• The Listener
  • North & South
  • Noted
  • RNZ

Finding the key to successful ageing

Midlife is the “black hole” of health research. But as members of the long-running Dunedin study turn 45, that’s about to change. 

Middle age. Our biological clock has ticked over from the summer of life into a darker, cooler autumn. Physically and mentally, we have passed our peak. In 20 years, it will be winter. We are 45.

From now on, our brains will start to shrink by about five per cent each decade. Our blood vessels will thicken, our blood pressure will increase and so too will our waistline. We’re already less fit than we were at 25, but now the speed of the decline will accelerate.

Apart from the obvious messages from the bathroom scales, we’ll be largely unaware of the extent and pace of the change within us. We might not worry too much about it while we still feel well. That awareness will dawn much later – way too late to change the trajectory of our lives.

Of the billions of dollars invested in health research, the bulk is spent at each end of our lives – what happens before birth and in childhood, and when we sicken and die, usually in old age.

But what is going on in our middle years? When do the subtle signs of the diseases that will eventually claim us first appear?

This year, the 1000 men and women who’ve been part of the Dunedin Multidisciplinary Health and Development Study since the early 1970s will undergo a battery of tests that will help to answer those questions. They are 45.

These investigations could not have been conceived of in the early years, when the first psychological and behavioural tests were done on the then three-year-olds playing with blocks in the Knox Church hall in 1975. They were simpler times, but in many ways, no less tumultuous. Whina Cooper led a land march from the far north to Parliament; Rob Muldoon led the National Party to an election victory. Jonah Lomu, Close to Home, and the Waitangi Tribunal were born.

This year, there will be the usual blood tests, heart tests, eye tests, hearing tests, dental tests and exercise tests. The study group will be questioned, over hours, about their social lives, their work lives, their love lives and their sex lives, their emotional health, mental health and financial health.

But for the first time, researchers will be getting inside their heads in a new way: each will have functional and structural MRI scans of their brains. It is possible a few might be showing the first signs of a degenerative brain disease they are unaware they have, says study director Professor Richie Poulton.

Study director Professor Richie Poulton.

The new data will provide a unique baseline of brain health for the group as they age. “Midlife disappears from a lot of knowledge bases. By the time a person gets old, those issues are not coming out of the blue. They’re a function of what happened really early, sure, but whether those early risks manifest as later risk is determined by what happens in the middle.

“The real engine room of what is happening could be the middle, but that’s kind of ignored, paradoxically, and we are in a great position to fill a huge hole in the literature.”

The Dunedin research, which has helped to transform international thinking about child health, could now have a similar influence on policy and treatment for people middle-aged and older, and further enhance the study’s global reputation and status.

In Why Am I?, a documentary series screened last year, eminent British professor Sir Michael Rutter – described as “the father of child psychology” – called it “probably the world’s most successful longitudinal study of a general community sample, ever”, making the Health Research Council decision to decline the study’s funding application in 2015 all the more inexplicable.

In 2014, Poulton, the study’s director since 2000, was named on a list of the top one per cent most-cited researchers in science. The 2016 Budget ring-fenced funds for the study, in large part to secure its future.

“This is a critically important study and it will continue to be important for the next 40 years,” says Sir Peter Gluckman, the Prime Minister’s chief scientific adviser. “It’s ahead of its time, because it’s continued to adapt to the science. Some of the most interesting stuff is now going to start to emerge as people age.”

Many other large groups of adults have been scanned over the years, but none whose lives have been so well documented since birth. “It has so many dimensions – emotional, biological and social. That’s its beauty.”

What the researchers will be able to do now, that others cannot, is to correlate the brain scans with the life course of the study participants. One of the tasks – a scratchy card-type game in which a correct guess can earn $5 – fires up the ventral striatum, part of the brain’s reward system; the test will measure just how strongly. Dysfunctional activity in the area is linked to impulsivity, poor financial planning, drug abuse and obesity.

The study has tracked the group’s financial planning decisions for decades, says Poulton. “We know who owns their own home, who has a retirement savings account and who has frittered away their money, so we’ll be able to do a valid assessment of whether anything about that area of the brain is also linked to real world outcomes and has implications for policy.”

The scans will measure memory, executive function, the fight or flight response in the amygdala, and the brain’s reward pathways. “All of these matter greatly for life success.”

