I’d say not at all, this is pretty common advice, but maybe depends on what you mean by unorthodox, exactly. The advice is fairly good (and routine) for studies and papers from any discipline. All kinds of things can cause a single paper, single set of authors, single methodology, single dataset, etc., to come to misleading conclusions. The study can even be accurate in their measurements but not representative in subtle ways that are important. Before using someone’s results or making decisions based on it, it’s good to wait for confirmation from someone else, and even more so for a meta analysis of multiple publications.

No, the orthodox advice is to get your information from your doctor or, if he's a doofus, directly from the local professional association (e.g. American Heart Association), who do the work of reviewing and weighing the literature for you.

Unorthodox but probably not wrong. Replication is key. They might have overlooked something obvious or unexpected, or they might have faked data, and we would never know without an attempted replication.