Conventional randomized controlled trials (RCTs) are rigorous scientific examinations of the tolerability and efficacy of potential new therapies.  RCTs define precise clinical end points, maximize adherence, minimize variance of clinical conditions, and are limited to relatively homogeneous patient populations through strict eligibility criteria and controlled conditions.  During RCTs, patients are managed according to study-defined protocols and typically undergo more intensive monitoring than in routine practice, with more detailed collection and recording of data than is typical in ordinary medical records.1 In addition, randomization in RCTs aims to minimize bias by balancing patient characteristics between treatment groups and thereby maximizes the likelihood that any observed differences result from differences in assigned study treatments.  Although RCTs can provide crucial information on the tolerability and efficacy of a treatment, certain aspects of RCTs limit the applicability of their results to all patients. RCTs typically comprise patient populations that are relatively homogenous and not identical to patient populations outside of RCTs.

The use of large-scale, highly selected patient populations in RCTs creates high statistical power, which can lead to the detection of statistically significant but small differences in patient outcomes.  Patient education and tightly controlled treatment administration are standard in many RCTs a nd may result in higher rates of treatment adherence and, therefore, improved response rates compared with those in routine clinical practice...