Daniel's asking the broad version this time. Not one disease, not one industry. The whole picture of what chemicals do to working people in twenty twenty-six, across the whole world, and how far we actually are from workers being able to know what they're breathing, know what would stop it, and get that protection without putting their job on the line.
And he's explicit about the power question up front. Workers want to stay healthy. Employers want the work done. Regulators only show up when those two things have already failed somebody badly enough to make a law.
Which is the part that never gets less bleak. The history of PPE is basically a ledger of people dying first and rules arriving after. Asbestos, benzene, silica. Every one of them followed the same sequence.
Silica's the one that still gets me. We've known since the nineteen thirties that breathing fine crystalline dust destroys lungs. And yet the mining safety rule in the United States went more than fifty years without updating the exposure limit. Fifty years. The rule finally got revised in twenty twenty-four and it's already delayed.
Fifty years is not a lag. That's an entire working life spent inside the gap between what the science says and what the law allows.
So let's put the scale of this on the table, because the numbers are hard to hold in your head. The most recent global estimates, for twenty nineteen, put work-related deaths at two point nine million a year. Roughly eight thousand people every single day. And here's the split that matters for Daniel's question. Only about three hundred twenty thousand of those are injuries. The other two point five eight million are disease.
Disease outpaces injury almost eight to one.
And within the disease pile, the stuff Daniel's asking about, hazardous substances, dusts, vapors, fumes, kills over a million people a year. One million seventy two thousand in twenty nineteen. That's the chemical exposure toll, just from the top categories.
So when people picture workplace danger, they picture the machine accident. The collapse, the explosion, the thing that makes the news. But the actual body count is slow. It's the lung that gives out twenty years after the job ended.
COPD alone accounts for over four hundred fifteen thousand of those deaths. And occupational exposure to vapors, gases, dusts and fumes probably drives fifteen to twenty percent of all COPD cases globally. That's not a small slice. That's a major driver of one of the world's most common chronic diseases.
And the cancer side?
Malignant neoplasms, nearly eight hundred forty three thousand work-related deaths. Asbestos is still the single biggest occupational carcinogen. The estimates disagree on exactly how many, but the more realistic number is around two hundred ninety thousand workplace asbestos deaths a year. The official global burden number was lower, around two hundred thirty nine thousand. The gap there is the undercounting problem.
Undercounting because the disease shows up decades later and nobody connects it to the job.
Right. A retired shipyard worker dies of mesothelioma at seventy one. The death certificate says cancer. It doesn't say the cancer was planted forty years earlier when he was lagging pipes. So the official count misses him. And the official count is what drives regulation. That's the loop.
A loop where the thing that would justify the rule is invisible because the rule doesn't exist to track it.
And the tracking itself is inconsistent. The ILO recognizes one hundred twenty exposure outcome pairs. The WHO ILO joint estimates only cover nineteen risk factors. So the recognized burden and the counted burden are nowhere near each other.
So we're probably low on everything. Which countries does the low estimate hurt most?
That's where the geography gets sharp. In high income regions, the leading work-related cause of death is cancer. Forty three to forty five percent of the disease burden. In low income regions, Africa, parts of Southeast Asia, the dominant killers are circulatory disease, respiratory disease, and communicable disease. Same work, different stage of the same story.
Because the cancer takes twenty years to show up, and in a place with lower life expectancy, you die of something else first.
The occupational cancer never gets a chance to be counted. Meanwhile the acute stuff, the poisonings, the infections, the lung failure, that's what shows up in the data. So the developing world's chemical burden looks like acute pesticide poisoning and respiratory failure, while the developed world's looks like tumor registries.
Let's talk pesticides then. That's the big one for agriculture in the developing world.
The meta-analysis on PPE and pesticide poisoning found that inadequate protective equipment use is associated with one point five seven times the odds of acute pesticide poisoning. So not wearing the gear nearly doubles your chance of getting poisoned. That's the clean statistical version of what Daniel's describing.
And the messy version?
Tanzania, a study of over a thousand cotton farmers. Forty eight percent reported symptoms of unintentional acute pesticide poisoning in the previous year. Nearly half. And only six percent sought formal medical treatment.
Six percent. So the symptom rate and the care seeking rate are almost in different universes.
And the reasons aren't mysterious. Illiteracy, poverty, regulation that's described in the Ghanaian research as ineffective. In Malawi, the same pattern. Farmers know the chemicals are dangerous. They don't have the equipment, the training, or the money to do anything about it. The threat is known and the protection is absent.
