Digital Health 14 February 2019
About six per cent of traffic accident victims will require long-term mental health support, sometimes for up to nine years, a study of health and insurance data has revealed.
“We found that early mental health intervention will reduce the need for long-term care, improving patient outcomes and reducing costs,” says Dr Nazanin Esmaili, Senior Data Scientist and Advanced Analytics Lead from the Digital Health Cooperative Research Centre and the University of Technology Sydney (UTS).
Road traffic accidents are the second leading cause of hospitalised injuries and injury-related deaths in Australia. Globally 20 to 50 million people suffer non-fatal injuries, and the cost is about three per cent of GDP.
“We focused on psychology services and analysed nine years of anonymised data provided by TAC Victoria,” says Dr Esmaili. Her research was recently published in PLOS One and was presented at the 29th European Conference on Operational Research in Valencia, Spain, in 2018.
Dr Esmaili categorised the mental health care requirements of patients into three distinct groups of care: brief, intermediate, and sustained. She found that the majority of patients will require a brief amount of care while approximately six per cent will require sustained levels of care for eight years or more post-accident. This led to the realisation that early intervention was key to reducing the length of time people need to access mental health services.
Other factors include age at which the accident occurred and the role of the patient in the accident.
The results have implications for other fields of health too.
“While we used psychology as the case study in this paper, the methodology we have developed can be applied to any health service,” Dr Esmaili says.
“By providing compensation agencies with the tools to predict ahead of time which claimants are likely to exhibit specific service utilisation patterns, an opportunity is created to design early, dedicated interventions aimed at improving the health outcomes of patients,” says Professor Massimo Piccardi, Head of Discipline, Signal Processing and Analytics at UTS and a co-author on this paper.
"This paper forms part of a broader piece of work using deidentified health data to both improve patients' health outcomes and better inform health providers, compensation agencies and insurers on how best to allocate resources to best support patients," says Associate Professor Frederico Girosi, CHief Scientist and Director of Education at the Digital Health CRC, and a co-author.
The next step for the researchers is to determine whether a patient using more than one health service, such as psychology and physiotherapy, has better or worse health outcomes than those who only use one health service. The group will present their findings at the Institute for Operations Research and the Management Sciences (INFORMS) International Annual Meeting, which will be held in Cancun, Mexico, in June this year.
Sign up for the DHCRC newsletter.