to create separate data warehouses that have been effective in managing large and diverse data sets and in supporting broad needs for information that are of a range and scale comparable to what is projected to be needed by CMS for decision support and business intelligence.³

Another factor that will need to be incorporated into CMS’s transformational efforts is the need for experimentation and pilots. Given the mandates of the Patient Protection and Affordable Care Act of 2010 and the pressing need to “bend the cost curve,” significant experimentation with various approaches to developing an efficient, integrated health care ecosystem will be needed. CMS will play a central role here, and it will need to balance its efforts in this area with the need to keep existing programs running smoothly. In general, incorporating flexibility into system and software architectures in the face of change will be critical, as any system developed now is likely to face even more changes in the future. Learning from other large-scale enterprises that have transitioned to new business models will be critical. The rapid change in the political and policy dialogues means that requirements for future CMS systems are not yet known or stable. And yet system and transformation efforts will have to be undertaken despite that uncertainty. Proceeding in such a way inevitably increases risks with regard to meeting requirements, or risks of outright failure as in the case of past procurements by a number of federal agencies as well as by organizations in the private sector. In short, CMS needs to leverage proven IT approaches and solutions that are most likely to be able to cope with ongoing significant advances in IT generally and with rapid evolution in health care IT and policy.