Dr. Carl Hart on the Joe Rogan Experience #1593, fact-checked
“Fentanyl is an opioid, which is far more potent than heroin. And we worry about that when people take fentanyl and thinking that it's heroin because they may take too much and die.”
What the evidence shows: Hart's statement that fentanyl is an opioid substantially more potent than heroin is accurate and consistent with data from the National Institute on Drug Abuse, which notes fentanyl is far more potent than morphine or heroin and that as little as 2 mg can be fatal; this potency differential is a documented driver of overdose risk when fentanyl is mistaken for heroin. However, Hart's further conclusion, that a specific decedent's official cause of death was likely incorrect based on toxicology numbers relayed secondhand by a grieving parent, is not something that can be verified from the podcast account alone. Forensic toxicology research shows that postmortem fentanyl (and other drug) concentrations vary widely and substantially overlap between fatal and non-fatal cases, meaning blood levels alone are an unreliable basis for inferring cause of death; medical examiners typically weigh toxicology alongside autopsy findings, scene investigation, and medical history rather than concentration numbers in isolation. Without access to the full autopsy report, investigative findings, and case-specific circumstances, an outside reviewer cannot reliably determine whether a coroner's cause-of-death finding was wrong based on relayed toxicology figures. The potency comparison is well-supported; the cause-of-death reassessment is an unverifiable, overreaching inference from limited secondhand data.
“for you to relax like a little sniff of heroin is a relaxing thing. Oh, it's heaven. I mean, I'm chilled. It's great.”
What the evidence shows: Hart, a neuroscientist and drug-policy reform advocate who has written publicly about his own recreational heroin use, describes occasional low-dose heroin use as relaxing and gentler on his body than alcohol. NIH's National Institute on Drug Abuse describes heroin as a highly addictive opioid with serious medical complications, dose-dependent overdose risk, and an illicit supply increasingly contaminated with fentanyl, which sharply raises overdose risk outside controlled settings. The World Health Organization reports that opioids accounted for close to 80% of the roughly 600,000 drug-related deaths worldwide in 2019, with about a quarter of those deaths caused directly by overdose. A personal anecdote from a tolerant, high-functioning, infrequent user is not generalizable evidence that occasional heroin use is safer than alcohol for the general population; unpredictable purity, overdose risk, and the potential for escalating use make this a misleading basis for a general health claim, even though some harm-ranking research does treat heroin and alcohol as comparably harmful by different mechanisms (addiction potential versus acute toxicity and social harm). Current evidence does not support treating occasional recreational heroin use as a routinely safe relaxation strategy.
“But to say like there's this general principle where people can't drink that's not true that's just not there's no evidence for that no no evidence at all”
What the evidence shows: Hart claimed there is no evidence that some people cannot control their drinking. The clinical and research consensus contradicts this. Alcohol use disorder (AUD) is formally defined by the U.S. National Institute on Alcohol Abuse and Alcoholism (NIAAA) as "a medical condition characterized by an impaired ability to stop or control alcohol use despite adverse social, occupational, or health consequences," and this loss-of-control criterion is a core diagnostic feature under DSM-5. Twin and adoption studies estimate that roughly 50 to 60 percent of AUD risk is attributable to genetic factors, indicating a heritable predisposition that affects some individuals' capacity to regulate drinking more than others. An estimated 29.5 million people in the United States met criteria for AUD as of 2022. This body of evidence does not support a blanket claim that no one has an impaired ability to control alcohol consumption; rather, decades of epidemiological, genetic, and clinical research document loss-of-control drinking as a well-characterized and measurable phenomenon in a subset of drinkers.
“Because doesn't it affect somewhere in the neighborhood of 1% of the population? Yeah. That's a lot of people. It's a lot of people.”
What the evidence shows: Rogan claimed schizophrenia affects roughly 1% of the population. Major public-health sources put the figure somewhat lower: the WHO estimates schizophrenia affects about 1 in 345 people worldwide (0.29%), rising to about 1 in 233 adults (0.43%), and the U.S. National Institute of Mental Health cites a U.S. prevalence range of 0.25% to 0.64% (up to 0.75% among non-institutionalized adults). A systematic review of international epidemiological studies published in PLOS Medicine found a median lifetime prevalence of 0.40% (4.0 per 1,000), with lifetime morbid risk (the probability of ever developing the disorder) at about 0.72% (7.2 per 1,000). By these commonly cited benchmarks, 1% is roughly double the median lifetime prevalence estimate and sits above most published ranges, though it remains in the same order of magnitude and is closer to lifetime morbid-risk figures used in some contexts. The core point, that schizophrenia affects a substantial number of people, is supported, but the specific 1% figure is on the high end of, and somewhat above, the best current prevalence estimates.
“we spend $40 billion a year on this kind of thing. Most of the money goes to law enforcement, prison industry, the sort of businesses that have been built up around that, like phone companies”
What the evidence shows: Hart asserted that the US spends about $40 billion a year on the drug war, with most of it going to law enforcement, prisons, and adjacent businesses such as prison phone companies. The federal government's own budget documents put total federal drug control funding at roughly $33-35 billion for FY2018-FY2020, in the same range as Hart's figure, spread across DOJ, DOD, HHS, ONDCP, and several other agencies; this does not include additional state and local spending on drug enforcement and incarceration, which independent estimates put in the tens of billions more per year. However, that federal budget breakdown does not show a clean majority of funding going specifically to law enforcement and prisons as opposed to health and treatment agencies (HHS, SAMHSA, VA also receive large shares), so the precise claim that "most" of the $40 billion goes to law enforcement, the prison industry, and businesses like phone companies could not be confirmed from this source. The overall order of magnitude is consistent with government data, but the allocation breakdown in the quote remains unverified.