Campbell's Law And Medicine

Another example of Campbell's Law: "the more any quantitative indicator is used for decision-making, the more it will be apt to distort and corrupt the processes it was intended to describe", from Reason

As goes Medicare, so goes the private insurance industry. Insurers, starting in the late 1980s, began the practice of using the Medicare fee schedule to serve as the basis for negotiation of compensation with the doctors and hospitals on their preferred provider lists. An insurance company might offer a hospital 130 percent of Medicare’s reimbursement for a specific procedure code, for instance.

The coding system was supposed to improve the accuracy of adjudicating claims submitted by doctors and hospitals to Medicare, and later to non-Medicare insurance companies. Instead, it gave doctors and hospitals an incentive to find ways of describing procedures and services with the cluster of codes that would yield the biggest payment. Sometimes this required the assistance of consulting firms. A cottage industry of fee-maximizing advisors and seminars bloomed.

I recall more than one occasion when I discovered at such a seminar that I was “undercoding” for procedures I routinely perform; a small tweak meant a bigger check for me. That fact encouraged me to keep one eye on the codes at all times, leaving less attention for my patients. Today, most doctors in private practice employ coding specialists, a relatively new occupation, to oversee their billing departments.

Another goal of the coding system was to provide Medicare, regulatory agencies, research organizations, and insurance companies with a standardized method of collecting epidemiological data—the information medical professionals use to track ailments across different regions and populations. However, the developers of the coding system did not anticipate the unintended consequence of linking the laudable goal of epidemiologic data mining with a system of financial reward.

This coding system leads inevitably to distortions in epidemiological data. Because doctors are required to come up with a diagnostic code on each bill submitted in order to get paid, they pick the code that comes closest to describing the patient’s problem while yielding maximum remuneration. The same process plays out when it comes to submitting procedure codes on bills. As a result, the accuracy of the data collected since the advent of compensation coding is suspect.