Study designs

A paper is not read front to back. This chapter gives a reading order, the question to ask at each stop, and a lookup table from "what you just saw" to "the page that explains it".

Where the line between prospective and retrospective actually falls, why an observed association is not a treatment effect, and how to write up a result that did not reach statistical significance without fooling yourself.

Why a case-control study yields an odds ratio and never an incidence rate, what matching actually buys, and how one real analysis nearly credited the right finding to the wrong cause.

What the statistics section of an RCT is doing, how to read one in order, and the parts textbooks rarely cover but real trials always have — interim analyses, early stopping, and the patient who was counted anyway.

Sensitivity and specificity are properties of a test, but the numbers you get depend on whom you enrolled and whom you verified — this chapter uses live figures to show two ways a good test is made to look better, and why the same test reports a sensitivity three times higher in one ward than another.

A prediction model does not ask which variable is associated with the outcome; it asks what this person's risk is — and that switch changes what counts as passing. This chapter walks through a real external validation where discrimination barely drops while calibration goes systematically wrong.

A national claims database does not exist for research; it exists for reimbursement. This chapter covers what that gap costs you, the limits specific to Taiwan's NHIRD, and a seven-rung worked example — including the rung that lands on the right answer by the wrong method.

Why a systematic review and a meta-analysis are not the same thing, what each box of the PRISMA flow diagram is defending against, why "the search returned nothing" and "nothing survived screening" must never be written as one sentence, and how a result that exists only inside a figure can push a whole review's conclusion off course.