Does adversity early in life change the function, structure and connectivity of the brain? Do these changes impact on the ability to prepare for successful ageing – or make us age at a different biological rate? Are the brains of successful, stable, emotionally positive midlifers differently wired or structured to those who’ve fared worse? If they’re addicted to drugs and alcohol, is it linked to over-active reward pathways? Do they have higher executive function – and is that why they’ve been able to put a handbrake on otherwise harmful behaviour or dangerous risk-taking? How good is their memory, the bedrock of the brain?

The study has published more than 1200 papers since it began, a number of which have challenged traditional thinking and led to health and social policy changes reflecting the importance of interventions early in life (see World Famous Findings, page 42).

Its later work, however, is revealing new insights in physical as well as psychological health in later years. In a 2015 paper, ranked by Science News the fourth most important scientific discovery of the year, the research team found the “biological age” of the participants at the age of 38 ranged from under 28 to 60, based on the way their bodies were functioning. The magazine said the work “tapped into the mystery that has long captivated scientists and the public”.

The 18 tests used to determine biological age included cardiorespiratory fitness, waist-hip ratio, blood pressure, body mass index, gum health and cholesterol. Study members who were biologically older perceived themselves to be in poorer health, and when photos were shown to university students, the people who were biologically older, also looked older.

The study’s associate director, North Carolina-based Professor Terrie Moffitt of Duke University, says it challenges the traditional theory of individual diseases and causes of death. “The theory is that as people age, one individual will have cardiovascular problems, one will have lung problems, one will have brain problems… and that each of those diseases needs to be studied one at a time for cures and treatments.”

The hypothesis of the newer field known as geroscience is that our organs age at the same pace, meaning a therapy or drug treatment could slow the process “with one silver bullet”, says Moffitt. “It’s an innovative way of thinking about ageing as the cause of disease. The potential is really science fiction, that one day, young people could take some kind of therapy in their 20s and 30s before their organs start to age.”

Duke University assistant professor Dan Belsky

The lead author on the pace-of-ageing paper, Dan Belsky, an assistant professor of medicine at Duke University, says although it was regarded by some in the media as an answer, “really what we found was a question. We were able to generate proof of concept that young, apparently healthy people are ageing at different rates, but precisely what those differences look like is a work in progress.”

The data being gathered at age 45 will give a better understanding of what the trajectories look like, and a key focus will be resilience. “How is it that some people experience adverse events in life – be it a car accident, infection, a job loss – that lead to an accelerated path of decline whereas others bounce right back?”

Belsky says it’s possible some people have a physiological resilience that is key to their successful ageing. From the new data, he says, they’ll be able to chart some of the dips and recoveries in the ageing rates of study members, “and that’s something we’re all very excited about”.

Logically, of course, it’s easy to think that different rates of ageing are caused by what we do to ourselves through our lifestyle decisions, rather than what our bodies do to us, but Belsky reckons it isn’t that simple. “It could be the case that when we observe a person in their 20s and 30s, they are exercising and eating healthily, but if they grew up under different circumstances, we could see a very different trajectory.”

He says we “presume” much of the variation can be attributable to behaviour, but it may be more important what behaviours occur when.

“What we are learning about the biology and physiology of ageing is that exposures that accumulate from very early life shape the trajectory of ageing from midlife to later life.”

Cardiovascular risk factors, although they don’t usually result in illness until midlife or later, are also detectable much earlier, says Dunedin researcher Dr Moana Theodore. She led a study which found that study members who had raised blood pressure at 38 had a number of risk factors in childhood, including being male, having a family history, being first born, and having lower birthweight. Those who had higher blood pressure were also more likely to have higher cholesterol levels. Higher body mass index was unsurprisingly associated with higher blood pressure, but the effect was particularly pronounced for those already in the high-risk group. “If you can identify things that can be changed, you can try to intervene younger.”

What the researchers don’t do, however, is intervene when they pick up risky results in study members. Unless there is “clear, obvious and imminent threat to their wellbeing”, there’s a policy of no feedback, says Poulton.

In blood pressure tests, the researchers don’t advise them of the results, unless their systolic measure is 180 or higher. Intervention at lower levels risks skewing the results. “We’re trying to capture the full range of vulnerability – normal, high, higher still. There are a lot of people walking around with a blood pressure of 140 or above who aren’t on medication and doing fine. They’re higher than normal clinical thresholds, otherwise you’d be intervening 20-25 per cent of the time because a lot of people as they age are heading to these points. Study members know this is the deal.”

