Physicians and Retirement

In 1905, Sir William Osler famously talked about the “comparative uselessness” of men [if there had been more than a few women in medicine at the time, he surely would have included them] older than 40 years and recommended retirement at age 60. ¹ Dr Osler was 55 at the time. It has been established convincingly that although crystallized intelligence remains fairly stable as we age, a decline in fluid intelligence begins during middle age. ² However, the timing and degree of cognitive decline vary greatly among people. Unfortunately, there are no good metrics to provide evidence-based guidance on when physicians should retire for cognitive reasons, although an approach to developing such information has been proposed. ³ It has been established that there is no relationship between an older physician’s self-assessment and objective measures of cognitive function. ⁴ It also has been established that older physicians tend to rely on first clinical impressions based on their experience, rather than on extensive testing; ⁵ but this tendency contributes to incomplete histories and data acquisition, faulty interpretation, deficient hypotheses and diagnostic error, ⁶ and jeopardizes patient care.

Retirement is important to all of us. Some of us are approaching “that age.” Other doctors practice with “senior” associates. So, issues of aging affect everyone practicing medicine, and they may affect our patients if aging issues are not handled well. That does not just mean ending the practice of older physicians before they start to make medical errors. It also means preparing transition strategies for physicians, practices/departments, and health systems. This issue is going to become more pressing. In 1950, 8% of the population of the United States was age 65 or older. In 2000, it was 12%; and by 2050, the number is expected to reach 20%. Life expectancy for males born now is 78 years (1/20 will live past 100); for females, it is 81.2 years (with 1/10 living past 100). ⁷

There are many publications on physician retirement and retirement planning, but one of the most helpful is a systematic review by Silver et al published in 2016. ⁸ They evaluated 65 English-language articles published between 1978 and 2015. Thirty-three came from the United States, and 32 had been published elsewhere. Their review crystallized a great deal of interesting information.

For example, in 1997, a 50-year-old was expected to work for 13 more years. In 2009, a 50-year-old was expected to work for 16 more years, until age 66. Physicians report that they expect to retire at age 60, but they actually retire at age 69 (3 years later than the general population). Mean planned retirement age for otolaryngologists is 67 years, ⁹ although current convincing data on otolaryngologists’ actual retirement age are not available. However, data are available for the years 1996 to 2000; and otolaryngologists in the southeastern United States during that time period retired at 63.5 years of age. ¹⁰

Many factors affect retirement in general. These include increasing life span, concerns about the sustainability of social security, and economic market fluctuations. Market fluctuations are especially important for physicians, many of whom have been in private practice and do not have pensions provided by an employer.

Several factors have been identified that influence delay in retirement among physicians. These include flexibility of work hours, intensity of work hours, work satisfaction, other career opportunities (or lack thereof), resource adequacy, sense of intrinsic self-worth, convenience, financial incentives, relationships with coworkers, length of training and late entry into the workforce, attachment to work and related strong work identity, and the 4% rule. The 4% rule says that a person will need approximately 25 times his/her annual expenses in savings/investments in order to retire with a comfortable lifestyle over a 30-year period (anticipating that investments will yield 4% per year). For many physicians, it is difficult to save enough money to afford to retire comfortably using this equation; and the 4% rule can be affected adversely by substantial market downturns, and by longevity greater than 30 years.

There also are factors that influence early retirement among physicians. They include work dissatisfaction, inflexibility, bureaucracy, electronic medical records, burnout, and desire for personal time. Gender is not a major factor. Other issues that figure into retirement timing include cognitive decline, physical decline, dexterity, frailty, and increased error rate (sometimes related to overreliance on first clinical impressions as noted above). Hence, experience can be an asset or a deterrent. If physicians remain thorough, diligent, and energetic, then experience should be beneficial. However, when physicians start relying on impressions at the expense of comprehensive evaluation, then problems arise.

Physicians are a microcosm of the general public. By 2026, it has been estimated that about 20% of physicians (Canadian) will be 65 or older. ¹¹ In the general population 65 and older, 13% have dementia, and 10% to 20% have at least mild cognitive impairment. ^12,13 Studies reviewed by Silver et al suggest that more than a third of physicians with concerns about competency have mild-to-moderate cognitive impairment.

What needs to be taken into account for retirement planning to be successful? The importance of finances is obvious. Sadly, it is not uncommon for physicians to be unable to afford to retire and maintain a lifestyle that they find acceptable. Physical changes also are important. Physical deterioration can impair not only medical practice but also the ability to enjoy an active life following retirement. The impact of the psychosocial dynamic of retiring from medical practice should not be underestimated. Much of the identity of many physicians is based in what they do and with whom they work. If physicians have developed no substitute and leave active practice for empty days, dissatisfaction (or worse) may result. Hence, it is helpful to plan a transition to retirement, and perhaps to make it gradual. Practices, institutions, and other physician-employment organizations can help this process, and they should be motivated to do so. Organizations benefit from allowing flexibility (change in hours worked, number of patients seen, complexity of cases, stopping surgery, etc). Excessive workload can lead to burn out particularly quickly in older physicians. Work barriers should be minimized, and agist stereotypes should be resisted. It is also helpful for organizations to offer retirement education and guidance, including access to financial advice. Organizations also are served well by creation of postretirement opportunities including peer support, teaching, mentoring, administration, and other nonclinical activities. This approach maintains access for younger physicians, patients, and administrators to the experience and wisdom of older physicians. Organizations should avoid mandatory retirement ages, and they also must recognize the limits of cognitive testing administered for the first time to older physicians. ³ Organizations that address these issues are not just being altruistic. They are responsible for many challenges associated with retirement including patient care continuity, maintenance of deserved reputations for expertise, and succession of physicians in the institution/hospital. Planning to manage these issues prospectively is far preferable to managing them emergently when key physicians leave.

Many other challenges accompany retirement including maintenance of a physician’s identity, interactions with a spouse, activities to fill the days (as much as desired and needed), long-term financial security (to the point at which the retired physician does not have to worry about money constantly), and others. Most physicians who retire and say that “it was a mistake” might have had a better experience if planning and support had been more thoughtful, and they might have continued to contribute to medical practice and patient care quality longer on a part-time basis than they did if organizational support, opportunities, and intervention had been utilized. Many physicians who are thriving in “retirement” are happily as busy as they ever were but with more control over their schedules. Many successful physicians are type A people and have high Grit scores! They would be unhappy and feel useless doing nothing, but when they recognize what makes them happy and fill their post-medical days with those activities and friends, then their “retirements” are successful. I have trouble thinking of people involved in so many activities as retired just because they are not seeing patients. Rather, I regard them as having transitioned into the next phase of their productive lives.

Retiring too late creates obvious problems such as an increased number of justified malpractice suits. However, retiring too early can be similarly damaging to society and the physician, although metrics might be less obvious. Everyone in medicine (regardless of our current age) should give considerable thought to this complex issue. It affects us all.