Searching for patterns that can save lives

It is not fair. Some people can smoke like a chimney all their lives, while others die from lung cancer despite never smoking. Although statistics rarely do justice to each individual, the current use of big data and registry research is generating some of the greatest revolutions in the treatment of people with diseases. Preben Bo Mortensen’s fascination with numbers and patterns, and his ability to ask the right questions, has resulted in him playing a key role in treatment for people with mental disorders worldwide.

“In the late 1990s, I got to be part of a committee, chaired by my co-recipient of this Prize, Merete Nordentoft, to come up with an actual national action plan for preventing suicide. To inform the committee, we did a study that showed that severe mental illness is a really dominant cause of suicide and that there are particularly vulnerable periods. This led to more systematic suicide prevention and that intervention at least coincided with a drop in Denmark’s suicide rates,” explains Preben Bo Mortensen.

The link with schizophrenia

Already from a very young age, Preben Bo Mortensen was deeply fascinated by how people function, so he decided to study medicine, but the high degree of rote learning required was not exactly his cup of tea.

“It was a bit like reading telephone directories: a stack of them every term, learn it by heart, throw it all away, take the next term, repeat, repeat. And, honestly, it was a bit boring. So that drew me towards research.”

During medical school, Preben Bo Mortensen also realized that getting a job would be difficult because unemployment among doctors was high at the time. So he decided that doing some research might improve his chances and happened to end in a situation that later created his career path.

“I stumbled into something called a prize thesis at Aarhus University. I had no idea that such tasks usually targeted a specific researcher, so I just gave it a shot, and I was lucky, because the researcher who was targeted decided not to do it. This became my opportunity to get into a whole new and exciting field.”

The topic was to study whether individuals diagnosed with schizophrenia, counterintuitively, have a lower risk of developing cancer than other people.

“It was like a total revelation. For the first time, I thought, during medical school, I was asked to think, come up with ideas, pursue goals I found interesting, read about topics that I chose because I thought they were useful and so on. And usually when people ask me: ‘How did it start?’ Well, I got into research while I was a medical student and then I sort of never left.”

A registry-related revelation

Preben Bo Mortensen not only won the Aarhus University gold medal for his thesis but also documented a special link. Today, cancer still stands out among people with schizophrenia as one of the few causes of death that is not hugely elevated. The reason is still unknown, but this work had triggered Preben Bo Mortensen’s interest in this field and made him realize that he was crazy about numbers.

The gold medal thesis also led to a position at the Department of Psychiatric Demography of Aarhus University led by Annalise Dupont. Denmark had introduced universal social security numbers (Denmark’s Civil Registration System) in 1968, and Annalise Dupont’s task was to create an electronic Psychiatric Central Registry based on the social security numbers. The Registry recorded data electronically already from 1969 – laying the groundwork for Danish epidemiological research on psychiatry to later take a leading international position.

“For my gold medal thesis, I was given the opportunity to work on a cohort going back 30 years with more than 6000 people with schizophrenia. At that time, I think I sort of just took these registries for granted. I got many questions: ‘Oh, are you still sitting up there watching numbers?’ and ‘Are you ever going to get out into the real world?’ And I certainly intended to start focusing on clinical work, which I actually also enjoyed very much. But Annalise Dupont told me that I should go out and see how they do stuff in other places before I decided whether to leave research altogether.”

Preben Bo Mortensen got a WHO fellowship and travelled to prestigious Johns Hopkins University in Baltimore, Maryland, where he was asked to give his first lecture.

“I was really starstruck and was feeling so guilty about 10 of the 6000 people with schizophrenia that we could not follow up. And then, after I had given my lecture and heard the reactions, it started dawning on me that maybe at least Annalise Dupont had been right, that we had something unique. So that was the first time I really realized that there is something totally unique here, and that was a new revelation for me.”

Strange seasonal phenomena

The potential of registry-based research was soon realized. In 1993, Preben Bo Mortensen published landmark results in the British Journal of Psychiatry, reporting, for the first time, statistics on mortality and causes of death for people diagnosed with their first episode of schizophrenia. The figures showed that people newly diagnosed with schizophrenia had a very highly elevated risk of suicide, especially in the first year.

“But we actually also saw more than a 50% increase in this already very high risk of early suicides during the 70’ies and 80’ies when half of the psychiatric beds had been closed down, so we speculated that the increasing suicide risk indicated some adverse effects of deinstitutionalization, which was also an early indication that something could be done about this problem.”

The study also confirmed markedly elevated all-cause mortality among people with schizophrenia and higher death rates from natural causes, except for cancer and cerebrovascular diseases.

