Several inconsistencies are inherent in the paradigm proposed in the position paper on the role of the future general internist [1]. Item 5 indicates that a general internist should be an expert equally effective in the office and the hospital. This seems an unobtainable goal. If managed care is to be the way we practice in the future, internists can expect to spend little or no time in the hospital practicing medicine. In fact, in many managed plans, an inpatient physician manager is now assigned to manage all inpatient rounds and orders; the attending internist makes social rounds only if he or she so desires. It is irrational to expect that expertise in the hospital setting would be maintained with such an arrangement.

Moreover, in managed care settings, the evaluation and management functions of general internists are only the foundation of their credentialing and their value to managed care organization. To perform adequately as a primary care physician, the general internist must also have substantial procedural skills, particularly in the areas of dermatology, orthopedics, gastroenterology, otorhinolaryngology, and office gynecology. Traditional internal medicine training has not stressed these procedure-based skills and needs to be retooled. Relying on the "learn as you earn" educational model is not satisfactory in a risk environment.

The paradigm that the American College of Physicians has created and promotes must be more aware of the need for more procedural skills for primary care internists. The paradigm must also be a policy sufficiently fluid to adjust to changes in reimbursement schemes and local medical standards of care.

Society has spoken, if only through the insurance companies, and the traditionally taught skills of the general internist must improve and evolve rapidly if internal medicine is to remain a viable career alternative for trainees.