There are several types: cohort studies follow groups of people over time to see who develops an outcome; case-control studies compare those who already have it with those who do not; cross-sectional studies measure everything at a single point in time. They are useful for studying large populations, rare effects or exposures it would not be ethical to assign.
Their great limitation is confounding. If people who use a product also exercise more or have higher incomes, any difference in their health may be due to that and not to the product. Statistical techniques adjust for known factors, but not for those nobody measured.
That is why, when an observational study suggests a benefit, the expectation is to confirm it with a randomised trial before treating it as proven.
“Is associated with” is not the same as “causes”. Telling these two phrasings apart avoids a large share of misinterpretations of the literature.
Related terms
Definition for educational and scientific purposes. It is not medical advice or a recommendation for use. NeoPeptidos products are sold labelled for research use only (RUO).