You get twelve blood markers back. Some are green, some are amber, one is red. Does that mean you're doing well or poorly? Is your health improving or declining? A single score that combines everything, adjusted for who you are, gives you a clear answer.
That's exactly what the Health Score does. It takes a full blood panel and compresses it into a 0 to 100 number, giving you an instant read on where things stand. But how does it actually work?
🧠The Health Score is not a diagnosis — it is a decision-support tool. It tells you where to focus your attention, not whether you are "healthy" or "unhealthy."
Three Layers of Logic
That score isn't a simple average of your markers, and it isn't built in a single pass. Three things happen before a number comes out the other end.
Layer 1: A marker-specific severity curve. Every marker is compared to its athletic-adjusted band, but "how far outside the band" doesn't mean the same thing for every marker. GGT can sit at several times its ceiling before that's treated as serious, because the marker itself tolerates a wide range. Hematocrit doesn't get that same room — a small rise above its ceiling already counts heavily, because thick blood becomes dangerous well before the number looks dramatic. Markers without their own tuned curve fall back to a generic rule: under roughly 2.5% outside the band reads as an early drift, and past roughly 15% reads as a substantial deviation. Markers with a dedicated curve — hematocrit, eGFR, ApoB and HbA1c among them — are graded on their own scale instead, which is why two markers sitting at the same percentage above their band can end up contributing very differently to the score.
Layer 2: Medical logic adjusts specific markers before they're combined. A handful of markers get adjusted by rules that mirror how a clinician reads them together rather than in isolation. Hematocrit is graded harder when it shows up alongside things that actually make thick blood dangerous — inflammation, elevated platelets, raised blood pressure, or an unfavorable lipid profile — and more leniently when none of those are present. A low HDL barely moves the score on its own, because low HDL in isolation isn't treated as a standalone cardiovascular risk; it counts for more once it's paired with an atherogenic lipid pattern or inflammation. And a low eGFR is discounted when Cystatin C reads clean, because creatinine-based filtration estimates run falsely low in muscular people — if Cystatin C isn't available to confirm that, the engine discounts less, not more.
Layer 3: Findings combine so one severe marker can't hide among normal neighbours. Within a body system, several mild findings still add up, but the system's score can never look better than its single worst marker — a ten-marker panel doesn't get an easy ride just because nine of them are fine. Across body systems, some carry more weight than others: kidneys and cardiovascular markers pull harder on the total than hormones do, because that's where irreversible damage in this population actually shows up. A single badly affected system is also capped from being diluted away by an otherwise strong panel, and a panel that measures too few of the right markers doesn't get a number at all — that's treated as "we can't assess this," not as a clean bill of health.
Context Changes Everything
How Markers Are Classified
Every marker in your panel gets a status. Those status names were changed in 2026 to remove clinical-directive language, so older reports use different words. Here's the current set and how each one affects your score.
The 2026 rename replaced 'Action' with 'Review' and 'Attention' with 'Monitor', to strip out language that told you what to do instead of describing what the marker itself is doing. Older reports may show 'Action' where you now see 'Review', and 'Attention' where you now see 'Monitor'. If you're comparing a 2025 report to a 2026 one, the statuses map directly — the scoring impact is unchanged, only the label is. Context and Descriptive are a different pair entirely: they were not renamed from something else. When you see 'Context', it means the marker is explained by your training, AAS use, or another factor and is removed from the score. 'Review' means it is abnormal and not fully explained, so it stays in and lowers your score.
OPTIMAL
SUBOPTIMAL
CONTEXT
DESCRIPTIVE
MONITOR
REVIEW
What Your Score Actually Means
The scale runs from 0 to 100, and each range corresponds to a real-world health category. Here is what each range looks like in practice:
78–100: Strong
No system is showing load. Where markers sit outside standard lab ranges, the context explains them. Most of the panel reads Optimal or Context, so this is a stable baseline to compare future panels against.
A score in the 78-100 range is a green light, but it is not a reason to skip your next draw. The Health Score is a snapshot, not a guarantee. If you are on a stable protocol and your score has been 85-90 for three consecutive panels, you can extend your retest interval to every 12 weeks instead of every 8. If you change your dose, add a compound, or have a major training shift, go back to 8 weeks. The score is most useful when it is stable. A sudden drop of 10 points, even from 90 to 80, is a signal to look at what changed. Did you add an AI? Did you switch from enanthate to propionate? Did you start a cut? The score will not tell you why, but it will tell you that something needs attention. Use it as a tripwire, not a destination.
