Collaborative overview ('meta-analysis') of prospective observational studies of the associations of usual blood pressure and usual cholesterol levels with common causes of death: Protocol for the second cycle of the Prospective Studies Collaboration
Iso H., Sato H., Chambless L., De Backer G., De Bacquer D., Kornitzer M., Ebrahim S., Whincup P., Wannamethee G., Wald N., Morris J., Knuimann M., Sweetnam P., Elwood P., Kronmal R., Kromhout D., Sutherland S., Keil J., Schnohr P., Jensen G., Grobbee D., Witteman J., Hames C., Aromaa A., Knekt P., Reunanen A., Tuomilehto J., Jousilahti P., Vartiainen E., Levy D., D'Agostino R., Silbershatz H., Thomsen T., Bengtsson C., Sharp D., Benetos A., Guize L., Goldbourt U., Yaari S., Murayama T., Tomita M., Nishimoto M., Staessen J., Criqui M., Davies C., Jacobs D., Blackburn H., Luepker R., Neaton J., Cox C., Ofstedal M., Weiss S., Cassano P., Sparrow D., Vokonas P., Tverdal A., Selmer R., Meade T., Garrow K., Cooper J., Menotti A., Spagnolo A., Tsuji I., Imai Y., Ohkubo T., Hisamichi S., Haheim L., Holme I., Hjermann I., Leren P., Ducimetière P., Richard J., Jamrozik K., Broadhurst R., Assmann G., Schulte H., Clarke R., Collins R., Donald A., Duffy S., Lewington S., MacMahon S., Peto R., Qizilbash N., Rodgers A., Sherliker P., Zhang W., Sorlie P., Garcia-Palmeri M., Barrett-Conner E., Langer R., Gillis C., Hole D., Nakachi K., Fang X., Li S., Buzina R., Kivinen P., Nissinen A.
Background: Many prospective studies have been reported on the associations of blood pressure and blood cholesterol with cardiovascular disease, but few have been large enough to provide reliable estimates of the nature of these associations in different circumstances or to characterize the associations with non-vascular causes of death. Moreover, almost all such reports have related risk to baseline measurements of the risk factor of interest, which can lead to substantial underestimation of the importance of the risk factor due to 'regression dilution' bias. Objective: By appropriate combination of data on individual participants from all such studies in a systematic 'meta-analysis', with correction for regression dilution, to characterize more precisely than has previously been possible the age-specific and sex-specific relevance of blood pressure and blood cholesterol levels to particular causes of death. Methods: Information has been made available on about 175 000 deaths (45% due to vascular disease and 20% due to cancer) among more than 1.4 million individuals followed for an average of 14 years in over 60 prospective studies of blood pressure and blood cholesterol (with data on high-density lipoprotein cholesterol available for a substantial subset of individuals). This represents approximately 95% of the data from all relevant studies.