Biohack Smarter: How to Test What Actually Works

Biohacking can be useful, but only if you treat it like an experiment instead of a shopping list of trends. In this episode, Dr. Hillary Lin explains rati…

Published April 5, 202541:06The Longevity Show

Dr. Lin's case for biohacking does not come from distrust of science — it comes from having worked inside it. She trained and ran research at Stanford and Columbia, and her frustration is with what that system structurally cannot do: research that never gets funded because it doesn't fit institutional priorities, trials designed around disease states rather than optimisation, and standard-of-care guidelines built on population averages that leave individual variability unaddressed. Biohacking, in her framing, exists to fill that gap — and should complement evidence-based medicine rather than replace it.

The episode's spine is a working method: the scientific method applied to a study population of one. Ask a specific question rather than a vague one. Do the reading. Form a testable hypothesis. Change one variable. Track. Conclude, then iterate. Around that sit the two disciplines that make it hold — a hierarchy of evidence for judging what you read, and a set of psychological failure modes for judging what you believe, starting with the illusion of causality that makes any intervention feel effective if you happened to improve while taking it.

The most useful section is where Dr. Lin turns the framework on herself and it does not go well. Weekly 24–36 hour water fasts for a couple of months delivered the fat loss and stable energy the literature predicts, then produced middle-of-the-night waking, agitation and — months later, on a delay the hair growth cycle enforces — noticeable shedding. She had also started four supplements at the same time, which is exactly the mistake the framework warns against, and so could not isolate the cause. She pivoted rather than pushed through, which is the actual lesson.

Before you watch

The illusion of causality is the engine of the whole pseudoscience economy: you take something, you improve, you credit the thing. Dr. Lin's sharpest example is not a supplement but a prescription — patients who ask for antibiotics for sinus infections and colds because antibiotics worked last time, when what actually worked was the passage of time. The absence of immediate side effects makes an ineffective remedy feel both safer and more effective than it is.
The vulnerability factors are researched rather than moralistic. Lower cognitive reflectiveness, susceptibility to magical health beliefs or medical conspiracy theories, low conscientiousness as measured in personality testing, and — importantly — prior negative experiences with conventional healthcare all raise the likelihood of accepting unproven claims. Motivational drivers matter too: uncertainty creates a hunger for control, and the wellness industry's language of empowerment is calibrated to that hunger.
There are five recognisable marketing tells. Cherry-picking, often a single small company-funded study while larger negative human trials go unmentioned. Testimonials that omit the concurrent diet and exercise changes doing the real work. Implied benefit language — 'supports immune health', 'helps boost energy', 'may promote longevity' — which states nothing falsifiable. Jargon without substance, like a serum claiming to 'activate telomerase'. And overgeneralisation, where one product allegedly improves brain health, digestion and inflammation at once.
Dr. Lin extends the critique to her own profession. Clinicians are subject to the same cognitive biases, including deference to senior authority figures, and she names pharmaceutical representative relationships as the clinical version of the influencer affiliate link — noting the most egregious practices, like expensive vacations, have been curtailed by regulation.
The hierarchy of evidence comes with its failure modes attached rather than as a clean ladder. Meta-analyses are only as good as their inputs and can oversimplify by pooling studies whose protocols aren't comparable — she uses the omega-3 cardiovascular meta-analyses, where dosage and baseline health differences produce genuinely mixed conclusions, as the example. RCTs eliminate bias but report averages under controlled conditions; the Diabetes Prevention Program produced groundbreaking findings while exposing how hard real-world adherence is. Observational work like Framingham established the cholesterol–heart disease link but cannot by design prove that lowering it prevents events.
Animal evidence needs to be read with the animal in mind. Lab mice are bred for genetic uniformity and kept on standardised diets in low-stress environments — ideal for isolating a mechanism, poor for predicting how a genetically diverse human population under real-world stress will respond. Yeast and worms give clean insight into fundamental ageing biology at a great distance from human complexity; dogs and non-human primates are closer but constrained by cost and ethics.
Dose scaling from animals is where enthusiasm usually breaks. Dr. Lin's example is resveratrol — the compound in grape skin that drove years of red-wine-and-longevity headlines — where scaling the effective animal dose to a human means a quantity of wine that would do serious harm long before any benefit appeared. Surface area scaling factors exist for exactly this reason, and she recommends running the arithmetic with a physician rather than by analogy.
Rapamycin is her illustration of strong animal data that still isn't a human answer. Lifespan extension has been replicated from yeast to mice to dogs, pointing to a conserved mechanism through mTOR inhibition, with mouse models also showing reduced tumour formation and less age-related weight gain. Human trials are still early, which is precisely the gap between compelling and established.
For anecdotes she offers a triage: who is telling it and what do they know; are they transparent about what else changed at the same time; do many independent people report the same thing; and is there a plausible mechanism. If a supplement supposedly resolved a chronic illness overnight, the mechanism question alone should stop you. Her risk framing includes a point people skip — most advertised interventions cost money, and cost is not harm-free.
Her own fasting experiment is the case study in getting it wrong. Weekly 24–36 hour water-only fasts produced real fat loss and unexpectedly stable energy, then night waking she could not resolve, a jumpy restlessness she attributes to the cortisol rise fasting produces, and hair shedding that only became visible months in because that is how long the hair growth cycle takes to reveal a change. The delay is the instructive part: an intervention's most important effect may not surface inside your observation window.
She had confounded her own experiment. To prevent malnutrition during fasts she had simultaneously started creatine, B12, CoQ10 and PQQ — any of which could plausibly touch sleep, and creatine in particular is known to affect sleep in sensitive people even as the research frames its wakefulness effect as a cognitive benefit. Combined with a pre-existing genetic tendency toward night awakenings, she could not isolate a cause. Her resolution was to pivot rather than persist: gentler time-restricted eating, and other routes to the same goals.

