Mandatory Testing Shows Cognitive Deficits in Late-Career Clinicians

About 13% show problems likely to impair their ability to practice medicine.

This article was first published on Tuesday, January 14, 2020 in MedPage Today.

By Judy George, Senior Staff Writer, MedPage Today.

Mandatory testing at one of the nation's top hospitals showed that nearly one in eight clinicians age 70 or older had cognitive deficits that were likely to impair their ability to practice medicine independently.

While 57% of 141 older physicians and practitioners who applied for renewal of hospital privileges at Yale New Haven Hospital demonstrated no cause for concern, the remainder faced yearly re-credentialing or further testing with possible outcomes that included proctored medical practice, resignation, or retirement, reported Leo Cooney, MD, a geriatric medicine professor at Yale Medical School, and Thomas Balcezak, MD, Yale New Haven's chief medical officer, in JAMA.

This may be the first paper to report the results of cognitive testing of late-career clinicians, noted Cooney.

"We found that 12.8% of clinicians who applied for renewal of hospital privileges had cognitive deficits which gave us great concern about their ability to practice medicine independently," he said. "None of these 18 clinicians had been previously brought to the attention of hospital authorities because of concern about their practice abilities," Cooney told MedPage Today.

"When we started this process, I thought it was a good idea, but I wasn't sure about its necessity," he continued. "I now believe that it is essential to review older clinicians. While older clinicians can bring a great deal of experience and expertise to the practice of medicine, we must be sure that they have the cognitive ability to solve problems and make appropriate judgments."

Every year, 20,000 more U.S. physicians turn 65, and, even though half retire by then, many continue practicing for years more, noted Jeffrey Saver, MD, of the University of California Los Angeles, in an accompanying editorial.

"U.S. policy makers are counting on these older physicians to do so to help mitigate the nation's growing physician shortage," he wrote. "Currently, an estimated 50 million to 70 million U.S. office visits and 11 million to 20 million hospitalizations each year are overseen by physicians older than 65 years."

Age-based testing of medical professionals is a growing and controversial trend: Scripps Health Care, Intermountain Healthcare, Stanford Hospitals and Clinics, and Penn Medicine are among hospital systems that have implemented cognitive screens for older practitioners.

"Information about the outcomes of these programs is difficult to find, but it is clear that these processes have been challenging," wrote Katrina Armstrong, MD, MS, and Eileen Reynolds, MD, both of Massachusetts General Hospital in Boston, in another editorial accompanying the paper. Stanford physicians rejected plans to use MicroCog, a 1-hour computerized test that assesses five cognitive domains, opting instead for a revised approach with rigorous peer review. And opposition by the Utah Medical Association to MicroCog led to a state law banning its use.

At Yale New Haven, all applicants for reappointment to the medical staff age 70 or older were required to have an objective evaluation of cognitive function as part of the 2-year reappointment process. A neuropsychologist developed a screening battery which was (and still is) kept confidential. The hospital medical executive committee decided that a pass/fail approach would not be used, but performance tests would be considered in the context of clinical privileges.

The battery consisted of 16 brief tests that encompassed information processing, visual scanning and psychomotor efficiency, processing speed and accuracy, working memory, concentration, verbal fluency, and executive function. Time to completion varied from 50 to 90 minutes. Applicants' ages ranged from 69 to 82 and averaged 74. Most (86%) were men and 89% were physicians.

Global cognitive status scores and domain-specific scores were compared with peer, population, and age-group norms. A medical staff review committee that included the previous and current chief medical officer of the hospital, a faculty geriatrician, and the neuropsychologist performing the exams reviewed test results and made recommendations to the medical staff's credentialing committee.

Of 141 applicants, 81 (57.4%) completed the testing requirements and continued the credentialing process; they would be retested in 2 years as part of the regular reappointment process. About 24% proceeded with the credentialing process but, because of minor abnormalities in test results, were scheduled for rescreening in 1 year. Several had limitations in specific domains and were re-tested; three of these applicants were determined to have significant problems and either retired or resigned from the active medical staff.

Eighteen clinicians (12.8% of the 141 tested) demonstrated cognitive deficits that were likely to impair their ability to practice medicine independently. None had been brought to the attention of medical staff leadership due to performance problems. All opted voluntarily to discontinue their practice or move to a closely proctored setting.

Screening programs are always challenging to implement, Armstrong and Reynolds observed. "Because the prevalence of the condition is generally low in the population being screened, false-positive screening results are common unless the test has near-perfect specificity," they wrote. Full assessment of clinical competence, even for medical students and residents, is inexact and evidence that increasing clinician age is associated with worse patient outcomes is "weak at best," they noted. Moreover, using age cutoffs for screening raises concerns about age discrimination, they added.

A policy like the one at Yale New Haven Hospital has several limitations, Saver pointed out. Because of confidentiality, impaired test performance cannot be correlated with poor medical practice, and independent individuals outside the institution were not involved in decision-making. In addition, the confidential nature of the test battery limited the ability to assess its validity.

Researchers and editorialists reported no conflicts of interest.