Case File 005: Why You Feel Off in Midlife Even When Your Labs Are “Normal”

There is one word on a lab report that I have learned not to take quite as literally as I once did.

Normal.

I like normal lab work. I am not looking for trouble, hoping something turns red on the report, or secretly disappointed when nobody finds a disease. Reassuring results are reassuring for a reason.

What started bothering me was what sometimes happened next.

You know something has changed. Perhaps your energy has dropped, your hair is shedding, you are suddenly cold all the time, your digestion is different, your sleep has gone sideways, your periods have become unpredictable, or your brain seems to have misplaced several functions you were fairly certain it used to perform without supervision. You describe the change, blood gets drawn, the numbers return inside their designated boundaries, and the explanation becomes:

“Everything is normal.”

That sentence sounds much more conclusive than the laboratory result actually is.

It took thyroid symptoms for me to really understand why.

A laboratory reference interval is a comparison tool. For many tests, laboratories establish or adopt a range based on values found in a defined reference population. Other tests are interpreted using clinical decision limits because evidence has tied particular thresholds to risk, diagnosis, or treatment decisions. Even before we get into individual biology, then, the familiar little green band on the report is doing more complicated work than the word normal suggests.

What it is not doing is describing the exact physiological territory in which you personally have always lived.

That distinction seems obvious once you say it out loud, but I am not sure most of us were ever taught to think about our lab work that way. We see a range, see our number tucked safely inside it, and understandably assume the question has been answered.

Sometimes it has.

Sometimes the test was ordered to investigate a particular possibility, the result makes that possibility much less likely, nothing else in the history raises concern, and everybody can reasonably move on.

But a value can also sit inside a population reference interval while being quite different from where that same person has historically lived. Researchers in laboratory medicine have been investigating individualized reference intervals for precisely this reason: within-person biological variation can be considerably narrower than variation across an entire population.

That idea is interesting, but it also needs guardrails. Personalized reference intervals are not sitting inside everybody’s patient portal ready for routine clinical use, and we do not yet have evidence that turning every historical laboratory result into an individualized range automatically improves outcomes. This is an emerging area of laboratory medicine, not permission to declare every numerical wobble a hidden disease.

Still, it changes the question.

Instead of asking only, “Is this result inside the range?” we can sometimes also ask, “Is this different from what is usual for me, and if so, does that difference matter?”

Those are not competing questions. They belong beside one another.

And that is the part I wish more of us understood before we arrived in the exam room.

Midlife makes the limitation of a single measurement particularly easy to see because reproductive hormones do not simply descend in a tidy line toward menopause.

Apparently nobody informed the ovaries that we would prefer a spreadsheet.

The menopausal transition involves changing menstrual patterns and changes in ovarian and pituitary hormones that are not uniform from woman to woman or even necessarily smooth within one woman. Longitudinal studies have found different trajectories for estradiol and FSH across the transition, while reproductive-aging frameworks place considerable importance on changes in menstrual-cycle patterns rather than pretending one blood draw can neatly stage the entire process.

That means a laboratory result can be completely accurate and still represent only the moment in which it was taken.

This is not an argument that hormone testing is useless. It is an argument for understanding what a particular test can reasonably tell us. A blood draw is a measurement, not a documentary film.

It also means we should resist the equally tempting explanation that every new symptom after forty must be perimenopause. Fatigue, poor sleep, brain fog, hair loss, anxiety, palpitations, changes in weight and altered mood have plenty of possible causes. Reproductive hormone changes may be part of the picture while anemia, medication effects, thyroid disease, sleep disorders, nutritional problems, stress, or something else entirely is also happening.

That matters because once we become suspicious of the word normal, it would be very easy to swing too far in the other direction and begin distrusting every reassuring result.

I do not think that serves us either.

The lab has information. Your symptoms have information. Your history has information. The sensible work is figuring out what those pieces mean together.

Thyroid was where this became more than an interesting concept for me.

It is also the perfect place to get ourselves into trouble, because thyroid symptoms have become internet catnip. If you are tired, cold, constipated, losing hair, gaining weight, having trouble concentrating, or generally dissatisfied with the cooperation level of your body, somebody somewhere is prepared to tell you that your doctor missed your thyroid problem.

