In a previous blog, we looked at how predictive analytics can be applied to health and safety. In this article, we discuss how predictive analytics can be used specifically from a wellbeing perspective.
When it comes to the use of big data in managing health and safety a lot of the initial focus has centred on the ability of the methodology to manage procedural variants and make predictions on workplace machinery and processes; however, a recent study in the US looked at how predictive analytics can be used to make predictions on more human factors.
The work from the Mason DataLab of George Mason University is working towards the creation of an analytics model that can increase the physical mental and social health of the population’s youth.
The ‘Data Analytics for Youth Risk and Protective Factors Project’ aims to identify factors that may directly affect healthy behaviours in the youth population measured. From this, better and more informed decisions could be made on structural and policy investment.
The motivations for the study are multi-pronged. The University has expressed a desire to meet the social responsibility it has for the student-faculty and wider youth population. Strategically, this could then support the Universities growth.
So far the findings from the predictive analytics tests have shown correlations between extra-curricular activities and teacher recognition. The correlation between these two factors are then related to other factors in the model. Theorised Conclusions are then drawn from the correlations which can then be used to inform Campus policy and investment.
Wider applications for the model in workplace health and safety
The data and information that has been gleamed for the study have already been put to good use with different initiatives being launched. The potential for the method to be applied to other areas is encouraging.
In regards to wellbeing at work, the model could be tailored for different industry and professions. This could then output data which could be used to help determine, for example the design of a workplace and the type of amenities that could best increase employee happiness and wellbeing.
Outside of the traditional confines of health and safety, predictive analytics is also being trialed in medical fields with a focus on how it can be used to deliver personalised and precision medicine to potential patients. Any improvements in how mental health is treated will have knock-on benefits for workplace health and safety.
Criticisms of potential models in predicting mental health
Despite the potential for predictive analytics to greatly improve the treatment of mental health there are some who have raised issues in its implementation. Firstly Becker et. al. (2018) state that extra care will need to be taken when using results of predictive analytics to determine treatment. Any potential errors in the data taken from predictive models, given that they are the basis for clinical treatment, will be far more impactful than other areas.
In addition to the concerns around predictive analytics in healthcare, wider issues of privacy have also been leveled at the ‘big data’ movement. Previous scandals concerning Cambridge Analytica and Facebook have impacted on the extent to which data online can truly be considered private. These obstacles will have to be overcome if predictive analytics is to truly become common place in the sphere of mental health and safety.
Becker, D., van Breda, W., Funk, B., Hoogendoorn, M., Ruwaard, J. and Riper, H., 2018. Predictive modeling in e-mental health: A common language framework. Internet interventions, 12, pp.57-67.
