Show me the study. And tell me what it means.

Absolutely. “A study says” should start the conversation, not end it.

I want you to know who was studied, what actually happened, and whether the result has anything to do with the decision you’re making. Otherwise, a citation can become a fancy accessory for an ordinary sales pitch.

One example: the percentage trick

In SELECT, major cardiovascular events occurred in 6.5% of the semaglutide group and 8.0% of the placebo group. You’ll see that described as a 20% relative reduction; the absolute difference was 1.5 percentage points during the study. Both are valid, but they paint different pictures if you hear only one.

Participants already had cardiovascular disease and overweight or obesity, without diabetes. That is useful information for those patients. It is not evidence that everyone who takes any GLP-1 gets the same benefit.

Three questions I ask

Was it designed to test cause and effect? A randomized trial answers a different question from an observational association or a few case reports.

Was the change meaningful? A small laboratory change is not automatically a noticeable improvement in health or how someone feels.

Does it apply here? The population, dose, treatment duration, comparison group, funding, and limitations all belong in the picture.

You’ll find direct links next to the research discussed in these answers. Some topics are better served by a medication label or official coverage policy than a clinical trial. I’ll tell you which kind of evidence we’re using.

You shouldn’t need to read a 30-page paper to understand one useful finding. Translating that finding without losing its meaning is part of the job.

Sources checked September 19, 2026. General education; your own care plan comes from your clinician.