Skip to content


Forecast the Future: Leveraging Data to Predict Home Care Worker Churn


Studies have shown that having at least 70 percent full-time employed nurses fosters continuity of care among many other benefits: reduced costs, better teamwork, and happier and more engaged staff. Under 70 percent, and the quality of client care can take a hit. Every time a home care worker churns or submits a resignation, it creates a series of challenges for both the organization and its clients:

  • Each client associated to that employee needs to be reassigned to a new care worker. This not only requires a recalculation of the best client/care worker match, but also an overhaul of the client’s schedule. 
  • Every time a client is reassigned to a new care worker, the level of care continuity decreases, which could have a direct impact on the quality of care the client receives.
  • Staff shortages force an agency to turn away potential new clients.
  • It raises the risk of missed visits, making existing clients unhappy and forcing the scheduling staff to scramble at the last-minute.
  • Lastly, assigning a new employee to that client creates additional strain on that employee’s workload, route and schedule.

Agencies that have higher retention rates than their competitors are more profitable and provide their clients better care.

The right technology can help prevent churn before an employee even entertains their departure — and can predict hiring needs before employees join an organization too. While this type of insight is only beginning to emerge, the picture is becoming clearer. 

What if you knew, based on data, that in the next two months you will need one mental health worker in a particular region whose shifts will fall on Mondays, Thursdays and Fridays? Or that three new PSWs will be needed to support an urban area? Predictive technology is the way of the future and it will enable home care agencies to hire proactively.

Technology also has the power to predict employee churn by identifying the specific factors — the indicators — that are linked to decisions by employees to leave. When large amounts of data can be studied by a software company with dozens of clients, machine learning can offer such predictions. How would your operations change, for instance, if you knew that in the next six months 10 skilled staff are likely to leave? It will soon be commonplace to use data to decode attrition rates and map these patterns to the future.

The AlayaLabs team is working on building reports for clients that would indicate if they have high risk of employee churn. To deliver that, we are testing the use of predictive analytics to base that churn forecast on employees’ workload, work schedule’s stability, employee travel distance, and more, to actually measure and highlight if specific employees have favourable working conditions. In a second round, it would be possible to assess the notes taken by the care workers during their visits and apply text mining and sentiment analysis. 

Overall, the best remedy for employee churn is to predict it — and prevent it — in the first place. The technology to address many of the issues currently faced by home care agencies is already here, with new advancements coming online every year to help provide even more insight into the pressing issue of attracting and retaining top talent.

Interested in learning more about how technology can help eliminate home care employee churn? Read the full article here.