Calley Means & Casey Means, MD on the Joe Rogan Experience #2210, fact-checked

What the evidence shows: Federal data broadly support the magnitudes cited, though the numbers are rounded up. NIDDK (drawing on NHANES 2017 to 2018) reports about 73.1 percent of US adults are overweight or have obesity (30.7 percent overweight plus 42.4 percent with obesity), close to the 74 percent stated; CDC's most recent cycle puts the combined figure at 72.4 percent. On diabetes, NIDDK reports roughly 11.6 percent of the total population had diabetes (2021) and that more than one third of adults (about 97.6 million, near 38 percent) had prediabetes, so the combined adult total lands near the 50 percent claimed once the two groups are added. The historical framing is also broadly supported: CDC long-term surveillance shows diagnosed diabetes was well under 1 percent when national tracking began in 1958, consistent with the roughly 1 percent in 1950 figure. Caveats: these totals lump in prediabetes, a category whose clinical significance is debated, and the historical and current figures reflect all diabetes rather than type 2 alone (though type 2 is roughly 90 to 95 percent of cases).

What the evidence shows: The autism figures match CDC surveillance data. The CDC Autism and Developmental Disabilities Monitoring (ADDM) Network reported overall ASD prevalence of 27.6 per 1,000 (one in 36) among 8-year-olds for 2020, with rates ranging from 23.1 per 1,000 in Maryland to 44.9 per 1,000 (about one in 22) in California, so the one in 36 and California one in 22 figures are correct. The one in 150 figure was the CDC's first ADDM estimate, based on 2000 and 2002 surveillance years, so that comparison is also accurate. The claim that 25 percent of men under 40 have erectile dysfunction is at the high end and not well supported: a peer-reviewed review of ED in young adults cites a US cohort prevalence of 14.2 percent and notes that estimates of up to 35 percent are outliers from broad-definition studies, while much of the literature reports 5 to 10 percent for men under 40. The autism claims are accurate, but the 25 percent ED figure overstates the more commonly reported prevalence.

What the evidence shows: The 77 percent figure is real: the Pentagon's 2020 Qualified Military Available study found that about 77 percent of Americans aged 17 to 24 would be ineligible to enlist without a waiver, up from 71 percent in earlier Department of Defense data. However, that total is driven by many combined reasons, not just obesity or drug abuse. A peer-reviewed analysis describes the 77 percent as reflecting ineligibility due to weight, substance use, or mental and physical health conditions, and notes that most young adults met several ineligibility criteria at once. Obesity alone accounts for roughly a third of the age group (a separate study using 2015 to 2020 data found 37 percent ineligible on the basis of high BMI), while drug and alcohol use, medical or physical conditions, mental health, aptitude, and conduct make up the rest. Casey Means correctly cites 77 percent but misattributes the entire figure to obesity or drug abuse, when those two factors account for only part of it.

What the evidence shows: Heart disease is correctly identified as the leading cause of death in the United States, but the specifics are overstated. In 2022 CDC recorded 702,880 deaths from diseases of the heart (about 21 percent of all deaths, ranked number one), not around 800,000. The 805,000 figure Means appears to invoke is the estimated number of Americans who have a heart attack each year, not the annual death toll. The claim that it is almost totally preventable also overstates the evidence: CDC estimated that roughly 34 percent of premature heart disease deaths are preventable, and while large studies attribute a high share of cardiovascular disease to modifiable risk factors (the PURE cohort attributed over 70 percent of CVD cases and deaths to modifiable factors), a substantial portion of deaths, including those at older ages, is not readily preventable, so almost totally is an exaggeration.

What the evidence shows: A 2024 University of New Mexico study (Garcia et al.), published in Toxicological Sciences, used pyrolysis gas chromatography mass spectrometry to analyze 62 human placental samples and detected micro and nanoplastics in every one of them (62 of 62, 100 percent), at concentrations ranging from about 6.5 to 685 micrograms per gram of tissue. Polyethylene was the most common polymer (about 54 percent), followed by PVC and nylon. The study confirms the claim that 100 percent of the dissected placentas contained microplastics. The authors note the health effects of these levels remain unknown, so the finding establishes ubiquitous presence rather than a demonstrated harm.

