As a surgeon, when I see patients referred to my clinic with a surgical problem such as a hernia or gallbladder disease, they usually bring their medical records from the referring primary provider. When I peruse the records before seeing the patient, it is more and more common—almost normal—for those records to contain a frighteningly long list of additional diagnoses. I expect to see an extremely fragile and unhealthy patient who might be a poor surgical candidate. Once I get to know the patient I am usually relieved to discover that they are fairly healthy, vigorous specimens.

Technically the diagnoses on such lists are legitimate. But very often the list is not a true reflection of the patient’s health or of any relevance to my clinical decision-making. Health care practitioners have been aware of this phenomenon called “ over-diagnosis” for years. And in hoping to rein in the explosion of spending that inevitably results when patients and doctors spend a third party’s money, the Center for Medicare and Medicaid Services (CMS) has implemented a new payment model that doesn’t help the over-diagnosis problem.

As technological advances make for more sophisticated laboratory and imaging tests, very small medical abnormalities are detected well before they become clinically significant. Meanwhile, many medical specialty panels have, over the years, recommended diagnosing and sometimes treating these abnormalities at earlier stages or lower values.

Yet over-diagnosis can lead to over-treatment. For example, millions of people today carry the diagnosis of diabetes or hypertension, and might be receiving treatment that offers more risk than benefits for their problem at its stage of diagnosis.

That’s where the federal government comes in. Congress passed the Medicare Access and CHIP Reauthorization Act (MACRA) in 2015. As a result, in 2017 CMS started paying individual and group practice providers using a new Merit-based Incentive Payment System (MIPS) under the Quality Payment Program. The Quality Payment Program has a second but similar scheme, called Advanced Payment Model, for Accountable Care Organizations and “Medical Home” models, which are basically patterned after the old closed panel HMOs of the 1980s and 1990s.

In both arrangements, data are collected by CMS on a variety of patient outcome measures through a Physician Quality Reporting System (PQRS) and providers are “scored” on an expected/​observed ratio. Depending upon their score, a provider is either given a bonus or is penalized financially at the end of each monitoring cycle.

CMS uses “Hierarchical Condition Coding” (HCC) to adjust the scores. The number and type of diagnoses attached to a patient are used to predict the expected outcome, complication rate, the degree of difficulty in treating that patient, and in this way affect the providers’ scores. The sicker the patient, the smaller the penalty for a bad outcome, and the greater the reward for a good one. In addition, more diagnoses means the provider can use a higher level “evaluation and management” (E&M) code in billing. Sometimes multiple diagnoses let the provider use multiple E&M codes to generate more charges for a single encounter.

All of this makes for better payment. It also makes for over-diagnosis.

An entire cottage industry has emerged to counsel providers on how to navigate and improve their income under these new rules. Many practices have hired experts in coding and billing to advise and guide practitioners when they enter information and diagnoses into the electronic record, and we doctors exchange charting and coding tips with one another in hospital doctors’ lounges.

If a bunch of boulders are placed in the middle of a stream, the water will still find its way—often around and between the rocks—downstream to its final destination. The same principle applies when regulating human behavior. Incentives drive behavior. Regulatory obstacles can make people adjust their course, but they still try to reach their ultimate destination. Medicare’s Quality Payment Program is just another cluster of boulders.