Last week I attended a conference on health policy at the University of Chicago, where I moderated a panel that examined implementation of the Affordable Care Act. For much of our time, the panel focused on Accountable Care Organizations. Panelists and attendees wondered whether ACOs would meet the same fate as Integrated Delivery Systems of the 1990s. Some in the audience mentioned that when it comes to integration, electronic medical records could be a game changer. EMRs could be used to monitor and reward cost saving decision making, for example. But most ACOs are still figuring out how to use EMRs for clinical decision making; their use in helping managerial decision making remains far off.
As more and more speakers expressed skepticism about the future of ACOs, a physician in the audience offered a truly fresh perspective, one that makes me feel much more optimistic. I never learned this physician’s name, so I will call him Dr. Yes. Before I summarize Dr. Yes’ argument, it is helpful to turn back the clock to the late 1990s, when IDSs were taking the health industry by storm. Perhaps the defining feature of IDSs in the 1990s was the integration of hospitals and primary care physician practices. This strategy failed in large part due to classic agency problems. In a nutshell, an agency relationship can fail because of incentive problems (the principal is unable to effectively motivate the agent) or selection problems (the principal employs the wrong type of agent.) IDSs suffered both. When hospitals acquired physician practices, they converted entrepreneurs into employees who resisted any kind of incentive payments. As employees, primary care physicians did not work as hard or show as much commitment to their practices. Moreover, those physicians most eager to give up their autonomy were those looking to dial down their practices and lead the “quiet life.” In these ways, IDSs experienced both incentive and selection problems, with devastating results.