An article in the Annals of Internal Medicine looks at the issue of physician cost profiling. (Annals Article) This is an important subject because many private payers utilize such profiling for physician feedback on appropriate use of health care services or for incentive programs and Medicare has considered a profiling initiative. The underlying assumption is that a particular patient’s overall health care use and spending can be attributed to a particular physician. That assumption is questionable in a system such as ours where a patient usually can and does freely go to a multiplicity of physicians.
The authors of the Annals article used data from Massachusetts commercial health plans to test the variability in attribution that would result from using different rules to decide which physician should be assigned responsibility for a particular patient. Twelve rules were tested, utilizing visits, costs, episodes and majority or plurality of each for attribution. As might be expected, there was a significant difference in the number of patients and the amount of cost assigned to a physician. These differences could result in a physician being referred to as high-cost versus average cost or low cost.
Unfortunately, it is not clear that any of these rules really capture the essence of the concern: which doctor is actually responsible for the patient’s care and consequently the spending on that patient. If there is not a high level of confidence in the attribution of cost and subsequent classification of the physician’s cost profile, use of the results for behavior-changing programs will be undermined.