Clinical trials and external validity

Although a number of clinical trials are available estimating the benefits of lipid-lowering therapies that include economic end-points, development of modeling methodologies are essential to extend results of those trials over time and to other populations. We reviewed the key issues to be considered when extending trial data to real-world situations. The availability of recent randomized controlled trials of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors in primary and secondary prevention has demonstrated the limitations of earlier modeling efforts to project benefits of lipid modification. The importance of risk stratification is demonstrated, particularly the importance of both LDL and HDL cholesterol either together or as a ratio measure. The selection of modeling methodology to extend benefits of a treatment beyond the end of a trial and over a lifetime is discussed. The relationship between benefits from lipid reduction and risk difference is described, demonstrating that for individuals with established coronary heart disease (CHD) and those older than age 58, benefits from lipid reduction are greater than those predicted from baseline lipid-related risk differences alone. The implications of these data for primary prevention of CHD in the elderly are discussed.