That is not what the evidence says.

TSH remains a very useful test for identifying primary thyroid dysfunction in most outpatient situations. A normal TSH is meaningful information. Fatigue, constipation, dry skin, weight changes, mood changes and brain fog are also maddeningly nonspecific; no individual symptom can establish hypothyroidism, and many people with similar complaints do not have thyroid disease at all.

So I am not interested in teaching people to wave away a normal TSH because their symptoms “feel thyroid.”

What interests me is the other part of the story.

One classic longitudinal study followed healthy people with repeated thyroid testing and found that each individual tended to operate within a narrower range than the broader group reference interval. The study was small—sixteen healthy men—so it would be ridiculous to use it as proof that every person has an undiagnosed thyroid condition hiding inside the laboratory range. But it demonstrated an important principle that has continued to interest laboratory researchers: population variation and within-person variation are not the same thing.

That helped me understand why previous results can sometimes be useful.

Suppose your TSH was 1.1, then 1.3, then 1.4 over several years, and later it is 3.5. Depending on the laboratory, all of those values may still fall inside its reference interval. That does not mean 3.5 is necessarily abnormal for you, and it certainly does not diagnose hypothyroidism. It does mean there is now a piece of longitudinal information that may be worth placing beside your symptoms, medications, health history, other thyroid measurements when appropriate, and everything else your clinician knows about the situation.

The old way I thought about labs was essentially binary: inside the lines, good; outside the lines, investigate.

I no longer think biology is always that accommodating.

One reason laboratory numbers acquire so much authority is that they are wonderfully tidy. A blood value fits neatly into a chart. “I have felt increasingly strange for six months” does not.

That doesn’t make the second observation worthless. It means it needs better description.

There is a substantial difference between walking into an appointment saying, “I’m exhausted,” and being able to explain that your sleep changed in March, you began waking several nights a week, your menstrual pattern changed around the same time, your daytime fatigue followed, and nothing obvious in your routine explains the shift. Neither account provides a diagnosis, but the second one gives the person trying to help you considerably more context.

This is the kind of self-advocacy I mean when I talk about advocating for yourself in front of the white coats.

I do not mean walking into the office prepared for battle. I do not mean ordering a suitcase full of laboratory tests because someone on social media drew a red circle around a biomarker. I also do not mean shopping for clinicians until somebody eventually endorses the diagnosis we arrived with.

I mean becoming a better witness to your own history.

If something has changed, write down roughly when it began. Describe what was normal for you before. Notice whether the symptom is constant or intermittent and whether anything seems to travel with it—sleep disruption, menstrual timing, medication changes, meals, alcohol, illness, stress, activity, or something else. If you have earlier lab results, keep them where you can actually find them.

You are not constructing a prosecution exhibit.
You are making the pattern easier to see.

That does not require a complicated tracker or six months of immaculate records. Before an appointment about a persistent symptom, it may be enough to write down four things: what changed, approximately when it changed, what was normal for you before, and anything else that changed around the same time. If you have older relevant lab results, bring those too. You are looking for context, not building a case for the diagnosis you already suspect.

This small amount of preparation can turn “I just don’t feel right” into something much easier to investigate. It gives the clinician a timeline, gives you something more reliable than memory under fluorescent lights, and sometimes reveals a pattern neither of you would have noticed from a single symptom or laboratory result alone.

That changes the conversation from “My labs are normal but I know something is wrong” to something much more useful: “These symptoms began around this time, here is how they have behaved, these are the changes I have noticed, and these are my previous results. What possibilities does that pattern raise, and what does the testing we have already done make less likely?”

Now everybody has something to work with.

And perhaps equally important, you are not asking the clinician to accomplish an impossible task: know the history of your body better than the person who has been living in it.

Clinical expertise and lived observation are different kinds of knowledge. Good medical care has room for both.

Once you discover that a laboratory reference interval is not a personalized definition of perfect health, the wellness world is delighted to sell you a replacement.

Optimal.

It is a very attractive word. Why settle for normal when we could be optimal?

