Skip to content

One million inpatient falls occur annually in U.S. acute care hospitals. Sitters, also referred to as Continuous Patient Aids (CPA’s) or safety attendants, are frequently used to prevent falls in high-risk patients. While it may make intuitive sense to use sitters to prevent falls, it does beg the question, what’s the evidence that they work?

We discussed with Drs. Adela Greeley and Paul Shekelle from the West Los Angeles Veterans Affairs Medical Center theirrecent systematic review published in Annals of Internal Medicine. Their review identified 20 studies looking at this issue (none of which are randomized trials). To sum up their findings, there were only two studies comparing sitters to usual care and they came up with conflicting conclusions (in one, the fall rate was lowered; in the other, it was not). In the other 18 studies, alternatives to sitter use were evaluated. The only thing that seems to have some evidence for was video monitoring (fall rates either stayed the same or improved, with a decrease in sitter usage).

We also talk about multi-component interventions and how we should think about them. One intervention that is sometimes included in multicomponent interventions are bed alarms, which we discussed in our very first GeriPal podcast. It’s also the podcast where we dreamed up the “anti-bed alarm” that would alert patients who haven’t gotten out of bed yet. Now that’s a fall intervention that that I can get behind.

by: @ewidera

Eric: Welcome to the GeriPal podcast. This is Eric Widera.

Alex: This is Alex Smith.

Eric: And, Alex, who do we have on our virtual studio?

Alex: In our virtual studio from Los Angeles. We have Adela Greeley who’s a hospitalist at the West Los Angeles VA, and is affiliated with UCLA School of Medicine. Welcome to the GeriPal podcast, Adela.

Adela: Thank you for having us.

Alex: And we have Paul Shekelle who’s a general internist, also at the West LA VA, and worked with our Chair of Medicine, Bob Wachter, on the patient safety report of which the paper we’re going to talk about today is the descendant about Sitters and Patient Safety Strategy to Reduce Hospital Falls published in Annals of Internal Medicine. Welcome to the GeriPal podcast, Paul.

Paul: Thank you.

Eric: We are very excited. We’re going to be talking all about sitters or CPAs or whatever you call them in your hospitals and do they help with falls. But before we go into that topic, we always ask for a song request. Do you have a song request for Alex?

Adela: Yes, we do. Fall On Me by REM

Alex: And why this song?

Adela: Well, one, I love REM and, two, whenever I’m admitting a patient, I think I always say, “Oh. Please. Please. Don’t fall on my watch.” Please. Don’t fall on me.

Alex: Please don’t fall on me. It’s great.

Alex: (singing)

Alex: Oh. Poor REM. I butchered that. Well, sorry, Michael Stipe. I didn’t have a full band behind me, and Eric didn’t learn the backup part.

Eric: Oh. Yeah. I was supposed to learn the back up part. You told me I had to learn the backup part. I totally forgot about that.

Alex: Well, hopefully somebody else will request REM in the future, and I’ll have another chance to do better justice to that. Terrific.

Eric: I will have another chance to forget that I was supposed to do something.

Alex: Terrific band.

Eric: We’re going to be talking about your Annals of Internal Medicine paper, Sitters as a Patient Safety Strategy to Reduce Hospital Falls. We’ll have a link to it on our GeriPal website. But before we talk in the paper, why did you get interested in this subject?

Adela: This was actually a subject that was chosen by central office at the VA to look at the effectiveness of sitters on falls. And so Paul was the lead in this. And then I was asked as a hospitalist who has used sitters a lot in my practice whether or not I would want to join the group to look at the evidence behind using sitters for falls.

Paul: Yeah. Let me expand on that just a little bit. We are one of four centers that the VA supports to specifically review the evidence on things that central office once reviewed on. These are things that usually come up from the field, questions that field practitioners have that for which the VA wants to make some kind of policy decision, but first they want an evidence review. This one had enough support from VA field practitioners to say, “Hey, let’s find out what the evidence about this strategy looks like, so that we can decide what we should do as a system about it.”

Alex: I’m sure many of us who practice in hospitals as we do are very familiar with the concept of a sitter and intuitively it makes sense. And I wonder if you could talk a little bit more about the rationale behind sitters and why they… If you know something about why they were instituted in the first place.

Adela: You know, we didn’t find a lot of background history as to when exactly the first sitter was used or how it came about. And I think, as you said, it is just intuitive that you would think if you have a patient who might be at risk of falling, having someone sitting there at the bedside would be the best way to prevent that patient from falling. But I don’t actually know how long ago…

Paul: I started in 1982. They were already being used in 1982. So I think that they, like the proto-sitter use, is lost in time.

