By Michelle Despres, Vice President of Product Management, One Call
When I say the words “artificial intelligence,” or the acronym, “AI,” where does your mind go? Do you immediately imagine machines and droids taking over the world? Or do you have a less dramatic outlook on the subject?
Artificial intelligence (AI) relies on the use of computers and machines, mimicking the problem-solving and decision-making capabilities of humans. AI comes in a variety of forms, including speech recognition, virtual agents for customer service, computer vision, and recommendation engines.1
Contrary to what popular science fiction books and movies depict, AI is not here to take over planet Earth or replace humans. Instead, it is coupled with our own critical thinking and natural abilities to make processes more efficient.
When we examine how AI is being used in the workers’ compensation industry, the benefits are undeniable. It can enhance service delivery, reduce errors, analyze vast amounts of data to predict trends, streamline claims management, and improve customer satisfaction. We are embracing and building trust with this technology at One Call to better serve injured workers – here’s how:
1) Reducing Delayed Recovery
AI powers our At-Risk Program – a data-driven program that identifies potentially at-risk/high-cost cases early on, helping to mitigate costs, reduce delayed recovery, and ultimately, drive better outcomes.
Over time, this iterative process has been refined with AI, leading to improved precision and identification of injured workers who would benefit from early clinical intervention:
- 2018 – Established initial at-risk “early warning” triggers
- 2019 – Developed first-generation predictive models using claim attributes
- 2020 – Incorporated optical character recognition (OCR) to extract content from provider notes
- 2021 – Deployed second-generation predictive models using claim attributes and OCR inputs
- 2022 – Incorporated natural language processing (NLP)
- 2022 – Deployed third-generation predictive models leveraging structured and unstructured data
The result – clients who use One Call’s At-Risk Program have experienced up to a 30 percent reduction in utilization outliers, driving lower claim cost and duration.23 While such models are a significant part of our program, we’re also seeing them incorporated into electronic medical record systems across the country. As a result, healthcare providers and clinicians are more equipped to use an evidenced-based approach for their patients, leading to better determinations on required care.4
2) Injury Prevention
While AI is instrumental in reducing delayed recovery, it can also assist with injury prevention.
For example, some tools use computer vision to record a worker as they actively perform a task. Red, yellow, or green markers on the individual’s joints indicate whether they are at high, medium, or low risk of injury. After the motion capture is performed, the application produces an easy-to-read risk report, including tips on how to mitigate the risk. With this multifaceted analysis, ergonomists and safety professionals can tailor the best solution for the task, individual, or jobsite.
Traditionally, clinicians and ergonomists made an educated guess when determining risk of injury. With the help of AI-powered computer vision, a lot of the guesswork is now eliminated, and time is saved by running multiple calculations automatically.
3) Patient Engagement
AI is also helpful in overall patient engagement. A great example is One Call’s sentiment scoring.
Using natural language processing (NLP), we can capture and analyze data from every interaction we have with our stakeholders – currently, that’s more than 30 million calls each year. The technology then generates an unbiased sentiment score for every interaction.
These unique insights enable our care coordinators to receive live feedback while engaged in calls. From there, our team can identify trends across our customer base – such as billing issues, communication breakdowns, and knowledge gaps – and proactively intervene with solutions that benefit everyone.
AI Isn’t Perfect
While AI can help solve problems, it doesn’t comprise critical thinking skills – that remains a human skill. AI uses algorithms to produce results. As we continue to refine these algorithms, AI appears to think on its own, but it’s just performing the task we’ve trained it to complete. The more input we provide to the algorithm, the greater the results or output.
AI solutions are continually evolving concepts that require thoughtful consideration. Algorithms rely on large datasets to optimize ingoing information and create sensible output. Commonly, organizations face challenges tied to competing business priorities and require IT resources, collaboration amongst stakeholders, and data standardization. Recognizing and understanding aspects of structured data (standardized text, numbers, and dates) and unstructured data (images, audio, PT notes, PDF files, etc.) helps ensure useful algorithms are created. Support from a great data scientist is also helpful to build high-quality machine learning models and effectively augment AI.
If you’re a little hesitant about AI, but you recognize its ability to make processes more efficient, I encourage you to look at the tools available and try something. Ask yourself: “What can AI do to support my work, my company, or my practice?” Whatever the answer is, AI can’t perform the task alone – it needs an intelligent, critically thinking human being like you to get the job done.
This is a sponsored post from WorkCompWire marketing partner One Call.
- IBM. (n.d.). What is Artificial Intelligence (AI)? Retrieved from https://www.ibm.com/topics/artificial-intelligence
- One Call national data. Reflects results from customers participating in One Call’s At-Risk program.
- IBM. (n.d.). What is predictive analytics? Retrieved from https://www.ibm.com/topics/predictive-analytics
- Loria, Keith (2023, December 1). What’s the Impact of AI on Physical Therapy? APTA Magazine. American Physical Therapy Association. Retrieved from https://www.apta.org/apta-magazine/2023/12/01/impact-ai-physical-therapy