That is not true. There are a lot of different designs that can provide you with good quality evidence. Furthermore, double blind randomized study, while providing you with highest quality of data, may not yield a useful data. In some circumstances it can even mislead you. For example, a study may show that COVID vaccine is effective because it is measuring antibodies at 2 months after vaccine administration and it would be a statistically significant result. However, antibodies may not play a significant role in immunity to Covid or may just disappear after 4 months. The last two statements, by the way, are just hypothetical. Either way, there are many examples in history of medicine where double blind randomized studies mislead physicians and/or patients.