At about the same time, Preben Bo Mortensen had the good fortune to meet two United States researchers, E. Fuller Torrey and Robert Yolken. They not only could see the potential in his work on registries but could also, with funds from the Stanley Medical Research Institute, enable new, more ambitious projects looking more closely for patterns related to who developed this chronic and severe mental disorder. This required a struggle, but the big breakthrough came in 1999.

“The most well-established risk factor for schizophrenia is to have a parent or sibling with the disorder, and ideally it would be great to be able to construct a psychiatric family history for everybody living in Denmark to get the denominators right. That is a tall order. I had been thinking that we could use Denmark’s Civil Registration System to actually link with relatives and then look up the relatives in the Psychiatric Central Registry, but back then getting these data was also extremely expensive. Then I was at a meeting about something related to registries where Mads Melbye from Statens Serum Institute was present and he had already done that, so we made an agreement that I could pay for a statistician who could work with him on their data.”

For the first time, the research could begin to quantify in a total nationwide population what had been said to be the causes of schizophrenia, and the researchers demonstrated the influence of genetics but also suggested that environmental factors have an unexpectedly large role.

“Although a history of schizophrenia among a parent or sibling is associated with the highest relative risk of developing the disease, curiously, we also showed that growing up in the city led to a twofold elevated schizophrenia risk. And people born in late winter or early spring also had a significantly higher risk.”

A very direct suggestion

At almost the same time, Preben Bo Mortensen joined a committee chaired by another expert in mental disorders, Merete Nordentoft. The task of this committee was to develop an action plan for preventing suicide.

“It was really curious that many experts in this field were saying: ‘This population group is very important, and it’s the young women, it’s the immigrants or it’s something else.’ At that time, no study in the world had been able to, at the same time, look closely at risk-factors such as mental illness, income and employment and demographic factors such as whether you are married or not, whether you have children or not.” This was necessary to get some idea of the relative importance of the risk factors.

To optimally inform the suicide prevention committee, Preben Bo Mortensen and his team performed a registry-based study that became the largest and most thorough ever. The study showed very clearly that mental illness is the predominant contributing cause of suicide in the population.

“However, it is not like you get a diagnosis and then you are doomed to a very high risk. There are phases: sometimes you feel better, sometimes you feel worse. You often go to a hospital when you feel worse, so actually suicidality increased more during hospitalization, but even more so just in the first days, weeks and months after discharge.”

The study, which was published in The Lancet, led to a very direct suggestion from the committee: that one aspect of preventing suicide should be more systematically asking people who are discharged from hospital whether they have suicidal thoughts but also closer follow-up just after discharge.

“It looks like this may have worked. Suicide rates stopped growing, but of course we will never know whether that was the cause. But I think that was probably the most direct example during my career where I thought that, just looking at the numbers, you can actually hope that this worked.”

Some hypotheses are wrong

The work of Preben Bo Mortensen and colleagues in building registry-based research infrastructure in Denmark really began to pay off in the following years. It resulted in many innovative studies, led by the many talented people working with him, proving existing hypotheses based on this infrastructure, whereas other hypotheses turned out to be wrong

For example, concern had been expressed about whether having an induced abortion potentially harms women’s mental health. Using Danish registry data, his team estimated the rates of first-time psychiatric contact for any type of mental disorder within the 12 months after the abortion or childbirth compared with the 9-month period preceding the event.

“We confirmed that the incidence of psychiatric contact was similar before and after a first-trimester abortion, so this did not support a hypothesis of an increased risk of mental disorders after abortion. Studies like this are important, regardless of whether they prove or disprove a connection. Ultimately, what we can do is give doctors and politicians new facts and possible tools to act on these facts.”

Treatment not effective enough

Other studies established important hypotheses about associations, such as an association between autoimmune diseases and subsequent risk of mental disorders, and the registry-based analysis also confirmed that mental disorders affect the subsequent risk of developing cardiovascular disease. People with severe mental disorders had a threefold risk of dying from heart disease 5 years after being diagnosed with it and also had a reduced likelihood of undergoing invasive procedures related to heart disease within these 5 years.

“Of course, these are just associations and numbers, but given the excess mortality and lower rates of invasive procedures, as well as fewer prescriptions of modern drugs for high cholesterol or blood pressure, we concluded that the treatment was not sufficiently effective. So we thought: maybe we can do something by paying more attention to whether we take blood tests and whether we actually try to implement state-of-the-art treatment.”