58–77: Moderate
One or two systems are carrying load beyond what context accounts for. This is the most common tier for anyone running above replacement doses. The value here is specificity: the findings say which system, and what would confirm it.
Below 58: Under Pressure
Several systems are showing load at the same time, or one marker sits far enough out that context can't account for it. This tier is reserved for corroborated patterns: a lipid drift confirmed by both ApoB and non-HDL, a kidney signal confirmed by eGFR and cystatin C together. A single dramatic number won't land you here; a pattern will. This is a report you should take to a licensed medical professional sooner than your next routine interval.
One common mistake is treating a single red marker as a crisis. A hematocrit of 55% with normal blood pressure and no other cardiovascular findings might be a Context marker if you are on 500 mg of test and live at altitude. The score will reflect that. But if the same hematocrit is accompanied by a blood pressure of 140/90, the score will drop hard because the two markers corroborate each other. The engine is designed to catch patterns, not isolated numbers. If you see a low score, do not panic. Look at which markers are flagged as Review. If it is only one marker, ask if the context is solid. If it is two or more in the same system, that is a real signal. The score is a summary, but the markers are the evidence. Always pull the evidence before you act.
Score Is Relative to Your Baseline
What Different Scores Look Like
To make this concrete, here are three hypothetical panels and roughly what they'd produce:
Score 91 — The Clean Runner
All markers Optimal or Context
Testosterone sits at 850 ng/dL on TRT, hematocrit at 48%, eGFR at 88, HDL at 38 mg/dL, and blood pressure at 118/74. Every one of those numbers lands inside its athletic band. Nothing here needs a conversation beyond confirming the trend holds at the next retest.
Score 64 — The Moderate Blaster
Mixed Monitor and Context markers
With 500 mg test/week on the protocol, hematocrit sits at 53%, eGFR at 72 (muscle artefact, cystatin C clean), HDL at 22 mg/dL, ApoB at 112 mg/dL, and blood pressure at 132/84. The eGFR leaves the score as Context. The lipid picture stays in, because ApoB corroborates the low HDL: that is the one specific thing this panel gives you to discuss.
Score 34 — Under Pressure
Multiple corroborated Review findings
Running 1 g of test with 600 mg of tren per week, this panel shows hematocrit at 59% and blood pressure sitting at 155/95. eGFR has dropped to 51, AST is 142, ALT is 98, and HDL is down to 12 mg/dL. Three systems are moving together here, and the hematocrit is not an isolated number. The blood pressure is tracking with it, and that combination is what actually loads arteries and kidneys. This is a report to put in front of a licensed medical professional promptly.
When your score drops below 58, the model is telling you that multiple systems are under load. This is the tier where you need to make decisions, not just monitor. For example, if hematocrit is 58% and blood pressure is 150/95, the score will be low. The next step is not another blood test; it is an intervention. That might mean a phlebotomy to bring hematocrit down, a dose reduction, or adding an ARB for blood pressure. The score gives you the urgency, but the markers tell you the target. If you are in this tier, do not wait for the next scheduled draw. Retest in 4 weeks after making a change. The score should move if the intervention is working. If it does not, the problem is not the marker you treated; it is something else. Use the score as a feedback loop for your actions.
How the Engine Avoids False Alarms and False Reassurance
Not every abnormal marker carries the same weight, and the engine uses a few distinct mechanisms to keep the final score honest in both directions:
1. Some markers simply count for more. ApoB, hematocrit, eGFR and HbA1c pull harder on the total than a mildly elevated ALT or a slightly low Free T3 does, because they answer more clinically consequential questions. This weighting is fixed per marker, not something the engine invents case by case.
2. The same abnormal result doesn't always mean the same thing. A mildly raised ALT or AST alongside a clean GGT reads as training stress rather than liver injury, and is discounted accordingly. A hematocrit that's only mildly elevated with none of its risk cofactors present is treated far more leniently than the same number alongside inflammation, high platelets or raised blood pressure — those are the combinations that actually make thick blood dangerous.
3. A panel that hasn't measured enough doesn't get a falsely reassuring number. If too few of the markers that actually answer a system's health question are present, that system — and potentially the whole score — is treated as unassessed rather than scored as clean. And on the other end, a small number of specific, corroborated combinations (like a severely reduced eGFR confirmed by a clearly abnormal Cystatin C) cap the entire score low regardless of how the rest of the panel looks, because no amount of other good news changes what that combination means.