AAPACN is dedicated to supporting post-acute care nurses provide quality care.

Home Health – Navigating Artificial Intelligence in Home Health Care and Hospice

Artificial intelligence (AI) is swiftly transforming the healthcare industry, and home health care and hospice agencies are just starting to see how this technology can improve operations. “The benefits are huge,” says Holly Chaffee, MSN, BSN, RN, principal chief executive officer for Care Central VNA & Hospice, Inc. “AI really improves the clinician’s satisfaction. It eliminates my staff’s biggest pain point by decreasing their time for charting and documentation, gives them more face time with patients, reduces their after-work hours, improves their OASIS [Outcome and Assessment Information Set] consistency, and helps ensure accurate reimbursement.”

This article discusses examples of two emerging AI solutions available for home health care and hospice currently used today, as well as tips for implementation, including getting staff buy-in, what to look for when deciding on AI tools and vendors, and best practices to ensure accuracy, according to Chaffee.

Two Emerging AI Solutions

Ambient Listening Tool

“We began at my agency using AI about two years ago with an ambient listening tool (a voice-to-text platform) that automatically generates the clinician’s visit note when they record it,” says Chaffee. “I was trying to solve the clinician’s biggest complaint, their documentation.”

Documentation was previously taking the bulk of a home health clinician’s day (and night). Chaffee describes the clinician’s new process with the ambient listening tool. “The clinician completes her visit with the patient and then she talks to the listening tool. The tool creates an in-depth summary of the visit. It generates a Subjective, Objective, Assessment, Plan (SOAP) note that addresses all of the patient’s problems. The clinician then reviews the note before it gets copied and pasted into the chart.”

AI Operating System: An EMR Replacement

Another new tool that Chaffee implemented into the hospice side of her organization was an AI operating system, replacing her existing electronic medical record (EMR) system.

Chaffee elaborates on this new system’s capabilities and benefits, saying, “The system is completely AI-based and does everything from getting the referral information all the way to discharge, including billing. It analyzes remote patient monitoring. It identifies the patient’s early signs of deterioration, their hospitalization risk, their fall risk, their non-adherence to medications, wound complications. Anything that a patient could potentially have wrong, it identifies all that and is automatically already in the system.”

Chaffee offers this example of its use: “If a patient is declining with chronic obstructive pulmonary disease or congestive heart failure, it picks up all those data pieces in the notes from A to Z in the plan of care and alerts the nurses to those concerns.”

One of the major benefits of transitioning to this operating system is its open-source capabilities, eliminating the cost of extra fees for integration with other vendors. “When you have an EMR, which we’ve been using it for many years, there they do updates. But then as regulations change, you need to add other kinds of vendors to integrate with your EMR, and those costs are astronomical, $10 to 15K for every software you have to integrate. With the operating system, there are none of those fees. It’s an open system where they can join and assist you without those big fees.”

AI Implementation Tips

Determine the Problem You Want to Solve

Chaffee described her first step when choosing an AI solution. “First, I identified the problem I was trying to solve. For me, that was staffing and retention. How do I address the staff’s biggest issue and what is it? I asked the staff and discovered it was documentation. Charting was taking too long, and extensive after-work hours were leading to burnout, so I realized I needed to find a vendor that was streamlining documentation.”

Clarify the Outcomes You’re Expecting

As the next step, Chaffee decided on the outcomes she hoped to see with the new technology. “I wanted shorter visits, the quality measures to improve, and the visit utilization to decrease,” says Chaffee of her expectations. “Those were things that we were all being measured on. And we have been able to accomplish that with AI.”

Assess the Security of the Platform

“Then I really assessed the compliance, the security, and the insurance that the AI company has,” says Chaffee. “I asked, ‘What did they have in place so we’re HIPAA-protected? How is our data protected?’”

The company should be able to provide information on whether their tool includes these features:

  • All regulatory compliance pieces for survey, where applicable
  • HIPAA-protected security
  • Insurance in case of a breach

Ask About Transparency and Responsiveness

“Make sure you have a transparent and responsive relationship,” says Chaffee. “The AI company should be picking up the phone when you have an issue or a question about how something works.”

So if the company isn’t responding promptly during the discussion phase, it may be best to look at another organization for the agency’s needs.

Pilot and Customize the AI Tool with Clinicians

Chaffee notes that building the tool with the developer and having the clinicians involved to discuss their needs is important. “Involve the clinicians from the beginning. You can’t sit in the administrative offices and make a decision to do this without involving them at the entry level. They can tell the company what they need to do, what they don’t need to do, and what’s repetitive and what’s not.”

Piloting the AI tool before full implementation is essential, says Chaffee. Four of her staff signed consents and then started implementing the AI tools into their workflows. She then reviewed their success after the trial, and three out of four of them had improved their workload and decreased their visit times.

Address Different Learning Styles During Implementation

“You have to know your staff and provide support to them when and how they need it,” says Chaffee. “If they need you to sit down and work with them face-to-face, then that’s what you need to do. If they can understand it just by looking at the online training, that’s great too. Every adult learner takes information in differently.” Ensure staff have what they need to be successful with the change, regardless of learning style.

Implement Processes that Safeguard Accuracy and Patient Care

“Our best practice is that AI is the draft. Then the clinician approves the final document and the plan for the patient,” says Chaffee. “Once they copy and paste what the AI creates into the chart, it reflects on their nursing and therapy licenses. This is your patient. And we drilled that into our staff. AI is just a tool to help them. They are ultimately the clinicians who drive care, and these are their words once finalized.”

It’s important to review any information generated by AI tools. Confirm with the vendor that they have review options in place before implementing a tool that integrates with your agency’s processes.

Conclusion

“I think other agencies need to understand AI is here to stay. If you don’t embrace the new technology, you may fall behind in your business,” says Chaffee. “Think back to the days when we charted on paper. I was a weekend nurse, and I had 15 charts I had to have documentation completed, everything wrapped up to put in the trunk of the VNA car on a Sunday night, so the nurses could then take the charts and visit their patients Monday morning.”

“Go back to that time and then remember the first computer that came out and the impact it had,” reminds Chaffee. “Technology is going to keep changing and our environment keeps changing, too. So don’t be afraid to embrace a change. Do it right. Do it step by step, knowing that no change is going to be easy in your organization. But once it’s implemented, staff might say, ‘I don’t know how I could do my job without this!’”

Are you looking for more strategies for AI implementation? Register for AAPACN’s free webinar series, The Executive Playbook: AI Strategies for Post-Acute Care Leaders.

This AAPACN resource is copyright protected. AAPACN individual members may download or print one copy for use within their facility only. AAPACN facility organizational members have unlimited use only within facilities included in their organizational membership. Violation of AAPACN copyright may result in membership termination and loss of all AAPACN certification credentials. Learn more.