Merely being in the study doesn’t seem to have altered the members’ lifestyles, despite the fact that logically, they’re probably more aware of what they should or shouldn’t be doing.

“I think people have gone completely down the wrong path on this,” says Poulton. “I know how important it is to do exercise. And to eat less than I do. I’m stacked to overflowing with that knowledge but do I do it? Do I do the right thing? Most of the time I do, but sometimes I do not. It’s incredibly difficult to get people to do the right thing, so thinking you can make a serious difference in how people act by interviewing them for eight hours every six years… it just doesn’t work like that.”

North Carolina Professor Terrie Moffitt

Moffitt says the age-45 assessments will add new tests to measure genetic changes in the immune system, which modify the function of DNA.

This should enable the researchers to better explain how childhood adversity – for example, abuse, neglect or losing a parent – causes life-long physiological changes that damage health. 

A 2007 Dunedin paper showed children who’d been maltreated in the first 10 years of their lives had clinically significant biomarkers of inflammation as adults. Chronic inflammation makes people more vulnerable to many conditions, including heart disease, cancer, diabetes and arthritis. An earlier paper from 2002, about growing up socioeconomically disadvantaged, showed that even if study members overcame that background to make a success of their lives, the damage to their cardiovascular and oral health remained.

“If you get exposed to those things in your childhood, it’s very hard to undo the damage,” says Poulton. “The consequences are long-lasting and far-reaching. When you’re young, certain ‘insults’ seem to rewire or reconfigure your hormonal systems, the way your physiology works, so that much further down the track it shows up as a much higher risk of developing physical health problems.”

“Psychologists and social scientists have for years published papers saying children who’ve lived with adversity develop depression or a drink problem and everyone would go, ‘Oh yawn,’” says Moffitt. “But once you said children who lived with adversity develop elevated inflammation markers that suggest 20 years from now they’re going to be first in the queue for a heart attack, policy-makers sit up and say, ‘We have to do something about childhood adversity.’”

Moffitt says brain imaging in less well-designed studies has reached misleading conclusions because they lacked context and background information. “There are lots and lots of studies that have claimed that early childhood adversity, low social class and lack of money causes poor brain health and poor brain development. But they do that research by getting 10-15 people who are in a depression clinic and asking them to recall whether they had an abused childhood. They put them in a scanner and come and say, ‘Look, childhood makes something go wrong with their brains.’”

They’ve been unable to say, for example, whether a pre-existing brain deficit might have contributed to child abuse. “Children become victims of adversity because adults prey on children who they think are vulnerable, and children who wouldn’t be believed if they told.”

She says much of the value of the Dunedin work has been in debunking “some pretty unscrupulous areas of science that were making wild claims based on retrospective analysis”, for example several recent studies that claimed obesity caused brain damage.

“Our data showed those who had poor brain health at three were more likely to have gained more weight as they grew older, so brain health was the cause, not the consequence. Preventing obesity won’t make everyone smarter.”

For Otago University neuroscience researcher Professor Cliff Abraham, co-director of Brain Research New Zealand, the back story the Dunedin study provides, along with blood samples stored since participants were 26, is invaluable. Abraham is investigating biomarkers in Alzheimer’s disease and says the blood samples from study participants will be checked against any that are found. “You can say, do these biomarkers appear much earlier? Are there predictors at age 45 and are there things about early life, education, drug abuse or whatever that might impact the presence or absence – or degree of presence – of a biomarker?”

Otago University neuroscience researcher Professor Cliff Abraham

But the Dunedin study is not just about biology and biomarkers. Its wide-ranging interviews have given candid insights into practically every part of the subjects’ lives, making the research findings “a manual on how to live a happy, healthy life”, says Otago’s head of psychology Professor Mike Colombo. “And we all want access to that. For some, we might have the knowledge too late, but that doesn’t mean you shouldn’t accumulate the knowledge. Because there is another generation behind them and another one after that.”

More than three-quarters of the Dunedin study members are parents – the first had a child in 1987 at the age of 15. While the women in the group are reaching the end of their childbearing years, some of the men are still to become fathers for the first time. The offspring now form part of the Next Generation study, which will examine how attitudes, lifestyles, health and behaviour are transmitted from parents to children.

