Conducting, Baseball, and Longevity

It seems to be a matter of conventional wisdom that conductors (of musical performances) live longer than most mortals, and that that their above-average longevity something to the fact that the occupation of conducting involves vigorous arm motions. Various writers have looked into the matter of conductors’ longevity, and have come to various conclusions about it. In the late 1970s, for example, a medical doctor named Joseph Atlas published an article on the subject, a news account of which is available here. According to the news story,

Atlas selected a random sampling of 35 major decades symphony leaders and computed their longevity at 73.4 years, compared with 68.5 for the average American male.

Then he arrived at a series of conclusions.

Among them:

– Gratifying, or happy, stress promotes longevity.

– Driving motivation and the sense of fulfillment that comes with world recognition help forestall the ravages of age.

Atlas defines gratifying stress as the opposite of frustrating stress, which is the kind that can lead to coronaries….

…As yet there is no scientific documentation to back him up. His conclusions are hypothetical, he says, “more anecdotal than statistical.”…

…Among the prime examples of longevity, Atlas cites, is Leopold Stokowski, who was active and vital until his death at 95.

“Arturo Toscanini lived an active life to the age of 89, Bruno Walter to 85, Ernest Ansermet to 86, Walter Damrosch to 88, Arthur Fiedler is 84.”

There are problems with Atlas’s analysis, but — at bottom — Atlas is right about the longevity of conductors, and probably right that “gratifying stress” enhances longevity. Whether vigorous arm movement has anything to do with longevity is another matter, to which I will come.

As for the problems with Atlas’s analysis,consider this passage from Robert P. Abelson’s Statistics as Principled Argument (1995):

 The longevity datum on famous orchestral conductors (Atlas, 1978) provides a good example [of a spurious attribution of causality]. With what should the mean age at their deaths, 73.4 years, be compared? With orchestral players? With nonfamous conductors? With the general public?

All of the conductors studied were men, and almost all of them lived in the United States (though born in Europe). The author used the mean life expectancy of males in the U.S. population as the standard of comparison. This was 68.5 years at the time the study was done, so it appears that the conductors enjoyed about a 5-year extension of life and indeed, the author of the study jumped to the conclusion that involvement in the activity of conducting causes longer life. Since the study appeared, others have seized upon it and even elaborated reasons for a causal connection (e.g., as health columnist Brody, 1991, wrote, “it is believed that arm exercise plays a role in the longevity of conductors.”

However, as Carroll (1979) pointed out in a critique of the study, there is a subtle flaw in life-expectancy comparisons: The calculation of average life expectancy includes infant deaths along with those of adults who survive for many years. Because no infant has ever conducted an orchestra, the data from infant mortalities should be excluded from the comparison standard. Well, then, what about teenagers? They also are much too young to take over a major orchestra, so their deaths should also be excluded from the general average. Carroll argued that an appropriate cutoff age for the comparison group is at least 32 years old, an estimate of the average age of appointment to a first orchestral conducting post. The mean life expectancy among U.S. males who have already reached the age of 32 is 72.0 years, so the relative advantage, if any, of being in the famous conductor category is much smaller than suggested by the previous, flawed comparison. (p.4, quoted here)

But the comparison is not as flawed as Abelson makes it out to be. Consider the excerpts of an talk given in 2005 by Jeremiah A. Barondess, M.D., then president of the New York Academy of Medicine:

I have had more than a glancing interest in this subject [longevity] for a long time. I was first attracted to it many years ago when I came across a squib in the newspaper to the effect that Leopold Stokowski, then about 90 years old, had been the subject of a complaint to the authorities by a young woman whom he had pinched. Morals aside, I thought the act reflected a certain energy on Stokowski’s part, and I found myself led into a rumination about the apparent vigor, and then the differential longevity of symphonic conductors. Stokowski, as it turned out, lived for 95 years, and gave his last concert at the age of 93 at the Vence Festival in France. Toscanini lived to be 90, Sir Thomas Beecham 83, and Eugene Ormandy 86. The more general question that emerged for me had to do with who, in any frame of life, lives a long time, and why. And, if the posit about symphonic conductors was correct, what was it about them or their activities that was operational?

Was it the music? There is some evidence that the right side of the brain is more involved in processing music than the left, and blood flow studies have shown that the same areas of the brain that respond to euphoria-inducing stimuli like food, sex and some drugs also respond to stimulating music. How this might have to do with longevity is admittedly obscure; connections between pleasure and longevity have not been clearly established….