Eric: Now there are some things that we do that may not have evidence and they’re really bad for patients, but what’s the harm of just having like a sitter? Somebody hanging out with someone making sure that they don’t fall.

Paul: Well, there’s a couple of potential harms. Okay? The first one is, is that these things are really expensive. And so if a hospital is having to spend X amount of resources on sitters. Okay? Well, if they offer a benefit? Great. Okay? But if they’re not offering a benefit? Then the harm is opportunity costs for that money to be spent on something else. All right? And then a second harm, which was not systematically reported on, in that there’s no systemic evidence, but we certainly came up anecdotal evidence, is that the sitter themselves can be harmed by the patient. And so there is an anecdotal body of evidence about things that happen to sinners from patients who are demented or aggressive or whatever it is. So it’s not a zero risk thing.

Eric: And are most sitters, do you think to try to prevent falls? Is that what they’re used for? Probably most commonly, I know there’s potentially other reasons that they may be used.

Adela: So in the studies that we found, the reasons for sitters were falls, [inaudible 00:06:16] and then also suicide risk. And when they were being used for [inaudible 00:06:20] or suicide risk, those patients were excluded from the study. So those are the three main reasons that we see sitters being used most often.

Paul: Yeah. And everything that we talk about in sitters going forward, it doesn’t have to do with patients that are having sitters because of suicide or because of wandering. Okay? That’s a separate kettle of fish that we’re not dealing with.

Alex: Can we talk about terminology before we dive into the article a little bit? What is the best word for sitters? I know at our hospital, we’ve started to call them CPAs for what is that Certified Patient Assistant? Is it okay to use the word sitter? It’s certainly the term that’s most familiar to most clinicians.

Adela: Yes. So as we came across in our review that some places call them sitters. Some places call them safety attendants. Some places would call them one-to-one specializing. And so we didn’t find one term that everybody uses or one term that anybody found to be offensive, if that were used.

Paul: Yeah. But we use the term sitters in the article because that seems to be the most commonly used term. But something that you brought up was a really good point. You use the word certified. Right? So that begs the question. What are the qualifications to be a sitter? And, at least in the articles that we looked at, they never specified exactly what you’d had to do to be a sitter. And, at least in a couple of them, they were volunteers. All right? But it’s conceivable that a different kind of position, a different kind of certification might result in different kinds of outcomes and what we actually filed.

Alex: And I wonder about if there’s, is there a pejorative sense of the word sitter, like babysitter? Is it infantilizing patients to some extent to use that term or the role of the person who is watching them?

Eric: I think the reason we changed it from sitters to CPAs is we actually don’t want them just to sit. We want them to do things to help first get this person activated, rather than just looking at their phone and sitting there.

Alex: Or watching the TV is my general experience is they have the TV turned towards them. And then I go into the room and turn off the TV many times during the day.

Eric: Yes.

Paul: Well, yes, that’s a good point. What are they supposed to do? I mean, this gets to the same issue that in my previous answers is that, at least in the articles that we looked at a description of what this intervention actually is, was usually nothing more than sitters. What they were told to do, what their job was, I mean, presumably they were told don’t let the patient fall. Don’t let them walk out and walk around the nursing station. But what the actual responsibilities are, it’s not detailed in any of these articles.

Eric: So let’s talk about your article. What, if you can just summarize what you did in your article?

Adela: So it was a systematic review that we did of the literature looking at the effectiveness of sitters on falls. We looked through… How many titles was it that we looked through?

Paul: A very large number.

Adela: And, you know, looked for anything where it was either sitters as the primary intervention to prevent falls or alternatives to sitters looking at falls specifically. And we found two articles that purely used sitters in comparison to usual care in the hospital. And all the rest of our articles were alternatives to sitters. And what we found was that there is very little published studies that actually looked at sitters versus no sitters. That the majority of we’re looking at sitters versus alternatives to sitters and of those, the alternative to sitters, the video monitoring was found to have the best and most consistent evidence that not only did it not worsen fall rates in the hospital, but it also decreased the sitter use significantly in the hospital.

Paul: Yeah. But just a couple of followups there. I think the distribution of articles that we found tells you how this is perceived in the US healthcare climate. Sitters are taken as a prima facie, on the face of it, good. Right? It’s a good thing. And so nobody’s trying to figure out whether sitters reduce falls. They’re all trying to figure out how to decrease the cost of this. And because every article was about decreasing the cost of sitters, sitters costs too much money. We have to figure out something that we can do differently. Ergo, we’re going to try whatever these various alternatives are and compare them to what sitters achieved.

Alex: And I find it fascinating that some… To learn about alternatives to sitters, could you go into a little bit more detail about what are these alternatives? What else can we do other than usual care?