Chapters

00:00

Actionable takeaways

Questions

What is rational biohacking and how is it different from regular biohacking?

Rational biohacking is self-experimentation run with the discipline of the scientific method rather than by trend-following. The distinction Dr. Lin draws is structural: ask one specific, testable question rather than a vague one; research it in credible sources first; change one variable at a time while holding sleep, diet and exercise as constant as you can; track outcomes systematically; then draw conclusions and iterate. She is blunt that the dismissive stereotype of biohacking is often earned — people stack five interventions at once, track nothing, and end up with a cupboard of supplements and no idea which one did anything. She has also treated patients who arrived after a 'miracle' intervention with side effects that took serious work to reverse.

Why does a supplement seem to work at first and then stop?

Usually adaptation rather than failure. Your biology recalibrates to a new input, so the initial effect fades — melatonin is the cleanest example, working well for sleep onset for the first few days and then losing its edge, which is why Dr. Lin restricts it to temporary circadian shifts like jet lag or shift work rather than ongoing use. The same pattern explains why GLP-1 doses are escalated month to month for continued weight loss and why strength training requires progressively heavier loads. The other explanations worth ruling out are the placebo effect, which is real and can produce genuine improvement, and a confounder — something else that changed at the same time.

How do I know if a health study is trustworthy?

Start with where it sits in the hierarchy — systematic reviews and meta-analyses above randomised controlled trials, above observational cohort and case-control studies, above case reports and expert opinion — but read each with its specific weakness in mind. Meta-analyses inherit the flaws of what they pool and can combine studies too different to combine meaningfully. RCTs report averages under controlled conditions that may not survive contact with real life. Observational studies show correlation and cannot establish causation, however large. Then check the red flags that cut across all of them: was it funded by the company selling the product, is the sample size large enough to support the claim, and has anyone else replicated it. A single flashy headline that nobody has reproduced is not evidence.

Can you trust animal studies for longevity interventions?

They are a starting point, and for longevity specifically they are unavoidable — human lifespan studies would take generations, so waiting for them means never applying anything while it is still relevant to you. But lab animals are bred for genetic uniformity and housed in standardised, low-stress conditions, which makes them excellent for isolating mechanisms and poor for predicting responses across a diverse human population. Dr. Lin's guidance is to weight animal evidence when it aligns with preliminary human data, look for replication across different models, and be extremely careful with dose translation — the resveratrol case, where the scaled human dose corresponds to an impossible quantity of wine, is her standing example of how badly that step can go.

What went wrong with Dr. Lin's own fasting experiment?

She ran weekly 24–36 hour water-only fasts for a couple of months. The upside arrived as predicted — measurable fat loss and, to her surprise, stable and even improved energy. The downsides accumulated: waking in the middle of the night and struggling to get back to sleep, a jumpy restlessness she attributes to the cortisol elevation that fasting produces, and eventually noticeable hair shedding. That last one only surfaced months in, because the hair growth cycle delays the signal. She had also started creatine, B12, CoQ10 and PQQ during the same period to avoid nutrient gaps, so she could not isolate which factor drove the sleep disruption — and she has a pre-existing tendency toward night awakenings. The shedding stopped once she discontinued the weekly prolonged fasts, and she moved to gentler time-restricted eating.

How do I talk to my doctor about biohacking without being dismissed?

Dr. Lin's approach is to make it a collaboration rather than an announcement. Be transparent about what you're trying and why, and bring the reading — a summary of peer-reviewed sources demonstrates you've thought about it. Speak in the terms clinicians are trained in: instead of 'I saw this on social media,' try 'I've read this supports mitochondrial function — are there risks I should know about?' Most usefully, ask for monitoring rather than approval: request periodic glucose, insulin and lipid testing around a fasting protocol, for example. Doctors engage far more readily when there's a concrete role in keeping you safe and generating data. If they raise a concern about an interaction with your medications, take it seriously. And if your primary physician isn't interested, build a wider team — longevity or functional medicine physicians, dietitians, physiotherapists — rather than going it alone.

The Longevity Letter