Protocol for prospective collaborative overviews of major randomized trials of blood-pressure-lowering treatments. World Health Organization-International Society of Hypertension Blood Pressure Lowering Treatment Trialists' Collaboration.
OBJECTIVE: To conduct prospectively planned overviews (meta-analyses) of the ongoing randomized trials of blood-pressure-lowering drugs. These overviews should provide reliable data about the effects of newer classes of blood-pressure-lowering drugs (such as angiotensin converting enzyme inhibitors and calcium antagonists) on major causes of cardiovascular mortality and morbidity for a variety of patient groups. METHODS: A registry of major ongoing or planned randomized trials (with more than 1000 patient-years of follow-up for each randomized group) of blood-pressure-lowering agents has been established. The principal investigators of each of these studies have been invited to collaborate in the project and to provide, upon completion of the study, a limited data set for inclusion in the overview analyses. The principal comparisons will be of newer versus older classes of blood-pressure-lowering drugs in treating patients with hypertension and of newer blood-pressure-lowering treatments versus untreated or less treated control conditions for a variety of other groups of patients with a high risk of cardiovascular events. Separate analyses will be conducted for the main drug classes and for major subgroups of patients defined by characteristics such as age, gender, blood pressure, diabetes, and history of renal disease, coronary heart disease or cerebrovascular disease. The principal study outcomes are stroke, major coronary heart disease events, heart failure, total cardiovascular deaths, total cardiovascular events and total mortality. RESULTS: In total 36 trials of blood-pressure-lowering treatments potentially eligible for inclusion in this project have been identified and agreement to collaborate has been confirmed by the investigators in 30 trials, with collaboration pending for six recently identified studies. The first round of analyses will be conducted in 1999 and will be based on total cardiovascular events observed among a total of about 64,000 patients, involving about 240,000 patient-years of follow-up. The second round of analyses will be conducted in 2003 on data from at least 195,000 patients and 899,000 patient-years of follow-up. By that time it is estimated that a total of about 8000 strokes, 12,000 coronary heart disease events and 23,000 cardiovascular events in total will have occurred. This should provide good statistical power to detect even modest cause-specific differences in the incidence of the main study outcomes. CONCLUSIONS: The combination of trial results in prospectively planned, systematic overviews both reduces random errors and avoids biases. As a consequence, this project should provide more reliable information about the effects of newer blood-pressure-lowering drugs than would any one study alone. The use of data from individual patients in these overviews will facilitate investigation of the separate effects of various drug regimens in treating members of major patient subgroups.