THEO MURPHY (AUSTRALIA) HIGH FLYERS THINK TANK
Preventative health: Science and technology in the prevention and early detection of disease
University of Sydney (Eastern Avenue Complex), Thursday 6 November 2008
Group B: Mental health
Chair: Professor Ian Hickie AM MD RANZCP FASSA
Ian Hickie is a consultant psychiatrist and is one of the first round of new NHMRC 2008 Australian Fellows. In October 2000 he was appointed as the inaugural CEO of beyondblue: the national depression initiative and from 2003 to 2006 served as its clinical advisor. He received the Australian Honours Award of Member (AM) in 2006. In 2007 was appointed to the Prime Minister's Australian National Council on Drugs and elected as a Fellow of the Academy of the Social Sciences in Australia. He is a founding member of the new National Youth Mental Health Foundation ('headspace'). In July 2008 he was appointed to the Federal Health's Minister's new National Advisory Council on Mental Health, and he serves on various mental health and policy committees. Over his research career, he has published over 220 peer reviewed journal articles, 20 book chapters and 30 educational materials.
As professor of community psychiatry at the University of NSW from 1997 to 2003 Ian's work focused on service reform and developing novel techniques to study neurobiological changes in affective disorders. He conducted a longitudinal study, which was one of the first to demonstrate clearly that changes in subcortical brain structures (particularly deep white matter abnormalities) were of functional and prognostic significance. Ian's research, clinical and health services development work focuses on expansion of population-based mental health research and development of international mental health strategies. He has the capacity to integrate clinical and neurobiological research and health service reform.
Thank you for the opportunity to speak. I do like to get the opportunity to speak, particularly to audiences where the issue of prevention is here before them. I hope some of the preventative health people are still here, because I think often there is an issue that some of the really big areas that are in desperate need of implementation of the knowledge that we have, somehow fall off the science agenda. They don't fall off the public agenda.
John Menadue, previous head of Prime Minister and Cabinet back in the 1970s, continuously asked the community, 'What's really important in health?' Three things continuously come out: child health, indigenous health and mental health. In terms of the investments in science and technology, public policy and everything else, it's kind of interesting to think of the extent to which that's actually reflected in what we do and what we recruit to, or whether we have a tendency to recruit to what we already have capacity in and that is supported and expanded by the knowledge that we have.
(Click on image for a larger version)
I am very fortunate to have recently been part of this development [Brain and Mind Research Institute] here at the University of Sydney, which is an integration of brain and mind sciences. It has been possible not only through the support of the University of Sydney – which has put aside such a large piece of real estate and moved some traditional areas such as cancer and cardiology – but also because of the investments of governments (the New South Wales and Australian), foundations, and people in the community actually wanting to know what is possible. It does worry me in the preventative health discussion that some of those issues have disappeared. Fortunately, we do share some of the risk factors and some of the interventions that will arise; but some are different and require considerable consideration. I know that Rob Moodie, the chair of the taskforce [National Preventative Health Taskforce], is very open to that.
One of the problems for us – and this is the advantage of this particular seminar – is that, unfortunately, mental health is such a large generic term. It's like saying 'physical health'. You know, it's so big and covers so many things that it becomes mysterious at a wider set of levels. We know this in the stigma area, we know this in the public policy area and we certainly know it in the science area. Unfortunately, in many ways a lot of the new understandings and new opportunities in prevention and early intervention we have – particularly in this area – are not translated into practice. We desperately need people like most of you in this audience, who are devoting aspects of their careers – serious career scientists to consider engaging in this. To do that, we have to have the capacity.
To pick up the point that Chris Goodnow and others have made, we're really lucky that this type of development has animal facilities on the top, basic science laboratories, NCRIS [National Collaborative Research Infrastructure Strategy] supported national imaging facilities, patients being treated for neurological and psychiatric disorders, public policy advocacy, new biochemistry labs and a totally new youth mental health facility. It has all of those on the one site. The idea is that the scientists need to interact if there is to be meaningful progress.