In any case, to return to symphonic conductors, the fact that that the sample was small and hardly random didn’t deter me much. Maybe it was just the successful ones who lived a long time. Maybe it was the music that did it. Maybe, if symphonic conductors really had preternatural longevity, it had something to do with waving their arms so much. That idea really intrigued me…. First of all, it’s plain that when people run they also move their arms a lot, so even if running is good for you, you may be able to get the same effect a lot more efficiently. And notice that arm waving is a form of upper body aerobic exercise, so the arms have a claim along that line as well, and, in any case, I found the idea that it might be better to play a little Mozart or Shostakovich and wave your arms in time with it much more congenial. Finally, in the case of symphonic conducting, an enormous amount of cognitive activity is involved, another element that has been linked to longevity.

Ultimately I felt more or less requited when I discovered a paper by Leonard Hayflick citing a MetLife study that involved 437 active and former conductors of major regional and community symphonies. The study started in 1956 and ended in 1975 when 118 of them had died, more than 20% at age 80 or older. The death rate for the entire group was 38% below that of the general population, and for conductors aged 50 to 59, a decade when stress and responsibilities are at their peak, the death rate was 56% less than that of the general population. I was somewhat disconcerted by a nearly simultaneous MetLife study that showed that corporate executives enjoyed longevity similar to that of orchestra conductors, punching a hole in the arm waving theory, though possibly not a definitive hole, since the study did not control for arm waving among the executives.

In any case, the conducting and arm waving thing had me hooked. The next thought, if you’ll forgive the expression, was that it might be interesting to compare longevity among baseball players who spent years in positions that involved a lot of throwing, and to compare them with those whose positions called for infrequent throwing. I tried to recruit to this question a bright young man who was taking a fellowship in general medicine with me, and he seemed interested. Accordingly (this was before every statistic in the world was available online, in fact before anything was available online), I provided him with the Encyclopedia of Baseball, thinking he could do the necessary with it. It contained data on everyone who had ever played professional baseball, the teams, the years and the positions played. The task proved too daunting for my young colleague, and it fell by the wayside under the pressure of other responsibilities, but, as evidence the idea wasn’t uniquely quirky, in 1988 a group at the University of Alabama published an article on the mortality experience of major league baseball players, in the New England Journal of Medicine. They assembled a cohort that included all players who had played their first games for a major league team in the United States between 1911 and 1915 and who survived at least until 1925. They had a cohort of 985 players to analyze, and successfully acquired follow-up information on 958 of them. Their average age at death was 70.7 years, the average year 1960. Infielders had the lowest overall mortality rate and catchers the highest; the differences were not statistically significant. Grouping all infielders may have blunted the study; it might have been better to compare first basemen, say, who throw relatively little, with pitchers or short stops. But there was an inverse association between standardized mortality ratios for the groups and the length of the player’s career; and being a baseball player in fact conferred a slight protective effect against death, with the cohort having only 94% of the deaths expected. It was most interesting to me that the data suggested that players who performed the best lived the longest, a fact that should bring some comfort to the accomplished people in this room. But my arm waving theory was not supported, at least by the gross categories established within the cohort. (“How to Live a Long Time: Facts, Factoids, and Descants,” Transactions of the American Clinical and Climatological Association, 116: 77–89)

It seems indisputable, based on the statistics cited by Barondess, that conductors and baseball players tend to outlive their peers. This leads to two questions: By how much do conductors and baseball players outlive their peers, and why do they and others, like corporate executives, outlive them? As Barondess suggests, the answer is not vigorous arm movement. If it were the answer, one would expect pitchers be longer-lived than other baseball players, and that is not the case.

But I am getting ahead of myself. Before considering what factors might yield a long life-span, I will present some statistics about conductors and baseball players.

