In addition to the seasonal regularities that are evident in the statistics for suicides there is a systematic relationship with latitude. The farther from the equator that humans live, the higher the overall rate of suicide, and the larger the difference in suicide rates between summer and winter. This latitudinal cline supports the notion that the annual rhythm of suicide is associated with photoperiod. The longer and brighter the days in the year, the more energy people have. If they haven’t overcome their depression, they are more likely to kill themselves then than at other times during the year.8
Photoperiod is one of the major environmental factors correlating with several seasonal statistics. Another player is temperature. General mortality, for example, also shows a seasonal rhythm that has two peaks every year, the major one in the coldest months of winter and a secondary peak in the hottest months of summer.9 These peaks are identical in the two hemispheres, unlike the seasonal rhythms that show only one peak, which are six months out of phase between the northern and the southern hemisphere. Interestingly enough, the turnaround point between the two hemispheres is not exactly at the equator but approximately five degrees north. The reason for the shift of this “biological” equator to the north is the asymmetrical distribution of landmass across our globe. Landmass is larger in the northern than in the southern hemisphere. The fact that the annual rhythms in vital human statistics reflect this biological equator strengthens the hypothesis that environmental rather than social influences form the basis of these rhythms.
Among the many vital human statistics, monthly birth rates show the most pronounced seasonality that systematically depends on latitude. Many researchers have investigated the phenomenon of seasonal birth rhythms, based on smaller databases and focusing on more limited geographical regions. The authors of these studies have concentrated predominantly on the fact that different numbers of children were born at different times of the year; hence they hypothesized that these rhythms correlate with factors associated with the time of birth—for example, with influenza epidemics. When I looked at the regularities in the worldwide database, it soon became apparent that any environmental model made more sense if associated with the time of conception, rather than with the birth date itself.
After I had tested numerous weather and climate factors, a clear picture emerged: what could best explain the influences on the seasonality of births (or rather conceptions) was an environmental model incorporating both photoperiod and temperature. The highest annual increase in conceptions was found to occur around the spring equinox in the statistics from all geographical regions, independent of latitude (except in equatorial regions—I’ll come to that later). However, at what point in time the annual conception rhythm reached its actual maximum depended on temperature, and therefore also on latitude. Conception maxima coincide with those times of year when the monthly averages of the daily temperature minima are around 12°C.
The biggest amplitudes in annual conception rhythms were found around thirty degrees of northern latitude, in countries like Jordan, Israel, Palestine, Lebanon, and several regions in northern Africa. In these countries, the difference between the lowest and the highest monthly birth rates used to be more than 60 percent. A reason for the large amplitude of the birth/conception rhythms in these regions might be that the optimal daily temperatures for conception occur close to the spring equinox, so that the two influences—photoperiod and temperature—act simultaneously, resulting in a conception peak at around the 22nd of March.10 With increasing latitude, the days with optimal temperatures occur progressively later in the year and so do the peaks of conceptions. The largest increase of conceptions still occurs at around the spring equinox, but the actual maximum is reached later.11
Whether the seasonality of human reproduction depends on environmental factors or is just a reflection of agricultural workload is still being disputed. One argument of the sociocentric advocates is the distinct difference in the shape of the conception rhythms between Central Europe and the Americas. Whereas the former shows only one pronounced peak (with a smaller one around Christmas), the latter has two more-or-less equally large maxima per year (a bimodal rhythm). One fact is often overlooked in this argument, however: similar bimodal rhythms are found in Eastern Europe. The differences in conception rhythms between Central and Eastern Europe (and the Americas) is far more likely to be associated with the respective climates. Central Europe is under the influence of the Gulf Stream, producing a rather temperate climate, whereas the Americas and Eastern Europe have a typical continental climate, with both hot summers and cold winters. Since these summer and winter temperatures are well outside the optimal range for conception, the rate of conception would decrease twice a year, resulting in a bimodal rhythm.
Irrespective of social or environmental factors influencing human reproduction, conception rhythms are apparently not caused by planning, since the seasonality in illegitimate births is in all countries we tested even stronger than those of legitimate births.12 The fact that conception rhythms in towns are weaker than in rural regions doesn’t help to distinguish between social or environmental influences because the seasonality in workload is stronger in the countryside, as are the influences of environmental factors like temperature and light.
So far I have described the systematic yearly cycle of conception rhythms and their geographical distribution only as they existed before industrialization drastically changed human life. Over the past hundred years these rhythms have almost disappeared, and the small remaining conception peaks (mostly no more than 1–2 percent above the annual mean) have changed their phase from spring/summer to autumn/winter. The typical increases with photoperiod and the peaks associated with the optimal temperatures have been replaced by a small peak at around Christmas. This remaining maximum could have purely social reasons—the longer nights and the usual holidays around that time of year lead to more time spent in bed, and thus could slightly increase the chance for a successful conception.