Adela: We kind of put it into four different categories. So one category, of course, was the video monitoring. Another category was creation of these constant observation units, where they would cluster four patients together into a room where they had one person directly observing them. Another intervention that was used was doing nursing assessment. It was basically taking the ordering away from the physicians. The physicians wouldn’t be the ones who would order the sitters. It would be through protocols that the nurses would use to determine who best needed a sitter. And then the fourth was a miscellaneous category where a lot of these studies would look at the need for sitters in their particular healthcare system. They would look at alternatives such as bed alarms. They would look at other alternatives that they could use that would benchmark with other hospitals, and put in place interventions to see what would happen to their fall rates while they were trying to decrease their sitter usage.

Paul: Yeah. Just to expand on that. This is one of the challenges with doing this kind of a systematic review is it’s not like a systematic review of, okay, well, which drugs for hypertension also have the best cardiac effects. So we’re going to compare beta blockers. We’re going to compare calcium channel blockers. We’re going to have diuretics where things are fairly homogeneous in those groups. Right? Almost all these things are kind of a one off idea that somebody had to reduce sitter use. And then we had to aggregate them in groups that made the most clinical sense to us, so that anything that had a video in it, went into the video category, even if they also had other things. And if they built a special room, but didn’t have a video, it went into that category. Anyway. So it was coming up with the categories within those categories. It’s still fairly heterogeneous.

Alex: I wonder if you could say a little bit more about video monitoring. I don’t know if either of you have clinical experience with this or have seen examples or visited hospitals that have this. I’m curious what this looks like and how similar it is to say ICU remote monitoring overnight with an ICU clinician sitting in one room monitoring multiple ICUs at the same time.

Adela: Yeah. I don’t have any personal experience with video monitoring. I just learned about it through reading all of these studies and how they were done, but I’ve never had a patient on video monitoring per se for falls.

Paul: Yeah. I mean, my sense from reading the articles was is that it’s not a remote thing. So it’s not like there’s a guy sitting in Nevada, who’s monitoring our care here in Los Angeles, but it sounds more like it’s somebody at the nursing station and they might have six monitors. Okay? And so there’s six patients and so they have six monitored patient rooms, and there’s quite a bit in the articles about like what kind of monitor and how it’s set up and how they get trained and all of a sudden kind of stuff. But then this one person monitoring these six monitors is taking the place of six FTE sitting in each of those rooms. Okay? So that’s sort of how the idea is, but it’s basically like closed circuit TV monitoring. Some of them also have microphones so that you can actually hear what’s going on. And some don’t.

Eric: So you said that there may be some evidence for that type of video monitoring?

Adela: Yes. So we found in all of the studies that we reviewed, the results were pretty consistent that the fall rates either stayed the same at the hospital. Some of them, the fall rates actually improved with the video monitoring and all of them reported significant decrease in the amount of sitter usage that they had. So we rated it as moderate quality evidence for the use of video monitoring.

Eric: Now, one of our very first podcasts that we ever done was on bed alarms. Reviewing one article that came out a couple of years ago on the use of bed alarms. Was that an Annals, too? I think that may have been an Annals.

Alex: I don’t remember. Maybe.

Eric: What are your thoughts on bed alarms? Did you look at them?

Adela: Did we look specifically at bed alarms or were they-

Eric: Yeah. Compared to sitter use or was there any studies that you found comparing those?

Adela: So they came up in some of the studies as an alternative to the use of sitters.

Paul: Yeah. But not by themselves.

Adela: But not by themselves. It was kind of a conjunction. That one of the interventions that they would employ would be using the bed alarm, but there was nothing that was specifically bed alarm versus sitter.

Paul: So these alternative often there… They’re almost always multi-component so it’s not like we put a wristband on them. We put a sign up saying they’re a fall risk. We put a video monitor in and Adela put together a really great table that’s in the article that shows what all the various co-interventions are. And so that’s really worth looking at, but it shouldn’t be assumed that these are simple, you know, beta blocker versus calcium channel blockers. They’re not like that.

Alex: So I want to get back to your main findings here, that two studies comparing sitters to usual care, and you found a very low certainty evidence of adding sitters reducing falls. And I want to ask, you’ve probably heard this phrase, absence of evidence is not evidence of absence. I’m not sure if I have that right. [crosstalk 00:17:01] Does that sound right?

Paul: Yeah. That’s the way it goes. Yeah.

Alex: To what extent is this a message that we have enough evidence here to say that sitter… There’s just a tiny, minuscule, poor quality signal that sitters may reduce falls. And to what extent do we say that we just don’t have enough evidence really to make a call on this one way or the other at this time, and we need better studies.