John McGrath, a good friend of Preben Bo Mortensen, found another possible association connected to the previous study showing that people born in the winter more often develop schizophrenia, suggesting that the reason could be that people in northern Europe lack vitamin D from the sun in the winter. An initial study did not find the expected relationships, but since the numbers were small and complex, the researchers did a larger and more rigorous study. This was possible because of another unique resource for research in Denmark: the Danish Neonatal Screening Biobank had been created by Bent Nørgaard-Pedersen and his successor David M. Hougaard at Statens Serum Institut, collecting and storing tiny amounts of blood on the filter papers used for neonatal screening of all children born in Denmark.

Preben Bo Mortensen started working with them, first on studies of early infections as risk factors for schizophrenia and then on John McGrath’s vitamin D study.

“That study, as well as a subsequent replication, showed that low concentrations of neonatal vitamin D are associated with increased risk of schizophrenia. Of course, you should always remember the low baseline rate of the disease, so most people will never be affected, but nevertheless we concluded that this could contribute to a sizeable proportion of cases in Denmark.”

The world’s largest study within psychiatric genetics

Similar to many of Preben Bo Mortensen’s other registry-based studies, John McGrath’s vitamin D study also stimulated the careers of many young researchers and kick-started new studies to understand how vitamin D status affects brain development and mental health. However, the many studies and associations linking mental disorders had also kick-started an idea in Preben Bo Mortensen’s head.

“I had had a dream for years of not only predicting and finding associations based on social and environmental markers, such as season, place of birth, nutrients and the like but also doing proper genetics. Of course, this was a moving target in 2000, because it was back when genomic technology was immensely more difficult and expensive and required higher input of DNA. Plus, I did not know anything about molecular genetics or laboratory work. Luckily, I knew people who did.”

In 2012, with the emergence of new genomic technologies, the time had come to realize that dream. With the largest grants ever awarded for psychiatric research in Denmark, the Lundbeck Foundation Initiative for Integrative Psychiatric Research (iPSYCH) was established by an interdisciplinary team of six principal investigators, including this year’s two Novo Nordisk Prize recipients. The purpose of iPSYCH is to find the causes and create the basis for improving treatment and prevention related to five of the most serious mental disorders: autism, attention-deficit/hyperactivity disorder, schizophrenia, bipolar disorder and depression.

“I believe that this is the world’s largest single study within psychiatric genetics, and for sure it is the only truly population-based one. In addition, we have this wealth of registry data covering multiple aspects of life for about 130,000 people as well as their relatives. The whole essence of the iPSYCH project is that we can actually measure something about the genetic susceptibility of individuals and combine this with many other things occurring to us from conception to the onset of disease.”

So far, the project has resulted in more than 740 publications. Progress in psychiatric genetics today is driven by international collaborations such as the Psychiatric Genetics Consortium, and iPSYCH has from its beginning been engaged in this. One of the first major successes was a 2014 article in Nature about schizophrenia-associated genetic loci, describing the largest genome-wide association study. The researchers analysed almost 37,000 cases of schizophrenia and identified 108 conservatively defined loci – 83 of which had not been reported previously. 

Distribute people in neat boxes

The population-based data from Denmark could be used to document the ability to begin to measure genetic liability meaningfully with biological data instead of being limited to using family history. Since then, the iPSYCH project has contributed importantly to the vast majority of the new major international studies unravelling the genetic underpinnings of major mental disorders.

Preben Bo Mortensen generally sees his work as an attempt to identify useful patterns in a complex web – a combination of biological building blocks plus the cumulative sum of life experiences. In his view, lifelong quarrels about genes and environment are therefore not very fruitful.

“Is this caused by genes? Is this caused by environment? I think this is a false dichotomy. Our knowledge is simply not detailed enough, and I seriously doubt that it is possible to usefully predict the future very exactly for any specific individual, based on a detailed mechanistic understanding of the processes leading to, say, an episode of psychosis. However, that does not worry me." 

According to Preben Bo Mortensen one can always speculate that, if we were clever enough, we could distribute people in these neat boxes in which they have a single cause for their condition or a treatment that would be perfect for this particular individual. 

"But I think what we’re looking at is really this complex and dynamic web of the evolving probabilities of something occurring to us, not a universe of deterministic causes. I think if you come from any other university background, you are perhaps interested in finding scientific truth in a more ideal sense. But, in my opinion, if you are a doctor, that is not what you do."

The goals are in a way more modest but not less important. 

"When you work with treatment or prevention, you decide ‘would it be better to do this, do that or do nothing?’. So, the purpose of my work is just to, in a world of uncertainty, help improve the chances of making the best decision possible.”

The 2020 Novo Nordisk Prize was awarded to Preben Bo Mortensen, Professor and Scientific Director, iPSYCH, Aarhus University and Merete Nordentoft, Clinical Professor, University of Copenhagen.