Relatively few observational (i.e., nonexperimental) studies have been conducted to examine the role of antihypertensive drug therapy in the primary prevention of coronary heart disease. To draw valid conclusions from experimental or observational studies, internal validity should be ensured. In particular, similarity of "extraneous" effects, of information, and of prognosis between the treatment groups compared in a particular study is needed. Because allocation of participants to antihypertensive drug therapy in observational studies is nonrandomized by definition, special efforts should be made to achieve comparability of prognosis, that is, to avoid "confounding by indication." Follow-up and case-control studies, the two main types of observational studies, also offer certain advantages over randomized clinical trials, particularly when different classes of drugs are compared. Although few valid observational studies on the efficacy of drug therapy for hypertension in the primary prevention of first coronary events have been published, the available data suggest that beta-blockers might confer greater protection than other drug regimens. However, more evidence is needed to confirm these findings. It is concluded that observational studies could play an increasing role in the assessment of the role of antihypertensive therapy in the primary prevention of coronary heart disease when the inherent potentials and pitfalls of these studies are appreciated.