Step 1: The question asks which observational study best tests the association between a risk factor and a disease, that is, which study most reliably points to a cause.
Step 2: A cohort study starts with exposed and non-exposed people who are free of the disease and follows them forward in time to see who develops the disease. Because exposure is recorded before the outcome appears, a cohort study clearly fixes the time sequence (cause before effect) and directly measures incidence and relative risk. This makes it the strongest observational design among the options.
Step 3: The ranking of analytic studies by ability to prove association, in decreasing order, is: systematic review and meta-analysis (overall most reliable), randomised controlled trial (best single study), cohort study, case-control study, cross-sectional study, and ecological study.
Step 4: Why the other options are weaker: a case-control study works backward from disease to exposure and is prone to recall and selection bias; a cross-sectional study measures exposure and disease at the same point in time, so it cannot establish which came first; an ecological study uses group-level data and risks the ecological fallacy. Hence cohort is the best option here.