Reputation

I liked economics in grad school because it explains our ‘rationale’ behavior. From an economic viewpoint healthcare is a transaction between principals (patients) who have hired agents (that would be physicians) to provide services (a healthcare outcome). The agent is hired for their greater knowledge to provide a service to the principal. Asymmetry of information is a basis for the principal agent relationship. The information physicians and patients have is different and influences how they transact ‘healthcare’. Informed consent is a good example of a ‘transaction’ where the parties have very asymmetric information. In fact, one goal of an informed consent is to reduce that disparity.

Asymmetric information results in several risks. For the more informed there is a moral hazard - the tendency to exploit its better information in a dishonest or undesirable way. For the less informed there is adverse selection - the tendency for bad products or services to be chosen. Decreasing the asymmetry of information reduces both of these risks.

Reputation is a traditional means of reducing information disparity especially with respect to moral hazard. Reputable agents (physicians) can command higher prices (at least in the unregulated marketplace) so cheating has a cost when it impairs your reputation. Reputational systems have evolved along with information technology. When there was little ability to distribute information, licensure and board certification were reputational measures; agents remarked upon agents. Patients were limited to commenting by word of mouth to their immediate social network. Increasingly information technology, especially the marketplace called the Internet, impacts reputation. Consider the recommendation systems of Amazon or Angie’s where users rate products or services. Amazon and Angie’s list broadcasts principal’s remarks about agents.

Interestingly, the newer recommendation systems are bidirectional. Uber allows riders to comment upon drivers and drivers to comment upon riders. Is there such a system in healthcare’s adjacent possibility? Two immediate possibilities spring to mind. First, the rise of wearable sensors, e.g. Fitbits or the new Apple watch, afford us the ability to know our patient’s habits and behavior more reliably. Second, access to meaningful electronic records would afford us a better understanding of our patient’s health history and interactions with other ‘agents.’

There are also unintended consequences because reputational mechanisms have a long shelf life. “If there are no second chances then second thoughts become even more important.” The agent concerned about their reputation will take fewer risks, we think twice before acting. As a result, there is increasing incentive, on the part of physicians, to avoid caring for high-risk patients. Alternatively, the increasingly reliable information about patient’s and their lifestyles afforded by wearable sensors will allow us to more carefully identify risk and choose which patients we care for and in what ways.

Lastly, information asymmetry creates pockets of ‘informational monopoly” and brokers arise to take advantage of that difference. Information brokers are now on the rise in healthcare. Hospital ranking can be found in US News and World Report, physician rankings at Health Grades. Reputation repair and consistency can be found through Reputation.com or Doctors.com. In this light, HCAHPS, as flawed as it may be, is a tentative solution to the problem of reputation. So are Hospital and Physician Compare.

Adam Smith puts the issue succinctly:

"The real and effectual discipline which is exercised over a workman is that of his customers. It is the fear of losing their employment which restrains his frauds and corrects his negligence."

 Information technologies make his thought increasingly true for both physicians-agents and patient-principals.