A data-driven approach to suicide prevention in Australia

Black Dog Institute is using social innovation to prevent suicide and support people to live full, contributing lives

Suicide is the leading cause of death for Australians aged 15-44 years. In 2017, there were 3,128 suicide deaths in Australia and for every suicide death, as many as 25 individuals will attempt suicide.

LifeSpan – an initiative of the Black Dog Institute – came into existence as a new approach to suicide prevention. Black Dog’s research had found that previous approaches to suicide prevention were fragmented and often not based on solid evidence.

The LifeSpan model was scoped in 2016 with funding from the NSW Mental Health Commission. The LifeSpan framework consists of nine integrated suicide prevention strategies, to be executed simultaneously. The strategies span the community, health, school, and emergency services sectors. Scientific modeling predicted LifeSpan would help prevent 21 percent of suicide deaths and 30 percent of suicide attempts.

After the initial scoping study, LifeSpan was rolled out to four trial sites in New South Wales (NSW) with funding from the Paul Ramsay Foundation. A further 12 pilot sites for integrated suicide prevention were funded by the Australian Government Department of Health and are being supported by the Black Dog Institute.

SAS, a long-time partner of Black Dog Institute, helped implement a data-driven approach to support rigorous evaluation of LifeSpan. “Without data, we wouldn’t be able to comment on whether LifeSpan has been effective or not,” said Dr Michelle Tye, Research Fellow, Black Dog Institute.

The first question concerned what to do with the data. LifeSpan was grappling with disparate data sets spanning myriad frontline emergency services, health services, universities and government agencies.

“The amount of data was an important consideration. When LifeSpan embarked on this project, we were terrified by how much data we had, what we were going to do with it and how do we get something meaningful from it,” said Dr Tye.

This is where SAS, working closely with LifeSpan’s team, has added value. “SAS has come in and set up the infrastructure to allow LifeSpan to clean, store, manipulate and analyse different data sets, while also providing ongoing support in terms of training and general assistance,” Matthew Phillips, LifeSpan Data Manager said. Enhanced use of data has enabled LifeSpan to track its impact across trial sites (Newcastle, Illawarra Shoalhaven, Central Coast and Murrumbidgee) over the next four years.

Measuring impact is pivotal. As a project funded through philanthropy and public money, LifeSpan is reliant on showcasing success to scale to new trial sites. One example of showcasing success is providing new insight into when and where suicides occur so that hospital resourcing can be assigned more effectively. This is aided by geospatial mapping using the Geographic and Resource Analysis in Primary Health Care (GRAPHC) platform at the Australian National University.

Secondly, suicide prevention and LifeSpan is a "long game," so driving sustained engagement is key. “We are not going to have outcomes for a couple of years. This means it is hard for the different stakeholders to stay engaged; however SAS’ data platform lets them see how the implementation is going, provides results and is helping to foster engagement in the project,” said Dr. Fiona Shand, LifeSpan Research Director, Black Dog Institute.

As part of LifeSpan, community helper training has been delivered to equip individuals with the skills to recognise, respond to and assist people at risk of suicide. This includes:

• QPR (Question, Persuade, Refer): 3,600+ community gatekeepers (including GPs) have completed QPR training.
• YAM (Youth Aware of Mental Health): Delivered to 5,500+ students at 66 schools across NSW trial sites.
• Primary Care: Programs equipping health professionals to identify and increase confidence in treating patients at risk of suicide.

Applying analytics to suicide prevention is already having wide-ranging outcomes, including unprecedented data quality. Scientia Professor Helen Christensen, Director and Chief Scientist of the Black Dog Institute, said:

“Right now, we have the best quality suicide data available in Australia, ever. It hasn’t been available at this level of detail before. Until now, we haven’t had the knowledge of where suicide incidents are occurring, when they are occurring or how, to the level of precision and accuracy that we do now.”

Available data includes media reporting of suicide within each of the LifeSpan sites, using a platform built in partnership with SAS. LifeSpan’s media monitoring will showcase social innovation at its finest. Text analytics and data mining will be applied to approximately 26,000 archived articles and 1,000 new articles each month. By late 2018, LifeSpan will be able to measure the evolution of media coverage of suicide, while also measuring whether LifeSpan has shifted perceptions and the language of suicide across the NSW trial sites.

Christensen is positive about the benefit this will have, adding, “It is possible to prevent suicide. We have the evidence; we just need to share and implement it. And data is the key to that.”