The children are being tested at the age of 15, and 300 have already been interviewed, says Next Generation study leader, Associate Professor Bob Hancox, and the results show they’re living “very, very complicated lives”.

“I think we still walk around with the notion of mum and dad and two or three kids, and recognise there are a lot of blended families when there’s been a relationship breakup. But we think of it as a once-only event.” The reality is much more complex. “Things keep changing for some children.”

A study published last year of 200 of the children born between 1991 and 1995, found they had an average of eight changes of address by the age of  15, and an average of eight different people had left or joined the family in that time. Only a quarter were still with both biological parents, while nearly 60 per cent were in a sole-parent family or in some form of “multiple resident” care. “Things change rapidly in these children’s lives – much more than we recognise in official policies or when we talk about family responsibilities.”

Hancox doesn’t want to label this as dysfunctional, saying some of the changes could be positive, and others negative. They could also be related to the fact the first children to reach 15 were born to relatively young parents.

What he does want to know is whether the changes make children more resilient and self-reliant, or predict problems ahead. And, if it’s the latter, what could break the cycle of risk?

The work has important implications for welfare and other policies, which assume there is relative stability in children’s lives. “We assume families look a particular way, and the reality is, they don’t.” One example is Working for Families, which requires applicants to be the principal child carer and to notify Work and Income whenever circumstances change.

Another arm of the Next Generation study has examined whether parenting styles are passed between generations. It’s found, probably unsurprisingly, that women, but not men, tend to parent the way they were parented.

The lead author in that paper, Californian child development expert Professor Jay Belsky (father of Dan) says it’s possible information about parenting in our childhood largely comes from mothers. “Another possibility is that, especially in the generation we are talking about but less so today, if you grow up as a little girl you expect to be a mother someday. It’s part of your script, implicitly if not explicitly, so maybe the parenting they get exposed to sinks in more deeply. Boys, especially of that generation in the early 1970s, didn’t grow up thinking, ‘I’m going to be a daddy someday’, so in a sense mothering might be more scripted, socially, and more subject to these kinds of influences than fathering is.”

Belsky, in research which was published in 2012, also examined whether the trans-generational influence of parenting styles weakens when people have children later in life, in their 30s, therefore putting more distance between their upbringing and parenting. Instead, he found the influence was almost as strong.

He describes the research as only the first step in understanding how future generations will function as parents, which he hopes will ultimately help to find ways of breaking cycles of violence, neglect and emotional abuse.

It takes a lot of time to study people well inter-generationally, Belsky says, and that’s where the Dunedin work has been so valuable. “Most of the [other research]work has been done quickly. We get a bunch of adults who are abusing their children and a bunch who aren’t and interview them about how they were raised. This is the retrospective fallacy. When you do that, you might find the ones who are abusing children are five times more likely to have been abused in their childhood. That’s studying today and asking about yesterday.”

The Dunedin study enables researchers to see what the children’s lives are like prospectively, in “real time”. “When you do that, you get a very different story. I have an identical twin and I’m sometimes amazed at his characterisation of my now deceased parents, because it’s different to mine. Our memories and recollections are selective, distorted and fundamentally fallible.”

Good parenting has been associated with many factors, perhaps most intriguingly, a good romantic relationship. “You can think of a good marriage as what some would call an emotionally corrective experience; it heals the wounds of a dysfunctional childhood to some extent.”

So what of the all-important love and sex lives of the Dunedin 1000, then? Judging by the research so far, for most study members, they’re pretty good.

Associate Professor Nigel Dickson

Associate Professor Nigel Dickson, who’s been studying the sexual and reproductive health and behaviour of the group since they were 18, says the data collected at age 38 found around 85 per cent of the men and women were in a relationship with someone of the opposite sex, and more than 60 per cent of these were marriages or de facto relationships. About half had been in the relationship for 10 years or more, and nearly 80 per cent for at least five years.

“I don’t think that’s terribly surprising,” says Dickson. “I think most relationships are lasting, although you hear about those that don’t.”

More interesting, perhaps, is the data on the quality of the relationships. About 85 per cent of the men and 80 per cent of the women described them as “very” or “extremely” physically pleasurable. While most participants were also very emotionally satisfied in their relationships, 20-25 per cent of men and women were in relationships that were only slightly, or not at all, emotionally satisfying, indicating that some people can still experience physical pleasure in a relationship, without a strong emotional connection.