What the evidence shows: The core science checks out. The US EPA estimates roughly 76.5 million pounds of atrazine are applied annually in the United States, so the 70 million pound figure is in the right range, and the European Union banned atrazine in 2004 (making it illegal there). Peer-reviewed toxicology confirms atrazine increases aromatase and aromatase activity, the enzyme that converts androgens such as testosterone into estrogens. The claim that the US buys atrazine from China and Germany is not supported: the ATSDR toxicological profile documents that atrazine is manufactured domestically at 24 US facilities (with the largest processing volumes in Louisiana, Alabama, Mississippi, and Missouri) and that historical import data showed only a negligible amount imported, so the country-of-origin framing is misleading even though the volume, EU ban, and aromatase mechanism are accurate.

What the evidence shows: The national survey of nutrition education in US medical schools (2008-2009 update) found that only 26 of 105 responding schools (25 percent) required a dedicated nutrition course, so roughly 75 percent did not, with students receiving an average of just 19.6 contact hours of nutrition instruction. That makes the 80 percent figure close but slightly high, and 'not one minute of nutrition' is an overstatement, since most schools still teach some nutrition within other courses even without a standalone required course. The 90 percent figure traces to a 2017 RAND analysis (echoed by the CDC) finding that about 90 percent of US health care spending goes to people who have one or more chronic or mental health conditions. Fact-checkers note that this describes spending on people who have chronic conditions, not spending that directly treats chronic disease, so phrasing it as costs 'tied to' disease is defensible while the stronger 'treats chronic disease' framing has been rated false. Overall the thrust of both claims is supported, with the specific percentages somewhat imprecise.

What the evidence shows: The general premise that roughly 7,000 steps per day yields meaningful health benefits is well-supported, but the specific figures are inflated. The largest evidence base, a 2025 Lancet Public Health dose-response meta-analysis pooling 57 studies and over 160,000 adults, found that 7,000 versus 2,000 steps per day was associated with a 14 percent lower risk of type 2 diabetes and a 38 percent lower risk of dementia, well below the claimed 40 to 60 percent for those conditions. The 40 to 60 percent range in that analysis applies mainly to mortality outcomes (47 percent lower all-cause and cardiovascular mortality), not to the diseases she named. Obesity and gastric reflux were not among the outcomes the meta-analysis examined at all. Her 3,500 steps figure also understates the US average: peer-reviewed estimates for US adults range from roughly 5,100 steps per day (pedometer) to about 6,500 or higher (accelerometer), with no source supporting 3,500.

What the evidence shows: The claim traces to a widely cited 2016 BMJ analysis by Martin Makary (referenced as Marty McCurry) and Michael Daniel of Johns Hopkins, which estimated more than 250,000 US deaths per year from medical error, a figure that would rank third behind heart disease and cancer. That estimate was derived by extrapolating preventability rates from a handful of older studies onto 2013 hospitalization totals, with no formal meta-analysis, and the difficulty of judging whether an adverse event was truly preventable is a recognized weakness. Competing analyses put the number far lower: a British study cited by AHRQ found only about 3.6 percent of inpatient deaths were potentially avoidable, translating to roughly 26,000 US preventable deaths per year, an order of magnitude below the 250,000 figure. The CDC has not adopted medical error as a death-certificate category or ranked it as the third leading cause. The underlying paper is real and influential, but its headline ranking is one of the most contested statistics in health-system research rather than an established fact.

