This article is sponsored by Caspar.AI. This article is based on a Senior Housing News discussion with Greg Robertson, Senior Vice President & Managing Partner at Notify, Alain De Leon, Vice President of Quality Care at Lutheran Life Communities, and Joe Anderson, Vice President at Caspar.AI. The discussion took place on July 21, 2022. The article below has been edited for length and clarity.
Joe Anderson: I’m going to pose a rhetorical question, you don’t need to answer this but I want you to think about what happened last night in Memory Care apartment number 32. Let’s say between the time you left the reception last night and the time you got here for breakfast in that 12-hour period, what happened in that apartment? If you run a community you may have a specific apartment number.
If you’re in a corporate office, think about one of your communities in one of your apartments. What happened in that apartment last night? How will it affect that resident’s behavior today? What could you learn from nighttime behavior that would impact the residents’ behavior tomorrow and going forward? Alain, what are the things that you think about in terms of residents and their activities at night? What concerns you?
Alain De Leon: Yes, what concerns me about activities is that a lot of them we don’t know, mainly because we’re not there. We don’t have the data for that. We don’t know what they’re doing in their rooms and the next thing you know they’re not behaving well during the day because probably they did not sleep at night. That’s one of the areas that are shortcomings for monitoring our residents at night. Another thing is there are a lot of behaviors at night that will indicate what will be their behaviors during the day.
A lot of their behaviors just like what Dr. Mastriani mentioned in the previous panel, the lack of sleep does affect behaviors of residents during the day. That’s the thing that we don’t know. Sometimes we know based on accounts from residents and sometimes we don’t. Those are mainly the things and also with regards to monitoring, we all know that staffing is an issue right now. Everybody is experiencing that nursing staffing per se and then, they mentioned as well during the night, we have the skeleton crew. We don’t have enough people to monitor residents at night. Going back to the question of what happened last night, nobody can answer, nobody can provide that information.
Anderson: I’m going to turn to Greg from Notify. Talk a little bit about how data that’s collected can get transmitted to staff members and in what forms. How can they then use that data?
Greg Robertson: I think that all of these technologies, we’ve heard a lot about interesting technologies today, just design and other monitoring tech solutions. All of that is great but what’s important is how is that delivered to the staff? How is that delivered to admins, but really the care staff that are on the floor? How does it integrate with the technologies that you’re already using? Does it integrate with your electronic health record?
In your case, if we notice that a resident is getting up more in the night than they had before, does that become a note in the electronic health record that’s then delivered to the caregivers? Can we send an alert if we recognize a fault? Where do we point those recognitions of changes in activity and how do we deliver those to caregivers in a meaningful way? There’s a lot of different ways that we can do that but the goal is to give the caregiver an easier way to communicate those needs to their staff.
Anderson: [We’ve been given the ultimate trust.] Having been a former operator, I understand that families bring their family members to us and they expect us to do 24 hours a day, 7 days a week, 365 days a year, what’s best for that resident. I don’t want to say we tend to forget the half of the day, but it is less important what happens when we’re not seeing the resident than the half of the day when we do see the resident.
Putting two and two together, that nighttime behavior with data that’s transmitted to the staff then becomes an opportunity for the staff to fulfill that goal. The ultimate trust has been given to you and to me to take care of those older people. Alain, if you had that information today about what happened in apartment 32 last night, what are some of the ways that you might be able to use that data to the betterment of the resident and their life in their Memory Care Community?
De Leon: We always say that if we knew back then what would we have done- differently now. With that data, a lot of things we can do first is staffing. If you knew that, a lot of things are happening between three to 5:00 in the morning, then we probably would staff more at that time rather than putting in the whole shift when you really need it between 3:00 and 5:00 AM.
Like we said, with the scarcity of staff, we need to work efficiently. We need to work smarter, not harder. We will not be producing any more nurses anytime soon. They’re slowly going away from healthcare. We need to work smarter with that. Next thing was, if we had that data, that could help us prevent any accidents in the community. If the patient falls down, we will be able to attend to the resident right away.
Also, one more is the changes in behavior. Of course, if you identify and you observe those changes in behavior, we as healthcare workers are responsible for acting on it. We are not just going to brush it away and say, “Well they had a rough night,” or whatnot. We will probably address that with their attending physicians because as a nurse myself, that is valuable data, valuable information that will be able for us to do our job more efficiently.
Anderson: That customer client community group is using the product now in other communities. We already know that care for certain residents has been altered to the benefit of those residents based on information that’s been learned about those residents at night. One of the things that we learned that’s especially true, and you all know this, is that the first few days that somebody moves into memory care is when we know the least about them.