That's the part that reframes the whole question. Daniel's asking how far we are from workers knowing the threats and having access to protection. And the Tanzanian number says the knowing isn't the bottleneck. Half of them are symptomatic. They know.
They're not confused about whether the pesticide is bad for them. They're choosing between poisoning and income. And in a lot of these places, the pesticide is supplied by the same buyer who sets the wage. The power asymmetry is the whole system.
Two point three billion people globally are overexposed to pesticides by some estimates. One point one billion exceed the safe intake tenfold. That's not a workplace safety failure. That's an agricultural model with the safety engineered out.
And now layer heat on top of it. This is the part I think is the most under-covered story in occupational health right now. Chronic kidney disease of non-traditional origin. It's showing up in sugarcane workers, in migrant workers in the Gulf, in tropical and subtropical agriculture. It's not a chemical in the classic sense, but it functions the same way. An exposure that destroys the body over time, concentrated in the people with the least power.
The Guatemalan study, what did they find?
Sugarcane workers, core body temperature above thirty eight degrees on fifty three percent of workdays. Above thirty eight point five on fifteen percent. And the heat strain was significantly associated with cross-shift kidney injury. These are young men, agricultural workers, developing kidney failure from heat and dehydration. The disease is chronic, it's occupational, and it's completely unregulated in most of the places where it's happening.
So the developing world's fastest-growing occupational disease is one that doesn't even require a chemical. It's just the working conditions.
And Nepali migrant workers in the Gulf states are facing the same thing with, quote, limited labor protections. That's the polite phrase for no one is coming to help you.
Let's flip to the developed world, because the picture there is different and somehow also the same.
The United States is where the contrast gets uncomfortable. We have the oldest, most elaborate regulatory apparatus in the world, and it's currently being hollowed out. OSHA has fewer than six inspectors per million covered workers. Down from nearly fifteen in nineteen eighty. About seven hundred forty federal inspectors for eleven point six million worksites.
Seven hundred forty people. That's a medium sized company. Trying to cover eleven million sites.
And the standards themselves are ancient. David Michaels, who ran OSHA, estimates ninety percent of the chemical exposure limits in the United States were developed in the nineteen sixties. Most workplace chemicals have no exposure limit at all. So even if an inspector showed up, there's often no legal standard to enforce.
And the fines?
Average serious violation, forty seven hundred dollars. Average fatality, about twenty thousand. For context, a company that kills a worker pays less than the cost of a used pickup truck. And in OSHA's entire history, only a hundred forty four workplace fatalities have been referred for criminal prosecution. Against nearly half a million on the job deaths.
So the enforcement mechanism is a parking ticket. And the parking ticket is optional.
And right now the direction is further rollback. The EPA proposed a rule earlier this year that reverses some of the Biden era chemical accident prevention requirements. The framing is harmonization. Aligning EPA rules with OSHA's. But Adam Finkel, who ran OSHA's health standards program, put it plainly. Under the guise of harmonization, it's always in the direction of picking the weaker one and making the stronger agency do worse.
Harmonization as a one way ratchet. That's a phrase that should be in a textbook.
The numbers back up the direction. Chemical accidents resulting in releases went from eighty three in twenty twenty one to a hundred thirty one last year. Accidents with injuries or fatalities went from sixty to eighty nine. That's a fifty percent rise in five years. And a watchdog group counts at least fourteen hundred sixty three hazardous chemical incidents since January twenty twenty one. About five a week.
Five a week. And the Longview thing in Washington state was the one that finally made the news.
May of this year. A chemical tank collapsed at a pulp mill in Longview, Washington. Eleven workers killed. One of the deadliest US workplace accidents in decades. And when the Associated Press dug into it, the site had a documented history of safety complaints, poor maintenance, toxic releases. A chemical engineering professor at WPI said it plainly. It is a site that has not been maintained, and that creates risk to employees and risk to the public.
So it wasn't an accident in any meaningful sense. It was a known hazard that was allowed to sit there until it killed eleven people.
And the company's response was to say the responsible course is to let the full investigation proceed. Which is what every company says when the investigation is going to find the thing everyone already knows.
The thing everyone already knows is that the maintenance wasn't done because maintenance costs money and the workers were the ones who paid for the savings.
That's the perverse incentive Daniel named. And here's the counterintuitive part. There are voices saying the states will hold the line. Jeffrey Laskin at Rutgers made the point that New York and New Jersey have their own rules, often stronger than federal. So the safety net might not vanish. It might just become a patchwork where your protection depends on which state you're standing in.
