The analytical model proposed by Glassman and colleagues [1] is an excellent example of how better definitions can help clarify what is really important to consider in developing decision rules. Although I agree that cost-effectiveness as a basis for clinical choices is questionable, I think that "medical necessity" is an important factor in clinical decision-making processes.

From a public health standpoint, medical necessity justifies the need for such actions as quarantining. It would be hard, for example, to require that children be immunized against most known infectious diseases if there were no medical reason for this immunization. Although it can be argued that herd immunity would not require all children to be immunized in order for a community to derive a benefit from such intervention, having all children immunized would be the right step toward eradicating the disease.

Second, as the authors note, what may be necessary may not be reimbursable by health insurance. This does not mean that the definitions used by health insurers are appropriate bases for defining medical necessity. Furthermore, as long as the courts are called on to define situations in which physicians are held accountable or liable for actions that may be considered negligent, medical necessity will remain a crucial consideration in prioritizing clinical actions.

Finally, it is understandable why medical necessity has lost its urgency as the management of chronic diseases becomes more common. In such cases, clinical outcomes are no longer as straightforward as matters of life and death. Rather, disease is measured by its impact on a patient's quality of life, which is uninsurable.