Interesting statement that I could interpret in different ways. A simplistic interpretation would be that scientists and trials produce evidence and patients produce stories. But I have an alternative interpretation:
Evidence is the fact, what is observered and measureable - although there may be error within the measurement. Stories are the discourse that subjectively select and interpet different combinations of evidence.
So a patients experiance is made up of facts (x happended, test said y) along with the story of how it fits together. The story is subjective because in is a framing for individual facts and often involves a choice of facts or an interpretation of them.
I don't see a clinical trial being a great deal different - although a little more planned,. Ok so things happen and are observed and various facts are created. The trial protocol then should form the subjective story around selected facts that is how the authors would like them to be interpreted. Concerns around trials exist because the subjective story should be formed before seeing the facts rather than afterwards. However, as we have seen from the pace trial this is often not the case.
We then have algorithms for combining facts (usually stats). These need to be very carefully chosen. In a patients situation we only have a few individual data points so the suggest the existance of an event rather than something that can be treated quantitively. The existance of experiance from multiple or many patients should let us make statements such as "at least x patients reported y". If we knew total number of patients with a particular disease it would allow us to express a lower bound. With a trial this should be done within a fixed framework but we should be very careful about statements being made. For example "trial x, with selection criteria y using randomisation process r processed results using algorithm a found ..." The choice of all the caveats in this stratement are part of the subjective story and should be stated. Sometimes they are wrong. For example it is simply wrong to quote the mean and standard deviation for the chadler fatigue scale hence the choice of the algorithm deriving results from the PACE trial is wrong and the results are therefore meaningless.
This leads me to believe that all the data from trials should be published so that others can process them in different ways.
I think it's very difficult, and maybe impossible, to distinguish 'stories' from 'evidence' or 'facts'. As you say, even high-quality medical trials contain a certain amount of 'story', or interpretation. But even hard 'data' is only created within a context.
Some 'facts' are more factual than others. Some evidence is more reliable than other evidence. A story, or hypothesis, or theory, or evidence, only becomes unquestionable 'fact', as far as an overwhelming number of people are concerned, when it has an overwhelming amount of unquestionable evidence behind it. Some 'facts' remain irrefutable for centuries, only for science to later find out that the 'facts' were wrong.
And one person's 'fact' is another person's dogma, fallacy, lie, or misrepresentation of the truth, etc.
We build a picture of evidence, with the more information that we have.
Personal 'stories' help to build a picture of the world.
People's individual experiences, and their interpretations of their experiences, is what forms people's opinions, and ultimately leads to how we construct our communities.
This is how politics works. Politics often reacts to the majority opinion. Opinions that are based on personal experience, or personal 'stories'.
I don't know if it's true, but I heard an interesting story on the radio the other day.
(I think they said it was originally written by a Financial Times journalist, and I haven't checked for the original article.)
It was a story about a competition in which competitors had to guess the weight of a bull. It also applies to guessing the number of jelly beans in a jar.
(There was a moral behind the story, but I'll just stick to this narrow interesting feature of the story.)
It said that, although it is rare for an individual to correctly guess the weight of the bull, or the number of jelly beans, if you have enough people involved in the competition, the average guess will be
very close to the actual answer.
So, this illustrates, that subjective evidence, or personal stories, when correlated from a number of people, can form an accurate view of the world.
It would be interesting to see if that story is true.
(There's also a major counter-point, and flaw, to this specific interpretation of 'group opinion' though, as I
never put my faith in the majority point of view The majority are so often very very wrong!)