Which is itself a kind of inequality. A worker in New Jersey gets benzene protections. A worker in a state with no plan gets the nineteen sixty eight standard and a prayer.
And the benzene standard itself is being reopened for comment right now. Methylene chloride too. The process is moving, just not in the direction of more protection.
So let me try to pull the riskiest occupations into focus, because Daniel asked for that specifically. Who's actually at the top of the chemical exposure list?
Agriculture is the biggest single pool. Pesticide exposure, heat stress, dust. Hundreds of millions of workers globally. That's the developing world's dominant risk.
And in the developed world?
Mining and quarrying. Silica, coal dust, diesel exhaust. Black lung never actually went away. The countertop fabrication industry is producing a new wave of silicosis in young workers cutting engineered stone. That's one of the ugliest current stories. Men in their thirties and forties with lungs that look like they spent sixty years underground.
Engineered stone is mostly silica. You cut it dry, you get a cloud of the exact dust that causes the disease.
And it's marketed as a premium kitchen surface. The worker who fabricates it is breathing the same stuff that killed generations of miners, but the product is new and the regulation hasn't caught up.
Construction is the other obvious one.
Construction is the cross-cutting risk. Asbestos in renovation, silica in concrete cutting, solvents in paints and adhesives, welding fumes. The construction worker is exposed to a rotating menu of hazards, often without any single one rising to the level of a named disease until decades later.
And the workforce is fragmented. Subcontractors, day labor, informal arrangements. The person who cut the concrete might have worked for three different companies that year.
Which makes the exposure history impossible to reconstruct. That's the undercounting problem again. The worker with occupational COPD from construction dust never gets counted because no single employer was responsible long enough to be identified.
Manufacturing, chemical production, waste handling. Those are the other obvious ones.
Healthcare. This one gets overlooked. Sterilizing agents, chemotherapy drugs, anesthetic gases. The operating room nurse handling cytotoxic drugs without proper PPE. The hospital sterilizer working with ethylene oxide. Healthcare workers face a genuine chemical burden that nobody thinks about because hospitals are supposed to be clean places.
Clean in the sense of sterile, not clean in the sense of safe for the person doing the sterilizing.
The sterilizer is the person who gets the concentrated exposure so the patient doesn't.
If I'm hearing the geography right, the developing world's burden is acute and visible. Poisonings, kidney failure, respiratory distress. The developed world's burden is chronic and delayed. Cancers, COPD, diseases that show up after the worker has left the industry.
The economic loss tracks the same divide. Work related disease and injury costs about six percent of global GDP. That's the economic loss from disability adjusted life years. In the Eastern Mediterranean region it's nine percent, in Africa eight percent, in high income countries four percent.
The places with the least money are losing the most of it to occupational disease.
The places with the most money are losing the least, which is why the incentive to fix it in the developing world is so weak. The workers who are dying are the ones whose labor is cheapest to replace.
That's the sentence that should make everyone uncomfortable. The market price of a worker's life is set by how much it costs to replace them. And in the developing world, that price is very low.
The regulatory intervention, which Daniel correctly identifies as the force that fills the gap, is exactly what's missing in the places where the gap is widest. Pakistan just announced a national occupational safety framework this month. That's a step. But it's a framework. The gap between a framework and an inspector with enforcement power is the same gap between a law and a worker who can afford to invoke it.
Let me ask the question Daniel actually ended on. How far are we from a world where workers can know the threats, know the protection, and have access to it without fearing their employer?
The knowing part is the most tractable. Safety data sheets exist. The ILO maintains a list of occupational diseases. The information is out there. But access to that information is stratified. A unionized worker in Germany has a safety representative who can explain the benzene exposure limit. A cotton farmer in Tanzania has a pesticide bottle with a label he may not be able to read.
Knowing is not just information. Knowing is literacy, language, time, and someone who can translate the label into what it means for your body.
The protection part is even harder. PPE is uncomfortable. That's not a trivial point Daniel raised. A respirator in a hot field is miserable. A chemical suit in a tropical greenhouse is a heat stroke risk. The worker who takes off the gear because it's unbearable is making a rational choice in the moment. The employer who doesn't enforce it is making a rational choice over the quarter. The disease arrives later and belongs to someone else.
The tension Daniel described, protecting your health versus keeping your employer on your side, that's the core. And in most of the world, the worker loses that tension every time.
Because the employer holds the job and the worker holds the body. And the body is the only asset the worker has that the employer doesn't already control. Until the disease takes it.
