The Value of Coercion: How to Move Health IT Forward

Sometimes in life it’s just necessary to put one’s foot down. This lesson, which most of us have learned as teachers, managers, parents, and (sadly) the adult children of our own parents, applies sometimes to innovation in health IT.

The lesson came home to us this month in shadow of disaster. The COVID-19 pandemic forced the government and health care providers to move from office visits to telehealth. This should have been done long ago, of course, not just to save patients from exorbitant parking fees but to save them from the contagious diseases borne by other patients.

I’m sure that after this crisis has lifted, patients will insist on continuing to use telehealth where appropriate. And we’ll learn over the coming months where telehealth works and where it doesn’t. In mental health, where practitioners insisted for decades that remote visits would be less effective than face-to-face ones, we’re finding that telehealth is often beneficial, perhaps due to changing norms in the surrounding society that makes online interaction feel more comfortable.

Could we coerce providers into making other beneficial changes? The government tries to do this all the time, and it certainly fails sometimes. One of the most disappointing findings in recent years has indicated failure of the Medicare 30-day readmission rule. I don’t know who thought up the magic number 30 or whether their concept was tested before rolling it out nationwide. But the article just cited, along with other evidence, reveal flaws in CMS’s assumptions.

For instance, a lot of hospital patients have comorbidities, so you might cure their heart problem but see them back again soon for a kidney problem. In addition, many patients have trouble following through on their treatment plans. Hospitals jump in to offer counseling and follow-up calls, but these aren’t always enough to overcome the difficulties faced by the patients, who may be isolated, homeless, or mentally disabled.

There are other reasons to pause before applying coercion. Take the pesky problem of information hoarding. I think this should have been solved ten years ago, and could have bypassed that expensive rhinoceros institution, the health information exchange (HIE), which got in the way of effective solutions for years. The Office of the National Coordinator (ONC) has gradually upped its requirements and incentives for data exchange, with considerable success, but again the COVID-19 pandemic reveals where we fall short.

My impression is that data exchange is lagging not because coercion would fail, but because the goal is defined incorrectly. Giving data to your doctor is useful only if the doctor uses it. Data exchange clearly reduces costs in specific situations, such as when it saves the patient from retaking a test. But doctors in general don’t have time to integrate large quantities of data about a patient and draw conclusions from patient histories.

So we need to provide a concrete use for data exchange in addition to the technical means to do so. Big data analytics may be able to derive meaning from data to support medical decisions. And this kind of evidence-based clinical decision support might be implemented by combining data sets from many providers, perhaps without the need for actual data transfers. Analytics could similarly be the key to unlocking the value in a lot of the measures required by CMS–measures that take up doctors’ time and annoy them to point where they consider quitting.

We thus end with the key insight that has to drive every technical innovation: an effective use stems from evaluating the whole value chain. In this case, that means ensuring a healthier patient (and other aspects of the well-known Triple Aim).

Our world’s health care system is starting to suffer from the biggest shock in its history. The casualties will include not only victims of COVID-19 but others whose medical conditions go unaddressed (telehealth notwithstanding). The government is taking the cautious route of reducing rather than increasing its demands on desperate health care systems right now. But we are learning that coercion is powerful. Without proper attention to consequences it can misfire, but when directed toward an unarguably necessary goal it can do wonders.

About the author

Andy Oram

Andy Oram

Andy Oram writes and edits documents about many aspects of computing, ranging in size from blog postings to full-length books. Topics cover a wide range of computer technologies: data science and machine learning, programming languages, Web performance, Internet of Things, databases, free and open source software, and more. My editorial output at O'Reilly Media included the first books ever published commercially in the United States on Linux, the 2001 title Peer-to-Peer (frequently cited in connection with those technologies), and the 2007 title Beautiful Code. He is a regular correspondent on health IT and health policy for He also contributes to other publications about policy issues related to the Internet and about trends affecting technical innovation and its effects on society. Print publications where his work has appeared include The Economist, Communications of the ACM, Copyright World, the Journal of Information Technology & Politics, Vanguardia Dossier, and Internet Law and Business.