In case I don't get to the end, which is highly likely because I get distracted by what I get interested in along the way, these are the key issues.
(Click on image for a larger version)
Mental health related disability is really big. If you take mental health alone, it's 27 per cent of all related disability costs in Australia. If you take the brain and mind sciences – which is a better way of looking at it – the combination of neurological disorders as they would classically be called, and psychiatric disorders, are actually 40 per cent of disability. We don't necessarily dominate premature death, although we are a big contributor there as well. We dominate the disability field. A terribly important issue for productivity in society is the way in which the sciences might contribute to better quality of life. We are in urgent need of shifting from a palliative care set of mentalities – a sort of maintaining disability – to an actually informed prevention and early intervention agenda. I am a little worried about the way in which the Preventative Health Taskforce has been constrained by government to primary prevention and on knowable risk factors in a rather narrow sense, because there are great buys across the early intervention agenda and, I think as Bruce [Armstrong] was showing, what might be called the secondary prevention agenda.
The mental health area is one that suffers from another really big issue. If there's one thing I'd like you all to do, it is to go out and talk about what you do. This area suffers most from a lack of science literacy. It's filled with opinion, 19th century ideas and non-science ideas, and what is known is poorly communicated. But the community interest in what we do is enormous. We do need, however, better science to drive what is going on.
I could spend a lot of time talking about genes and environments and everything else, but the other big issue for us is really one of timing. It is understanding not just how the mechanistic bits that add up to risk factors interact, but also how they interact in critical periods. There are big opportunities – as with many of the other diseases that will be discussed – in the foetal and neonatal areas. What's happening in the womb and beyond and what happens to maternal health; there are critical issues in childhood. However, it is not all over and done with in childhood. There are critical opportunities in the adolescent phase, particularly with regard to frontal lobe development in the later adolescent phase and, fortunately for everyone in this room, there are big opportunities as you age.
(Click on image for a larger version)
One of the problems about being defined as a complex disease – and I think this was emphasised by Paul [Zimmet's] talk this morning as well – is that there are complex pathways to outcomes which can result in a throwing up of one's hands because there is not a simple mechanistic intervention; you can't simply stop smoking and expect a whole lot of diseases to go away. It will be very interesting to see what the taskforce actually does about diabetes, given its complexity. It does not mean that there are not key things to be done, however. I was discussing this yesterday with someone in Canberra, having spent the day there. One has to get over the complexity thing and talk about the key things that are to be done. Obviously, the better the science behind them, the more likely they are to be done.
Fortunately, in our area we are now beyond many of the arguments that you will still see on a daily basis in our newspapers, which are either a kind of genetic neurodevelopmentalism – 'You're one of those people born with that thing, fortunately not like the rest of us' – and that is a bad thing. That has a terrible history, of course, in the eugenics movement and elsewhere, of people who are not as good as other people. On the other hand, you have people who believe that there is no biology in the areas in which you work; it is all due to social factors operating in very bizarre ways. We need to be engaged in that debate to show what we now know a lot about that.
(Click on image for a larger version)
This is the summary of everything else to come. We know a lot about gene–environment interactions. We have some very good specific sets of issues between life events and the serotonin transporter; the risk of cannabis in psychosis. In particular, the risks there are with drug use and psychosis are ones that have immediate public health potential impact, even if they are somewhat unpalatable for those of us who have grown up in a cannabis-tolerant generation.
We know a lot about brain changes during illness onsets. They have been profound in their sets of meanings – in both the psychosis area and some of our own work in the depression area – to show that the brain is actively changing during illness onset, and a key opportunity exists for intervention during that period. It is not simply a developmental disorder. We know, and are unapologetic about, the extent to which providing treatments in secondary intervention has produced a very important set of concepts that has actually reduced suicide; that we may change the course of illness through early interventions, which is best tested so far in the psychosis area; and, particularly in later life, that we are moving to stages where preventative and early intervention strategies are becoming very realistic targets and are starting to be assisted by biomarker development in that area.