Then the employer hires someone else.
The thing I keep coming back to is that the regulatory apparatus in the developed world was built because the market failure was so total that even the people who owned the factories eventually had to accept limits. But the limits are now being rolled back in the name of efficiency, while the global burden keeps rising. So the direction of travel is not toward the world Daniel's asking about. It's away from it.
The United States is deregulating while the developing world is still waiting for its first real regulation. So the two halves of the world are moving in opposite directions and neither one is arriving at safety.
The undercounting means the pressure to act is always weaker than the reality. If the official number is two hundred thirty nine thousand asbestos deaths and the real number is two hundred ninety thousand, that missing fifty thousand people are the reason the next rule gets delayed. The people who would have been the evidence for the rule are the ones who aren't in the data.
The question of how far we are, the honest answer is, further than the data suggests. Because the data suggests a million chemical exposure deaths a year and the real number is probably higher. And even the official number isn't moving the needle.
The one thing I'd want a listener to hold onto is the eight to one ratio. For every worker killed by an accident, eight die of disease. The accident is visible, it's investigated, it's regulated. The disease is slow, it's invisible, and it's mostly ignored. That's the whole story of occupational chemical exposure in one ratio.
The second thing is that the worker already knows. The Tanzanian farmer with symptoms knows. The countertop fabricator with the cough knows. The knowing isn't the problem. The problem is that knowing doesn't come with any power to act on it.
That's the part that keeps me up. The information is there. The protection exists. The gap is entirely in the power to use them without losing the job.
Hilbert: I was in a fertilizer plant in nineteen seventy four. The ammonia line. They gave us these rubber gloves that were too big, and when you took them off there was this residue. Nobody said what it was. We just wiped it on our pants and kept going.
Hilbert: The guy next to me, Frank, he used to say the smell was the job. Like if you couldn't smell it, you weren't working hard enough. He died at fifty eight. Lung thing. His wife said it wasn't the plant, it was the cigarettes.
Hilbert: But Frank didn't smoke.
Hilbert: The thing about the respirator being uncomfortable, that's true. In the plant, nobody wore them. They were in a bin by the door. You'd put one on if the foreman walked by, then take it off when he left. The foreman knew. He'd done the same job ten years earlier.
Hilbert: What I notice now is the labels. You buy a can of anything and it's got warnings in six languages. But the guy mixing the stuff at the plant, he's got the same can with the same label and nobody reads it to him. The label's for the person who buys one can. Not for the person who opens a hundred a day.
Hilbert: I don't know what the answer is. But the idea that the worker doesn't know, that's not right. Frank knew. He just couldn't do anything with it.
The label being for the consumer and not the worker. That's the asymmetry in physical form.
It connects to the right to know gap. The consumer gets the warning because the consumer can sue. The worker gets the warning but can't act on it because the worker can be replaced. So the same information has completely different weight depending on who's reading it.
Frank knew the smell was the job and he was right. It was the job. The job was killing him and the smell was the proof.
The part about the foreman having done the same job is the part that gets me. The enforcement is supposed to come from the person whose job it is to enforce. But that person is also a worker. He's also choosing between his health and his job. He just has a slightly different job.
The entire hierarchy is built on the same trade. Everyone from the guy on the line to the foreman is making the same calculation about their own body. The only people not making that calculation are the ones who never go near the ammonia line.
That's why the regulator has to be outside the hierarchy. Because everyone inside it is compromised by the same incentive. The inspector who doesn't need the job is the only one who can tell the truth.
We have seven hundred forty of those for eleven million worksites.
Hilbert: Frank's wife got a settlement. Eighteen thousand dollars. That was nineteen ninety something. She bought a car.
Hilbert: I don't know what that works out to per year of his life.
That's the number that should be on the wall of every regulatory hearing. Eighteen thousand dollars for a lung.
The car probably lasted longer than the settlement.
The one thing I'd leave with is this. The world Daniel's asking about, where workers can know the threats and access protection without fear, that world is not a technology problem. The respirators exist. The data exists. The knowledge exists. The gap is entirely in the power to use them. And the power gap is widening in the places that used to have the strongest protections.
If the last fifty years of silica limits are any guide, the gap between what we know and what we enforce is measured in working lifetimes, not in legislative sessions.
The question is whether the next fifty years look any different. The data says no.
This has been My Weird Prompts.
Thanks to Hilbert Flumingtop for producing.
If you want to weigh in on this one, email us at show at my weird prompts dot com.
We'll be back soon.
