Skip to main content

Burnout seems to be the latest buzzword when it comes to employee wellbeing, appearing everywhere from Forbes, HR publications, and LinkedIn, to the office. The World Health Organisation (WHO) classifies employee burnout as a ‘syndrome conceptualised as resulting from chronic workplace stress that has not been successfully managed,’ with symptoms like exhaustion, increased mental distance from one’s job, or feelings of negativity towards their career, and reduced professional productivity. All of which spells bad news for your business. Burnout is officially a major global health challenge according to WHO, HR teams are suffering and desperately trying to put measures in place after the fact.

But how do we get to the root of the issue? How do we tackle burnout using preventative measures rather than as a reactive response? The answer could be in wearables that give businesses a continuous stream of health data that can be monitored for any deviations to baseline information and provide accurate predictions regarding an employee’s mental and emotional state. These predictions could be used to trigger interventions in the form of digital notifications with directions to wellbeing support or potentially CBT depending on the severity of the issue, allowing employees to adapt their emotional state prior to it getting out of control.

“I am positive about using wearables to accurately predict burnout. On the one hand, the development of wearable sensors and IoT have enabled us to easily collect our physiological data, on the other hand, multi-modal machine learning and affect computing have equipped us to mine the collected data and extract the deeply hidden latent features, so to generate predictive outcomes. There are several pieces of research that have shown the feasibility of using objective indicators taken by wearable sensors to predict burnout.”

Dr. Yongqiang Cheng, Latus Health Advisory Board member

The future of predicting burnout is around the corner with the implementation of wearables, however, there are several concerns from the employee side about privacy and data security due to being continuously monitored. Implementing this practice would also require time and precision from businesses to ensure positive sentiment towards rollout amongst employees. By outlining a strategy to effectively implement wearable technology, could this be the next best tool for HR and business leaders?

According to a study by Kaczor, where wearables were used to predict physician stress levels alongside self-reporting of stressful events and emotional states, “the best algorithm to detect stress compared to baseline data utilised by the data collected twenty minutes prior to a reported stress episode (accuracy of approximately 70%), this suggested that wearable sensors can detect stress before it is reported or even recognised by the individual.”

Kaczor expands on this to discuss a potential paradigm that could be used, “our data could be applied with an automated alert function using our stress detection algorithm. This could be an important tool for helping physicians become more aware of their emotional state and use them to initiate personal and institutional interventions aimed at decreasing stress in physicians.” Whilst Kaczor’s study relates directly to physicians, the same logic could be applied to anyone working within high-stress sectors, or even for just generally monitoring employee health and wellbeing. Ultimately this can be used to reduce employee absenteeism and presenteeism, it also allows businesses to implement preventative strategies when it comes to employee wellbeing rather than the reactive ones that are currently in place and doing nothing constructive to solve the employee burnout crisis. The current measures are simply performative measures that businesses are using to make it look like they’re doing something about the crisis, rather than actually solving the problem, which has inherent benefits for the business.

According to a study by the IEEE, “a generic omnibus machine learning model trained to predict mood is inherently limited in the performance it can obtain,” and therefore individual differences need to be accounted for through multi-task learning (MTL), which will allow each person to have a model customised to them. By personalising the way we approach employee wellbeing, businesses can reduce unnecessary employee turnover and boost employee wellbeing before reaching the extent of absolute burnout.