USask PhD student uses computer modelling to better understand and predict suicide

Rifat Zahan, a University of Saskatchewan (USask) graduate student, wants to better understand the complexity of suicide and self-harm behaviours through her research in USask’s College of Arts and Science.

As a PhD student in the college’s Department of Computer Science, Zahan is developing computational dynamic simulation models to study suicide and self-harm behaviours at both the individual and population levels. She is also using a machine learning algorithm to help predict suicide attempts and death by suicide.

For Zahan, suicide is an issue of both academic and personal interest.

“Mental health and suicide are two deeply connected public health phenomena,” said Zahan, who was born and raised in Bangladesh.

“The culture I am coming from still suffers from different stigma associated with mental health and, especially, suicide. I have had friends, relatives and colleagues who were either directly or indirectly impacted by self-harm behaviours. In my community, I try to strongly advocate for mental health and well-being. I also got training back home on positive mental health. Given that, I personally care about mental health and suicide.”

Prior to coming to USask, Zahan obtained an undergraduate degree and a master’s degree in applied statistics at the University of Dhaka. A high-achieving student, she received several scholarships for academic merit from the Government of the People’s Republic of Bangladesh and began working at the International Centre for Diarrheal Disease Research in Bangladesh upon graduation.

Zahan then decided to continue her post-secondary education in Canada, where she earned a Master of Science degree in biostatistics at USask’s School of Public Health in 2016. She is now completing her PhD under the supervision of computer science faculty member Dr. Nathaniel Osgood (PhD) in the Computational Epidemiology and Public Health Informatics Lab (CEPHIL), where technology is used to address public health challenges.

Through the CEPHIL, Zahan has had the opportunity to connect with world-class researchers from the United States, Canada and Australia, as well as decision-makers from local, provincial and federal governments. For example, she is working with the Sax Institute in Australia to develop a dynamic model of identifying suicide clusters among youth and to assess interventions to prevent suicide. She is also working with researchers from the Public Health Agency of Canada to develop a national-level suicide system dynamics model.

Zahan said there are already numerous intervention strategies in place aimed at preventing suicide in Canada and Australia, but studies looking at interventions within the context of systems science are limited. The benefit of computer modelling is that researchers can look at suicide incidence and prevalence in real time and observe the impacts of adjusting specific variables.

“Systems science models have long been used in infectious disease modelling, consumer behaviour, health-care delivery, operations research and business, but using such a tool in the domain of suicide is very limited,” Zahan said. “That’s why I chose to use a systems science approach to model the complex system of suicide.”

Zahan’s work with the Public Health Agency of Canada is designed to implement a machine-learning algorithm that can analyze multiple factors, such as people’s consideration of and planning around suicide, the type of methods people may use to attempt suicide and self-harm hospitalization, while also considering uncertainties arising from real-world situations. Such machine learning tools can ultimately help researchers predict, with higher accuracy, the incidence and prevalence of suicidal behaviour in a population.

“This would allow public health researchers to assess different suicide prevention interventions on the model and address ‘what-if’ questions policy-makers may have for future use in Canada,” she said.

Through her work with the Sax Institute in Australia, Zahan is looking at the need for a youth-specific suicide intervention, particularly in the context of social media platforms. The youth suicide rate continues to increase in Australia despite various interventions already in place, so the study aims to assess the impact of a co-designed social media intervention known as #chatsafe on clusters of youth suicides and suicide attempts, using a dynamic modelling approach grounded in systems science.

“The study models individual self-harm and suicidal behaviours and how people interact among themselves about such behaviours. In this simulated model environment, a person’s tendency to have riskier suicidal behaviour may be impacted by how their network communicates about suicide,” Zahan said.

“We created this simulated environment to study the dynamics of both the social media communication and communication among friends and family about suicide. The #chatsafe guidelines is a social media campaign that delivers content involving safe communication about suicide on social media. The dynamic model that I am developing will assess the impact of this preventive intervention among youth suicide and attempt clusters.”

As Zahan continues to work on these research projects and on her PhD dissertation, she is also active in groups and activities on the USask campus. For example, she has served as the president of the Computer Science Graduate Council and as a steering committee member for Research Fest. She is currently the president of the Graduate Students’ Association.

Among numerous academic achievements, Zahan has also been honoured with two awards for her leadership and volunteerism at USask: The Citizenship Award from the Department of Computer Science and the Excellence in Leadership Award from the Bangladesh Students’ Association.

Prior to the COVID-19 pandemic, Zahan travelled to Florida to attend a summit organized by the American Foundation for Suicide Prevention and the International Academy of Suicide Research. The experience deeply impacted her, and she wrote a travelogue about the trip in her first language, Bengali. A conversation that Zahan had in Florida with a man whose sister died by suicide reaffirmed Zahan’s decision to dedicate her life to “research that makes positive changes in the world.”

“It is always important to know if the research we are doing is impacting or going to impact the world,” she said. “While working on my doctoral work, I got to see how powerful dynamic models are well received by decision-makers all around the world. When I see that my work is being used by decision-makers like governments, senior researchers and suicide specialists, it’s quite rewarding.”