Most of the distance between a headline and reality can be closed with six questions. None of them require you to understand the statistics.
1. Was it in humans?
Mouse studies are where most longevity headlines originate, and most compounds that extend mouse lifespan do nothing measurable in people. This single question filters more noise than the other five combined.
2. How many participants, and for how long?
Twelve people for eight weeks is a pilot study. It generates a hypothesis. It does not establish anything, no matter how the abstract is worded.
3. What was the actual endpoint?
There is an enormous difference between "reduced a biomarker" and "reduced heart attacks." Surrogate endpoints are cheaper and faster to measure, and they frequently fail to translate into outcomes anyone cares about.
4. Compared to what?
No control group means no conclusion. An uncontrolled study measures the passage of time, regression to the mean, and the placebo response, and attributes all three to the intervention.
5. Relative or absolute?
"Reduces risk by 50%" sounds decisive until you learn the risk went from 0.2% to 0.1%. Relative risk reduction is the number that makes press releases; absolute risk reduction is the number that should inform your decision.
6. Who funded it?
Industry funding does not invalidate a study, but it shifts the prior. Look for pre-registration, independent replication and whether the authors sell the thing they studied.
The hardest habit
Applying this consistently means occasionally concluding that something you wanted to be true isn't well supported. That discomfort is the method working, not failing.