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
Rapporteur: Dr Nick Glozier
Associate Director for Population Health Sciences, George Institute for International Health
Nick Glozier is a consultant psychiatrist with a focus on public health psychiatry, comorbidity and early intervention, particularly in high prevalence conditions. He has interests in cross-cultural health, behaviours and disability, having worked in areas such as mental health services and autism in South Asia, post natal depression in Ethiopia and epilepsy in Australia. Previously he worked at the Institute of Psychiatry in London and with WHO, primarily in the area of disability. His research interests in developed countries are in psychosocial and work-related disability research, particularly stigma and discrimination. He helped develop work stress prevention programs both with the UK government and large multinational organisations.
Nick is currently working in areas of novel risk factor identification for the trajectory of mental ill health and disability in high prevalence conditions, depression and anxiety and work and education, and comorbid physical illness (for example, young stroke survivors), epilepsy and those with cardiovascular risks. He has also conducted controlled trials in ameliorating environmental risks for poor outcomes rather than focusing upon health service and treatment delivery.
As you can imagine, we had a wide-ranging group. As Ian Hickie pointed out earlier, one of the questions in mental health is: what is mental health or, rather, what is mental illness? We had a group that covered childhood onset illnesses; adolescent illnesses (or those that have an adolescent onset); those that have an onset in later life; and generally cognitive disorders or those that have a neuropsychiatric presentation. I am going to try to use the matrix that we were asked to assess.
(Click on image for a larger version)
The first is the concept of 'omics. The consensus appeared to be that this is less of an immediate issue in mental health. The biologists in our group suggested that currently our models are comparatively poor in this area. There needs to be much more development of animal models in our particular area.
Of genomics, there is more of a use. A couple of people flagged that, despite the limitations of GWA – genome-wide association approaches – one of the things that we often look at is identifying biomarkers or other kinds of markers for risks. The other recent appreciation by those who work in the field is that in psychiatric or mental illness we deal with trajectories: disorders come and go, they fluctuate and people get worse and they get better. The majority of the serious disorders are chronic, even lifelong disorders. We need risk prediction tools and the ability to use these genome-wide associations to predict risk: risk of onset, remission and recurrence. So the question is: can we link those genome-wide associations with particular biomarkers which can be valuable in many ways, such as in understanding the pathophysiology of mental disorders, predicting prognosis and response to treatment, and possibly helping early detection of illness in high-risk patients?
The area of biomarkers to link with the 'omic approaches has expanded rapidly in the past decade. Some of these biomarkers have been identified for some time. For example the P300 event wave potential and schizophrenia. New ones are published monthly such as the exploration of energy metabolism proteins, such as VGF.
One contributor flagged the use of pluripotent stem cells to try to identify specific biological markers in our field. Linking biomarkers and behaviour is the hallmark of modern imaging with improvements in neuroimaging (for example, functional imaging under task stress, MEG, PET and SPECT). New behavioural markers are being established, such as identifying gaze avoidance as a predictor of autism, even at only 12 months. We really need to be exploiting and linking these kinds of markers together and across the quite disparate fields that we sometimes work in and actually how we collaborate across those. Identifying the technology that could help these disparate groups assimilate information in different fields will be a key task
Much of this will have implication for psychiatric nosology and we need to be looking at some of our phenotypic assumptions, particularly as we want to look at associations with co-morbidity and, again, at the trajectory of the kinds of illnesses we have. I think that brings us to an area that other people talked about, which is the concept of a biological repository or a biolab, which we would endorse. Understanding the differential uses in our area is complex; for example differential protein degradation in stored brains.
(Click on image for a larger version)
With mental health and gene-environment interactions, one of the things that strikes me is the enormous amount of available data in Australia. However, it is often in quite large cohort studies that are not necessarily linked together – the Queensland twin registries, LSAC [Longitudinal Study of Australian Children], HILDA [Household, Income and Labour Dynamics in Australia] and the Victorian Adolescent Health Cohort Study – to name a few. We have lots of these cohort studies with lots of data that could really do with collaborating and linking to try to identify the risks and the trajectories of these kinds of illnesses. Improving the technology to facilitate linkage with routine data sources such as MBS [Medicare Benefits Schedule] and PBS [Pharmaceutical Benefits Scheme] would enhance this.
How do we pull these things together? How do we take a concerted approach? In particular, how do we link those cohorts to the kinds of biomarkers and other markers that we have talked about early on so that we can use them and maybe generate consent later on in life for going back to the data that we have already collected on these individuals as they grow and develop?
