Hello, former clinical lab scientist here. A couple thoughts.
there are several major manufacturers of laboratory analyzers that run thyroid tests. Even if you are getting results that seem off from more than one lab, that doesn’t mean it can’t be an assay issue as they could be using the same analyzer/reagent kit. There are checks and balances on both the manufacturer side and lab side to prevent issues, but it’s still not unheard of to get a lot that performs oddly. Without feedback from the clinicians it can be hard sometimes for labs to detect and pinpoint.
Might help to ask around what manufacturer and assay is being used for the thryoid tests at each of the labs where you’re seeing this. If you can reach out to a lab director (PhD or MD level) with specific pts and examples of results that don’t seem right, they might be able to look into it more closely. At least at academic places if you can provide good details and just come at it seeking help (not accusing the lab of screwing something up) the PhD lab director types especially seem to really like trying to figure these things out.
Second, definitely ask about biotin. It interferes with immunoassay tests which includes most thyroid tests. The direction of the interference depends on the assay design. The trick is, biotin can be called a lot of different things and so patients might not even realize they’re taking it, or something that contains a lot of it. So you have to ask about all kinds of supplements and stuff for hair and nails or concentration boosting (it’s being marketed as a cure all for soooo many things). You might just want to have them bring in anything OTC supplement they’re taking and check the label. A former, colleague of mine did a lot of work on biotin interference. It’s kind of wild how much ppl wind up taking these days and how much it screws up testing.
Other supplements can do wonky stuff too so if there’s something new out there circulating around or some new internet fad, could cause issues but probably would be more difficult to pin down.