There is a specific kind of frustration that comes from being told your results are "all normal" while you still feel like something is off. The number sits inside the parentheses on the page, the box is unticked, the visit ends. What almost no one explains is that "normal" on a lab report is not a statement about your health. It is a statement about statistics — about where most people in the lab's reference group happened to land. The gap between that and the range where your risk is actually lowest is the single most useful idea in reading your own bloodwork, and it is the idea this piece is built around.
Where the "normal" range actually comes from
The reference range printed next to your result is manufactured with a formula, not a health assessment. A lab (or the test's manufacturer) measures a reference group — ideally at least 120 people screened as healthy — and defines "normal" as the central 95% of their results, from the 2.5th to the 97.5th percentile. For data that clusters in a bell shape, that is roughly two standard deviations on either side of the mean. The consequence is arithmetic, not medical: by construction, the highest 2.5% and the lowest 2.5% of results are labeled abnormal, and everyone in between is called normal.1 Nothing in that process asks whether the people in the middle are thriving.
Two problems follow directly from the method. First, the reference population is not always as healthy as it sounds. Some labs derive ranges from routine patient results and rely on statistics to filter out the unwell, and even screened groups can include people on medication or with undetected disease — older TSH reference groups, for example, included people with unrecognized thyroid disease.12 When you benchmark yourself against that group, "in range" means "similar to the people who were sampled," which is a lower bar than "well."
Second, the range is a snapshot with no memory. A ferritin of 35 ng/mL is inside almost every lab's normal band. A ferritin that fell from 95 to 35 over eighteen months is a trend worth investigating, but the report cannot see the fall — only the single point. This is why the same number can be reassuring or alarming depending on context the lab never captures. Our complete normal range chart lays out those reference bands marker by marker; this article is about the space between those bands and where your risk is genuinely lowest. For the wider set, the biomarkers that actually matter walks through each one. If a flagged value has you worried, is my blood test result actually bad? explains how to read the flag.
Normal detects disease; optimal minimizes risk
It helps to name what each range is actually for. A reference range is a screening tool. Its job is to catch the person whose result is so far from the pack that disease is likely — the glucose that screams diabetes, the TSH that confirms overt hypothyroidism. For that job, a wide 95% band works well, because you only want to flag the clear outliers and avoid alarming everyone else.
An optimal range asks a different question: across everyone who is technically "normal," where does the risk of future disease, or the burden of current symptoms, sit at its lowest? That range is usually narrower, and at its best it is derived from outcome studies — laboratory scientists call these clinical decision limits — rather than from the spread of results in a reference group.1 The two ranges answer two genuinely different questions. Normal asks, "Is there disease here right now?" Optimal asks, "Am I in the position least likely to develop disease later?" A result can pass the first test and fail the second at the same time, and that overlap is exactly where the phenomenon of feeling unwell with a clean report lives. We wrote a whole piece on normal labs when you feel terrible precisely because that gap is so common.
Five markers where the gap is widest
The abstract idea becomes concrete the moment you look at specific markers. Five stand out because the distance between the lab's normal band and the commonly cited optimal band is large and well studied, even where experts still disagree about where the line belongs.
TSH. Most US labs report thyroid-stimulating hormone as normal up to roughly 4.0 to 4.5 mIU/L.34 Yet the National Academy of Clinical Biochemistry noted years ago that when you strip out people with hidden thyroid disease, the disease-free reference population sits substantially lower — more than 95% of healthy people are below 2.5 — and some experts have argued the upper limit belongs closer to 2.5.2 Many functional-medicine clinicians target somewhere near 1.0 to 2.5. That means a value of 4.2 can be reported as normal while still sitting in the zone the thyroid gray area covers in detail — associated by some research with a higher chance of progressing toward overt hypothyroidism.5 Other endocrinologists counter that people in this zone have not been shown to benefit from treatment.4 The lab passes it; the debate does not.