Dickson was taken aback at how few participants identified as gay or lesbian – less than two per cent – and the amount of fluidity in reports of sexual attraction to the same gender. He does think, however, that the idea that 10 per cent of people identify as gay – a figure touted since Alfred Kinsey’s research in the 1940s and 1950s – is a “gross exaggeration”. Only four men (0.9 per cent) and six women (1.3 per cent) said they were in current same-sex relationships. For three of the men and all the women, these had lasted a year or more.

Nearly a quarter of the Dunedin women, although regarding themselves as mainly heterosexual reported having “some” sexual attraction to other women. Only five per cent of the men reported some same-sex attraction.

While the confidentiality accorded the participants means their responses are believed to be honest and accurate, Dickson acknowledges the possibility that some might not want to divulge their sexual preferences.

Men tended to be more disapproving of sex between men, with about a third saying it was “always” or “mostly” wrong, compared with only 10 per cent of women. When asked about sex between women, however, only 10 per cent of the men thought it was wrong.  It’s possible the influence of porn played a part. “Hetero men probably don’t want to watch men having gay sex.”

At age 38, nearly a quarter of the Dunedin study members reported infertility problems. A quarter of the women and nearly a third of the men had not had a child. Of those, half said they still wanted to become parents. Dickson says the age 45 assessments will give a clearer picture of fertility. “We should be able to answer questions about the impact of that delay on your chances of having children.”

Despite the ostensible stability the relationship data reflects, the study group findings reflect the country’s high rate of mental disorders, most commonly anxiety and depression. Staggeringly, the interviews taken at intervals between the ages of 11 and 38 revealed just 171 of the 1000 had never reported having a mental disorder at some point.

“I think it’s incredibly surprising,” says Moffitt. “We bet among ourselves they would be smart and rich if they’ve never had any mental disorder. But they are not. We have a mystery here. They didn’t have better parents – some had been maltreated or lost a parent in the early days. They just seem to be really emotionally sturdy.”

She says researchers initially thought the “stable” people weren’t responding honestly, or were trying to portray themselves “as better than they are”, but interviews with people who knew them well confirmed the findings. “They agree these people had never had a problem with drink or depression, never talked about suicide and never seemed anxious.”

Moffitt and her husband and fellow Duke University researcher, Avshalom Caspi led a study that helped explain some of the mystery in a ground-breaking but controversial paper from the Dunedin study, published in 2003, which showed that a particular gene variation was associated with a greater risk of depression in people who’d experienced similar adverse life events. They found people with the short version of the serotonin transporter gene were more likely to become depressed than those with the long version. The findings were voted the second most important scientific breakthrough in the world – in any branch of science – in 2003.

Moffitt says the finding was one of the most exciting moments of her 30-plus years with the Dunedin study. “I was standing at the computer when the analysis came up on the screen – my husband was doing all the programing. We’d had a hunch, but just to see it there was so amazing, and it was amazing because the vast majority of what we find is something your grandmother knew – like smoking is bad for 13-year-olds, or women with breast cancer tend to get depressed, or you’re better prepared for retirement if you buy a home early. We can show all these things empirically but people go, ‘Didn’t we already know that?’

“But this was a thing that nobody already knew. I’ll never forget that day. Both of us were completely over the moon. We were in London, at King’s College, and we phoned Richie in the middle of the night in New Zealand.”

The Dunedin study had been one of the first in the world to start collecting DNA, in 1996, when the technology was still in its infancy. The finding was their vindication. “We said, ‘Do you remember all that DNA we collected and how people made fun of us at the time?’ They said, you don’t even know what you’re going to use it for, and it’s unethical if you don’t know what you’re going to do with it,” recalls Moffitt.

“The idea of getting DNA from a group of ‘free range people’ with really nothing wrong with them was quite remarkable at the time. We just knew it was going to pay off.”

In 2014, Moffitt, Caspi and Poulton made another important contribution to mental health research, co-authoring a paper that described what they called the “p factor” as putting people at greater risk of illness.

