WHAT GOT CUT AND WHY

Milk Thistle (Silymarin)

Milk thistle was the longest-tenured candidate I tested before cutting. The published evidence base for silymarin in non-alcoholic fatty liver disease is reasonable ², so the inclusion logic was: if it helps liver function in NAFLD, it might help liver function in acute alcohol exposure. After about six weeks of paired testing at 200mg silymarin, it made no difference I could see against my no-supplement nights. That matched the Cochrane finding that milk thistle has no demonstrated effect on alcohol-related liver outcomes specifically ², so I cut it. (More on this comparison: DHM vs. Milk Thistle.)

N-Acetylcysteine

NAC was on the list because of its well-established role as a glutathione precursor and its clinical history in acetaminophen overdose. The 2024 binge-drinker clinical trial that found NAC ineffective at reducing total hangover scores in humans ³ aligned with what I saw -- at the supplement-typical 600-800mg dose, no detectable effect on my morning-after data. Higher doses (in the gram range) might have done something but were not practical for an oral formula format. I went with intact glutathione delivery via SAG instead. (More on this: DHM vs. NAC.)

Two more NAC studies are worth knowing. In a randomized crossover trial, NAC made no overall difference to hangover scores, though the women in it did better on some symptoms than the men⁵. In rats given alcohol every day for a month, NAC given alongside it lowered liver-enzyme and oxidative-stress markers⁶, which is the animal-to-human gap again. The bigger reason NAC did not make it is that it is a buffer strategy: it gives the cell more raw material for glutathione and assumes the damage is already under way. DHM is not a glutathione precursor at all; the case for it sits upstream, in how fast alcohol and acetaldehyde are cleared, where the animal results are mixed. DHM vs. NAC has the full comparison.

Prickly Pear (Opuntia ficus-indica)

OFI is the ingredient with the famous JAMA hangover study ⁴. I tested it at the published 1,600 IU dose in standalone runs and saw a small but inconsistent effect on inflammation-related symptoms (nausea, dry mouth) but no effect on the cognitive or fatigue side. The mechanism is anti-inflammatory rather than metabolic, which sits downstream of where DHM intervenes. With DHM already in the formula for the upstream side, the OFI effect was redundant in my data. Cut. (More on this comparison: DHM vs. Prickly Pear.)

B-Complex Vitamins

The B-vitamin pitch is appealing -- alcohol depletes thiamine and B6, deficiency is plausible in heavy drinkers, supplementation should help. I ran B-complex pairings at typical multivitamin doses for several weeks. Removing each of these ingredients did not change the morning-after score in any direction I could detect, in my own data at least. (the B-vitamin cut was the one that surprised me most, because the deficiency-replacement story sounds so persuasive in supplement marketing.) The likely explanation is that I am not B-vitamin deficient at baseline, so supplementation has no headroom to operate.

The Other Cuts

Alpha-lipoic acid, ginger, turmeric (curcumin), magnesium beyond electrolyte amounts, and a basic electrolyte mix all got cut for the same reason: removing them made no difference I could see against the no-supplement nights, at the doses I was willing to use. The 2005 systematic review of all hangover-prevention RCTs reached a similar across-the-category conclusion: most candidates fail to show clinical effect at human doses ¹. My personal results converged with that literature finding.

Why the Cuts Mattered

Every cut freed budget for the ingredients that survived. The cumulative weight of the twelve cut ingredients was the budget that funded the 1,500mg DHM dose in the final formula -- which would have been impossible to include at that dose if the formula had retained the maximalist starting list at any reasonable retail price.

For the elimination logic that drove these cuts, see The Elimination Method.