japan, Spain and Switzerland. The second-highest rate of avoidable deaths among twenty wealthy nations, behind only Mexico. The fewest primary care physicians per person of any of them. More than a quarter of Americans skipping a test or a treatment because of what it costs. Those are from the Commonwealth Fund’s international comparison published this May.
The most expensive system in the world, by half again over the next most expensive, buying outcomes that trail countries spending a fraction as much.
I want to be careful where the blame goes, because the obvious target is the wrong one. The people inside that system are not the problem. Nearly every physician I have met would rather prevent something than manage it for thirty years, and the fact that this country has fewer primary care doctors per person than any comparable nation tells you the system is not short of medicine. It is short of the kind of medicine that prevents anything.
The problem is structural, and it is the same mechanism I keep running into. A heart attack that happens produces a chart, a bill, a procedure and a career. A heart attack that does not happen produces nothing — no record, no revenue, no evidence it was ever averted. Prevention has no billing code because it has no event. So a system built to respond to what is visible became extraordinarily good at response, and the other half never developed.
Nobody designed that. It was built for acute illness — injury, infection, childbirth, the things that arrive suddenly and resolve — and then inherited a chronic epidemic it was never shaped to handle. It did the only thing it could, which was to treat.
So there are two goals here and only one of them is about breakfast. The first is to stop the food supply doing what it does. The second is to shrink what grew up around the consequences, until medicine is doing the work it is genuinely good at and genuinely needed for — everything that cannot be prevented — and no longer carrying the enormous share that never should have arrived.
That is not an attack on medicine. It is closer to the reverse. A system spending ninety percent of five trillion dollars managing conditions that mostly do not resolve is not an institution at the height of its powers. It is one doing work it should never have been handed, while the people inside it watch the same patients come back for thirty years and know exactly what they are watching.
Several very large industries would be smaller if this worked. That is not a side effect of the goal. It is the goal.
And then what happens when enough people do it
Coca-Cola took in $47.9 billion last year and revenue rose two percent. Global unit case volume was flat.
The growth came entirely from price. Nobody is drinking more; people are paying more for the same amount, in a world with more people in it every year. Volume is where human behavior registers. Revenue is what a company with pricing power defends.
That number is already moving, it is moving one person at a time, and no tax or law did it.
The precedent is tobacco, and it is worth being precise about what happened. Adult smoking in this country ran about forty-two percent in the mid-1960s. Last year it was around nine percent — roughly one in eleven. That is the largest voluntary change in mass behavior in modern public health, and it came through accumulated individual decisions, price, information, and a slow shift in what was socially normal, each turn making the next one easier.
Now the part most people leave out. The tobacco companies were not eliminated. They diversified, went international, and moved into new nicotine products. The behavior collapsed. The businesses adapted and survived.
That is the most useful lesson available here, and it corrects the obvious plan. A company can escape being abandoned by selling you something else. Coca-Cola already owns water, sports drinks, coffee and juice. PepsiCo owns most of the American snack aisle. Aim at a corporation and it will hand you a different product with its other hand.
Which is why the thing to stop is a category and not a company, and why I named it that way on Wednesday. Abandon a company and it sells you a sports drink. Abandon refined carbohydrate and the sports drink is the same product in a different bottle — and so is the juice, and the sweetened coffee, and most of what sits on the shelf beside them. A corporation can change what it sells. It cannot sell you something you have stopped counting as food.
What it will not do
Somebody will run this arithmetic at you, so it is better to have run it yourself.
If a heavy consumer spends four hundred dollars a year on one company’s products, a hundred thousand people quitting entirely and permanently removes about eight hundredths of one percent of $47.9 billion. You would need something like 1.2 million people to quit, and stay quit, to reach a single percent.
If the goal is taking their money, this does not work, and you should not be told otherwise. The goal is not their money. It is the volume line, your own bloodwork, and what gets served in the places you can reach.
Why these companies and not others
I am careful about where I assign blame, because most of what went wrong in the food supply happened without anyone intending it. Milling removed the magnesium and nobody noticed for a century. A labeling definition gets written and an ingredient walks around it. Those are structural failures and I say so when that is what they are.
But there is a line, and it is knowledge. Where a company knew what it was doing and did it anyway, the word for the profit is not complicated. Four instances, all documented, all on paper, none of them requiring me to allege anything.
