Dr. Rhonda Patrick on the Joe Rogan Experience #773, fact-checked

What the evidence shows: This matches Matsumoto et al. (Journal of Nutrition, 2016), a randomized, double-blind, placebo-controlled trial of 55 metabolic-syndrome patients (71% men, not all men; diagnosed with metabolic syndrome rather than a formal atherosclerosis diagnosis, though the two are closely linked) who took 2,400 mg/day of aged garlic extract or placebo for about a year, with coronary plaque tracked by CT angiography. The trial found no statistically significant difference between groups in total coronary plaque volume change (0.3% vs. 1.6%, P = 0.13). The only statistically significant benefit was a reduction in low-attenuation plaque, a specific vulnerable-plaque subtype (-1.5% vs. +0.2%, P = 0.0049). The widely circulated 80% figure for overall plaque reduction is not a statistically supported result from the trial itself, which explicitly found no significant effect on total plaque burden. The study, dose, and population are real, but the 80% total-plaque claim overstates what the peer-reviewed results actually showed.

What the evidence shows: Patrick said eight clinical trials led the FDA to add a warning label to ibuprofen specifically because chronic use doubled the risk of both stroke and heart attack. In July 2015 the FDA did strengthen its warning that non-aspirin NSAIDs, a class that includes ibuprofen, naproxen, and diclofenac, can cause heart attack or stroke, but the change applied to the whole drug class rather than to ibuprofen alone, and the FDA and reporting on the decision stated the risk can begin within the first weeks of use, not only with chronic use. That decision drew on evidence including a large meta-analysis of individual participant data from randomized trials, published in The Lancet in 2013 by the Coxib and traditional NSAID Trialists' Collaboration, which pooled data across hundreds of NSAID trials (280 placebo-controlled and 474 head-to-head), not eight. That meta-analysis found high-dose ibuprofen was associated with roughly a doubling of major coronary events specifically (rate ratio 2.22, 95% CI 1.10-4.48, drawn from 22 ibuprofen-comparison trials), but explicitly found no statistically significant increase in stroke risk for any NSAID, and the broader major-vascular-events rate ratio for ibuprofen (1.44, CI 0.89-2.33) was also not statistically significant. So the claim is directionally correct that regulators strengthened NSAID cardiovascular warnings based on accumulated trial evidence, but the eight-clinical-trials count is unsupported (the underlying evidence base ran to the hundreds of trials), the warning covers the NSAID class rather than ibuprofen specifically, and the twofold figure held for major coronary events at high ibuprofen doses but not for stroke or for vascular events overall.

What the evidence shows: A 2009 randomized trial (Kuptniratsaikul et al., 107 knee osteoarthritis patients) did compare 2 grams/day of a curcumin extract against 800 mg/day of ibuprofen over six weeks and found the two similarly effective and safe for pain and function, with ibuprofen showing a small edge only for stair-climbing pain. However, that trial used a standard Curcuma domestica extract, not Meriva, the proprietary curcumin-phosphatidylcholine (phytosome) complex Patrick names. A published trial of Meriva specifically (Belcaro et al.) used a much lower curcumin-equivalent dose, about 200 mg/day, not 2 grams/day, and had no ibuprofen comparator arm at all. A separate Meriva pain-relief study compared it to nimesulide and acetaminophen, again not ibuprofen. No identified published trial tested 2 g/day of the Meriva phosphatidylcholine complex head-to-head against 800 mg/day ibuprofen. The specific dose comparison Patrick describes appears to conflate findings from two different curcumin formulations and studies; the underlying general point that some curcumin preparations can match NSAID pain relief in knee osteoarthritis is supported by separate literature, but not as stated for Meriva at that dose.

What the evidence shows: Public Health England ran a 2016 Change4Life Sugar Smart press campaign stating that UK children aged 4-10 consume roughly 5,500 sugar cubes (about 22kg) of sugar a year, a figure the campaign compared to the average body weight of a five-year-old; this was an annual/cumulative statistic, not a literal daily-intake-equals-body-weight claim. Separately, a peer-reviewed analysis of the UK government's National Diet and Nutrition Survey data puts average daily free-sugars intake for children aged 4-10 at 54.5 grams for boys and 49.9 grams for girls, closely matching the 50-grams-a-day figure Patrick cites. WHO recommends free sugar intake stay below 10% of energy, with a conditional target of under 5% (roughly 25g/day on a 2,000-kcal adult diet), so UK children's measured intake runs well above that ceiling. Patrick's on-air recollection conflates the daily-intake statistic with the campaign's separate annual/body-weight comparison, but the core 'about 50 grams a day' figure for the average UK child in this age range is well-supported by government-linked national dietary survey data. Status: mostly accurate, with the daily-versus-annual framing garbled in the retelling.

What the evidence shows: Patrick states that alcohol-induced damage increases the risk of traumatic brain injury by roughly tenfold. The tenfold figure that actually exists in the peer-reviewed literature describes a different pairing of factors: a 1995 study in Neurology (Mayeux et al., n=236) found that carrying the ApoE4 allele together with a prior history of head injury was associated with a roughly tenfold increase in the risk of developing Alzheimer's disease, compared with a twofold increase from ApoE4 alone, and no increased risk from head injury without ApoE4. A 2021 living systematic review and meta-analysis in the Journal of Neurotrauma (n=2,593) found that ApoE4 carriers have only modestly worse functional outcomes after sustaining a TBI, not tenfold worse, with an odds ratio of 1.39 for a favorable outcome in non-carriers versus carriers. Separately, ApoE4 carriers appear more vulnerable to alcohol-related neurotoxicity through impaired neuronal repair and oxidative stress, but no source located establishes that alcohol-induced brain damage itself raises the risk of sustaining a traumatic brain injury by a factor of ten. The claim as stated appears to conflate the well-documented ApoE4-plus-head-injury-history tenfold increase in Alzheimer's risk with a distinct and unsupported claim about alcohol directly raising TBI risk.

What the evidence shows: Patrick claimed that ALA, the plant-derived omega-3 fatty acid, must be consumed at roughly 33.5 times the dose of preformed DHA/EPA to raise brain DHA/EPA to the same level. Peer-reviewed research confirms that human conversion of dietary ALA to long-chain omega-3s is very inefficient: a widely cited review (Gerster 1998) reports conversion to long-chain metabolites at approximately 6% for EPA and 3.8% for DHA on a typical high-saturated-fat background diet, and notes that a diet high in n-6 PUFA reduces conversion by a further 40-50%. A 3.8% ALA-to-DHA conversion rate implies roughly 26-fold more ALA is needed to match a given amount of preformed DHA, in the same order of magnitude as the 33.5x figure cited. A more recent review (2022) confirms that dietary ALA intake reliably raises circulating/tissue ALA and, to a lesser extent, EPA, but generally has little to no effect on DHA levels, supporting the broader point that preformed DHA/EPA intake is needed to reliably raise brain omega-3 status. However, neither source, nor any other identified in this review, independently corroborates the specific 33.5-fold figure; conversion-efficiency estimates vary across studies by diet composition, sex, and genetics, so published ratios cluster in a range rather than at one precise number. Status: the direction and rough order of magnitude of the claim (ALA requires a much larger, roughly 25-to-50-fold, dose than preformed DHA/EPA for equivalent brain levels) is well-supported, but the specific '33.5 times' figure could not be independently verified against a single consistent published source.