How does Dr. Sater decide which tests to order?
Written and medically reviewed by Zahraa Sater, M.D., M.P.H.
Reviewed August 12, 2026 · Next review due August 12, 2028
Short answer
Three questions decide it. Would the result change what I do? Is the condition likely enough here that a positive result would mean something? And what is the plan for each possible answer, including the borderline one? A test that fails all three adds cost, anxiety and follow-up testing without moving the diagnosis anywhere.
Every test is decided by three questions. Would the result change what I do? Is the condition likely enough in this person that a positive result would mean something? And what is the plan for each possible answer, including the borderline one that sits just outside the range? A test that fails those questions is not neutral. It generates repeat draws, scans and referrals, and it costs the patient both money and sleep.
Would the result change management
The first filter removes most of what patients read about online. If a levothyroxine dose will not change whichever way a test comes back, the test is not worth drawing. Reverse T3 is the clearest case: the American Thyroid Association states that in healthy, non-hospitalised people, measurement of reverse T3 does not help determine whether hypothyroidism exists and is not clinically useful. No value it can return alters the prescription, so nothing is gained by knowing it, and a number outside a range invites a treatment that has not been shown to help.
Is the condition likely enough to test for
The second filter is probability. A test performed in someone unlikely to have the condition returns mostly false positives, which is why screening rules exist at all. The US Preventive Services Task Force concluded in 2015 that the evidence is insufficient to assess the balance of benefits and harms of screening for thyroid dysfunction in nonpregnant, asymptomatic adults. That is not a statement that thyroid tests are useless. It is a statement that testing everybody is a different act from testing someone with symptoms, a family history, or a previous abnormal result.
What is the plan for each possible answer
The third filter is the one most often skipped. Before a test is ordered there should be an answer ready for high, low and borderline. A morning cortisol drawn with no plan for a mildly raised result leads to a dexamethasone suppression test, then an adrenal scan, then an incidental finding, then another referral, in a patient who never had Cushing's syndrome. The NIDDK notes that Cushing's syndrome is hard to diagnose because fatigue and weight gain have many causes, which is precisely why the sequence has to be planned rather than improvised after the fact.
| Commonly requested test | The decision |
|---|---|
| Free T4 with TSH | Ordered in anyone on levothyroxine, or where the TSH does not fit the symptoms |
| TPO antibodies | Ordered once when Hashimoto's is suspected, not repeated to track the disease |
| Reverse T3 | Not ordered; the ATA states it is not clinically useful in healthy outpatients |
| Fasting insulin with glucose | Ordered on the same sample when insulin resistance is the actual question |
| Morning cortisol | Ordered when there is a clinical reason and a plan for a borderline result |
| IgG food panels, hair mineral analysis | Not ordered; neither changes an endocrine diagnosis |
| Thyroid ultrasound | Ordered for a palpable nodule or an abnormal finding on other imaging, not as a screen |
If you want a specific test, ask for it and say why you want it. A reasonable clinician will either order it or tell you which of the three questions it fails. What you should not accept from anyone is a test ordered with no stated reason, or a result reported to you with no stated next step. Both are how a workup turns into a treadmill.
The clinical detail
Sequencing rules that follow from the three questions. Cortisol is drawn between 8 and 9 a.m., because a result without a draw time cannot be interpreted. Fasting insulin is only interpretable alongside fasting glucose from the same sample, since HOMA-IR requires both. Total testosterone in men is drawn between 8 and 10 a.m. on two separate mornings before hypogonadism is diagnosed. Prolactin is repeated before acting on a single mildly raised value, because venepuncture stress alone can raise it.
Biotin at supplement doses interferes with streptavidin-biotin immunoassays and can falsely raise free T4 and free T3 while falsely lowering TSH, so the supplement history forms part of deciding whether a result is real before deciding what it means.
Written and medically reviewed by Zahraa Sater, M.D., M.P.H.
Reviewed August 12, 2026 · Next review due August 12, 2028
Sources
- Thyroid Function Tests, patient information on which tests are useful. American Thyroid Association
- Screening for Thyroid Dysfunction, recommendation statement. U.S. Preventive Services Task Force
- Cushing's Syndrome. National Institute of Diabetes and Digestive and Kidney Diseases
- Thyroid Nodules, patient information. American Thyroid Association
- Clinical Practice Guidelines We Can Trust. Institute of Medicine, National Academies Press
This library is educational and is not individual medical advice. Read our Editorial and Medical Review Policy.