Sometimes that data that pops up at night might indicate there’s a UTI or some other issue that needs to be dealt with. You know that quicker than waiting for a few months to have some incident happen. The other thing that we know and have learned, memory care residents can’t speak for themselves in some cases, or they can’t speak accurately for themselves so it’s hard to get information or feedback, but as they move in, we can learn quickly from the data that gets provided through the system which then gets pushed out through Notify and other sources. Greg, what are some of the things that can be retained from the data that’s gathered at night?
Robertson: I think that again, technology, and it’s been said in multiple sessions here is, technology is a tool. It’s never going to replace the love of a caregiver and the interaction of a caregiver, but they are very valuable tools. We heard in an earlier session with the SafelyYou people where a resident was on the ground every morning. What they thought was an unobserved fall, but what it really was because they were using the technology, they were able to see that this resident was actually getting on the ground to get their embroidery tools.
That’s the type of thing that overnight we don’t see residents or caregivers are not with the residents in the room, they’re not observing these activities. The data that you can capture from monitoring solutions, from motion detection or all the other things that people have been talking about is really valuable to understand, “What does a person’s normal night look like?” Do they normally get up five times a night to go to the bathroom? If they do, great, that’s what they do. If Joe doesn’t and you get up a couple times a night, then we want to start notifying staff.
Just picking up on the normal activities of the daily life of those residents in the normal nighttime activities specifically when the staff is low and we’re not observing those residents and then being able to deliver any discrepancies in those when they’re needed are the types of things I think are most valuable with monitoring.
Anderson: That leads into another aspect which is related to staffing. As we all suffer through shortages of staff and turning over of staff and so on, sometimes we leave a nighttime staff member and say, “Visit room 32 at three o’clock every morning.” How do you know they did? There is data that can be gathered that shows when there’s an entrance from the outside.
If you don’t see that three o’clock entry every morning, you know that that staff member didn’t deliver what they were supposed to do on behalf of that resident. It’s not that you want to use it as a watchdog, but it is an outcome that comes from collecting the data. The other area that’s important at night, especially when staff is not in and out of the rooms or seeing residents all the time is fall detection and alerts. One step is predictive, which is great.
We know that there’s a risk here. What happens when a resident falls and they are unable to reach a pull cord, or they don’t know what the pull cord is. The best solution I ever saw at this point was a sticky note in a memory care community in a resident’s apartment that said, “Don’t forget to put on your pendant.” Think about it. Who’s going to detect the fall, how are they going to detect the fall? Who’s going to get there and how fast are they going to get there? We can push messages through Notify to staff within a matter of minutes that there has been a potential incident in apartment 32. Again, I ask you to think back about apartment 32 last night. What happened and what do you know about it?
Robertson: I think just to add to that a little bit, it’s not the negative things that you want to capture either. We want to capture the positive things that the caregivers do. It’s a great tool to use if they don’t go on their 3:00 AM rounds, but I also want to say, Joe, you did a great job responding to this event that happened last night. With the staffing problems that we have, Alain you just talked about and everybody knows, engaging the staff is a big piece of this.
We all talk about resident engagement, that’s vitally important, but also engaging the staff and letting them know that they’re doing a great job, because as an earlier presentation said, if you’re working in memory care, you love those residents, or you’re not going to be working in memory care very long and recognizing that the good work that they do and using technology to do that I think is also a key piece of what we’re all trying to do.
Andereson: Alain, along the lines that Greg is talking about, how can you see this data informing your management of staff during the day or impacting your training or helping staff members realize the benefit of their work, feeling satisfied about their work?
De Leon: As a nurse the satisfaction of a nurse is to see your residents, your patients doing well. If the satisfaction of us nurses, and CNAs is to see that we have prevented a fall of a resident. Those are the main things that really puts a pat on the back of our nurses and CNAs. Then going back to what can we do about that data? It made me think about when my mother-in-law used to live with us.
If we knew that my mother-in-law wasn’t sleeping at night, my husband and I could have done something to my mother-in-law, rather than now she’s in a long-term care facility. We could have adjusted how we would approach her. We could have brought her to her attending physician more frequently just to monitor that change in behavior because we check for UTIs, which will be prompted but there wasn’t. I think there’s a lot of benefits to having AI in our setting. Like what the other panelists have mentioned, based on the data that you receive, it’s how you act on it and how you process it. It’s not just data and it’s there. It’s useless if you’re not going to do anything about it, but to start, we need to have the data first before we can act on it. I believe AI is one of the things that are up and coming, and I think it will stay as we go into the future. I think it will be part of nursing care, part of how it will drive how we take care of our residents.