(Click on image for a larger version)
If you want to go to misunderstandings, every day you can pull out the newspaper and find that drugs 'don't work'. That means that they are equally as good as a lot of psychological interventions and only have benefits at more severe levels. It doesn't mean that the drugs don't work; it just means that there are equally good sets of therapies, but it gets interpreted as 'they don't work'. My personal favourite still is: 'It's all your parents' fault.' I'm never sure whether it is all your parents' fault genetically, environmentally or what. But it has two meanings: 'It's your parents' fault and you shouldn't be treated,' which is an interesting kind of nihilism as well – 'You should have treated the previous generations; bad luck if you've got it'; and the notion of actually blaming people continuously for the outcomes in these particular sets of areas. They are really difficult areas for us to get around.
(Click on image for a larger version)
Importantly for us, the face of mental health problems is a disability and it is chronic disability of the young. One of the difficulties for us in joining up all the risk factors with everything else related to ageing, cardiovascular disease and diabetes – where a lot of the risk factors come home to roost in mid or later life – is that our traumas have come home to roost for many people quite early in life. So this person has a chronic disease and has already had 15 or 20 years of illness; they are already going to be isolated, non-productive and out of the economy by the time they are dealt with; but it is very unlikely that the health system has dealt with them at all. We only spend about seven per cent of the health budget; a great majority of young people, particularly young men, with mental health problems never receive treatment. In fact – I will show you this at the end – we have just had another national mental health survey. In 1997–98, we treated 38 per cent of those with a disorder; last year, in 2007–08, we treated 35 per cent. We actually went nowhere in increasing access to treatment, and those who stay outside of treatment are young, male and have whole sets of other accumulated health problems.
(Click on image for a larger version)
The distribution of disability with age is very important to us. There is a big set of disabilities that come home to roost in early adulthood. So 60 per cent of disability of 15- to 34-year-olds is due to mental health and substance abuse. There is a large chunk in childhood, part of which contributes and part of which arises at birth. There is a large chunk associated with ageing, and that is the important set of the traditional neurodegenerative disorders and brain injury disorders. Interestingly, this is one in which the total rate of disorder and the age of onset of that disability is actually already moving. Prevention is already having an effect due to a lot of what the cardiovascular health factors have been in midlife. So some of the things that we were seeing at 60 and 70, we are not seeing or we are seeing at 70 or 80, and we are seeing them move. So we are already beneficiaries of other people's preventative health in the ageing movement and we need to maximise that as we go.
(Click on image for a larger version)
The big challenge for us is going to be what we do here. We already have key early intervention strategies and we are looking for additional prevention strategies. But we are – in total disability terms – eight out of 10 of the major causes of premature death and disability in young Australian men, and six out of 10 in young Australian women.
(Click on image for a larger version)
Actually, this is the Everest for us: how do we get rid of this peak in the 15- to 25-year age group? The lives of many of the people I work with in the clinical area are wasted in phenotypic discussions and debates about how many different phenotypes are here. For those of a more generalist view, you might think there is something common going on during that particular period that may be relevant.
(Click on image for a larger version)
We now know from neuro-imaging that some important things are going on, particularly in the maturation of frontal lobes in these later periods, and that is really interesting. That is the peak age of onset. That is also the peak age of exposure to some particular factors – particularly drug use and the onsets of illnesses, which themselves are toxic – but also the peak period for opportunity for intervention or for maximisation of other factors which may be beneficial: social interaction, education et cetera. So the neuroscience is informing and matching with some aspects of the epidemiology and should start to drive public policy.
(Click on image for a larger version)
In the basic sciences areas, really informed sets of information about brain development are now contributing – particularly in terms of synaptic development and then synaptic pruning – over the childhood and then the adolescent period. The extent to which the illnesses we see are a failure of synaptic generation in the first place or accelerated synaptic pruning have become critical issues for us.
