I have doubts about the use of "opinion" in vinika's question to the extent that it forms part of the answer.
So as soon as you see or hear: "my opinion is"; "in my opinion"; or "the general opinion is" then the alarm bells should start to sound saying "is this good science or even science at all?"
Scientists make statements they do not issue opinions (although these are allowable in discussion as to the verity of a statement).
To detect bad science needs hard-working analysis of what you are expected to believe. This belief should not claim to be the absolute truth, merely the best explanation of the latest chemico/physico/medico data.
The opinionator should be able to begin by explaining where the opinion originated and it's rational basis i.e. the starting hypothesis. This should exclude any "self-evident truths" but be based on observation or derived from what is now accepted as fundamentally proven beyond reasonable doubt (to date).
The evidence for testing the hypothesis should be stated and be practicable for others to carrry out. This testing should be used to gather data and the tests should be constructed to DISPROVE the hypothesis not to confirm it.
After the exhaustive testing by many different individuals or groups, without any disproof, the hypothesis is then elevated theory.
Although a theory can be built upon, once estalished to the satisfaction of all the experts in the field, (there will always be rebels which is not a bad thing but they then have the gargantuan task of overturning the theory). Failure to establish a theory in this way but still proclaim it true is bad science.
Continual criticism is the bedrock of good science and refusal of this criticism is bad science if science at all.
I know a bit of stats but will not dwell on this except to warn about reading graphs. Do the ordinate and abscissa (y,x axes) both start at the same fair point. So often the vertical axis is chopped to make the trend seem steeper.
Daniel Fynberg was a great science teacher. In explaining that in nature, notably the electron, neutron and proton etc occurred as whole entities he cited the statement: "Statistics tells us that the average family has 2.5 children". He then challenged his students to show him half a child! Hence stats can be meaningless in the real world.
I'd like to deal with clinical trials and studies but maybe later.
Enough for now,
SIQ.