MedisGroups

To the Editor. —  As a physician practicing in Pennsylvania, I read with growing alarm the recent article by Drs Iezzoni and Moskowitz.¹ The authors point out that MedisGroups severity weights are highly influenced by the use of specialized diagnostic technologies. They conclude that this fact might make patients admitted to teaching hospitals, where such procedures are more commonly used, seem more ill than analogous patients admitted to community hospitals.The influence of such technologies also might affect the severity-adjusted mortality rates of patients in nursing homes and reflect on their treating physician. Patients with moderate to severe Alzheimer's disease who are admitted to hospitals are less likely to receive high-technology procedures. For example, such a patient is much less likely to undergo endoscopy for upper gastrointestinal tract bleeding because of surrogate (ie, family member) refusal to give consent, even if the attending physician wishes to use such an