(Click on image for a larger version)
Medical schools, like the one here, used to teach this rather neuronalism view: you have all the neurones you had at birth and you just deteriorated. It didn't really matter what happened. You might have been lucky enough to get educated or fed along the way, but your brain existed in a box outside of the world. Nothing could be further from the truth. No matter what genes you take into life, no matter what happens in terms of your other vulnerabilities, there are clearly opportunities for dealing with both the degeneration process and the importance of pruning processes in these key periods.
Some of you might also consider what happens with ageing. This is a rather old set of slides and this may well be changing itself for those who stay in education and employment – hopefully, with the world financial crisis, you won't retire; you'll continue.
(Click on image for a larger version)
We now also have some really good animal models developing, such as how immune activation in pregnancy in rat models gives rise to onsets not in the juvenile period but in the adolescent period. In fact, by a judicious use of interventions at its critical phase, they may be delayed in terms of the pathology and the onset. We have some interesting animal models developing that we would not have predicted in the past, and we have to start to test some of these ideas.
(Click on image for a larger version)
It is similar in relation to psychosis and in relation to dementia. We now have good matches between genetic models and animal models, so we can move on to that mechanistic level, which Chris [Goodnow] was talking about, to test the relative benefits of different sets of interventions.
(Click on image for a larger version)
We have new techniques for looking at the brain during active phases. This is from Richard Barnardi, one of our colleagues at the Brain and Mind Research Institute, looking at microglial activation; what the immune system is doing during the onset of psychosis. It is something that was not thought to be relevant previously and not able to be imaged previously, but it actually may lead to biomarker development.
(Click on image for a larger version)
Down the human service delivery end, technology is also important. We are never going to have sets of health professionals on the ground dealing with people in these age groups in the traditional medical paradigm. This is Professor Helen Christensen from the Australian National University, who is one of the world leaders in the development of action intervention programs and their evaluations. They are highly cost effective, highly deliverable and probably work as both prevention and early intervention programs.
(Click on image for a larger version)
Gene environmental reactions in unpacking them are going to be critical in our area.
(Click on image for a larger version)
What is interesting to us is not the unique genes; it is the common genes and the ways in which we are seeing the sets of neurodevelopmental and neurodegenerative genes. Other big processes are where glucocorticoid, circadian and other biochemical and neuro-immune groups of genes become relevant.
(Click on image for a larger version)
Of course, what becomes relevant then is: what are they interacting with? We now have lots of proof of principle of relationships between different genotypes and key environmental exposures, whether they are maltreatment in childhood, likely events in early adulthood or the use of cannabis. There is a whole range of others that were just picked somewhat randomly from recent literature and ones that I am interested in. These will give you some idea of the flavour, that we are moving on to a better understanding of some of the key environmental factors and what they interact with to determine what we see.
(Click on image for a larger version)
We see a lot where there are strong or common environmental effects. For many of the big disorders we are talking about, you have big environmental effects running – ones that change the world we see and ones we might contribute to changing – where there is high individual variability and others where there is undoubtedly strong inherited or neurodevelopmental effects, but there are still environment agents to be unpacked. That is becoming clearer and clearer in those areas – less of a search for unique genes and more of a search for the unusual environmental factors that may be making contributions.
(Click on image for a larger version)
In a good review of this area by Caspi and Moffitt in July 2006, they signal an important switch in our area, which is getting off the genome, getting back into the environment and trying to work from the environment back to what actually goes on.
(Click on image for a larger version)
So we have spent a lot of time trying to make the gene-disorder type connection. They suggest that there is a much more interesting place to start. We already know a whole lot of environmental risk factors; we need to combine that with what we know about the genotype and try to combine it with the neural substrates to better understand what the range of disorders that we actually see is.
