Agent Orange on Trial

by Brian Dunning
January 21, 2025

When I found this episode in my lap, it was like being handed a grenade with the pin already pulled. The question of Agent Orange's toxicity to people exposed to it has been considered settled science for more than half a century, and probably beyond any meaningful need for skeptical review. But the fact is that the more you look into the countless studies — including ones performed in very recent years — the more you find that many of them are contradictory. It's also a political hot potato: some conservatives often tend to defend chemical products from large corporations, and some liberals tend to see such products as inherently immoral. When I began the research for this episode, I honestly had no idea where it was going to take me; the one thing I knew for sure is that it would put me squarely in the center of a minefield.

During the Vietnam War, it was determined that massive spraying of defoliants could accomplish two things: first, to attack rural Vietnamese crops and force fighters to move to the cities; and second, to defoliate the jungle to deprive the NVA and Viet Cong of cover. So from 1965 to 1971, over 11 million gallons of Agent Orange were sprayed on croplands and on jungle warfare areas, from specially equipped Fairchild C-123 transport aircraft. And, as we all know, countless Vietnam soldiers were exposed to some degree, then came back home stricken with injuries believed to have been caused by the exposure.

On the US Veterans Administration website is a page that lays out the requirements for vets to receive benefits based on their presumptive exposure to Agent Orange during the Vietnam War. It lists 19 diseases that it says are caused by Agent Orange, and then a pretty detailed list of places and dates in which vets must have served. If you served in one of those times and places, and if you have a disease listed on the schedule, you are automatically entitled to certain benefits intended to fairly compensate victims without the need for lawsuits.

There are a billion details that I'm going to spare you, but the summary is that I did my due diligence here. The places and dates for vets to have served in is, in my opinion, a very good (though not perfect) reflection of where Agent Orange contamination is worst. For example, forward air bases from where the herbicide was loaded onto the C-123 planes which sprayed it are the most heavily contaminated places, and that's accurately reflected in the schedule.

However, after even more due diligence, I can't say the same thing for the schedule of diseases. Of the 19 that the VA lists, only one seems to have unambiguous support in the academic literature. That one is chloracne, which we'll talk more about in a moment; but most notably, all eight of the cancers listed on the schedule fall flat when compared against the literature. It's an obvious incongruity when most of the diseases that entitle victims to compensation don't match the diseases caused by the alleged agent; so let's take a detailed look at the history of how we got here.

Agent Orange was in use in the United States since its development in 1940 as an agricultural herbicide. Twenty years later when the Vietnam War effort faced a need for defoliants, nine large chemical companies were contracted to begin mass production of Agent Orange under the Defense Production Act of 1950. These nine infamously included both Dow and Monsanto.

Agent Orange was one of seven recruited into service, the so-called "rainbow herbicides" — each designated with a different colored stripe painted on the 55-gallon drum. There was Agent Purple, Agent Blue, Agent Green, Agent Yellow, Agent White, Agent Pink, and Agent Orange. They were the Reservoir Dogs of Vietnam.

Agent Orange is a 50/50 blend of two common agricultural herbicides used in the United States, called 2,4-D and 2,4,5-T. 2,4-D remains one of the world's most popular herbicides today, while 2,4,5-T went out of production in 1985. Either by itself is relatively harmless to people and probably could have been used in Vietnam with no ill effects. But the problem with 2,4,5-T is that its production can also produce dioxin in varying amounts, depending on a number of production factors. The nine companies producing Agent Orange had to meet unprecedented high production demands, so they needed to turn to cheaper source materials, faster reaction times, different temperatures, and relaxed quality control standards. Every barrel of Agent Orange in Vietnam had some random unpredictable amount of dioxin in it — and it is the dioxin that's responsible for all the claimed injuries suffered by both sides. Specifically, TCDD dioxin — the very worst kind.

Now if you're an experienced Skeptoid listener, you might be remembering back to episode #921 on the Seveso dioxin disaster[[/episodes/4921]] in Italy in 1976, considered one of the worst cases of TCDD dioxin contamination in history. There were 750 residents inside the highest area of dioxin concentration; of these, 193 people (mostly children) came down with chloracne, a long-term skin condition similar to cystic acne, known to be caused by exposure to dioxin or other halogenated aromatic hydrocarbons. The single worst case took seven years to fully recover. After 15 years, epidemiological studies found no increased rates of any other conditions. Chloracne is one of only two health effects of dioxin supported by strong evidence, the other being transient mild hepatotoxicity (temporary mild impairment to liver function).

The most-dioxin-poisoned person in history was Ukrainian politician Viktor Yushchenko, victim of a Russian assassination attempt in 2004. He was poisoned with a massive dose of TCDD dioxin and was found to have 50,000 times the normal level of dioxin in his system. He quickly became alarmingly stricken with chloracne so bad that he scarred up, leaving him permanently disfigured. Yushchenko's case was so uniquely severe that scientific papers were written about him and he provided tissue samples for years; and his case was instrumental in guiding our knowledge of dioxin's long term toxicokinetic effects in the body. Other than the chloracne, he suffered no other long-term effects.