The trouble is that narrowing a range and giving it a more aspirational name doesn’t automatically make it more scientific. For some biomarkers, evidence supports particular targets or clinical decision thresholds because those numbers are associated with meaningful outcomes. In other situations, an “optimal range” may represent an expert preference, a particular functional-medicine framework, an extrapolation from incomplete evidence, or simply a number that has been repeated often enough to acquire an air of authority.

So I do not want to replace blind faith in normal with blind faith in optimal.

I want to know what the number means.

Where did the range come from? What population was used? Is this actually a reference interval, or is it a clinical threshold? Does being outside it predict anything important? Has changing the number been shown to improve an outcome we care about? And does the evidence apply reasonably well to the person sitting in the room?

Those questions are considerably less exciting than declaring that everybody should have a TSH of precisely whatever number is trending this month.

They are also more useful.

The part I keep coming back to is the quiet leap that can happen between two statements:

The tests we ran were normal.

and

Nothing is wrong.

They are not necessarily the same statement.

Sometimes normal testing really is reassuring enough to close the question that prompted it. Sometimes it removes a frightening possibility from the table. Sometimes symptoms improve and there is nothing more to investigate.

Sometimes, though, the first round of testing simply tells us where the answer is not.

That is still progress.

If the symptoms continue, the next step does not have to be panic, distrust, or an expensive fishing expedition through every laboratory test available to modern civilization. It may simply mean returning with a clearer history and asking what the original tests actually evaluated, what they reasonably ruled out, what other explanations fit the pattern, and what would justify another look.

That is what changed for me.

I stopped expecting one laboratory range to settle questions it was never designed to answer, but I also stopped treating my observations as if they should outrank legitimate medical evidence.

I want both.

I want the objective test. I want the clinician who understands disease, differential diagnosis, and the limitations of the testing. I also want the history of what has changed in this particular human being, because nobody else arrives at the appointment carrying that information unless the person living it brings it along.

So when I see the word normal now, I am grateful for what it can tell me.

I just do not ask it to tell me more than it knows.

The question is no longer simply, “Are my labs normal?”

It is:

What did these results actually tell us, what has changed for me, and is there still a question worth answering?

That is a much better place to begin.

• A laboratory reference interval is a useful comparison tool, but it is not a personalized definition of how your body normally functions.
• Previous results can sometimes add valuable context because within-person biological variation may be narrower than variation across a population.
• A single hormone measurement may provide only a snapshot during the menopausal transition, when reproductive patterns are changing over time.
• Symptoms associated with thyroid disease are often nonspecific. A normal TSH is meaningful information and should not simply be dismissed because symptoms persist.
• Tracking when something changed, what your baseline was, and what seems to accompany the change can make a medical conversation more useful without turning symptom tracking into self-diagnosis.
• “Optimal” ranges deserve the same scrutiny as “normal” ones. Ask where a target came from and what evidence supports it.
• Self-advocacy works best as participation: your observations and your clinician’s expertise should inform the same investigation.

Title	Link
The Role and Limitations of the Reference Interval Within Clinical Chemistry and Its Reliability for Disease Detection	https://pubmed.ncbi.nlm.nih.gov/38481978/
Personalized Reference Intervals: From Theory to Practice	https://pubmed.ncbi.nlm.nih.gov/35579539/
Narrow Individual Variations in Serum T4 and T3 in Normal Subjects: A Clue to the Understanding of Subclinical Thyroid Disease	https://pubmed.ncbi.nlm.nih.gov/11889165/
Executive Summary of the Stages of Reproductive Aging Workshop + 10: Addressing the Unfinished Agenda of Staging Reproductive Aging	https://pubmed.ncbi.nlm.nih.gov/22344196/
Menstrual Cycle Hormone Changes in Women Traversing Menopause: Study of Women’s Health Across the Nation	https://pubmed.ncbi.nlm.nih.gov/28368525/
Hypothyroidism: The Difficulty in Attributing Symptoms to Their Underlying Cause	https://pubmed.ncbi.nlm.nih.gov/36814580/
Reference Intervals in Value-Based Laboratory Medicine: A Shift From Single-Point Measurements to Metabolic Variation-Based Models	https://pubmed.ncbi.nlm.nih.gov/40817855/