The modern decline in the seasonality of human reproduction can be associated with the onset of industrialization in each respective country—for example, earlier in Germany than in Spain. The next graph illustrates this development. It shows a continuous record of monthly birth rates in Spain over almost the entire last century. Important social events like wars were only able to perturb this immensely regular rhythm as mere blips. After the Second World War, the amplitudes of the rhythm became much smaller. But it was not until the 1960s that the rhythm started to become highly irregular and also shifted its peak from late spring to early winter. Franco had launched a political campaign around that time to massively industrialize rural Spain.
Again, these developments cannot distinguish between the social and the environmental explanation. Although industrialization drastically decreases the seasonality of the workload, it also means that humans are increasingly shielded from environmental factors. The strongest support for environmental rather than social hypotheses is provided by the observation that the two factors—photoperiod and temperature—have lost their influence in a systematic way. The influence of photoperiod disappeared first. In the early days of industrialization, people stopped working outside and began working inside, which shielded them from daylight. They still were exposed to seasonal temperature fluctuations, however. An environmental model using only temperature as a predictor can fully explain the annual conception rhythms of this era. Only much later were people shielded with comparable efficacy from the influence of temperature by both central heating and air-conditioning. It was only then that human reproduction almost completely lost its seasonality, except for the aforementioned (social) blip around Christmas.
A record of monthly birth rates in Spain over almost the entire last century. Reprinted from T. Roenneberg (2004). The decline in human seasonality. Journal of Biological Rhythms 19(3):193–195.
One could argue that the contraception revolution brought about by the birth control pill coincides with the observed lo
ss of rhythmicity in human reproduction. Yet the introduction of the pill had astonishingly little influence on the seasonality of reproduction—only on its level. This is not too surprising because contra- and proconceptive planning are both part of sexual behavior, which is seasonal, as the annual rhythm in condom sales clearly shows.
We don’t yet know the underlying mechanisms for the seasonality in human reproduction. Is there seasonality in both male and female sexual behavior? Is it that sperm production or sperm agility is seasonal, as studies have shown, or that eggs are more or less fertilizable? Or is there seasonality in how successfully a fertilized egg survives the first weeks of pregnancy?13 The following finding supports this last possibility: although some illnesses like schizophrenia are only diagnosed in early adulthood, the births of these patients are seasonal—showing an even stronger amplitude than those in the general population. Many of these illnesses have a genetic basis, which could already have an influence on the development of these patients as embryos. If embryonic development were seasonal, then these patients would have a higher chance of being born at certain times of the year than at others, which could explain the annual rhythm in their birth rates.
It is quite remarkable how many aspects of our lives are influenced by industrialization and its consequences, isolating us increasingly from natural signals. To allow organisms to accommodate and anticipate the natural changes across day and night but also across seasons, evolution has developed internal clocks. One might argue that an industrialized human doesn’t need these clocks anymore since we live in a society that is not only 24/7 but to some degree also 24/7/365. Yet we shouldn’t forget that life in any form is constantly under attack. Where the attackers of our ancestors included big enemies (like saber-toothed tigers, wolves, or bears), our modern selves are still under attack, only by much smaller but just as dangerous enemies like bacteria, fungi, or viruses—and these enemies are still seasonal. An appropriate timing system in our body would therefore still be useful—at least for our immune system. But our internal clocks are only effective if we provide them with sufficient information: with sufficient differences in daily light and darkness, or with sufficient contact to changing photoperiods and temperatures.
Tom’s experiment made Barbara and Gerry live strictly by photoperiod, and neither of them developed a cold that winter.14 You probably guessed the rationale behind the pink glasses, the filter sheets, and the special light bulbs. After dusk and before dawn, they aimed to shield Gerry and Barbara from that part of the light spectrum that reaches our timing system most effectively (the blue parts of the light spectrum).
23
Professional Selection
The conference dinner had long been over. About thirty neurosurgeons had gone to a jazz bar in the old part of town to celebrate a successful meeting. The core of the party consisted of three neurosurgeons—among the best in the world—who had known each other for decades. Their frequent joint appearances had earned them the nickname “the three-pack.” Many of those who attended these conferences regularly had learned over the years to follow wherever the three-pack led—especially when Drs. Fergusson, Skinter, and Lafayette went to a jazz bar.
A quartet (bass, saxophone, piano, and drums) was playing swing on a small podium in a corner of the room opposite the long bar. The group of surgeons had moved several of the small tables together between bar and podium. By now they were the only patrons left in the bar. When the musicians stopped for a break, Fergusson made eye contact with Skinter and Lafayette. “What do you think? Shall we go for it?” Lafayette’s face lit up with almost childish anticipation, but Skinter looked, as always on these occasions, far too tired to respond adequately. Fergusson ignored Skinter’s familiar lack of enthusiasm and got up to chat with the musicians, who were having a drink at the bar. When he returned, he nodded to Lafayette and Skinter and the three went up onto the podium. They almost had to carry Skinter to the piano, where they dropped him onto the stool and placed a triple espresso on a small table beside him. Fergusson climbed behind the drums, and Lafayette picked up the bass.
