The hypothesis that bundle synergy exists has not been formally tested. Observational studies examining outcomes before and after bundle implementation are not appropriate proof-of-concept demonstrations or substitutes for prospective randomized trials. Furthermore, with multiple therapeutic interactions across heterogeneous patient populations, conclusions about the safety and efficacy of specific bundled interventions are not readily tenable. Consider, for example, the failure to recognize a lack of efficacy and possible risks of Xigris (activated protein C; Eli Lilly and Company) use in patients with sepsis. Relatively recent withdrawal of Xigris by the US Food and Drug Administration occurred despite years of its inclusion in sepsis bundles with little hint of the rather dramatic lack of efficacy during observational follow-up. Additionally, an intervention that produces a positive effect in a particular group of patients cannot be extrapolated to another group or to all patients with a similar condition. Contrary to expectation, bundles might dilute rather than enhance the benefits of specific treatment elements when combined together. Most alarming is the concept of all-or-none bundle compliance. CMS, IHI, and other quality organizations suggest that if all the elements of the bundle are not met, no credit should be given for any of the elements. In other words, credit for delivery is all or none. There is no scientific data to support this notion; indeed, the before-and-after studies that investigated the 6-h sepsis bundle strongly contradict this idea. Nolan and Berwick’s4 assertion that “the movement to all-or-none performance assessment is an important milestone in the journey to high quality health care” may not translate when high-quality clinical evidence is being packaged with other interventions that are unproven or harmful. Furthermore, bundles are, in essence, consensus packages that are not continually updated as evidence changes. Indeed, the 6-h sepsis bundle and the VAP prevention bundle have not been updated by the IHI since originally published in the mid 2000s. Thus, the pressure to comply with bundles may accelerate the very situation that bundles are trying to correct: outdated, potentially harmful care.