What the evidence shows: The two figures count fundamentally different things, so the comparison is not apples to apples. The roughly 10,000 US number (traced to a 2013 Pew analysis) includes all substances that may contact food, such as packaging materials, processing aids, and self-affirmed GRAS ingredients, whereas the ~400 EU figure counts only the numbered E-additives. The European Food Safety Authority states that more than 300 substances (315 approved before January 2009) are authorised as numbered food additives in the EU, but these numbered additives are only a fraction of the total substances permitted for food use in Europe, which also includes flavourings, enzymes, and processing aids not captured by the count. The underlying concern is real: under the US GRAS system, manufacturers can self-determine a substance is safe and market it without notifying or being reviewed by the FDA, and analyses estimate roughly 1,000 such decisions were never reported to the agency plus thousands more ingredients never reviewed. The framing that the US simply allows 25 times more food chemicals than the EU is not supported, though the difference in regulatory philosophy and oversight is genuine.

What the evidence shows: The 67 percent figure traces to a 2021 JAMA study (Wang et al.) that analyzed NHANES dietary data for US youths aged 2 to 19 using the NOVA classification: the share of total energy from ultra-processed foods rose from 61.4 percent in 1999 to 2000 to 67.0 percent in 2017 to 2018. So the number is correctly stated and well sourced for that period. The one caveat to the word now is that more recent NHANES surveillance (CDC data through 2023) shows youth ultra-processed intake has edged down to roughly 62 percent, so 67 percent reflects the 2017 to 2018 peak rather than the very latest data. The core claim that ultra-processed foods make up about two-thirds of children's calories is accurate.

What the evidence shows: The underlying direction is real (the U.S. is by far Novo Nordisk's most important market and its GLP-1 drugs carry very high U.S. list prices), but the specific figures are wrong or inflated. At its March 2024 peak Novo Nordisk was the 12th most valuable company in the world, at about $604 billion, not the ninth, and it was never ranked ninth globally; it was Europe's most valuable company and the second-largest drugmaker behind Eli Lilly. The $1,600 per patient per month figure is above actual U.S. list prices: the highest-priced GLP-1, Wegovy, lists at roughly $1,349 per month, and U.S. prices, while far higher than in peer nations, do not reach $1,600. CMS and CBO estimates put the Medicare cost of expanding anti-obesity coverage at roughly $24.8 to $35.5 billion over about a decade rather than any single per-patient figure like the one cited. The 90 percent of profits from the U.S. claim overstates U.S. dependence: North America Operations is roughly half of Novo Nordisk's sales, not 90 percent.

What the evidence shows: Dr. Fatima Cody Stanford is an obesity medicine physician at Massachusetts General Hospital and associate professor at Harvard Medical School, and her own published disclosures list Novo Nordisk among the companies she has served as a paid advisor or consultant. A 2024 correction in Nature Reviews Gastroenterology and Hepatology states she has served as a paid advisor/consultant to Novo Nordisk (alongside Eli Lilly, Pfizer, Boehringer Ingelheim and others). In 2023 Senator Chuck Grassley formally raised her financial ties, noting she had been paid tens of thousands of dollars by makers of weight loss and obesity drugs, including the producers of semaglutide and Ozempic (Novo Nordisk), in connection with her appointment to the federal Dietary Guidelines Advisory Committee. The claim that she was significantly funded by Novo Nordisk is consistent with her disclosed consulting relationship and the payments documented by public scrutiny.

What the evidence shows: The two population-and-share figures are close to correct. WHO reported over 7 million confirmed COVID-19 deaths globally as of early 2025, and the United States has recorded roughly 1.2 million reported deaths, so the US share of reported global COVID deaths is about 16 to 17 percent, while the US holds only about 4 percent of the world population (roughly 335 million of about 8 billion). The US-versus-Japan per capita comparison is directionally right but understated: cumulative reported COVID deaths run near 3,600 per million in the US against roughly 600 per million in Japan, a gap closer to six times rather than three. So the framing is accurate on the 16 percent and 4 percent figures, and Japan did fare far better per capita, though the true per capita ratio is larger than the three times stated (note that reported deaths undercount true mortality, more so outside high-income countries, which inflates the US share of reported deaths relative to its share of estimated excess deaths).