Says Poulton: “It showed you can boil down mental illness into a core vulnerability, where neurodevelopment in the brain has not been ideal. From that spring two or three more granular breakdowns. One is a tendency to be depressed and anxious, another is the tendency to act out, and another is a tendency to develop extreme forms of thought disorder – schizophrenia, for example.

“People have spent a lot of time trying to parse up anxiety disorders into various types, but it looks like they all derive from a substrate of extreme excitability, or over-active fight or flight response, and what form it tends to take is conditioned by the experiences you have as you grow up.”

The five or 10 per cent of people with the deepest level of vulnerability, says Moffitt, might start off with anxiety or ADHD, develop conduct problems and then become depressed and later develop an even more serious disorder.

“Then you have people who stay in the ‘shallow end’ of that dimension and they have separation anxiety from their parents when they’re small, phobia in their teens, then get depressed, then have a drink problem, then get depressed again and then have generalised anxiety. They never develop a serious disorder but always have rumbly stuff going on internally.”

In February 2015, Poulton was appointed chief science adviser in the Ministry of Social Development. The job allows him to take the Dunedin study findings out of medical journals and into policy papers.

He says the research can help governments accurately predict, from an early age, the 20 per cent of the population that ultimately accounts for 80 per cent of the country’s health and social costs. Policy-makers refer to it as the “seven Datsuns in the driveway” phenomenon because of the number of government agencies interacting with one family. The work is helping to answer big social questions, about the interplay of physical and mental health, and how societal issues “get under the skin to cause physical rot”, and implementing the knowledge gained to talk not about individual clients, but population wellbeing.

“This is what Prime Minister Bill English describes when he talks about the social investment model,” says Poulton. “You take the presenting population today and segment it to identify the highest cost group over the long term, then throw all your resources into them to reduce the forward liability to the government and taxpayers. My dream would be to see when I retire, and before I croak, how they spend that money being guided by the research that we and others have produced. That would give me a real kick.”

Given the important impact of childhood poverty on future health, and the Dunedin research which has emphasised the effects last well into adulthood, the government’s relatively supine response to the issue is even more difficult to understand.

But Sir Peter Gluckman says fundamental change is happening and the Dunedin study is helping to inform that, bringing the scientific evidence driving the thinking behind initiatives such as the new Ministry for Vulnerable Children. “Bill English has had half of Wellington working on social investment, which is all about addressing the consequences of a poor start to life.” 

For Poulton, it’s about selling the evidence to back the policies “at a moment in time. To get the vitality, the reaction, the air, you have to talk their language. I don’t talk about life course disadvantage, I talk about forward liability and everyone listens and then you put data on a screen and they go ‘Oh, my god.’”

When policies change as a result, the credit will be largely due to the unique legacy of 1000 people born in one Dunedin hospital 45 years ago. We know almost everything about them, but, because of the study’s strict rules of anonymity, practically nothing at all. It’s these men and women, Poulton says, who are the real heroes.

Nature via Nurture?

If any research has debunked the nature or nurture argument, it’s the Dunedin study, by showing it’s not an “either-or”, and explaining how previously unrecognised interactions between the two can shape our destiny.

Since the sequencing of the genome, scientists have been able to trawl through hundreds of thousands of genetic markers to find out if combinations of variants make people more or less susceptible to some physical or psychological traits.

These genome-wide association studies have thrown up clusters of variations associated with conditions as diverse as macular degeneration and heart disease. In 2014, researchers from King’s College in London identified variations which in combination were linked to educational achievement, intelligence and family socio-economic status.

Last June, a provocatively-titled paper, The Genetics of Success, from the Dunedin study found higher “polygenic scores” – those with more of the variations – predicted adult success even after adjustments for educational achievements.

Scores followed a bell curve range – a few had many of the variations, a few had very few, and most had scores in the middle. Those with the highest scores became more successful, not only in their schooling, but also in their economic and professional lives. The success depended only partly on their educational achievement. While children with higher scores had grown up in families with higher socio-economic status, “even for children born into socially disadvantaged circumstances, higher polygenic scores predicted upward social mobility”.

Children with higher scores were more likely to say their first words at younger ages, quicker to begin communicating using sentences and stronger readers. They performed better in IQ tests and rated more highly for being friendly, confident, co-operative and communicative, but didn’t reach physical developmental milestones ahead of their peers and were not physically healthier.

