One intervention, every measured benefit and result that did not improve. See who it moved for, where it stayed neutral or went the other way, and the next hypothesis Phoenix can validate.
Phoenix APOE4 Research · Reviewed by Kevin Tran, APOE4/4 carrier and founder · Data through 2026-08-23 · How to read this research
How to read this
Observational research: real APOE4 carriers tracked over time, each compared with their own earlier self. Members chose their interventions, with no randomisation or placebo group. Everything else they changed is shown beside each result.
Read the full methods and limitsPhoenix has measured Magnesium threonate against 2 exact outcomes for APOE4 carriers. The same two magnesium-threonate cases nudged HbA1c lower, but the public engine does not publish their before-and-after values. Two clean curated cases lowered homocysteine by 28.8% and 31.9%; these are two cases, not a response rate.
| Outcome | Direction | What we observed |
|---|---|---|
| HbA1c | Positive individual result | The same two magnesium-threonate cases nudged HbA1c lower, but the public engine does not publish their before-and-after values. |
| Homocysteine | Positive individual result | Two clean curated cases lowered homocysteine by 28.8% and 31.9%; these are two cases, not a response rate. |
Phoenix has measured Magnesium threonate against 2 exact outcomes for APOE4 carriers. The same two magnesium-threonate cases nudged HbA1c lower, but the public engine does not publish their before-and-after values. Two clean curated cases lowered homocysteine by 28.8% and 31.9%; these are two cases, not a response rate.
The same two magnesium-threonate cases nudged HbA1c lower, but the public engine does not publish their before-and-after values. Support: 87 logged members, 2 members with repeats, 2 observations, 4 measurements.
Two clean curated cases lowered homocysteine by 28.8% and 31.9%; these are two cases, not a response rate. Support: 142 logged members, 33 members with repeats, 2 observations, 4 measurements.
Search by health goal or intervention to see what changed for real members, including results that moved the wrong way.
Magnesium threonate · 2 outcomes shown.
HbA1c: Decrease
Cases: 2
The same two magnesium-threonate cases nudged HbA1c lower, but the public engine does not publish their before-and-after values.
Magnesium threonate
✓Positive individual resultThe same two magnesium-threonate cases nudged HbA1c lower, but the public engine does not publish their before-and-after values.
Little published human data on this link yet.
HbA1c moved lower in the same clean window; public before-and-after values were not printed.
What else changed: nearby starts included B-vitamin intake not captured.
Next study: repeat HbA1c after a dated Magnesium threonate window.
This result is under review: Not published alongside the homocysteine finding above: magnesium threonate has only 6 days of confirmed exposure before the after-draw in a much longer window.
Homocysteine: Decrease 3.0 µmol/L
Cases: 1
Two clean curated cases lowered homocysteine by 28.8% and 31.9%; these are two cases, not a response rate.
Magnesium threonate
✓Positive individual resultTwo clean curated cases lowered homocysteine by 28.8% and 31.9%; these are two cases, not a response rate.
Little published human data on this link yet.
Homocysteine
Before10.5After7.5µmol/L-28.8%
139 days between measurements
Full stack credited: Vitamin B12 (lead), Magnesium threonate (co), Folate (undated), B Complex (undated).
What else changed: no nearby starts were recorded.
Next study: No human magnesium-threonate homocysteine study was found; B-vitamin exposure is the main alternative to separate.
No human magnesium-threonate homocysteine study was found; B-vitamin exposure is the main alternative to separate.
This result is under review: Not published as a finding under the minimum-exposure rule: magnesium threonate has only 6 days of confirmed exposure before the after-draw in a 791-day window, and the other plausible co-drivers are undated and untagged. The other member in this pair keeps this result.
The 43-page study: what members tried, how their bloodwork and wearable results changed, and what else changed alongside them.
Turn evidence into your next test
476 members have already tried, measured and logged. Join Phoenix and your next blood draw joins the same Navigator you are reading.
Full research index
Choose an outcome, like LDL cholesterol or sleep, or an intervention, like training: both paths open the same measured records.