TSH

What it actually measures

Put a hand on the front of your throat, just below the Adam's apple. Underneath is a small butterfly-shaped gland that sets the pace of nearly everything your body does.

Here is the part that trips people up. TSH is not made there. It is made by the pituitary, a gland at the base of your brain, and its job is to tell the thyroid how hard to work.

So the number on your report is not your thyroid hormone. It is the volume of the instruction being shouted at your thyroid. When the pituitary sees plenty of thyroid hormone in circulation, it goes quiet and TSH falls. When it sees too little, it shouts louder and TSH rises.

That inverse relationship is why the test confuses people. A high TSH generally means a sluggish thyroid, not an overactive one. High number, low function.

It is a genuinely clever piece of design, and it is a good screening test. It is also one measurement standing in for a chain with several steps, which is where its limits come from.

The range nobody agrees on

Here is the tension though.

In 2005 a paper in the Journal of Clinical Endocrinology and Metabolism argued that the upper limit of the TSH reference range should come down substantially. The reasoning was specific and worth understanding: the reference populations used to build the range had included people with undetected thyroid disease, and once those were excluded the vast majority of genuinely healthy people sat below about 2.5 mIU/L.

That argument is why a great many practitioners prefer to see TSH between 1 and 2, and why you can be told your 3.8 is perfectly normal in one office and worth watching in another.

The counter-argument is also real and belongs here. Narrowing the range means labelling a very large number of people as having a thyroid condition, many of whom feel fine and would never progress to anything. Later reviews have pushed back on how much a mildly raised TSH actually predicts in an otherwise well person, particularly with age, since TSH drifts upward normally as people get older.

Both positions are held by serious people. What that means for you is practical rather than philosophical: your TSH result is a reasonable place to start a conversation and a poor place to end one.

The thing TSH cannot see

Now here is where I have to tell you something that reframes the whole test, so listen close.

TSH cannot show you autoimmune thyroid activity.

The most common cause of an underactive thyroid in the developed world is autoimmune, where the immune system gradually damages the gland. That process can run for years before it moves TSH far enough to be flagged. What reveals it is antibodies, and antibodies are a separate test that has to be asked for.

The numbers on this are worth knowing. In the Whickham survey, a community study that followed people for twenty years, those with thyroid antibodies and a normal TSH went on to develop overt hypothyroidism at roughly 2.1 percent per year, and the risk was considerably higher when a raised TSH and antibodies occurred together. Antibodies are also common in people whose thyroid is currently working perfectly well.

So a normal TSH and a positive antibody result is a meaningful finding: it says your thyroid is coping now, and it tells you what to keep an eye on. You only get that finding if somebody orders the antibodies.

What moves it

Up: autoimmune thyroiditis, by far the most common cause. Iodine deficiency, and also iodine excess, which surprises people. Some medications, including lithium and amiodarone. Recovery from a serious illness. Age, since TSH drifts up normally over decades. Being tested early in the morning, since TSH follows a daily rhythm and peaks overnight.

Down: an overactive thyroid. Slightly too much thyroid medication, which is a common and easily corrected finding. Early pregnancy, where hormonal changes suppress it. Occasionally a pituitary problem, in which case the whole chain reads oddly and needs proper investigation.

A note on timing. Because of that daily rhythm, drawing at different times of day can shift a result enough to matter near a decision threshold. Consistent morning draws make your results comparable to each other.

The honest hedge

I am not a doctor and I am not diagnosing anyone.

Thyroid is one of the most over-attributed topics in wellness writing. A great many symptoms get laid at its door, and it genuinely is responsible for some of them. It is also true that plenty of people are told their thyroid is the problem when their sleep, their training load, or their stress is doing most of the work.

What I would say plainly: TSH alone is a third of the question. Measuring the whole panel is how you find out whether the thyroid is actually involved, rather than assuming from a single number in either direction.

And if you are already on thyroid medication, do not adjust it based on a page on the internet. Dose changes belong with the prescriber who can see your full picture and your symptoms.

So. What to actually do.

Tonight, at no cost. Find your last panel and look at what was actually run. If TSH is the only thyroid line on it, you now know the size of the gap.

When you test, go wider than TSH. A thyroid panel with free T4 and free T3, plus antibodies, in one draw. Running them together is what makes any of them interpretable.

Draw in the morning, consistently. Same rough time each visit, so your results can be compared to each other rather than to the time of day.

If antibodies come back positive with a normal TSH. That is useful information rather than bad news. It means monitoring on a sensible interval instead of finding out years later, and it is a conversation worth having with a good functional medicine doctor as well as your regular physician.

None of this has to happen this week. But tonight you can look at whether anyone has ever measured your antibodies.

You are not a slow metabolism. You are a signalling chain with several links in it, and the test almost everyone gets measures only the first one.

Your body is not broken. It is blocked. And for a great many people who have been told their thyroid is fine, the block is simply that nobody ordered the other two thirds of the panel.

Go look at what was actually run.