Artificial Intelligence Could Help Case Managers Improve Efficiency and Outcomes
March 1, 2024 6 minutes read
By Melinda Young
EXECUTIVE SUMMARY
Healthcare and case management might benefit from artificial intelligence (AI). The technology could improve documentation, translation services, and accuracy.
- Case management jobs could be affected, so leaders should be part of the decision-making process about the purchase of AI technology and how it is used.
- The American Medical Association calls AI “augmented intelligence” and emphasizes how it can assist in healthcare.
- AI can anticipate when a patient is ready to leave the hospital and help with decisions about which level of care is needed.
Artificial intelligence (AI) is poised to take over the fields of media and marketing, banking, legal services, and programming. It also is used in the healthcare field, including case management.
That poses the question: Will AI replace case managers?
“Does that weigh heavily on my mind at night? Yes, it does,” says Lisa Bednarz, LCSW, ACM-SW, CMAC, the regional director of case management for RWJ Barnabas Health in New Brunswick, NJ. “My way of looking at this is we have two options. We can be a part of the process of implementing AI into our work, or it can happen to us. I want to be part of the process. I don’t want to be a bystander bulldozed over.”
Healthcare in 2024 is in an outcome-driven environment with value-based work, Bednarz notes. “I don’t see us going away from that even with the adoption of AI,” she says. “I see it as a tool with clinical decision support, as well as a way to mitigate the staffing shortages we’re seeing because people don’t want to come to work in a hospital right now.”
The American Medical Association (AMA) House of Delegates reframes AI by using the term “augmented intelligence,” saying it is focused on AI’s assistive role and how it can enhance human intelligence, but not replace it.1 An AMA study revealed that 65% of physicians see at least some advantage to using AI in their practice. Only 38% of physicians use AI in practice, mostly for documentation, translation services, and diagnosis assistance. Physicians’ chief concerns are that AI could affect the patient-physician relationship and patient privacy.1,2
The main goal of AI in healthcare is to eliminate human error. “I want to make it as easy as possible for my team to make the right decisions when they’re caring for patients,” Bednarz explains. “It’s about making those decisions easily, surfing information they need as fast as possible.”
Potential Benefits
Potential benefits of using AI in case management include:
- anticipating where a patient will go when they leave the hospital and what level of care they will need;
- anticipating when the patient will be ready to leave the hospital.
“These are two big opportunities with predictive analytics,” Bednarz says. “The next frontier is matching the individual to the right service they might need when they leave.”
The current case management practice is to give patients a list of options for their next level of care, such as a skilled nursing facility or outpatient rehabilitation services. “You give someone a list with quality scores, and the person needs to make a decision that is not individualized to the patient,” Bednarz explains.
A more personal option would be to hand the patient the name of a facility with the best outcomes for patients with similar needs. “We really want to be outcome-driven now in our current health environment,” Bednarz notes.
Precision case management, assisted by AI, is expensive and unlikely to be adopted by health systems that are under cost constraints. “In our current landscape, we don’t have excess money to pay for that,” Bednarz says. “There are many things that could make the patient experience better and improve outcomes, but we — the healthcare ecosystem — can’t afford them.” This likely will change, which is why case management leaders might want to learn more about the services AI offers.
As new AI options are developed for case management, the experts — those on the frontlines — need to provide information that will help AI develop the best answers. “Leadership needs to know what piece of information they’re looking for and what would change the way case managers handle a case,” Bednarz explains. “I’ve been in administration for a decade. I need to know from the frontline which information would change the way they provide care and what piece of the puzzle is missing.”
Case managers could be a part of that conversation. “You’re saying, ‘I think we have an opportunity to look at this differently, and I think leadership appreciates that viewpoint,’” she says. “I never would say ‘no’ to an opportunity.”
The opportunity could be to participate in an end-user work group that provides part of the roadmap for the future of using AI in case management.
“Hospitals are really focused on shared governance now, so include innovation as part of the shared governance structure,” Bednarz suggests. “For people to buy into change, they need to be included in it.” Although case managers experience time constraints, it would be helpful to build in time for their participation in these implementation and innovation efforts, she adds.
Some healthcare leaders already use data analytics to match patients to programs, but social services are behind on digitization, Bednarz says. “Social services is the last frontier to digitize and share data, and we’re a little bit behind in using historic data for predictive analytics,” she notes. “There is a great opportunity for automation and robotic process automation to help apply for services for patients, which takes an administrative burden off of frontline workers.”
For instance, if a patient needs food stamps, and there is a long application form, the right technology could use a bot that reads the patient’s record, fills in the application, and submits it, she explains. This scenario is not that far down the road.
“It’s a very doable concept,” Bednarz says. “The rate-limiting factor is that all of our social services tend to be state-based, so when you’re developing these programs, you can’t do it on a large scale. You have to create 50 different codes instead of one code for a federal program.”
Focus on Care Transitions
This is why emerging AI solutions are focusing on transitions of care instead of social determinants of health. “The solutions we need — the ones focused on social determinants of health — have a complexity that you need to ingest data from outside sources, including the government. That’s challenging,” Bednarz explains. “With transitions of care, we have a lot of the data we need internally.”
Bednarz has worked on creating a data-driven way to assess patient complexity. One solution she addressed is to use patients’ ZIP codes as a feature to represent social determinants of health because those often are tied to where a person lives.
“What I’m excited about as we continue to obtain patient-level social determinants of health information is I will no longer have to use ZIP code as an indicator,” Bednarz says. “I can look at a patient’s needs on their level.”
As hospitals collect more data, this will improve care transition predictions. “We know from evidence that predictive analytics and AI are more accurate than humans, and why wouldn’t we want to apply that accuracy to taking care of people?” she asks.
But there is bias in AI, and the people developing programs that use AI need to mitigate bias, she adds. The choice is not binary. It is not using no AI or using completely autonomous AI, where it makes decisions and provides care.
“This is a spectrum. Most of the AI we implement is meant to support clinicians and not make decisions for them,” Bednarz explains. “It’s to help them make the best possible decisions.”
REFERENCES
- American Medical Association. Augmented intelligence in medicine. Updated on Nov. 28, 2023.
- American Medical Association. AMA Augmented Intelligence Research: Physician sentiments around the use of AI in health care: Motivations, opportunities, risks, and use cases. November 2023.
Artificial intelligence is poised to take over the fields of media and marketing, banking, legal services, and programming. It also is used in the healthcare field, including case management. That poses the question: Will artificial intelligence replace case managers?
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content