(Click on image for a larger version)
With screening and early detection, something that was flagged is that we need to refer the work on perinatal markers. A federal government agenda is looking at perinatal health, maternal and child health. We understand some of the phenotypic markers there, but we understand very little about the biological markers in this particular area.
Again, with screening, what are the risk factors? We are quite good, for instance, in screening in psychosis and Australia has led the world in this area. There are still enormous amounts of work to be done. The concept of the more prevalent mental illness as chronic illnesses and understanding that these are the conditions that cause the greatest health burden from a societal viewpoint is quite a new concept. Mental illness used to be seen as discrete illnesses. There is still an enormous amount of work to be done in looking at risks and markers for different trajectories of illness. How we funnel the resources and collaborate with the different fields in that area is absolutely vital.
We need to start translating some of the approaches from other disorders. Algorithms for risk prediction are routine now in cardiovascular disease. Similar approaches can be explored, for example, in cognitive disorders. Hypertension, we know, is an enormous risk factor for later cognitive decline. But we need to be translating some of this knowledge into developing similar algorithms to allow for screening for early detection and possibly early intervention in this group.
How do we roll out these kinds of things in the high-prevalence group disorders –substance abuse, anxiety and depression areas – that we are not tapping into currently and that really need to be on people's agenda?
(Click on image for a larger version)
Prevention: we do know a number of things about how we prevent mental illness. Resilience is not simply the obverse of vulnerability, and all of the above approaches can be applied to identifying and enhancing resilience markers and behaviour. What we don't do very well is the way we translate these kinds of things and, in particular, how we can possibly use the new technologies to translate these interventions and look at stepped care approaches in this particular area. Using technology to enhance and broaden the appropriate delivery of behavioural interventions has started, for example, delivering cognitive behavioural therapy through the internet. Can this be extended, for example, to improving parenting early on in life, which may ameliorate a whole range of conditions and certainly have longer term benefits? Are there other ways of delivering that by using new technologies, such as mobile phones or PDAs?
Another area in terms of prevention that we are very, very keen on – which came up again and again – is identifying the key risk groups, particularly in allocating the resources to those key risk groups. A very recent meta analysis has just highlighted how interventions to prevent depression reduce the number needed to treat dramatically if delivered to indicated groups.
How do we expand the interventions from other arenas? How do we expand those interventions in the cardiovascular disease into our area? Also, how do we expand the interventions we know, say, in adults, and take them further up stream? How do we use technologies to intervene? For instance, people were talking about using, in children and adolescents, the lifestyle interventions we already use in adults. Participants talked again and again about how we use technology for screening and for interventions.
(Click on image for a larger version)
We were very aware that, when we get the mental health field together, there are a lot of elephants in the room, and we need the key linkages which can be provided by technology and the technologists. Economics: a number of people have mentioned behavioural economics. How successful it is to use economies to change people's behaviour? Urban design came up and also – very important for us – linking in with other people's health agendas and co-morbidity.
(Click on image for a larger version)
So there are key tasks: linking with other agendas in insurance, private healthcare and other industries. What came up several times was the idea of people taking control of their own lives a key agenda for chronic disease prevention in a resource limited environment. How do we work with people having individualised health – using technology to target interventions through pharmocogenomics for instance? How do we link our agendas into people's own agendas and use technology to enhance that? How do we look at providing incentives to change current practices, both with health care delivery and also individuals? How do we do those things?
We fully supported the concept of biobanks and biological data repositories that we can then possibly link – Graham Brown's ideas early on of some of this healthcare de-identified data that we already have – across other health related conditions. Certainly it would be very important for things like mental health, insurance and workers comp.
(Click on image for a larger version)
In summary, we focussed on the use of technology to provide markers for risk and resilience identification, targets for individualised treatment and enhancement of our behavioural standards. Technology can enhance collaboration between us and different people working in our field, and the use of technology as a mechanism to deliver healthcare and screen for health risks and map healthcare over people's illness trajectory, will be key.
A particular agenda for us, of course, is that mental health and disadvantage go hand in hand. Any use of technology to ameliorate disadvantage should also have knock on effects for mental health. This is a target we really need to be concentrating on as we move this agenda forward. The concept of linking population-based and biological data was fully supported. We think concerted federal coordination of that is needed to overcome the numerous barriers to technology implementation. This has been flagged again and again during our time here.