“A one, a two, a one, two, three, four,” and off they went on a journey of the coolest swing, blues, and boogie-woogie ever heard outside of a concert hall. The triple espresso seemed to kick in, carrying Skinter along despite it being so much past his usual bedtime. He was one of the best jazz pianists around, if one of the least known—few people outside of the neurosurgical world had ever heard of him or had ever heard him play. But within that circle he was famous for both his innovations in neurosurgery and his heavenly command of the piano. If it had been several hours earlier, he definitely would have brought the house down and would have made sure that the bar’s regular quartet, though excellent themselves, would have retired for the night. He could, without doubt, have made it to the top of the crop in the jazz world. But one just cannot live two full lives within one lifetime.
So far I have discussed the science and history of clock research. Although the stories at the beginnings of chapters wrap the science in fiction, the content of every chapter is based on peer-reviewed, published fact.1 In the remaining two chapters, I take the liberty of engaging in more speculative thoughts about body clocks. In this chapter I will discuss how chronotype and the individual need for sleep could influence people’s careers, or even their behaviors and personalities.
This chapter’s case story was inspired by a colleague of mine who is both a gifted neurosurgeon and an extremely talented jazz pianist. He could have chosen either profession and would have been very successful in both—that is, if he hadn’t been such an extreme early chronotype. He would not have survived the long concert evenings, having to give his best performance way past his peak abilities and preferred bedtime. He has definitely made the right choice in his career because he has no difficulty whatsoever in giving his best performance in the operating theater at 6 or 7 A.M. The early working hours of doctors suggest that this profession is submitted to a selection pressure favoring early types. In addition, doctors often have to cope with sleep deprivation, especially at the beginning of their careers, which potentially has consequences for patients.2
A few years ago, Harvard sleep researchers investigated the medical errors made by young interns. They compared the error rates in two types of work schedules: the traditional, extended schedule and a so-called intervention schedule. During their study period, the Harvard group investigated a total of 2,203 patient-days involving 634 admissions and found that interns made about 36 percent more serious medical errors on the traditional schedule. Disturbingly, more than half of these were not intercepted before reaching the patient.3 The results of this study are perhaps to be expected. It’s not surprising that someone who works for up to seventy-two hours without noteworthy rest makes more mistakes, but what seems like common sense is worthless without a quantitative basis provided by a properly designed study.
After the results of that study were published, decision makers reluctantly started to implement slightly better work schedules for interns. Before the Harvard sleep researchers published their data, no one was able to change the system, even though everyone believed that grueling schedules must produce more medical errors. Why is this? First of all, we are dealing with a very special portion of society. Doctors have gone through a long university education, their responsibility concerns the life or death of a patient, and they remain among the highest-ranked professionals on our ladder of social recognition. It seems that this elite system actively exerts a selection pressure that ensures the survival of only the fittest. Besides, those who make the decision about a young doctor’s working hours went through exactly the same extreme work schedules when they were interns themselves and don’t see why this tradition should be changed.
Physicians are a prime example of professional selection involving the capacity for short sleep and top performance in the early hours.4 Teachers fall into this category as well. To discipline, reach, and teach between twenty and forty child
ren or teenagers at 8 A.M. requires teachers to be in full command. Being a teacher is one of the few professional situations where individuals remain in the same environment (except for the interruption of going to university) from childhood to retirement. This suggests that they felt comfortable with this environment before they made their career decision. It seems unlikely, therefore, that they are particularly late chronotypes, or even were late chronotypes as teenagers, and thus may belong to the early chronotypes in our population.5 Note that all of this is purely hypothetical. We need to conduct epidemiological studies into professional selection, and doctors and teachers would certainly be a good starting point.
The selection for early types and short sleepers may be a general prerequisite for the top professions in our society, for managers and decision makers who have to give their best from early in the morning until late at night. Can you imagine a manager or a politician who needs an average amount of sleep and is an extreme late type? Such an individual might never make an appointment before 11 A.M. A late type and long sleeper would also have a higher statistical chance of falling ill, since lack of sleep challenges the immune system.
A recent study recorded the sleep duration and sleep quality of about 150 healthy subjects for two weeks before the participants were quarantined in a clinic and given nasal drops containing a rhinovirus. The researchers then looked for the development of clinical symptoms of a cold. Participants who slept seven hours or less were almost three times more likely to develop a cold than those who slept eight hours or more. The influence of sleep quality was even stronger than that of sleep duration. Individuals reporting their sleep quality to be less than 93 percent of the possible optimum had a five-and-a-half-fold chance of catching the cold compared with individuals who rated their sleep quality to be 98 percent of the optimum (or higher). Of course the researchers made sure that this result was not caused, for example, by the fact that subjects slept badly because they already were fighting off an illness during the baseline period.6 Thus, poorer sleep quality and shorter sleep duration in the weeks preceding exposure to a rhinovirus were associated with lower resistance to illness. This result strengthens the notion that decision makers are more likely to be early types and short sleepers, simply by selection. Their sleep behavior efficiently protects them from falling ill, thereby making them overall more successful compared with late types and long sleepers, who would regularly suffer from social jet lag if they had to comply with a decision maker’s daily schedule.
Internal Time: Chronotypes, Social Jet Lag, and Why You’re So Tired Page 20
