The evidence

Choudhry and colleagues published a systematic review on the relationship between clinical experience and the quality of health care. Despite significant heterogeneity of the studies reviewed, over half reported decreasing performance with years in practice for all outcomes assessed. The rest of the studies found decreasing performance for some outcomes but not others, and some found no association. The authors concluded that physicians who have been in practice the longest may be at risk for providing lower quality care. This publication provoked a flurry of letters to the editor, mostly negative.

Blasier in a recent symposium on the problem of the aging surgeon concluded that age causes deterioration in physical and cognitive performance. He pointed out that aviation has a statutorily mandated retirement age of 65 years and the United Kingdom has mandated retirement of surgeons from the Public Health Service at 65 years and from private practice at 70 years. There has been no progress in a similar direction in the United States as not all studies are consistent. Drag and colleagues showed that 78% of surgeons aged 60 to 64 performed within the range of younger surgeons on computerized cognitive testing; 38% of surgeons aged 70 and older compared favorably with the younger surgeons. The authors concluded that older age does not inevitably preclude cognitive proficiency. Several reports from the University of Michigan including a 2007 publication from Waljee and Greenfield concluded that age alone was insufficient as an indicator of performance among surgeons. Instead, volume of cases, complexity of cases and specialty training have a significant impact. Even so, this group of surgeons and psychiatrists in their most recent 2010 publication concluded that the observed variability in performance “suggests the need for formal measures of objective cognitive functioning” to inform the retirement decisions of surgeons.

Orthopedic surgeons have also advanced an opinion regarding clinical competence and aging. Green has made a comparison to driving an automobile, although testing actual surgical skill is more complex. He does cite the AMA’s Ethical Guidelines in stating that when “failing physical or mental health reaches the point of interfering with a physician’s ability to engage safely in professional activities, the physician is said to be impaired.” Methods to objectively measure that physical or mental health is where the problem lies. Hummer concludes that the American Academy of Orthopaedic Surgeons should simply embrace some objective assessment of technical skill to help surgeons define when aging has eroded their performance sufficiently to cease operative patient care. The Neurological Surgeons Consensus Conference on transition to retirement also heard data on the physiologic effects of aging and their variable effect on surgical performance. The conclusions included a preference by participants on using local hospital credentialing processes and clinical proctoring rather than government-mandated age regulations in assuring physician competence.

Katz has reviewed issues of aging in the practice of anesthesia. He reported that the average age of retirement among American anesthesiologists was 64.1 years, with a range of 49 to 77 years, although this was more than a decade ago. A number of the physiologic effects of aging impact anesthesia, including changes in vision and hearing. There is 1 report that 40% of anesthesiologists older than 65 were unable to detect 1 or more of the standard equipment alarms. Ability to recover from night time practice during scheduled cases the next day also deteriorates with age, although this is not unique to the provision of anesthesia care. This fact impacts emergency department physicians who spend a great part of their careers working at night. In an interview for Emergency Medicine News, Henry and Goldberg acknowledged the impact that aging has on the emergency department physician but suggested a number of strategies to compensate. They opined at the time of the interview that there was little objective data in their specialty on the impact of aging on the provision of quality care.

There is even less information for the obstetrician and gynecologist. Purdon et al commented in 2002 that most physicians in our specialty do not plan for career transitions. Although they acknowledge the impact of aging in physician impairment, there is no cited data among obstetrician gynecologists; rather, they advise us to enter retirement in steps and stages. Of interest, they quote Thomas Percival (1792) who advised physicians to consider retirement “when an unconscious decay in faculties first experienced as the wonted confidence of their peers” was apparent.

So why is there so much variation in the published clinical data of the impact of physician aging on performance? Some insight may be gained from the study of aging and cognitive performance in general. Information processing theory suggests that there are 2 modes of cognition: analytic, or the active collection and analysis of data to make decisions, and nonanalytic, or experiential, wherein decisions are based on prior experiences. In the latter mode, long-term memory in which elaborate schemas are stored, allow the aging physician to compensate for declines in rapid analytic ability until the interaction of both modes is overwhelmed by the required “cognitive load.” Deliberate practice over a period of years is required to build these schemas and expertise, and deliberate practice can continue contributing to nonanalytic cognition. Studies of experts in fields other than medicine suggest that rigorous deliberate practice can maintain expertise well into the seventh and eighth decades. Accordingly, although mental efficiency (working memory, processing speed for analytic cognition) may begin to decline after age 40, an individual may apply domain-specific and experienced-based knowledge to perform well in problem-solving throughout the life span. Although issues related to physical stamina and other technical abilities (hearing, vision, eye-hand coordination, reaction time) as may be required to perform in the surgical and intervention specialties are not addressed by information processing theory, this type of insight may explain the significant variation observed in the descriptive clinical studies of aging physician performance described previously. At the end of the day, however, Durning and colleagues conclude that screening for cognitive deficits in older physicians may be warranted, and that such formal scrutiny is clearly where the profession is heading. They go on to state that development of objective and valid tests would be of great use, as they would permit capable aging physicians to continue to practice at a high level. Eva in his review has reached similar conclusions, although also emphasizing that individual differences and variability in performance exist among aging physicians, undermining any prescriptive mandatory retirement policies. Further, he points out that declining analytic ability does not necessarily lead to poor performance; indeed, he cites studies in which nonanalytic approaches to problem solving can be superior to analytic approaches. He concludes with a recommendation for external support (ie, attempt to decrease cognitive load and other distractions), deliberate practice, and education and testing (including peer review).