What the evidence shows: It is true that the FDA collects industry user fees, so the underlying point that this is a legal, publicly documented arrangement (not a hidden conspiracy) is fair. But the 75 percent figure does not describe the FDA as a whole. According to an HHS Office of the Assistant Secretary for Planning and Evaluation issue brief, user fees made up about 46 percent ($2.9 billion) of the FDA's total budget of $6.2 billion in FY2022, and 66 percent of the narrower human drugs program budget (an NIH National Academies volume reproduces the same figures). A figure near 75 percent applies only to the specific prescription-drug review program (PDUFA), not to the agency overall, so applying it to the entire FDA overstates the share by roughly 25 to 30 points.

What the evidence shows: It is accurate that Kraft Heinz reformulated two Lunchables products (Turkey and Cheddar Cracker Stackers and Extra Cheesy Pizza) with more protein and whole grains to meet federal standards and made them available through the USDA-administered National School Lunch Program starting in the 2023 to 2024 school year. The seven billion figure is overstated and misattributed: it describes the entire National School Lunch Program, not Lunchables, and even then the program provided about 4.8 billion lunches in fiscal year 2024 (roughly 4.7 billion in fiscal year 2023), not seven billion. The framing also implies Lunchables now supply that volume, when in fact they were a small optional item; by November 2024 Kraft Heinz pulled both products from the program, citing demand that did not meet its targets and following Consumer Reports testing that flagged sodium and heavy-metal levels. So the deal was real but the meal count is inflated and wrongly tied to Lunchables specifically.

What the evidence shows: Head-to-head evidence does not show exercise demonstrably beating antidepressants plus therapy. A systematic review and network meta-analysis of randomized trials in non-severe depression found no significant difference between exercise and antidepressants (standardized mean difference −0.12, 95% CI −0.33 to 0.10) and no treatment superior to another, though all beat controls, and exercise had higher dropout (relative risk 1.31). A large 2024 network meta-analysis (218 studies) reported that some exercise forms (walking or jogging, Hedges g −0.63) had point estimates numerically comparable to or larger than SSRIs alone (g −0.26) and CBT (g −0.55), but rated confidence in nearly all results as low to very low due to lack of blinding and high risk of bias, and positioned exercise alongside, not clearly above, medication and psychotherapy. So exercise appears roughly as effective as antidepressants for milder depression, but the claim that lifestyle without drugs or therapy produces demonstrably better outcomes than combined drug-plus-therapy care overstates what the trials show.

What the evidence shows: The underlying premise is correct: PCOS is the most common cause of anovulatory infertility (roughly 90 to 95 percent of anovulatory women seeking fertility treatment have PCOS), and insulin resistance is central to it. But the specifics are overstated. The 2023 International (ESHRE-partnered) PCOS guideline concludes there is no evidence to support any one type of diet over another and names no ketogenic diet; the first-line steps are general lifestyle change and, pharmacologically, letrozole for ovulation induction, not a 12-week keto protocol. Recent systematic reviews find ketogenic diets produce short-term reductions in weight, insulin resistance, and some reproductive hormones, but rest on few, small, short RCTs and stop short of calling keto the most effective intervention, with fertility data too sparse to pool. No evidence was found that most European countries subsidize a keto diet for PCOS infertility. In sum, the claim's framing about PCOS is accurate, but keto being the single most effective, guideline-backed, subsidized intervention is unsupported.

What the evidence shows: The 221 million figure traces to Salami et al. (JAMA Cardiology, 2017), which reported that US statin use rose to 39.2 million individuals filling 221 million prescriptions in the 2012-2013 period, up from 134 million prescriptions in 2002-2003. That number is accurate for its source but is roughly a decade old rather than a strictly current annual count, and more recent IQVIA-cited estimates put statins at over 200 million prescriptions in 2020. The second part is solidly correct: CDC and NIH mortality data confirm heart disease has been the leading cause of death in the United States for over a century and remained number one in 2023, with 680,909 to 683,491 deaths. Overall the quote is roughly accurate, though it presents a 2012-2013 prescription total as if it were current.