Vitamin D. Here the organizations themselves disagree, which is instructive. The Endocrine Society's 2011 guideline defined 25-hydroxyvitamin D deficiency as below 20 ng/mL, insufficiency as 21 to 29, and sufficiency as 30 and above,6 while the Institute of Medicine (now the National Academies) set the bar for bone health lower, at 20.7 The Endocrine Society's own 2024 update then dropped target levels altogether, citing no clear evidence for an optimal level.8 NHANES survey data has repeatedly shown a large share of US adults sitting below 30.9 So a level of 28 might not be flagged by a lab whose lower cutoff is 20, even though it falls under the 2011 Endocrine Society sufficiency line. The practical takeaway is captured in our breakdown of a vitamin D of 30 ng/mL: a "normal" result and an "optimal" result are not the same finding.
Ferritin. Lab lower limits for ferritin can drop to around 12 to 15 ng/mL, and the WHO uses cutoffs in that region to define outright iron deficiency.1011 But the symptoms people associate with low iron — persistent fatigue, restless legs and, less consistently, hair shedding — are frequently reported at ferritin levels well inside the "normal" band, which is why some clinicians look for a more comfortable buffer above 50.121314 A value can be normal by the lab's floor and still leave someone symptomatic, the situation our ferritin 30 explainer walks through.
Fasting glucose. The American Diabetes Association defines normal fasting glucose as under 100 mg/dL and prediabetes as 100 to 125.15 But risk does not switch on cleanly at 100. Research has associated higher-normal fasting glucose — the high 90s — with a greater future likelihood of developing diabetes compared with the low 80s, even though both are reported as normal.16 That is the reasoning behind treating a fasting glucose of 100 as a nudge rather than a non-event.
hs-CRP. High-sensitivity C-reactive protein is perhaps the cleanest example because the categories are explicit. The AHA and CDC have described cardiovascular risk tiers as low below 1.0 mg/L, average from 1.0 to 3.0, and higher above 3.0.17 Many lab reference ranges, however, only flag hs-CRP when it climbs above 3.0 or even higher. So a value of 2.5 can read as "normal" on the report while sitting in the middle-risk tier the elevated hs-CRP discussion unpacks. Same number, two very different messages.
Optimal is not universal — and that matters
It would be a mistake to walk away thinking optimal ranges are hard cutoffs that override your doctor. They are not. Optimal targets shift with context in ways a single number cannot express. Hemoglobin and ferritin references differ by sex.1811 Kidney and some hormone markers drift with age, so a value that is optimal at 30 may be unremarkable at 70.1920 Pregnancy rewrites thyroid targets entirely.21 Medications, acute illness, recent exercise, hydration, and even the time of day can move a result.22 And much of the functional-range literature remains genuinely debated — different expert groups draw the optimal line in different places, as the vitamin D disagreement shows.
That is why the honest framing is that optimal ranges are a lens, not a verdict. They are most useful for two things: deciding which "normal" results deserve a closer look, and giving you a more informed conversation with a clinician who knows your full picture. A number outside an optimal band is a reason to ask a better question, not a reason to diagnose or treat yourself. Bloodwork sits squarely in the territory where getting it wrong has real consequences, so caution and professional input are part of the method, not an afterthought.
How to read your own results through this lens
The practical version of all this is a short discipline you can apply to your last panel this week. Start by scoring each marker against two ranges at once: the lab's reference band, and, where one exists, the outcome-based cut-off from bodies like the ADA and the AHA. When a value lands inside the first but outside the second, you have found a gray-zone result — technically normal, plausibly worth attention. Those are the numbers to raise with your doctor and, where appropriate, to act on through the ordinary levers of sleep, movement, nutrition, and follow-up testing.
Then watch the slope, not just the point. The most informative thing about most markers is the direction they are moving across panels, and no single report can show that. A B12 of 420 means one thing if you have held there for three years and something else if you were at 600 last year. Keeping every panel — Quest, LabCorp, or a direct-to-consumer service — in one place lets you see drift before it becomes a flagged abnormality. If you want a step-by-step method for working through a report this way, our guide on how to read blood test results is the companion to this framework, and a biological age estimate is one way to see how several of these markers combine into a single trajectory over time.
Reading each result against both a lab range and an optimal range, connecting markers across the report, and holding the whole trajectory in one timeline is exactly the gap Merios was built to close. But the tool matters less than the habit: once you stop reading "normal" as "optimal," your own bloodwork starts telling you a far more useful story — not just whether you are sick today, but whether you are in the best position to stay well.