I compiled a list of 152 conductors born after 1800 and before 1930 who are prominent enough to have Wikipedia entries. I obtained names of candidates for the list from this page at a site known as knowledgerush (which displays obnoxious banner ads), and from two lists at Wikipedia (one of 19th century conductors, the other of 20th century conductors). Of the 152 conductors, 18 are still living. The earliest year of birth of a living conductor is 1919. I therefore focused on the 112 conductors who were born in 1918 or earlier,* inasmuch as the inclusion of conductors born after 1918 (among them 18 living ones) would bias the analysis by understating the longevity of conductors born after 1918. Here is a plot of the 112 conductors’ ages at death, by year of birth:

The linear relationship between age at death and year of birth (dashed line) is statistically insignificant, but it roughly parallels the rising trend of life expectancies for white males aged 40 (the green line). (I used life expectancies for white, American males, given here, because there are no non-whites or females in the sample of 112 conductors.) In words, a person who — like almost everyone in the sample — had become a conductor by the age of 40 was very likely to outlive the general run of 40-year old white males, and to do so by a wide margin. By 1918, that margin had shrunk to about 6 years, but it was still large enough to say that conductors enjoyed unusually long lives. The trend line, however weak statistically, suggests that conductors will continue to enjoy unusually long lives.

But, as I have said, arm-waving probably is not the key to conductors’ long lives. The evidence for that assertion is found in an analysis of the longevity of baseball players. Using the Play Index (subscription) tool at, I compiled lists of deceased major-league players who either pitched at least 1,000 innings or played in at least 1,000 games in one of the following positions or groups of positions: pitcher, catcher, second base-third base-shortstop, first base-outfiield. I chose those groupings because pitchers use their arms intensely and often every few games; catchers use their arms somewhat less intensely than pitchers, but more often than other players; the second base-third base-shortstop positions involve less intense and frequent arm motions than pitching and catching, but more frequent (if not more intense) than the first base-outfield positions.

The total number of players in the sample is 1,039, broken down as follows: 592 pitchers; 41 catchers; 178 players at second, third, or short; and 228 players at first or outfield. A regression on age at death yields the following:

Age at death = 57.1 + 1.02 x number of seasons in major-league baseball – 0.004 x number of games played + 6.05 if played primarily (at least 90% of games) at 2b, 3b, or SS + 5.17 if played primarily at 1b or OF + 2.90 if played primarily at catcher.

The P-values on the intercept and coefficients are 1.7E-178,  7.89E-12, 0.05, 0.02, 0.05, and 0.33, respectively

What about pitchers? The positive coefficients on the non-pitching positions imply a negative coefficient on “pitcher.” The correlation between “pitcher” and “age at death” is negative and significant at better than the 1-percent level. The difference between the average age of pitchers at death (68.4 years) and the average age of other players at death (71.3 years) is statistically significant at the 1-percent level.

In sum, pitchers do not live as long as other players. And catchers, though they live longer than pitchers, do not live as long as other non-pitchers. So much for the idea that longevity is positively related to and perhaps abetted by vigorous and frequent arm motion.

What about the longevity of baseball players in relation to that of the population of white males? I derived the following graphs from the Play Index and the table of life expectancies (both linked above):

I chose 1918 as the cutoff point for ballplayers because that is the last year in the sample of 112 conductors. (As of today, only 15 of the thousands of players born in 1918 or earlier survive** — not enough to affect the comparison.) Before I bring in the conductors, I want to point out the positive trend for longevity among ballplayers (indicated by the heavy black line), especially in relation to the trend for white, 20-year old males. The linear fit, though weak, is statistically robust, and it reflects the long, upward rise in ballplayers’ longevity that is evident in the scatter plot.

I now add conductors to the mix:

For the period covered by the statistics (birth years from 1825 through 1918), conductors enjoyed a modest and significantly insignificant increase in longevity, indicated by the dashed black line. By 1918, ballplayers had almost caught up with conductors. The trends suggest that, on average, today’s MLB players can expect to live longer than today’s conductors. Conductors, nevertheless, seem destined to live longer than their contemporaries in the population at large, but not because they (conductors) wave their arms a lot.

If the secret of a long life is not a lot of arm-waving, what is it? I return to Dr. Barondess:

[W]e’ve heard for years that the best way to live a long time is to pick long-lived parents, and there is increasing evidence that the pace of aging is to a significant degree genetically determined, but environmental influences and personal behaviors are clearly also of great importance. Scandinavian studies have calculated the heritability of average life expectancy to be 20 to 30%, with environmental differences accounting for at least 70% of variation in age at death among twins. And studies of 7th Day Adventists suggest that optimizing health related behaviors could yield up to 25 years of good health beyond age 60 with a compression of morbidity toward the end of life. The authors of that study suggested that when it comes to aging well there is no such thing as the anti-aging industry’s free lunch. I think a better suggestion might be that a really good anti-aging maneuver is no lunch, in light of other studies connecting undereating with extension of life expectancy….