Besides this and the hepatotoxicity, dioxin has a number of suspected effects including various cancers, reproductive and developmental problems, damage to the immune system, and endocrine disruption — suspected meaning there's insufficient data to provide a clear determination. Yet, when you read virtually any study about Agent Orange or dioxin, it almost always opens with a statement of fact that dioxin is a known cause of a long list of diseases including a number of cancers. At the top of this show I stated that the studies are often contradictory, and this is because when you look at studies examining the relationship between dioxin and any one specific disease (other than chloracne and hepatotoxicity), the majority of them have negative findings.

As an example, let's talk about prostate cancer. Prostate cancer is universally included in the list of conditions that Agent Orange is "known" to cause.

Here's a systematic review of previous research, published in 2024 in the journal Current Urology titled "How Agent Orange impacts prostate cancer risk, pathology, and treatment outcomes." Here's a snippet from the abstract, which finds men diagnosed with prostate cancer don't die any younger if they were exposed to Agent Orange, which suggests that Agent Orange did not impact the course of their disease:

…The literature seems to indicate that there are no significant differences in survivorship between exposed and unexposed veterans who are given a diagnosis of [prostate cancer]. Given these heterogeneous outcomes, the evidence does not encourage a significantly different approach to the diagnosis and management of [prostate cancer] for veterans exposed to [Agent Orange].

But then, conversely, here's a 2008 study of Vietnam veterans in Northern California published in the journal Cancer and titled "Agent Orange exposure, Vietnam War veterans, and the risk of prostate cancer." It found the opposite:

Individuals who were exposed to Agent Orange had an increased incidence of prostate cancer; developed the disease at a younger age, and had a more aggressive variant than their unexposed counterparts.

And just to make it more complicated, here's another 2024 study published in Acta Oncologica titled "Agent orange exposure and prostate cancer risk in the million veteran program." The Million Veteran Program, part of the US Department of Veteran Affairs, describes itself as "the nation's largest biorepository of Veteran data and is one of the most diverse cohorts of any genetic research program in the world."

…Associations with age at diagnosis of any [prostate cancer], metastatic [prostate cancer], and death from [prostate cancer] were assessed... On univariable analysis, exposure to Agent Orange was not associated with increased [prostate cancer], …metastatic [prostate cancer], …or fatal [prostate cancer]... Similar results were found when accounting for genetic risk. Agent Orange exposure history may not improve modern [prostate cancer] risk stratification.

I'll interpret that for you: The basic finding is that knowing someone's Agent Orange exposure history does not enhance prostate cancer risk prediction, but factors such as race, family history, and genetic predisposition may play a larger role. So we can't determine that Agent Orange makes you any more or less likely to get prostate cancer. This is basically what you'll find when you compare the diseases commonly attributed to Agent Orange to the literature: the newer and better the study, and the longer we've had to assess the epidemiology, the weaker and more tenuous the connection becomes.

So let's make it even more confusing. It can be rightfully argued that most people who receive compensation for Agent Orange exposure never developed chloracne, and to conclude that this proves they never had a harmful exposure to Agent Orange. However, this would be in error. Chloracne is absolutely associated with dioxin, but primarily to a high-dose, acute exposure to it; like Viktor Yushchenko or the victims at Seveso. Conversely, a chronic low-dose exposure to dioxin is much less likely to cause chloracne. Dioxin is a persistent organic pollutant that accumulates in the body, mainly in the fat tissue, and builds up over time. Its half life can be up to a decade. So a lack of chloracne does not at all prove that someone did not have chronic exposure to Agent Orange which would ultimately prove to be harmful.

However, the reverse is also true: having dioxin in the fat tissue does not indicate exposure to Agent Orange. All of us have dioxin building up in our fat tissue all the time. Dioxin is found naturally in the environment, and the leading vector of dioxin in humans is the consumption of animal fats such as meat and dairy. An average person today who never had any exposure to Agent Orange is likely to have a similar dioxin level in their fat to a veteran who had huge exposure during Vietnam. If a Vietnam vet gets a dioxin-related disease today, there's no medical way to know if it can be attributed to Agent Orange during the war or if they've just been living normally in the world ever since.

This is a really hard subject on which to maintain our objectivity, as most of us consider the Vietnam War to have been an injustice, and just as big an injustice to send American soldiers there to be killed and injured, and also an injustice to dump all that Agent Orange onto the Vietnamese people, combatants or not. The challenge to all of our critical thinking skills here is to balance our morals with our duty to good science. Consider this: If there was solid evidence that our previous conclusions on specific diseases attributed to Agent Orange were wrong, would you advocate that veterans' benefits for those diseases be terminated? I don't believe I could, despite my very public commitment to data-driven policy.

Instead, I'll end this episode in perhaps the best possible way: To all Vietnam veterans, thank you for your service.

By Brian Dunning
Follow @BrianDunning