Opinion editor's note: Star Tribune Opinion publishes a mix of national and local commentaries online and in print each day. To contribute, click here.

•••

My 97-year-old blind mother slipped into dementia last year and now requires full-time assistance in a memory care unit. Her long-term care insurance carrier insists on three-month care updates for continued payment, even though she will never improve. I empathize with Kim and Kinsley Munson, the examples in Sunday's editorial on prior authorization ("Don't let red tape burden patients," March 31).

Despite my personal experience, I am opposed to legislation that exempts PA for categories of medical care. The Legislature does not have the expertise to make medical management decisions. Too often lawmakers write laws based on anecdotes that result in disastrous unintended consequences.

For example, the proposed law exempts cancer therapy from PA on the assumption any delay will be deadly. The wrong therapy can be even more deadly.

From 2005 to 2018 I managed the UnitedHealthcare oncology department. My team noticed a progressive surge in the usage of erythropoietin, a drug that prevented the need for blood transfusions in chemotherapy patients. In the era of transfusion-related HIV infections, this medication was a lifesaver, but HIV testing was commonplace by 2005, so manufacturers promoted the therapy to alleviate fatigue. The drug had a dark side. When it was administered to patients who already had reasonable blood levels it could cause strokes and heart attacks. It also accelerated cancer recurrences in several studies. The FDA issued a black box warning in 2007 stating that no one with a hemoglobin level above 12 (a measure of anemia) should receive the drug.

Our research found that in 300 consecutive patients receiving erythropoietin more than one-third were being treated incorrectly. A PA was started immediately. The Medicare program copied our program six months later because of the safety issues.

In another example, UnitedHealthcare policy paid for any chemotherapy drug that was recommended by the National Comprehensive Cancer Network (NCCN), a coalition of the best cancer centers in the United States. The company did not use PA, but it did review the claims after treatment to ensure that they matched the NCCN guidance. For four successive years, approximately 7% of those treatments were not compliant. Those patients were getting the wrong drugs — losing precious time and an opportunity for a response. It is important to note that these errors occurred in almost every oncology office in the nation; they were not caused by a few bad actors.

In collaboration with the NCCN, the insurer built a digital version of the guidelines. It was used as a required PA and granted instant approval if the requested drugs matched NCCN guidelines. Error rates dropped to zero.

PA prevented errors for thousands of oncology patients. The erythropoietin program also saved millions of otherwise wasted dollars. The current exclusion for cancer proposed in the bill would have blocked these two programs.

The legislative process is not designed to make medical decisions. More than two decades ago, the Minnesota Legislature mandated coverage for bone marrow transplant treatment of breast cancer based on an unconfirmed clinical trial demonstrating miraculous results. Insurers had refused to pay for the procedure until a confirmatory study was completed. Several years later, the National Cancer Institute proved that the transplant procedure shortened survival compared to standard therapy, but coverage for this harmful procedure is still mandated in Minnesota.

I believe PA is overutilized for many of the reasons stated in the recent editorial, but solutions based on anecdotes rather than data will fail. Patients would be better served by government creating collaborations with payers, physicians and the health department to identify the best topics for PA. The group could begin focusing on the few procedures with high error rates, measuring the actual impact of the PA, and then deciding whether the PA should continue or be shuttered. Exempting physicians from PA won't correct the underlying errors or overutilization. Similarly, there is no justification for unnecessary PA — it is expensive and time consuming for patients, physicians and payers. Payers should produce the data to justify each review. PA is a tool to be used sparingly for good purposes that are identified by data rather than legislators or payers informed by anecdotes.

Dr. Lee Newcomer, of Wayzata, is a retired physician. The opinions expressed in this article are not intended to represent any UnitedHealth Group policy or opinion.