It might have been something your mother said. The events that trigger the decline of mental health are difficult to predict and managing their outcomes is equally fraught.
New methods being developed, using precision data management, promise to assist healthcare workers and policymakers to overcome some of the worst declines in the mental health of unemployed young people.
"There is no 'one-size-fits-all' strategy to assist young people who are 'not in employment, education and training'," said Nick Glozier, professor of psychiatry at the University of Sydney's Brain & Mind Centre.
"We have found that an alcohol intervention strategy is more likely to be useful for 25-year-old males than it is for 16-year-olds (male or female) and an overprotective mother is more likely to lead to male youth disengagement – so working with both mother and son is important.
"Latest techniques in data science allow us to deliver individualised intervention strategies," he said.
"Negative attitudes from friends, your mother's opinion, or the area you live in have a surprisingly strong influence on outcomes as well as the individual things such as depression, previous criminality and drug use, which we could look at before."
But isn't there the danger of profiling people along the lines of race or class?
"There is aways a danger in any technology for it to be misapplied," Professor Glozier said. "That's why we work with ethicists to ensure our research avoids these pitfalls."
Untangling these myriad influences to discover the best preventive treatments is at the heart of how this approach to data is applied in health sciences.
A new centre launched on Thursday at the University of Sydney aims to apply these methods across the board in the physical, life and social sciences.
We are told we are in the age of "big data" – petabytes of information being produced daily that can shed light on our complicated lives for policymakers.
However, Professor Hugh Durrant-Whyte believes the granularity we need to make meaningful decisions within this tsunami of digital information can be lost.
The new multimillion-dollar research centre he runs at the University of Sydney will look at what he calls "small data". Finding the particular pathways to enable meaningful decisions based on the data science is what the Centre for Translational Data Science aims to do.
"Data is the currency of the digital age and is changing how we live. It will be an age of wonder if we succeed in exploiting information to our best advantage including excelling at analysing relatively 'small' data sets," said Professor Durrant-Whyte, the director of the centre.
Employing a dedicated team of 15 data engineers, the centre is already working on 30 projects in fields as diverse as geology, mental health, cancer patient management and criminology.
Traditional data science mines information looking for problems to solve, Professor Durrant-Whyte said.
"What we do is find problems, then use data science to develop appropriate methods to deal with them."
He said that in mental health, traditional methods of research could manage only a handful of factors at a time. "If you want to look at lots of factors–- taking in family history, drinking habits, social practices, employment, socio-economics – you need to manage the data well," he said.
"Our focus on small data and finding the pathways for outcomes will allow us to find not just any model, but the best model that can be used for predictive purposes."
Whether untangling the factors influencing mental health or sorting through the pentabytes of geological data to build useful 3D models, it's a process that takes computing grunt.
The launch of the centre comes just as the university is rebooting its supercomputer, Artemis.
The computer will undergo an upgrade to more than 4300 compute cores and three high-memory nodes with six terabytes of memory and 64 cores each to provide dedicated capacity for cutting-edge simulations in cosmology and informatics.
Professor Geraint Lewis, an astronomer in the school of physics who is helping lead the Artemis project, said: "We are instituting the Sydney Informatics Hub to provide solutions across the university. The goal is to build expertise in informatics and data, and weave it through the university, in terms of research and teaching, and exposing students to new skills."
Though by no means the biggest or fastest computer in Australia, Artemis will be useful for work done by researchers involved with the new data science centre and in training for use in bigger and faster cloud computing.
Professor Durrant-Whyte said some of the projects to be managed through the centre are:
- Developing models to predict patterns of unreported domestic violence
- Real-time data management to incorporate information for breast cancer care, including predicting whether patients will be readmitted
- Predicting outcomes for the health of pre-term babies by continuous monitoring of oxygen levels
- Precision medicine at the Charles Perkins Centre, including data analysis of the human genome, or cell protein production to assist prevention of metabolic diseases such as type 2 diabetes
- Tracking political behaviour with researchers at the US Studies Centre
- Prostate cancer treatment management at the Royal North Shore Hospital
