THE star you were born under influences the person you become. Not something you expect to hear from scientists but, incredibly, it seems to be true. There is firm evidence that the time of year you are born affects not just your personality, but also your health, specifically your chances of developing serious mental illness. But don't expect to find clues in your horoscopes. The star in question is the star we were all born under - the sun.

Being born at certain times of year gives a small but significantly increased risk of problems such as depression, schizophrenia and anorexia nervosa. The question is no longer if the seasons affect mental health, but how. Pinning this down could yield vital clues on how to intervene to prevent mental illness.

The effect was first noticed as far back as 1929, when Swiss psychologist Moritz Tramer reported that people born in late winter were more likely to develop schizophrenia. We now know that for people born in the northern hemisphere in February, March and April, the risk of developing schizophrenia is between 5 and 10 per cent greater than for those born at other times of the year. The effect has been replicated numerous times over the decades and is far from trivial. According to a study carried out at Aarhus University Hospital, Denmark, the increased risk of schizophrenia that comes with a winter birthday is almost twice the increase in risk linked to having a parent or sibling with the disorder.

Season of birth seems to be linked to other conditions, too. A recent study of more than 25,000 suicides in England and Wales found that 17 per cent more people who had committed suicide had birthdays in April, May and June than in the rest of the year. That's late spring and early summer in the northern hemisphere. Similarly, people with anorexia in the northern hemisphere are 13 per cent more likely to have been be born between April and June than in other months. That doesn't necessarily mean that it's best to be born later in the year, however. Autumn birthdays are associated with an 8 per cent increase in the likelihood of suffering panic attacks, for example, and a small but significant increase in alcoholism in men.

Finding out just how the changing seasons change the risk of certain psychiatric conditions is proving a daunting task for epidemiologists. Until recently, the leading hypothesis on why winter and spring births increase the risk of both schizophrenia and anorexia centred on the health of a mother during pregnancy. Perhaps a viral infection in the mother at a key stage of the baby's development could somehow interfere with brain development or damage early brain tissue. Several studies in the 1980s seemed to have found spikes in the number of schizophrenia cases in people born in the months following outbreaks of viral diseases such a flu and measles. But several large studies, in particular one in 1999 by Stephen Miller at the University of Georgia, Athens, which looked in detail at patient records for 750,000 people, failed to find a similar correlation between epidemics and rates of schizophrenia.
Seasonal fertility

While the idea that infections can affect the development of a fetus's brain still has some support, more recent explanations have concentrated on the more direct effects of the seasons on the mother: how much sunshine a pregnant woman is exposed to, for example, or the effects of temperature.

The hormone melatonin, which regulates the sleep-wake cycle and is suppressed by sunshine, could play a role during gestation or early life, says consultant psychiatrist Emad Salib of Peasley Cross Hospital in St Helens, Merseyside, UK. Salib was lead author on the study linking birth in late spring to suicide, published in the British Journal of Psychiatry last year. Several other studies have found that suicide itself also follows a seasonal pattern, peaking in the month of maximum daylight. Salib reasons that as people born in April, May and June were conceived in July, August and September, high melatonin levels during crucial months of development could prompt changes in the brain, and then stimulate the act of suicide later in life.

Beth Watkins, an eating disorder researcher at St George's Hospital Medical School in London, has suggested a more subtle reason for the season-of-birth link to anorexia, which follows a similar pattern to that for suicide. Her idea is not that the seasons cause changes in the fetus, but rather that seasonal effects allow babies vulnerable to the condition to be conceived and born only at certain times of the year. People with anorexia are eight times as likely as the general population to have a parent or sibling with the disorder. Often that relative is the mother, and this got Watkins thinking. Was there something about overly thin mothers that might vary by season? "Their actual fertility is on a knife-edge," she says, and babies born in the months most strongly linked to anorexia were conceived in July to September - which follow the northern hemisphere's warmest months. Could the higher temperature allow an anorexic mother to conserve just enough energy to tip her into a fertile state?

Watkins and her colleague Kate Willoughby looked at a sample of nearly 400 women in the UK with anorexia and other eating disorders. In the UK, only in the summer months of July and August does the average monthly temperature tend to rise above 15 °C, and in keeping with Watkins's hypothesis, significantly more people with anorexia had been conceived during these warmest months. They then collected data on 200 patients in Australia living in and around Sydney, where average temperatures drop below 15 °C only in the winter months of June, July and August. Sure enough, they found that fewer had been conceived in these cooler months. Finally, they looked at a sample of people with anorexia from Singapore, where the temperature remains constant at over 25 °C all year round. There they found no season-of-birth effect at all.