Adolescents with higher scores had higher aspirations as secondary school students and were more likely to seek work overseas. At 32 and 38, they were better at managing money. While all this sounds like a recipe for life success, higher scores didn’t mean people were more satisfied with their life.

The paper’s lead author, Duke University-based assistant professor Dan Belsky, says the genetic measures explain just a tiny variance in outcomes in people’s lives. “Nothing we have done suggests anyone is doomed before they start.”

He’s now looking into those children with very low scores who bucked the trend and achieved highly. “We are trying to understand what these apparent errors in genetic prediction might look like and what they mean. It could mean there are other genetic factors that are very helpful. Alternatively, the more exciting prospect is that we will discover, in examining those individuals’ lives, that there are resources that have been made available to them, or opportunities they’ve had that enabled them to achieve success through other means.”

Belsky, the father of children aged two and nearly six, says he wouldn’t get them genetically tested for these scores. “The kinds of predictions we can make from genomes are simply not accurate enough to be informative in that way.”

From the age 45 data, he’ll be trying to better understand why people who are socially successful often live longer, with better physical health and less cognitive decline.

Dunedin molecular geneticist Professor Stephen Robertson says information from the DNA samples should become more valuable as the cohort ages, to help give more personalised advice. “I think people want to know what their individualised risk factors are. I don’t really want to hear that I should lay off alcohol and smokes, go for my run every day and eat whole grains. I would rather know that in fact I can look at my family history and know I’m unlikely to die of a heart attack because no one in my extended family has, that I’ve got a touch of cancer there I need to keep an eye out for and dementia doesn’t run in my family. Surely that information, when rendered more precisely, is going to help me lead a more educated life about what risk I expose myself to, rather than the topdressing approach where we should all have an even smattering of puritanism to keep us healthy.”

Robertson says Poulton is always challenging him to come up with ideas for ways in which the Dunedin data can be used to find more gene-environment interactions. Instead of scanning the genome for common variants linked with disease, the data from the study enables researchers to turn that process on its head and search for those linked to phenotype, or characteristics.

“Even though we can sequence genomes, the scale and biological relevance of the variation we discover is still something we are struggling to comprehend. In a small cohort, you run the risk of actually swamping your potential signals by trying to measure everything in that category, when the vast bulk of it is not going to be relevant. You squander your opportunities simply because you consider too many irrelevances to spot the gold nuggets statistically.”

But the Dunedin team can take a variant they believe is relevant and ask which characteristics measured since day one correlate with that genetic factor.

Robertson says he wants to understand what makes people “susceptible to things that lead to them crumbling at the edges”.

Loss of cognitive function probably worries people more than physical frailty, and “we don’t understand how to control the mental decline”.

 Thirteen Dunedin findings that made headlines – and a difference. 

  • Having wider retinal vessels is linked with lower IQ scores in childhood.

  • Criminal offenders fall into two main groups – those whose anti-social behaviour is limited to their adolescence and others who are life-course offenders. Life course (or persistent) offenders, were more likely to show problems soon after birth or during their pre-school years, and were difficult to parent, easily stressed and, even at an early age, likely to take out their distress on others. Adolescent-limited criminality was linked to the “maturity gap” between their physical and mental development. 
  • Self-control in childhood is more important than socioeconomic status or IQ in predicting adult physical health, wealth, life satisfaction, addiction, crime and parenting of the next generation. 
  • Boys maltreated in childhood are more likely to turn to violent and antisocial behaviour if they have a weakened version of a gene that controls production of an enzyme called MAOA. MAOA breaks down key neurotransmitters linked with mood, aggression and pleasure. Abused children with normal MAOA genes were no more likely to develop behaviour problems than those who weren’t abused. 
  • Injury research from the study was instrumental in ensuring that thermostats were introduced into hot water cylinders, the length of electric jug cords being shortened and safety mats being laid under playground equipment. 
  • Young, heavy users of cannabis had increased risk of psychosis, and cognitive declines culminating in the loss of eight IQ points early in midlife. The likelihood of a psychosis developing in adulthood depends on a particular “vulnerability” genotype. 
  • People exposed to childhood violence show accelerated erosion of their telomeres, which cap and protect the ends of chromosomes. Shortened telomeres have been linked with more rapid ageing. 
  • Not only do children in poverty suffer from health issues at a greater rate than their peers who do not live in poverty, but the ill-health has lifelong effects. This is true even for those who spent their early years in poverty but were not poor as adults. Blood tests showed that study members who were abused or neglected as children had the highest levels of inflammation, making them more susceptible to conditions such as depression, arthritis, heart disease and cancer. This created an evidence base for new initiatives to use anti-inflammation medication for treatment-resistant depression. 
  • People who had suffered similar life adversities were more likely to suffer depression if they had a short version of the serotonin transporter gene, rather than the long version. In 2003, this was voted the second most important scientific breakthrough – in any branch of science – in the world. 
  • Of apparently normal children who reported delusional beliefs and hallucinations at age 11, a third were later formally diagnosed with schizophrenia-type symptoms, suggesting the disease can be identified much earlier than currently. 
  • Domestic violence is fairly evenly split between women and men, although women are more often injured. This research resulted in the reversal of legislation in the US that provided government funding only for mandatory treatment of men and excluded couples therapy. 
  • The width of blood vessels in the eye may predict brain health years before the onset of dementia and other deficits. Having wider retinal vessels was linked with schizophrenia and lower IQ scores at age 38 – and in childhood. The work has implications for better early diagnosis, and perhaps innovative treatments. 
  • Childhood and adult ADHD may not be the same disorder. This has led to calls to review treatment policies, particularly regarding medication.