(Click on image for a larger version)
In a lot of the disorders, such as major depression, we already see an unpacking of issues like brain-derived neurotrophic factor and its relationship to hippocampal volume and potential interaction with things like glucocorticoid stress to impair a particular function like cognition within the symptom complex of depression. In fact, the biology of this area has skipped along very rapidly, largely unknown to those outside the field.
(Click on image for a larger version)
But we need to continuously flick back into the public policy domain unashamedly and ask, 'Well, what do we know and what can be done?'
(Click on image for a larger version)
We share a lot of risk factors with some that have been identified in the drug and alcohol field. We can work out what proportion of DALYs [disability adjusted life years] associated with illicit drugs and alcohol are shared with a lot of those areas and, undoubtedly, several would be beneficiaries if significant things were done.
(Click on image for a larger version)
However, we also contribute to those areas. Mental health is a risk factor to those areas. It is estimated that half of adult substance abuse disorders are preceded by mental health problems, so actually we are a prevention of that risk factor for further real health.
Then there are issues related to violence and circumstance that are particularly relevant. If you think about childhood sexual abuse and think nothing can be done, nothing can be further from the truth. The issues around child sexual abuse have changed amazingly in the last 20 years and they continue to change rapidly under social forces, and we will be beneficiaries of those changes.
At many levels, some of the primary prevention is going to be in the broader social domain. But there are big issues, like social cohesion, which are critical. Technology is not necessarily a problem. For example, the use of the internet by young people in a group we work with – the Inspire Foundation's Reach Out – uses technology to join people up and to increase cohesion, not necessarily to isolate them. Technology used in the wrong way, of course, may lead to isolation in that particular area.
There are big things for us to say in the alcohol and drug domain. The delivery of cognitive behavioural skills not only as interventions but also as preventions, is now happening via internet technologies in appropriate young people.
(Click on image for a larger version)
In terms of intervention, the use of any health models and changes in service models will be critical. We have very good Australian derived evidence for teaching parenting in simple ways, the right response to anxiety in childhood well worked out and delivered, increasingly using e-health and combined community technologies. The communications technologies assist everyone to do the same thing. They can share information across GPs, teachers and parents to have the same response to an anxious child.
(Click on image for a larger version)
It is important to treat some of the disorders to prevent other disorders. Despite all the debate about ADHD and the constant reporting – for example, like the Daily Telegraph in Sydney, that we are medicating and mind-controlling children – there is very good evidence, again just published last week, about proper stimulant use in childhood reducing cigarette smoking, alcohol and drug use in adolescence. Some of the people who go on to use them in these environments clearly have prior problems that have contributed; dealing with a childhood problem to reduce another adult risk factor.
(Click on image for a larger version)
It is the same for depressive symptoms in childhood. If you want to deal with one of those pathways to obesity, you may have to deal with the mental health of children at a prior period. Again, there was another good study just published last week looking at the association between childhood mental health factors and other health outcomes.
(Click on image for a larger version)
But it isn't all for young people. I am the father of twins and I like this idea that twins go on forever and they take their genes with them. In some work done with Nick Martin and the Australian Twin Registry, we've been looking at different genes at different ages in relation to late-onset depression. Different environmental factors, different sets of genes and different opportunities may present themselves with ageing.
(Click on image for a larger version)
We also need to think about change in what we do clinically. One of the biggest issues for us in mental health is that symptoms at low levels are rapidly associated with disability.
(Click on image for a larger version)
You do not have to be more than a bit off to have trouble giving or listening to a talk like this, before it doesn't make much sense. When you start to have a number of problems, you can go off quickly. However, that can be very difficult to differentiate at a phenotypic level from anyone else sitting in the audience. With the issues of biomarkers and markers of risk, if we're going to have early intervention strategies, who we are going to intervene with and interfere with, becomes a big issue. In the absence of biomarkers, we are really struggling in this area. What we need are trajectories – not one-off biomarkers, but markers over time – that tell us the key to a young person who is on the wrong trajectory and who would justify an intervention.
