You get your bloodwork back, and the number is printed next to a bracket — a low end, a high end, and a note that says "normal" if you land inside it. It's easy to read that as a verdict: inside the brackets, nothing to worry about. That's a reasonable-sounding assumption, and it's often wrong in a specific, well-documented way.
What a reference range actually is
A reference range isn't a definition of health. It's a statistical description of a reference population. The standard method — endorsed by the International Federation of Clinical Chemistry's committee on reference intervals — sets the lower and upper limits at the 2.5th and 97.5th percentiles of results from a group of people considered "healthy" for that particular test, which by definition puts the bottom and top 2.5% of even a healthy population outside their own lab's "normal" range (Ozarda et al., 2018). The Merck Manual's own description of the process is blunter still: reference ranges reflect the middle 95% of a reference population, not an outcome-based definition of what's good for any one person (Merck Manual, Laboratory Reference Ranges).
Two things follow from that directly. First, "in range" was never a personalized claim — it's a statement about where you fall relative to a group, and how that group was selected (age, sex, activity level, geography, sometimes fasting status) shapes where the brackets sit as much as biology does. Second, being at the 6th percentile and the 94th percentile are both "in range" and can describe very different underlying pictures, especially for markers where risk rises gradually across the range rather than only past a hard cutoff.
A concrete example: ApoB and LDL cholesterol
The clearest illustration of this in practice is the standard cholesterol panel. Most panels report LDL cholesterol (LDL-C) — the mass of cholesterol carried by LDL particles. What actually predicts cardiovascular risk more precisely is apolipoprotein B (ApoB) — a direct count of the atherogenic particles themselves, since each particle carries exactly one ApoB molecule regardless of how much cholesterol it's carrying (Sniderman et al., 2024).
Most of the time, LDL-C and ApoB tell a consistent story. But they don't always agree — a pattern called discordance — and when they disagree, the discordance itself carries independent risk information. In the ATTICA cohort, tracked over 20 years, participants whose ApoB ran disproportionately high relative to their LDL-C carried meaningfully elevated cardiovascular risk even when their LDL-C alone looked unremarkable — and the reverse discordance pattern (low ApoB relative to LDL-C) was associated with reduced risk, independent of the absolute LDL-C or ApoB level on its own (Giannakopoulou et al., 2025). This kind of discordance is common enough that a substantial share of people with a "normal," in-range LDL-C carry a disproportionately elevated ApoB — driven especially by elevated triglycerides, small dense LDL particles, or metabolic syndrome features (Sniderman et al., 2024).
None of this means LDL-C is a useless number, or that a "normal" LDL-C is meaningless. It means a single in-range result on one marker doesn't rule out an elevated-risk pattern that a second, related marker can reveal — and ApoB isn't part of the standard panel in most routine bloodwork, so most people never see the number that would tell them whether they're in the discordant group.
| What's true | What it doesn't mean |
|---|---|
| A reference range describes the middle 95% of a reference population | It does not describe what's optimal for any one person in that population |
| LDL-C and ApoB usually agree | Agreement isn't guaranteed — discordance happens often enough to matter |
| An in-range LDL-C is a genuinely useful data point | It is not, by itself, proof that particle-level risk is low |
Why this is easy to miss
Standard panels are built for broad screening, not personalized risk stratification — that's a reasonable design choice for population-level medicine, but it means the panel's "normal" flag is answering a narrower question than most patients assume it's answering. A flag that says "in range" is really saying "consistent with the reference population we compared you to," not "optimal for you specifically" or "rules out an elevated-risk pattern that a related, non-standard marker would catch."
What to actually do with this
A few practical takeaways follow directly from the research above:
- Treat "in range" as a real but limited data point — it tells you where you fall against a reference population, not whether that's optimal for you.
- For cardiovascular risk specifically, ask whether ApoB (or non-HDL cholesterol, a reasonable proxy where ApoB isn't available) has ever been checked alongside LDL-C — discordance between the two is common enough to be worth ruling out, not a rare edge case.
- Where a marker sits within its range matters, not just whether it clears the bracket — the 6th percentile and the 94th percentile are both "normal," and they're not the same thing.
- Bring the actual number to the conversation with your doctor, not just the flag color — "in range" and "optimal for me" are different questions, and only one of them is answered automatically by a standard panel.
When to actually check in with someone
Nothing here is a substitute for a conversation with a doctor or qualified healthcare provider, and no single marker — ApoB included — should be read as a diagnosis on its own. If you're looking at a panel that's technically "in range" but something about your risk profile, family history, or how you feel doesn't add up, that discrepancy is exactly the kind of thing worth raising directly, by name, at your next appointment.
