Biocracked
Approach: conceptual metaphor. Palette: D, bold color field.

A single glass beaker-shaped water pitcher sits on a saturated mustard-yellow seamless backdrop, photographed straight-on at eye level. It is almost empty, with only a thin line of clear liquid at the bottom, standing for drained B12. Beside it, a clean white ceramic head-shaped bust, with no face detail, glows with warm light, as if it were full. A small electric green ribbon is draped over the pitcher handle as the only green accent. Lighting is soft, diffused studio light from the upper left, with gentle, crisp shadows falling to the right on a glossy surface that gives a faint reflection. The mood is curious and paradoxical, a puzzle in which something is depleted yet the outcome looks better. The composition is minimal, with the objects on the right third and generous empty yellow space on the left. Photorealistic, with no text.
Longevity

Metformin Drains B12, Yet Diabetic Brains Scored Higher. Why?

Dr. Synapse · · 4 min read

Metformin has a well-known downside: it lowers vitamin B12. B12 is a nutrient your brain needs, which makes this awkward for a drug that longevity researchers keep eyeing as a possible neuroprotector. A new NHANES analysis in Diabetes, Obesity and Metabolism asked whether the B12 dip cancels out any brain benefit. In this dataset, it didn't.

The findings are intriguing, but they are observational. They show an association, not proof that metformin protects the brain.

The paradox: lower B12, better scores

The researchers, from Texas Tech University, analysed NHANES 2011-2014 data on 587 US adults aged 60 and older with type 2 diabetes. Of these, 408 took metformin and 179 took other non-insulin diabetes drugs. The comparison group matters. Comparing metformin users with other diabetes patients is more informative than comparing them with healthy people, because everyone in the sample has the same disease.

The headline numbers:

  • Serum B12: 569.3 pg/mL in metformin users vs. 825.3 pg/mL in the comparison group (p = 0.026)
  • Digit Symbol Substitution Test (DSST): 47.5 vs. 39.5 (p < 0.001) in unadjusted comparisons. The DSST is a common test of processing speed, attention and executive function.

So the metformin group had less B12 but higher cognitive scores.

After adjustment, the signal shrank but held up

The raw gap overstates things, because metformin users may differ from other patients in age, education, kidney function and disease severity. After full adjustment, the metformin group scored 3.75 points higher on the DSST. The 95% confidence interval ran from -0.06 to 7.55 (p = 0.053), which is borderline. With multiple imputation for missing data, the estimate was β = 3.49 (p = 0.006).

That is suggestive, not definitive. The primary adjusted result sits right at the edge of statistical significance, and the authors describe their approach as exploratory.

The B12 plot twist

The obvious worry was that B12 depletion would offset metformin's benefits. The team tested whether B12 and methylmalonic acid (MMA) mediated the link between metformin and cognition. MMA is a metabolic marker that tends to rise when functional B12 status is poor. They ran four statistical frameworks: bootstrap mediation, parallel mediation, structural equation modeling and weighted quantile sum regression.

About 95.1% of the association was direct, meaning it was independent of B12 and MMA. In other words, B12 and MMA did not statistically explain the cognitive difference.

This does not show that B12 doesn't matter for the brain. It shows that in this sample, B12 and MMA levels weren't what linked metformin use to DSST scores. The data are also cross-sectional, so they can't show how B12 and cognition change over time in people on long-term metformin.

Who seemed to benefit most

The association wasn't uniform:

  • Ages 60-69: The benefit was concentrated in this "younger-old" group (interaction p = 0.003).
  • Lower DSST quantiles: People with weaker baseline cognitive performance showed more of the difference, which suggests cognitively vulnerable subgroups may be where the association is strongest.

One caution applies to both patterns. Subgroup findings are more likely to be statistical flukes, and they need replication.

Why it can't prove metformin boosts the brain

Be skeptical of any "metformin makes you smarter" headline built on this study.

  • It's observational. NHANES is a snapshot. It can't establish that metformin caused better scores.
  • Confounding is likely. Doctors may prescribe metformin to patients who are healthier overall, with better kidney function, milder disease and fewer complications. Those factors also affect cognition.
  • Causal mediation methods have limits. The authors call their analysis exploratory, and these methods depend on assumptions that observational data can't fully verify.
  • There's one cognitive test. The DSST is useful but narrow. It doesn't measure dementia risk.
  • Everyone in the sample has diabetes. The results say nothing about metformin use in healthy people for longevity.

Practical takeaways

Dose and timing: The paper doesn't analyse dose, duration or timing, so there is no evidence-based number to extract. Metformin is prescription-only, and any dosing decision belongs with a physician.

Who this is relevant for: Older adults with type 2 diabetes who are already prescribed metformin, especially those in their 60s. The finding is mildly reassuring. It doesn't suggest the B12 drop is wiping out cognitive benefit, though it doesn't make the drop harmless.

Who it is not for: Healthy biohackers. Nothing here supports taking metformin as an off-label brain or longevity booster. The study didn't test non-diabetic people.

Monitor your B12 anyway: Metformin lowering B12 is established, and B12 deficiency can cause anemia and neuropathy. If you take metformin, ask your clinician about periodic B12 testing. MMA is sometimes used when B12 results are borderline. Supplementation, if needed, should follow lab results and medical advice.