(Click on image for a larger version)
So we've moved clinically to what we call a staging model clearly based on the cancer model, not just a bunch of illness phenotypes or comparing this bipolar disorder with that psychotic disorder, but pathways to illness with different sets of interventions and then tying in nonspecific less-toxic interventions at different phases.
(Click on image for a larger version)
That has led to a really interesting set of earlier intervention strategies – that is where a lot of the money is in our area at the moment – looking at not only traditional agents but also novel agents and focusing on not just symptom reduction but also neuroprotective strategies.
(Click on image for a larger version)
This is an area where, if we had more people involved right across the basic science and right across the particular sets of opportunities, we will make significant advances. The key is for us to be in things like the Preventative Health Taskforce – and not just as a separate thing but also with it strongly depending on the accurate communication of the science and technology on which it is based. Thanks.
Discussion:
Question: George Mellick from Griffith University, Brisbane. A lot of these things that you're talking about are time dependent, and developmentally dependent. That is one thing in science that we find really difficult to model, particularly at a cellular level or in animal models. Do you have any comments about how we might do that?
Ian Hickie: The advances in animal models are terribly important to us. For example, work that Iain McGregor does at this university looking at cannabis and ecstasy exposure and its effect on the cognitive functioning of rats depending on time of exposure, whether it has transient or prolonged effects. That is an excellent example of using animal models to clarify some of these issues. The whole synaptogenesis issue of what sort of prenatal insults impair original synaptogenesis or affect the latest post-preputial synaptic pruning, need to be worked out in animal models.
For key aspects of the pathophysiology the development of good animal models has been really important. They are now mirrored on neuroimaging and other examples from humans. The phenotype may not be exactly the same, but there is strong overlap in a lot of the neuroanatomical or neurophysiological correlates to give one some confidence that they may be relevant. So this scenario with animal models is very important to us, particularly now understanding that the mouse phenotype is somewhat different from the rat phenotype, and manipulating genes or genetic abnormalities within mice.
Question: Chair. It seems that the cell biology of isolated cells is of less interest to you than…[indistinct]
Ian Hickie: No. If Professor Max Benner was here – he's the person I am really interested in working with – he would totally disappear into P2X7-dependent ATP mechanisms within cells on the way they drive cellular architecture within the nervous system. So I think there are a number of things going on at different levels. He would constantly argue to go back to the cellular level, and for the intracellular level to come up with new products to derive this. But at the same time they need to be related to or have some sorts of sets of association. The P2X7 one is a good example of where the polymorphism has just been associated with depression and bipolar disorder in population samples. It is not a polymorphism that we would have thought would be relevant in psychiatry; it is an ATP in infection-generated response systems. So you find these sorts of things in a set of association studies and somebody gets on with the basic biology of it.
Question: Paul Zimmet. I think Mike Daube will join me in saying that we have to look at new areas now (from June) from which to benefit. So you have put a very passionate case, which has been heard. I just wanted to mention: as part of this obesity–diabetes epidemic, apart from the sorts of natural genetic and other environmental things, we now have a new way through the psychotropic drugs and such like. I just wonder if you want to comment on this being a growing area of the burden of obesity and diabetes and dis-leukopaenia receptive.
Ian Hickie: Like all great medicine in our area, there are winners and losers – and we have been winners. There is a particular slide I will show you about the effect of olanzapine on preventing some early brain changes in people with psychosis.
[Slide Hickie 42]
There are amazing changes – if it's ever replicated – on preventing grey matter loss in early schizophrenia. However, olanzapine would be the worst diabetes-generating drug we've ever invented. So this leads to the issue: what would be the appropriate use of such a drug? Should it only be used for a short period? Clearly, we will need better compounds which have whatever it is that olanzapine does – it is strongly debated whether it has a BDNF-dependent effect or some other growth factor effect – that gives it its therapeutic effect without putting at longterm risk of premature cardiovascular disease.
