The weakest form of published clinical evidence is also, historically, one of the most useful. A case report describes what happened to one patient. A case series describes a handful. Neither has a control group, neither can estimate how often anything occurs, and neither can establish that a treatment caused an outcome. And yet the recognition of new conditions, of unexpected drug harms, and of rare presentations of common diseases has repeatedly started exactly there.
Its job is to raise a hypothesis, not to test one. A clinician notices something that does not fit the expected pattern and puts it on the record so that others can look for it. That is a legitimate and necessary function. The failure mode appears when the report is read as though it were a study: a single striking recovery becomes evidence that a treatment works, when the same report is equally consistent with spontaneous improvement, regression to the mean, a wrong initial diagnosis, or the effect of something else the patient was taking.
A case series of twelve patients who improved tells you nothing until you know how many people received the same treatment and did not improve. That denominator is exactly what an uncontrolled series does not have. This is why "we have seen remarkable results in our practice" is not a claim that can be checked, and why series with dramatic conclusions so often fail when someone finally runs a controlled trial.
The familiar pyramid — systematic reviews above randomized trials, trials above cohorts, cohorts above case series — is a rule of thumb about susceptibility to bias, not a ranking of importance. A large, badly conducted trial can be less informative than a careful observational study, and for effects that are enormous and immediate, no trial was ever required. The useful question is not "what level is this?" but "what specific bias could produce this result if the effect were not real?"
When coverage of a condition rests on a case report, the honest description is that something was observed once. When it rests on a series, the honest description is that something was observed a few times, without comparison. Those are real starting points. They are not conclusions, and treating them as conclusions is how a rumor about a condition acquires a citation.