  • Study members with more brothers and sisters are less likely to have asthma and allergies than only children. Children brought up on farms, and those in urban homes with pets, are also less likely to develop allergies. It’s consistent with the hygiene hypothesis – the amount of dirt, or microbes, a child is exposed to helps protect them.

Star Survey had Small Beginnings

Dunedin study founder Phil Silva didn’t think the research would last 45 months, let alone 45 years.

When he launched the Dunedin study in 1971, Silva never dreamed the participants would still be having assessments more than four decades on. “We had no staff, no accommodation and no money. At first, it was ‘let’s just survive for a year’.”

He’d snared $1500 in funding, and relied on the goodwill of volunteers, from receptionists and data processors to medical graduates and professors of psychology. “We thought, ‘This is a crazy idea. We can’t do it. It hasn’t been done before. But well, we’ll try it.’”

The makeshift group were housed in the old (condemned) Presbyterian manse and Sunday school hall at Knox Church.

Silva had come up with the idea of a prospective study after being asked to examine a group of 250 babies with perinatal problems who’d been subject to high-tech interventions in the late 1960s. At the time he was a psychologist in the Department of Education, but soon joined Otago University as a scientific officer in the Department of Paediatrics.

“The big thing to come out of it was that there were an awful lot of kids with problems – vision problems, hearing problems, language problems – that went undetected. It was too small a number to study properly, and not set up to do that, so we decided we needed a new study.”

The babies in the Dunedin study, born at Queen Mary Hospital between April 1, 1972 and March 31, 1973, weren’t examined at birth, and nor were their parents; the first real testing took place at age three, when 1037 children were enrolled in the study. That 1000 remain in the study group – 37 have since died – is one of the abiding achievements of the research team.

Silva says with the benefit of hindsight, the children would have been tested and their parents interviewed before age three, “but there are limits to what you can ask families to do”. He also wishes there were better and more reliable measures of nutrition. “I also would have liked to have known much more about the development of the study members’ understanding and use of money.”

He says the latter was something then-Health Minister George Gair, who took a keen interest in the study during his tenure in the 1980s, was interested in finding out. “He said it wasn’t a health imperative, it was a life imperative.”

Silva says he never got the money to allow the research to be done, but he’s sure it will resurface, as study members reach retirement age.

He’s justifiably proud of the study’s achievements, some of which are largely unrecognised. For example, the Well Child health and development record, which replaced the height and weight-focused Plunket book in the early 1990s, was largely the result of the Dunedin work.

North & South asked Silva to rate various life influences on a scale of 1-10 according to their importance to our life course. Here are his rankings: 

  • Wealth – or lack of it: 6
  • The genetic legacy of your parents: 7
  • Learning difficulties such as dyslexia, hearing or vision impairment: 7
  • How your parents raised you: 8
  • Schooling and teachers: 8
  • Luck: 8
  • Physical or mental abuse: 9


This was published in the February 2017 issue of North & South.

Follow North & South on on Twitter, Facebook, and Instagram and sign up to the weekly e-mail.