There are some data connecting a specific region on chromosome 4 to the longevity of centenarians and nonagenerians, and a number of longevity genes have been discovered in yeasts, worms and fruit flies. So apparently there are gerontogenes, or longevity-enabling genes, and the genetic contribution to longevity is being investigated with increasing enthusiasm….

There’s been a good deal of research activity, and perhaps even more in the public prints in recent years, with relation to diet and longevity, especially caloric restriction. These effects were first demonstrated in the 1930s, when it was shown that laboratory rats on limited diets live about 40% longer than normal and are resistant to many chronic diseases typical of aging. These studies have been replicated in yeasts, fruit flies, nematodes, fish, spiders and mice, and there are hints that the effect may also hold true for primates. Recent research on the mechanisms underlying these phenomena has shown that the effect of caloric restriction is tied to genetic factors….

Numerous mechanisms have been suggested without great clarification to this point, but it does appear that life lengthening through caloric restriction is not primarily related to retardation of disease processes, but rather to slowing of primary aging processes, and this is related to restriction of calories rather than specific nutrients….

On the other hand, specific nutrients may impact disease processes themselves…. One study suggested that pizza intake had the potential to reduce cardiovascular risk, presumably because of the tomato sauce component, and despite the cheese.

A number of other foodstuffs have been thought to enhance health prospects, including nuts, for their resveratrol content, organo-sulfur compounds in garlic and onions, and various carotenes. Cocoa, flavanol rich, is thought to be good for you; the makers of Mars Bars are working hard on this. So are blueberries, high in antioxidants, as are raspberries, cranberries and strawberries….

One study from Rome considered alcohol consumption and its effect on longevity. The study suggested that drinking 4 to 7 drinks per day, roughly 63 grams of alcohol, a dose some might think heroic, led to a two-year edge in life expectancy, but drinking more than 10 drinks was negatively associated with longevity. These drinks were 97% wine, primarily red, high in resveratrol content. Other studies have suggested that 250 to 500 cc. of red wine a day is associated with a diminished risk of macular degeneration, Alzheimer’s disease and cognitive deficits….

Several studies … have found that social networks among humans are important predictors of longevity, including participation in formal organizations, contact with friends or relatives, and so forth. In one study of African American women aged 55 to 96, those who were extremely isolated in a social sense were more than three times as likely to die within a five-year period of observation, an impact unaffected by the use of community senior services. A search for the effects on longevity of living as a recluse or a hermit produced no results, I imagine because follow-up would be difficult, but on the other hand a number of additional papers about socialization in humans turned up. One suggested that providing social support may be more beneficial than receiving it. Mortality was significantly reduced for individuals reported to be providers of support to friends, relatives and neighbors, and emotional support for spouses.

In a study from Columbia University, the impact of marital closeness on survival was examined in 305 older couples. Closeness was defined as naming one’s spouse as a confidant or as a source of emotional support, versus not naming, or being named by the spouse on at least one of the two dimensions, versus not being named. Husbands who were named by their wives as confidants or supports, but did not name them, were least likely to have died after six years. Compared with them, husbands who were not named by their wives as a confidant or source of social support, or did not name their wives, were from 3.3 to 4.7 times more likely to be dead. The results among wives showed a similar pattern, but a weaker one….

Studies of personal histories have illuminated some personality factors that may bear on longevity. One important investigation is the Nun Study, organized by David Snowden in 1986. In this longitudinal study of aging and dementia, he was looking at a convent community of nuns aged 74 to 106, retired from careers in a variety of sites, many of them as teachers. They tended not to drink or smoke, had similar diets, income, and quality of health care and had an active social network. Snowden examined short biographical notes written at an average age of 22, on entry into the order. These suggested that positive emotions in early life were associated with longevity, with a difference of nearly 7 years between the highest and lowest quartiles of positive emotion sentences. That is, positive emotional content in early life autobiographies was strongly associated with longevity six decades later. There was some sense that positive response patterns, or more rapid return to a positive outlook after negative events, serves to dampen the physiologic sequelae of emotional arousal, such as heart rate and blood pressure changes, and presumably also hormonal responses. In a word it’s best to be cheery, or at least positive.

Further to the effect of optimism and pessimism as risk factors for disease, Peterson and his group studied questionnaires filled out by 99 Harvard graduates in the classes of 1942 to ’44, when they were about 25 years of age, and then determined physical health from ages 30 to 60 as measured by examination by physicians. Pessimistic explanatory style, the belief that bad events are caused by stable, global or internal personal factors, predicted poor health at ages 45 through 60 even when physical and mental health at age 25 were controlled for, across an array of diseases ranging from gout to diabetes, kidney stones to hypertension. The correlations increased across the life span, from age 30 to 60.