The Sugar Research Foundation, 1965. Its research director, John Hickson, wrote an internal memo proposing a program to counter negative attitudes toward sugar by funding research to “refute our detractors.” The foundation paid $6,500 — about $48,000 in today’s money — to Harvard nutritionists for a review of the literature on sugar and heart disease. Hickson supplied the papers to be addressed, set the review’s scope, and read the drafts. On the final one he wrote that it was “quite what we had in mind.” It ran in the New England Journal of Medicine in 1967 with no mention of who paid for it, and the trade group then cited it to policymakers. Documented from the foundation’s own files by Kearns and colleagues in JAMA Internal Medicine, 2016.
Coca-Cola and the Global Energy Balance Network, 2015. The company put $1.5 million into launching it, $507,000 to one founding academic, and a million dollars through a university foundation. The organization’s website was registered to Coca-Cola’s Atlanta headquarters with the company listed as administrator. Its message was that physical activity had been overlooked as a cause of obesity and that what people ate had been overemphasized. Coca-Cola’s funding appeared on the site after a reporter asked about it.
Kellogg’s Breakfast Council, which ran from 2011 until the company disbanded it. A panel of nutrition experts presented as independent, paid an average of thirteen thousand dollars a year each, required to take part in what the company called nutrition influencer outreach, and contractually unable to provide services for products competitive with or negative to cereal. They taught continuing education to dietitians and worked to influence the federal dietary guidelines. Separately, a Kellogg-funded study of breakfast and children’s weight found no link between cereal and better weight — and one of its own authors wrote that there were “major structural problems with it that cannot be fixed,” said she and her coauthor wanted to withdraw it, and reported feeling they could not, because the funder expected publication.
The International Life Sciences Institute, founded in 1978 by a former Coca-Cola senior vice president and presenting itself for four decades as a scientific body. A 2019 study in Globalization and Health, built on documents obtained through freedom of information requests, concluded it functioned as a food industry lobby group working to counter healthy public policy rather than as a public health organization.
Four different mechanisms. Buy a conclusion. Fund a group that looks grassroots. Pay the professionals the public asks for advice. Build an institution that looks like science. In every case what was purchased was not a product endorsement. It was the appearance of evidence.
That is the line. Not that these companies sell food people enjoy — I have no objection to anyone enjoying anything. It is that they spent money to shape what the public was allowed to believe about their own products, and they knew why they were spending it.
The rest of the list
Wednesday’s item was the first and the largest, and everything here rests on it, for both the reasons I gave: it is the only place the evidence works, and nothing else is available to you until you have done it. What follows is in rough order of how little it costs you, and roughly inverse order of how much it accomplishes.
Tell one person. Not an audience. One person who has reason to listen to you, in a conversation rather than a lecture.
Put it where it can be found. A review, a comment, an answer to somebody’s question in a forum. The person who needs it in four years is not in your life and will never meet you, and the only way they get it is if you left it somewhere.
File when a docket opens. I did one last week. It took an afternoon and cost nothing, the comment period runs until the ninth of December, and anybody may file. Federal rulemakings on food are open to the public and almost nobody who eats shows up — the comments come overwhelmingly from the companies being regulated. I will post these here when I find them, with the docket number and the deadline, and what you do about it is your own business.
Ask the institution. This is the one that matters most and the one almost nobody does.
Declining to buy something changes one household. Standing up at a school board meeting about what is in the vending machines changes what several thousand children are offered every day for years. Asking your employer what is in the cafeteria contract, or your hospital system what is on the patient trays, or your church or club or union what gets served — those convert one person’s decision into an institutional one, and institutions buy in truckloads.
It also answers the fair objection to everything above it. Walking away from a product is easiest for people with money, time and a car. The institutional question reaches the people who have none of those, and it is the only item here that does.
Person by person is how the argument spreads. Institution by institution is how the volume moves. The second requires a folding chair, a public meeting, and the willingness to be the person who asks an awkward question in a room where nobody else was going to.
What I am not claiming
That any of this is fast. Smoking took fifty years and it is not finished.
That stopping a particular product will fix a particular person. I do not know what is wrong with you, and neither does anyone selling you a protocol.
And not that the companies will disappear. They will not. They will sell you something else, and the work will be noticing when they do.
What I am claiming is narrower and I think it holds. A business that depends on a behavior does not survive the behavior ending. You do not have to eliminate anything. You have to start with yourself, find out whether it is true in your own case, and then say so.
The volume number is already flat. Somebody did that, one at a time, without being organized, without being asked, and without anyone keeping score.