Robertson: Not to beat a dead horse, but I think how you deliver that data is very important. You already have more to do than you can possibly do. When you go into a building and say, “Oh, I’m going to give you this dashboard and it’s going to show you all these great things,” and you’re going to say, “Well, I don’t have enough time to look at what I have right now.” It’s important to integrate technology into what your workflow is and to have a customizable solution that can integrate to the tools you’re already using because giving you something else to look at is not helpful, but integrating it into the things that you’re already doing is very helpful.
Anderson: So the data that can be collected at night can be driven through an existing system?
Robertson: An existing electronic health record, the existing nurse call system, the existing communications tools that the staff use, and tying those into things that you’ve already invested in and you’re already using are really the ways that these types of AI and everything else are going to really get into the workflow of a nurse. We don’t give them another device. We don’t want to give them another website, another password, and all of those things. Integrating it into what you’re already doing is a key, in my opinion.
Anderson: How do you suggest talking with families about the privacy concerns and the data, and how that’s balanced with the need for care?
Robertson: I would suggest that the conversation be about how we live up to the trust that you’ve put into us, that we want to put every possible tool in the toolkit to work on behalf of your mom. This is what’s best for your mom. We’re not using cameras, we’re HIPAA compliant. We’re not identifying mom by name, but we’re collecting data and the data informs her care plan.
The other line of discussion may be asking them what kind of information they’re able to get from their mom. They may say, “Mom doesn’t talk, or I don’t get–” Well, that then becomes the answer to the family, which is we can get the data to speak on behalf of your mom. It’ll tell us that she’s getting up six times a night and walking around the apartment, but not going in the bathroom. The data will tell us that she’s having restless nights of sleep.
The data will give us something to go to her physician with, and perhaps work on her medications. It’ll give my care staff data that they can work with your mom tomorrow. Wouldn’t you agree that having that data is a good thing? Other than that, there are always going to be privacy issues or questions, and obviously, it’s going to come from the family. It’s not going to come from the resident.
De Leon: One thing that we approach, not just for AI, but for anything that we will introduce to our residents is in the approach of need. Maslow’s Hierarchy of Needs. We have the physiological, and then you can have the safety needs. Usually, in nursing, we go to the safety needs. This is what is safe for your mom. If we identify that this is safe for your mom or for your loved one, usually they are very accepting of that because, for nursing, we also believe that safety supersedes everything.
If this is to save your mom’s life or to save her from falling, usually you get a good buy-in from the family for that, if you approach it towards safety needs. At this point, we can’t prevent falls, or we can identify or address falls right away if we look at it in that sense because like what you said, they don’t have cameras on these Casper AI.
It’s really just how many times and the data is very confidential or very private. That’s just a tidbit of how we approach things to do that.
Robertson: I think including the family and the technology is a big key to that too. How you keep the data safe is the company’s job, it’s HIPAA compliance, it’s being comfortable with who you’re working with, but introducing the technology and letting the family use it as well. Making a family part of that communications or a part of that reporting. Having the ability to do that could also alleviate some of those concerns.
Anderson: Let’s say there’s a provider that is very interested in the technology, but is not ready to buy it or deploy it. Could you spend a little bit of time telling us what could be done today in a building so something like this could be adopted and deployed say two years from now? If somebody’s planning a renovation or a new build, how complicated is it to plan for something like this for the future?
Robertson: I think it really depends on the technology that you want to implement. A lot of the stuff we’re talking about doesn’t require a lot of infrastructure in a building, but if you’re doing additions, it’s really a lot of internal cabling and things like that for technologies in the future. What we’re talking about is already using existing infrastructure, but I think two years from now, it’s hard to know if you need wires anymore. It’s a tough question to answer.
Anderson: At the current time, you can use a combination of replacing light switches with the GE Jasco Switch, plus a certain number of sensors, and then a device that will read the data that’s collected, assuming that it’s a fairly new building and that there’s reasonable connectivity in that building. If you’re running a 150-year-old building that’s being redesigned, there’s going to be some difficulties, but if it’s less than 25 years old, probably not. You could do an apartment in two hours tomorrow to be outfitted. There are things coming now, wireless that will be better in the next couple of years.
Robertson: It’s a great point. I think there are still lots of buildings just based on the reaction that don’t have great Wifi, but I think that one thing that needs to be covered right away, because the potential residents and the future residents moving into these buildings are going to require that. I think of my parents, if they can’t connect to Netflix in their room, they’re not moving.
Caspar.AI provides safe, healthy and engaged living experiences for residents of retirement communities. The system is customized for luxury residential, assisted living, and memory care communities. To learn more, visit: https://caspar.ai/.