With regard to the impact of cognitive activity on optimism, health, and possibly life expectancy, there is good reason to believe, as Guy McKhann and Marilyn Albert have pointed out, that the phrase “use it or lose it” applies. Maintaining one’s mental abilities is made easier through a variety of activities, including reading, doing crossword puzzles, learning ballroom dancing, using a computer and going to lectures or concerts. Studies have shown that in rats an enriched environment that includes exercise, toys, mirrors, tunnels and interaction with other rats strengthens connections between cells in the hippocampus and even increases the rate at which new cells are born. The idea of rat fraternization may be counterintuitive, but somewhere here there may be a link with the academic parable expressed in prior talks by Dick Johns on how to swim with sharks. Fraternization with rats, has, I think, a weaker set of academic projections, but I pass it along for what it’s worth.

Related to the last point is evidence of

[a] strong inverse correlation between early life intelligence and mortality … across different populations, in different countries, and in different epochs.”[3][4][5] Various explanations for these findings have been proposed:

“First, …intelligence is associated with more education, and thereafter with more professional occupations that might place the person in healthier environments. …Second, people with higher intelligence might engage in more healthy behaviours. …Third, mental test scores from early life might act as a record of insults to the brain that have occurred before that date. …Fourth, mental test scores obtained in youth might be an indicator of a well-put-together system. It is hypothesized that a well-wired body is more able to respond effectively to environmental insults…”[5]

A study of one million Swedish men found showed “a strong link between cognitive ability and the risk of death.”[6][7][8][9]

People with higher IQ test scores tend to be less likely to smoke or drink alcohol heavily. They also eat better diets, and they are more physically active. So they have a range of better behaviours that may partly explain their lower mortality risk.—-Dr. David Batty[7]

A similar study of 4,289 former US soldiers showed a similar relationship between IQ and mortality.[7][8][10]

The strong correlation between intelligence and mortality has raised questions as to how better public education could delay mortality.[11]

There is a known inverse correlation between socioeconomic position and health. A 2006 study found that controlling for IQ caused a marked reduction in this association.[12]

Research in Scotland has shown that a 15-point lower IQ meant people had a fifth less chance of seeing their 76th birthday, while those with a 30-point disadvantage were 37% less likely than those with a higher IQ to live that long.[13]

Here is my take on all of this: Conductors, baseball players, and corporate executives (among members of other identifiable groups) tend to be long-lived because they tend to be physically and mentally vigorous, to begin with. Conductors must possess stamina and intelligence to do what they do.To rise in the corporate world, one must be capable of working long hours, putting up with a lot of stress, and coping with many complex issues. And, contrary to the popular view of athletes as “dumb,” they are not (as a group); in fact, intelligence and good health (a key component of athleticism) are are tightly bound.

Moreover, conductors (who make music), ballplayers (who play a game) and corporate executives (who attain high status and high incomes) are engaged in occupations that yield what Robert Atlas calls “gratifying stress.” And, as persons who usually enjoy above-average incomes, they are likely to enjoy better diets and better health-care than most of their contemporaries.

The finding that pitchers do not live as long as other ballplayers supports the view that “gratifying stress” fosters longevity, whereas “frustrating stress” may shorten a person’s life. Pitchers, uniquely among ballplayers, are credited or charged with the wins and losses of their teams. And pitchers, as a group, win only half the time — an ungratifying outcome. Further, pitchers with consistently bad records do not last long in the big leagues, and end their careers having won less than 50 percent of the games for which they were held responsible (i.e., with a won-lost record below .500). Accordingly, more pitchers end their careers with losing records than with winning records: In the history of the major leagues, from 1871 through 2011, there have been 6,744 pitchers with a career record of at least one loss; only 29 percent of them (1,935) had a career won-lost record better than .500.

I conclude that occupation — conducting, playing professional baseball, etc. — is a function of the main influences on longevity — mental and physical robustness — and not the other way around. Occupation influences longevity only to the extent that increases it (at the margin) by bestowing “gratifying stress” and/or material rewards, or reduces it (at the margin) by bestowing “frustrating stress” and/or exposure to health-or life-threatening conditions.

*   *   *

The footnotes are below the fold.
Continue reading