I Took Five Health Tests That Didn’t Agree on Anything.
You all know I love me some health data. So when I had the chance to stack up a few epigenetic “biological age” tests, a DNA test, a standard blood panel, and an ultrasound, I said yes to all of it. My assumption was probably much like yours if you’ve been keeping an eye on all the diagnostics out there - run enough tests and you get a clearer picture of your health.
Nope. I got a hot mess.
How old am I, really?!?
I started my diagnostic journey with the metric everyone is talking about right now: biological age. For the record, I’m 45 years old chronologically (the same vintage as Roger Federer, Beyonce, Britney Spears, Zlatan Ibrahimović, and Serena Williams, for those keeping count). I took three biological age home tests at exactly the same time a few months ago. Here’s what the results showed:
TruDiagnostic’s OMICm clock: 33.7, eleven years younger
TruDiagnostic’s Symphony Age (same report, same blood sample, same day): 45.1, bang on my age
Generation Lab’s SystemAge: 49.8 (same day blood draw, different sample): almost five years older, in their “accelerated aging” bracket
Two clocks, built from the same sample, disagreed by over eleven years. One says I’m aging backwards, the other says I’m right on schedule. A separate company, different biomarkers entirely, says I’m aging noticeably faster than average. That’s a 16-year spread on the exact same question: how old is my body, really?
Here’s the main problem - these clocks are trained on different cohorts and built for different jobs (some predict mortality, some organ-specific decline, some mostly generate a shareable report). Based on these reports, it is quite clear to me that “biological age” isn’t ready for commercial prime time yet…
DNA vs Epigenetics
The third test that I did using a saliva sample was a nutrition related DNA test from Nutrigenomix. This DNA test read my actual genetic code (which is fixed for life and pretty accurate). It’s like a blueprint. Among other things, it told me that I carry variants, called a SNPs: one that make me sodium-sensitive, one that raises diabetes risk if I skimp on whole grains, and one linked to a weaker antioxidant enzyme (SOD2).
By contrast, the biological epigenetic tests from above read something else entirely. Rather than read the blueprint, they read the sticky notes and highlighter marks of decades of living called DNA methylation. Methylation is a chemical tag (-CH3) that turns the actual genes up or down without changing the DNA sequence itself. It’s shaped by exercise, sleep, stress, diet, alcohol, pollution, emotional trauma, and pretty much anything else that shapes your day-to-day. And, unlike my genetic blueprint, the methylation patterns will change over time depending on exposure.
So did my genes and methylation patterns line up?
Nope. Even the “problem areas” didn’t line up
Ok. Some things were in agreement (but I’m not convinced its due to anything more than statistical chance). Where everything aligned: my DNA flagged elevated diabetes risk if whole grain intake is low. My epigenetic metabolic markers independently came back in the single-digit percentile, flagged “critical” (yikes!). My DNA also flagged elevated bone-nutrient needs, and skeletal was epigenetic risk #2 (good thing I strength train). My genes flag sodium sensitivity, and cardiac was epigenetic risk #1.
Except then I layered in my actual routine bloodwork and checked the measurement that diagnoses blood sugar and metabolic problems in real time: fasting glucose and HbA1c. Glucose: 88 mg/dL, very solid. HbA1c: 5.1%, also great. So the gene flagged diabetes risk, the epigenetic report agreed and calls my metabolism “critical,” but the clinical marker says I’m fine. Two data sources told the same scary story but the test with real clinical validation behind it said I was doing everything right. W.T.F.
The overlap fell apart in a few other places too. My DNA flagged elevated inflammatory (IL-6) activity and a weaker antioxidant enzyme (SOD2), exactly what you’d expect epigenetics to confirm. Instead my epigenetic inflammation and oxidative defense markers came back solidly normal.
The two epigenetic reports didn’t even agree with each other on where to look. Generation Lab ranked cardiac, skeletal, immune, auditory (indoor cycling, anyone?), and nervous as my top five risks. TruDiagnostic’s TruHealth, from the same day, flagged metabolic markers as the standout “critical” concern and barely mentioned cardiac. Company A says I’m a walking disaster - heart, bones, immunity, hearing, nerves - ready to enter assisted living at 45. Company B says metabolism is the five-alarm fire. What is an average person supposed to do with that?!?! (Just for fun, I ran all of this through Claude too and it was also pretty confused, which is saying something).
None of these tests caught the thing that actually mattered
Here’s the twist that I’ve reported on before. Buried in an unglamorous standard metabolic panel from my regular doctor was an important medical finding: my bilirubin has been slowly climbing across three separate draws over the years (1.4, then 1.6, then 1.9 mg/dL, against a normal ceiling of 1.2). All other liver markers were fine, just my bilirubin changed over time.
Not one DNA or epigenetic test flagged this. Not the age clocks, not the “hepatic system” score which supposedly was testing my liver, not the genetics report. It took a boring CMP blood draw to catch it, and a follow-up liver ultrasound (an ultrasound tech and a radiologist, no algorithm) to confirm my liver looks great. Turns out I have Gilbert’s Syndrome - which is an actual genetic disease!!! And yet none of the gene based tests caught it.
The tests with the shiniest branding and the dramatic graphics missed the one number actually trending in a direction worth watching that was being driven by my actual genes. The unsexy, decades-old blood test, and a primary care doctor with a sharp eye, caught it immediately.
So what do I actually do with all this information?
You all know I do this for a living and even I have absolutely no idea. Some of it is genuinely interesting, and the directional signals (pay attention to metabolic health, cardiovascular fitness, bone density) are probably worth watching, even if the projected “age” numbers aren’t. But, honestly, I would have been paying attention to them anyways. I’m not sure the jump scare from the tests was worth the stress.
Here’s what I will say, standard, simple, clinical labs are still your anchor - boring, validated across millions of patients, and repeatable. A genetic or epigenetic score isn’t a guaranteed clinical outcome. Mine agreed often enough to be interesting but disagreed often enough that I won’t lose sleep over them (although, I’ll admit, that 55 year biological age score made me want to cry). We all need to keep in mind that the wellness tests are hypotheses, not diagnoses. If several point at the same system, that’s worth tracking over time. But if they contradict each other, which they did everywhere in my case, you probably want to take them with a generous grain of (Himalayan sea) salt.
If you’re drawn to this kind of testing, and I get the appeal, my takeaway is to enjoy the data, but hold it loosely, and don’t let a very expensive, glossy PDF talk you out of trusting the boring blood draw your doctor already ordered.