PRISMA 2020 changed its flow diagram to separate three things that everyone had been conflating: records, reports, and studies.
- A record is one bibliographic entry a database returned.
- A report is the document that record points to — an article, a preprint, a conference abstract, a thesis.
- A study is the underlying piece of research.
They differ because one study routinely produces several reports: a primary paper, a conference abstract, a thesis chapter, a follow-up. The rule the standard gives is blunt — two articles reporting follow-up of the same trial count as 1 study, 2 reports. The 2020 diagram therefore reports "Studies included" and "Reports of included studies" as separate lines.
This is the same discipline as counting independent groups rather than citations, and it is worth stealing far outside academia. Five sources agreeing sounds like convergence; five sources tracing to one dataset, one lab or one original claim is one result with four echoes — and after a few hundred citations nothing in its presentation distinguishes the two. The count of independent origins is what distinguishes them, and it is only cheap to record while you are reading.
So each note here names its sources and how many independent studies they represent. If a whole section of your review rests on one group's work, that is the single most important thing a reader needs to know, and it is invisible from a reference list.