Skip to content

PCMS Member Corner: Metrics


There is no perfect way to pay for the work a physician does. Fee for service payments can lead to increased utilization and a flat salary can lead to poor productivity. The search for better payment strategies has led insurers to design complicated systems for physician reimbursement with unclear benefits and an explosion in the administrative burden.

This has led to the adoption of metrics concocted by administrators who feel the metrics are essential to the business of medicine. Some of those metrics are clearly important (like ER visits or hospitalization rate if they are correctly risk-adjusted), many are neutral, but unfortunately some adopted metrics are actually harmful.

One of my daily frustrations is dealing with the risk-adjustment of patients. Medicare relies on the “Risk Adjustment Score” (RAF), and it would be hard to design a worse scoring system. A computer program assesses what diagnoses are associated with increased healthcare spending and if a patient has one of those diagnoses, the patient is expected to incur greater costs, and consequently the system allocates more money.

Clearly, though, no one with any meaningful medical experience has gone through the list to make sense of it and validate it. Many diagnoses that would not be expected to increase health care costs are considered “significant” (e.g. calcified pulmonary nodule, senile purpura), while many that do increase risk are not included (too many to list). For example, one of my patients who was missing an arm from an old farming accident did not receive “credit” for the “loss of an arm” diagnosis, but “loss of a hand” did increase his RAF score. I had to chart the “loss of hand” diagnosis and addend it with a note clarifying he is missing the entire arm, not a good use of anyone’s time.

The way to game the system is to make people look sicker by including more diagnoses that increase the RAF score, which obviously happens. However, when an administrator looks at the (flawed) data and sees that people with higher RAF scores now have reduced hospitalization rates, they congratulate themselves, thinking they are making a difference, when the results reflect manipulation of a poorly devised system. The system also reduces the time and attention physicians could have used to focus more on the care of their patients.

The use of patient satisfaction scores as part of physician reimbursement has always been a bad fit. Everyone knows the best way to increase patient satisfaction is to conform with the patient’s expectations even if not medically indicated, such as prescribing antibiotics for acute viral infections, benzodiazepines for anxiety, and narcotics for pain. A study published in the Archives of Internal Medicine in 2012 called “The Cost of Satisfaction” looked at how patient satisfaction in about 52,000 people correlated with outcomes. The conclusion was that “higher patient satisfaction was associated with less emergency department use but with greater inpatient use, higher overall health care and prescription drug expenditures, and increased mortality.”

Specifically, it was associated with a 26% increase in mortality. There was another larger study that looked at the same concern in 2019 and found that the increase in mortality was only significant in female patients.

We need more health care administrators to work with physicians to align the financial incentives of the business of medicine in ways that improve patient care, not undermine it. This will only happen if we present a united, informed voice and continue to advocate for our patients and for each other. Thank you for supporting the Pima County Medical Society.

Written by Dr. Roy Loewenstein, PCMS Board President
July 2023


Back To Top