Medical Necessity: Unexplained Clinical Variation in Care

I am a physician who writes and edits guidelines designed to assist in determining appropriate utilization of clinical resources. In a nutshell, the issue that pertains to today’s topic, medical necessity, is “unexplained clinical variation in care.”

What I mean by that is the fact that patients with similar clinical features and issues are treated very differently depending on variables unrelated to severity of illness, risk of deterioration, or clinical need.

What varies is the amount or intensity of care (testing, inpatient hospital care, procedures, etc.). This variation is called “unexplained,” as it persists even after taking into account items such as patient age, socioeconomic status, and illness details.

Importantly, a consistent finding is that this variation in the intensity of care is not associated with improved patient outcomes. That is, despite general belief to the contrary, more care is not necessarily better care.

This sort of variation has been identified across all manner of clinical entities, treatments, and variables. Care provided varies rather profoundly, for example, by geographic region in the U.S.. In general, more care, and a higher intensity of care, is rendered in the Northeast than in the West, for example.

Furthermore, this variation can be found within geographic regions, according to physician specialty and practice location, and variation exists even within individual groups of physicians. For example, unexplained variation exists between hospitals in the same or similar settings, and between individual clinicians within a single hospital or practice. This variation is not random, in that the same geographic areas, specialties, and individual doctors are found to provide more resource-intensive care than their counterparts.

What sort of variation do I mean, and how does this relate to medical necessity?

Important aspects of care to measure include those decisions and interventions that carry high cost and potential for risk of harm. An early measure was inpatient length of stay. A more recent measure has been the inpatient admission rate. For example, patients seen in the emergency department for the same reason and with similar clinical features are admitted to the hospital for inpatient care at rates that can vary significantly.

The cost ramifications of the admission decision are straightforward. Less appreciated is the consistent finding that being a patient in a hospital is quite risky, and therefore should only be considered when the benefit (that is to say, need) clearly outweighs the risk of harm. Study after study has found that somewhere in the neighborhood of 4 percent of hospitalized patients experience a preventable harm (for example wrong medication, wrong dose, hospital-acquired infection, etc.).

With this background, the importance of medical necessity becomes clear. Simply leaving it up to individual clinicians has resulted in the variation seen. At the same time, it is in no way a simple matter to standardize which patients need which type or amount of care.

For unexplained clinical variation, an implemented response is the expectation that clinical decisions and interventions (or at the least, payment for these interventions) be justifiable, that is, defendable according to some mutually accepted standard (in other words, documentation of medical necessity).

Various clinical tools, such as the MCG evidence-based guidelines, have been used by involved parties (for example, payors and auditors) to assist in the determination of when the clinical documentation supports a defined threshold of “medical necessity.”

It is crucial that whatever standards are applied, they be clinically “right,” that is, neither overly strict nor lenient, and seen as unbiased by all parties involved. An important means by which to achieve this standard and level of acceptance is to be strictly evidence-based. This entails the difficult process of searching for the best evidence, expertly interpreting the evidence, and incorporating new evidence when appropriate.

Correct usage of guidelines is likewise important. For example, the MCG guidelines are intended to supplement and support clinician-based decision-making, not replace it. They are designed to be used as guidance, not interpreted as inflexible rules. Our guidelines are very specific and detailed when the medical literature allows, and at the same time acknowledging of the “gray areas” of decision-making when the evidence is not as clear.

In either case, the guideline content is used to not only set a standard for how to determine severity of illness or need for a procedure, but also to provide a common set of key moving parts within any given clinical situation that should be documented and described.

It is through this consistent, appropriate use of evidence-based guidelines that the central, chronic issue of unexplained clinical variation can be recognized and addressed. Identification, determination, and documentation of medical necessity are the active ingredients in any attempt to reduce unexplained clinical variation in care.