The bottom line

In a nationally representative sample of older adults with type 2 diabetes, metformin users had lower B12 but better processing-speed scores. B12 and MMA didn't statistically explain the gap. It's an interesting hypothesis-generating result for researchers studying metformin and brain aging, but it needs prospective studies and trials before anyone treats it as a cognitive intervention.

Original study

Read the original research on pubmed.ncbi.nlm.nih.gov →

This article summarizes published research for general information. It is not medical advice.

DS

Written by

Dr. Synapse

Dr. Synapse writes for Biocracked: biohacking science, longevity & human performance.

Related Articles

Approach: conceptual metaphor, in palette D (bold color field).

A saturated mustard-yellow backdrop holds seven identical wooden-handled stopwatches in a staggered row, shot from a slightly raised three-quarter angle. Five stopwatches have run on and show tarnished, cloudy glass faces. The two in front have clean, glossy glass, and each carries a small electric green (#00e676) ribbon tied to its crown. The faces carry no numerals or readable marks. Soft, even studio light from a large overhead softbox gives gentle, short shadows. Shallow depth of field blurs the back of the row. The mood is playful but analytical, with crisp contrast between the dull watches and the bright pair. The image suggests that extra time and good time are not the same thing.
longevity

All 7 Longevity Compounds Extended Lifespan. Only 2 Kept Worms Functional

BY DR. SYNAPSE ·

Seven well-known longevity compounds all extended the lifespan of male roundworms in a new GeroScience study, but only two kept the animals functional late in life. Sulforaphane and metformin were the only treatments that improved late-life mating success, a proxy for reproductive healthspan. The other five added days without preserving that function, showing that lifespan and healthspan can be decoupled. This is strictly worm-only research from C. elegans, not evidence for humans. Still, it argues for screening longevity compounds on function, not survival alone.

Approach: Macro science, with a Bold color field palette.

Extreme close-up microscopy-style image of a single microglial cell with fine, branching processes, rendered with the real texture of fluorescence-stained tissue: translucent, slightly granular cell body, delicate ramified arms reaching across the frame. Shot from straight overhead through a microscope objective with shallow depth of field, so the outer branches soften into blur. The cell is a luminous, warm ivory-white with a few small electric green (#00e676) points of light along its branches. Behind it is a saturated, even magenta backdrop with subtle tissue grain. Lighting is crisp and backlit, with gentle glow around the edges and no harsh shadows. The mood is curious and precise, with a quiet sense of renewal. No text or labels.
longevity

Mouse Study: NR Shifted Aged Brain Cells Toward a Younger State

BY NOVA CHEN ·

In a new GeroScience study, 22-month-old mice given nicotinamide riboside (NR) for eight weeks did better on memory and nest-building tests. Their brain immune cells also looked more like those of young mice. This was a mouse study, and it does not show the same benefits in people. The dose was 400 mg/kg, far above what humans take, and the abstract reports only male mice. Here is what changed in the microglia, what the study can't tell us, and what the findings mean for NR users.

Approach: Macro science. Palette: B, warm.

Extreme macro photograph of raw berberine: a small heap of vivid yellow-orange crystalline powder with needle-like crystal clusters, shot at a low 30-degree angle with a shallow depth of field. Beside it, a dry sliver of Berberis root bark shows its bright yellow inner wood and fibrous grain. The surface is a smooth dark walnut board. Low golden-hour side light from the left makes the crystals glint like amber glass and throws long warm copper shadows across the wood. One tiny electric green highlight reflects off a glass spatula edge in the soft background blur. The mood is quiet and investigative, with rich saturated honey and burnt-orange tones and no blue cast. Photorealistic, microscopy-like texture, no text or labels.
metabolism

Berberine vs. Metformin: The Pilot Trial Behind HbA1c 9.5% to 7.5%

BY HELIX WARD ·

A small 2008 pilot trial put berberine head-to-head with metformin in newly diagnosed type 2 diabetes, and the glucose-lowering effect looked similar. In the berberine group, HbA1c fell from 9.5% to 7.5% in three months. A second trial in 48 people already on treatment saw HbA1c drop from 8.1% to 7.3% and HOMA-IR fall 44.7%. But these were small studies in people with diabetes, and about a third of participants had GI side effects. Here is what the data show, what they don't, and what to know before you think about berberine.

Approach: conceptual metaphor. Palette: bold color field.

A single ripe, glossy red tomato-colored apple sits at the center of a saturated mustard-yellow seamless backdrop, with a neat row of five identical white capsules lined up beside it on the same surface. The apple is the clear winner of the frame. One capsule has rolled slightly away, cracked open and spilling a tiny pile of pale powder. Shot at a low three-quarter angle with a shallow depth of field, so the back capsules blur softly. Lighting is a large, soft key light from the upper left, giving crisp but gentle shadows with warm yellow bounce. A thin sprig of electric green (#00e676) leaves rises from the apple stem as the only green accent. Mood: confident, playful and skeptical, a clean verdict that real food outweighs pills. No text or labels.
longevity

Zombie Cell Supplements? 27 Human Trials Say Diet Beats Pills

BY NOVA CHEN ·

Supplement labels love to promise they'll "clear zombie cells," but a new systematic review of 27 human trials and 3,811 participants suggests the evidence is thin. Calorie restriction produced the most consistent drop in senescence-linked inflammatory signals. Omega-3 was the most promising supplement, though the evidence was limited and mixed. Telomere length and the classic p16/p21 markers mostly failed to move. Here's what that means for what you eat and what you buy.

This page is available in