Automation has been improving back-office tasks in recent years, but 2025 will be the year when artificial intelligence (AI) reshapes clinician home visits.
 
With ambient listening and other new tech features, AI is taking on the most time-consuming tasks so the clinician can stay focused on the patient and not entering results on the tablet.
 
In the 2025 DecisionHealth Trends Survey, 76% of the 37 respondents said ensuring clinicians provide detailed, timely documentation was one of the biggest barriers to success under PDGM.
 
The industry’s leading tech firms and innovative startups are putting fresh attention on how to address these issues head-on. In some cases, they’re significantly cutting the time for start-of-care visits that just seem to get longer these days.
 
New AI-powered tools that have arrived on the market in just the past few months include Netsmart’s Bells AI and WellSky’s SkySense. They join tools from Skypoint AI, Roger Healthcare and others.
 

Ambient listening going beyond taking notes

At growing tech firm AutoMynd, they’ve built clinician-focused tools to streamline several key points in the visit. Most notably, ambient listening technology that can complete the OASIS forms using the conversation between the clinician and the patient.
 
The technology is currently in large-scale use by three agencies, making up about an 8,000-patient census. Early next year, they'll be launching more partnerships with a national EMR and a top home health agency.
 
AutoMynd founder Rohit Shetty has been working for Bayada for the past three years, currently serving as director of automation services. 
 
He built more than 100 automations for the agency, mainly focused on back office and administrative tasks in revenue cycle and payroll. About two years ago, Shetty started working on an independent project to help clinicians in the field.
 
An early prototype was able to generate a visit summary, but Shetty knew that a tool specific to home health needed to do more. A summary of the conversation still left the clinician to fill out the OASIS.
 
“We had to really push ourselves harder to make a home health specific assessment documentation,” he says. “So that it’s not just a regular summary of your meeting.”
 
With ambient listening, the AI can capture the conversation and generate the clinical documentation for you.
 
“You can think of it like advanced speech recognition,” Shetty explains. “It not only transcribes for you, but also intelligently identifies that when it hears ‘vitals’ in the conversation, that information should parse into the vitals section in the assessment.”
 
With AutoMynd, the clinician sits the tablet on the couch or coffee table and goes on with the assessment and the tablet collects all of the information and prepares the documentation for review when you’re done with the session.
 
The clinician crosschecks everything, so something that would have taken 30 minutes to document manually now only takes five minutes to review.
 
Shetty notes that human review is an important part of the process.
 
“With everything that we do, we think clinician can never be replaced,” he says. “You are simply cutting down the waste which has built up overtime in this tedious process of delivering care. Back in the day, we didn’t burden our clinicians with 20 pages of documentation. Care delivery is getting compromised because the clinicians are spending roughly 40% to 50% of their time looking at their device.”
 

Tools reaching beyond ambient listening

Beyond accurate and timely assessments, medication reconciliation is another persistent issue that clinicians can struggle with.
 
Another tool that AutoMynd offers lets the clinician line up the medications on the counter, take one picture and then have the medication list populated for review.
 
And there’s help before the clinician even gets to the house. In a survey of 300 clinicians, AutoMynd found that they spent roughly 10 minutes per visit awkwardly sitting in the driveway as they reviewed the patient chart before entering the home.
 
During the drive between visits, a feature called Martha can help clinicians prepare, Shetty notes. You can ask for a history of present illness, demography or a last visit summary.
 
“The ability to get this information means they can simply walk in the door without spending a lot of time outside,” he says. “It is something that brings an additional level of efficiency.”
 

AI is working for managers, too

AutoMynd’s effort goes beyond building out documentation and quality tools for the clinician. The technology also includes features for clinical managers and quality reviewers so they can monitor what steps may have been missed in a visit or track issues.
 
Managers are able to review all aspects of a visit. For example, if vitals were missing, it’s noted in the system. It’ll also note if there was a change in the patient’s status or medication. 
 
“We want to make sure, beyond just clinicians, that clinical managers can spend time identifying quality issues with the documentation,” Shetty says.
 
This can also help reviewers identify why certain answers were selected in the assessment, cutting down on the time-consuming back-and-forth between clinicians and managers when there are questions about a particular score on the OASIS.
 
Several agencies included in our 2025 Trends Survey noted they were just starting to consider AI tools. Shetty recommends that agencies find a vendor they can work closely with as the technology continues to develop over the next few years.
 
“What we have learned over the last two years speaking to more than 50 agencies and understanding the process, is that everybody does things differently,” he says.
 
Agencies are going to have different documentation needs, so a one-size-fits-all approach isn’t going to work, Shetty notes.