Watkins accepts that what appears to be a temperature effect could just as well fit some kind of explanation based on prenatal exposure to sunlight, now a leading theory in explaining the season-of-birth effect in schizophrenia. In 2005, John McGrath and his colleagues at the Queensland Centre for Mental Health Research in Wacol, Australia, reported that the higher the latitude, the greater the excess incidence of schizophrenia among people born in winter. Between 30° and 60° north, the excess is about 5 per cent, at latitudes above 60° north it is 10 per cent.
A sunny disposition

Not convinced by the idea that winter infections are the cause, McGrath focused on the effects of a seasonal lack of sunlight. He was intrigued by a study in Brazil by Erick de Messias of the Maryland Psychiatric Research Center in Baltimore, which found a raised incidence of schizophrenia in people born three months after the rainy season. Then there are the unexplained high levels of schizophrenia among children of dark-skinned people who have emigrated from equatorial countries to northern ones. McGrath's own work went on to link schizophrenia in Queensland, Australia, with winter birth dates - but only every three or four years in a cycle that seemed to be in sync with the climate phenomenon known as El Niño, which can lead to particularly cloudy, rainy weather. A lack of sunlight during pregnancy could explain all three, he reckons.

One way sunlight is known to influence the body is by stimulating the production of vitamin D, which fetuses need for brain development. Trouble is, in northern latitudes, there isn't enough sun around in the winter. In London, at 51° north, it's hopeless, says McGrath. "You just can't make vitamin D in winter even if you lie naked all day."

McGrath suspects vitamin D deficiency might link birth date to an elevated risk of schizophrenia. Working with neuroscientist Darryl Eyles, he has established that the brains of rats deprived of vitamin D during gestation have abnormally large lateral ventricles. People with schizophrenia also have enlarged lateral ventricles. The rats with enlarged lateral venticles run around more than is normal, and this behaviour can be rectified by the antipsychotic drug haloperidol, which is thought to block the neurotransmitter dopamine and is used to treat schizophrenia. The link between vitamin D deficiency and neurodevelopmental abnormalities is a "done deal", says McGrath, at least in rats. Does that mean the same applies to humans? We're still working on that," he says.

Other researchers like the idea, but are equally cautious about making the jump to humans. "It hasn't been tested yet," says Ezra Susser, an epidemiologist at Columbia University, New York. Robin Murray of the Institute of Psychiatry, London, agrees that McGrath may be onto something. "But as yet he doesn't have evidence linking vitamin D in mothers to offspring with schizophrenia."

Whether or not this simple explanation underlies any or all of the seasonal effects in mental health, the ultimate goal is to pin down the molecular mechanisms underlying these seasonal trends and, potentially, to fix them. McGrath speculates that since vitamin D can turn genes on or off in the cell nucleus, a lack of it may somehow leave an affected individual with a lack of key proteins, and that this is what puts them at risk of the disorder.

Jayanti Chotai at Umeå University in Sweden is working on the broader question of how the activity of genes that code for certain mood-related neurotransmitters might be altered. He and his colleagues were particularly interested in three variations of common genes: the tryptophan hydroxylase gene (TPH), which limits synthesis of serotonin and has been suggested as a cause of unipolar depression and suicide; the serotonin (5-hydroxytryptamine) transporter gene (5HTT), which has been linked to depression and bipolar disorder; and the dopamine D4 receptor gene (DRD4), one version of which has been linked to psychosis.

Chotai figured that if there is a particular gene variant that predisposes a person to a mental illness, and something about the season of that person's birth or gestation switches that gene on, then you will find an overabundance of that variant in patients with that particular illness born in certain seasons. In a study of 950 patients with either schizophrenia, unipolar depression or bipolar disorder, that's what he and his colleagues found. Men with depression showed season-of-birth effects for the serotonin gene - more were born in January - as did men with bipolar disorder for the TPH gene. Women with schizophrenia showed season-of-birth effects for the dopamine gene, with an excess born in October and fewer in April. Confusingly this is different from the well-established effect for schizophrenia itself - though this is hardly surprising since there is more to schizophrenia than one gene.

Chotai has also linked season of birth with levels of serotonin, dopamine and noradrenalin, and mapped them onto personality traits such as "novelty seeking" and "reward dependence" and even the tendency to be an early bird versus a night owl. He got some intriguing matches. People with low levels of the serotonin metabolite 5-HIAA, for instance, were more likely to have been born between February and April. So are people who suffer from mood disorders. Novelty seekers are more often born in April or May. Chotai claims that his findings show that season of birth affects basic personality characteristics by influencing the expression or activity of genes that modulate the neurotransmitters for mood, although he doesn't yet know what the key environmental factors might be.

There are many more unknowns, not least exactly when the crucial change takes place between conception and birth. Researchers studying the season-of-birth effect do not, however, see their work as a predictive tool, and certainly don't want to scare would-be mothers into altering their family planning. The reassuring fact is that most winter babies escape schizophrenia, and spring babies by and large do not grow up with eating disorders. What we can hope for is that we may some day understand the biological basis for this seasonal blip in the incidence of mental disorder. And if we can do that, we may then be able to fix it.

From issue 2588 of New Scientist magazine, 27 January 2007, page 40-43