Adela: I think it is that we really do need better studies. The two studies that came out, as we said, they were, first of all, they were both from Australia and the fall rates there are five times… Their baseline fall rates are five times what they are here in the United States. And they were volunteers that were used. They weren’t there 24 hours a day. What would be wonderful is if we could get a pure randomized controlled trial of sitters versus usual care, but I don’t know if that’s actually possible here in the United States, because sitters are such an ingrained part of what our usual care is now in the hospital. That as soon as we take away a sitter, I think we start adding or doing things to try to prevent them from falling. I don’t know if we’d ever get the pure sitter versus usual care.

Paul: Yeah. You make a great point. That is the absence of evidence here enough to say that the things don’t work and we make the point in our Annals article that we don’t think that’s the case. Okay? I mean, the rationale is so compelling that two studies that reached conflicting conclusions, aren’t enough to jettison the whole thing over the side. However, I would say that the ease with which studies have been able to take things, take them away, and put in alternatives suggests that it’s not going to be a ginormous effect. Okay? Cause it’s pretty hard to take away something that’s a ginormous effect, unless you are substituting something that has an equally ginormous effect, and not see an adverse consequence of it. And so a lot of these studies that did this, we did this, we did this, and they didn’t all reach conclusions that showed that they were able to reduce sitter use much, but they didn’t really increase the fall rates. Okay? So my guess is it’s not a ginormous effect.

Eric: Yeah. I find it interesting because I also think about the use of sitters. Individuals get sitters, and they got to have them discharged generally within 24 hours of discharge, if they’re going to a nursing home because no nursing home wants to take a patient with a sitter. So magically they’re not falling in those 24 hours often. It would be interesting to do a study. What the time points that sitters are discontinued and seeing fall rates, both pre and posted of, and magically. I think, that 24 hours before discharge, magically, these sitters disappear.

Alex: Right. And the nurses know this. I mean, they know, Oh, we’re trying to get this patient out. So nurse is more attentive. They have the patient come out in the hallway and sit there, you know.

Adela: Right. Right.

Adela: Are they moving them closer to the nurses station to keep an eye on them? Are they wheeling them out into the hallway when they’re eating meals? Are they making sure they have better sleep hygiene or something to prevent them from trying to get up out of bed.

Paul: Those are great points. And I want to make sure that your listeners understand that one of the take home messages from our article, isn’t go out there and blow a million dollars into a video monitoring setup. Even though that is the thing that had the most number of published studies, there are definitely good, low tech examples of getting in and doing all these kinds of little things. And we talk about one article in particular, that we reviewed in detail, which showed pretty dramatic changes in decreasing the need for sitters without changing fall rate. And it’s all that kind of stuff you were just describing.

Eric: And do you remember what are some of the things they did in that article to help decrease fall rates? Decrease sitter use?

Adela: Yeah. That was the Adam’s article. They did a lot of benchmarking. I think they used chair alarms, bed alarms. I think they had… They revamped their ordering system for sitters where you had to justify why you were having a sitter. They did all sorts of little things that they added every little step over time. It wasn’t just one big bundle. They would look at an intervention. They would see if the sitter rates were going in the right direction, that the fall rates were going in the right direction. And then they would recalibrate and add another intervention and it was a nicely done study.

Paul: Yeah. And so the point there is not really, at least from my perspective as a quality improvement patient safety initiative kind of guy, it’s not exactly which things they picked. It’s how they got to those things. So they did a lot of work talking with frontline staff, trying to figure out why are people falling? Okay? What can we do to change this? And so they fed into a whole bunch of different things. They had a multi-disciplinary team working with it. And then they started adding these things, iteratively, testing each one out until they came up with what they finally thought was their most effective package. And that meant that a package may look different at your hospital than our hospital. It may look different at a third hospital, but it’s the process of how they got there that shows what I think can be done for a lot of different types of safety and quality issues.

Alex: I have two more questions. The first is relates back to that first GeriPal podcast about bed alarms and Eric and Ken Kavinsky came up with this notion that we should have the anti-bed alarm.

Eric: Oh yeah. That was a good one. That’s my favorite intervention.

Alex: If you’re in bed for some period of time without getting out of bed, then the bed alarm goes off telling you it’s time to get out of bed. Right? And I wonder if there’s some similar sort of spin it on its head notion around sitters. Can we re-conceptualize them as people who are doing all of these important investigative things like fall investigator or anti-fall investigator. I don’t know since it’s sort of empower them or empower the nurses or empower the team taking care of them to prevent falls in all of these different ways that you mentioned, Paul.

Paul: Yeah. That’s a great suggestion. I don’t know. At our hospital, we have something called a wound care team, and so you admit the patient for whatever, for heart failure. But if they got… The wound care team comes around and looks at everybody, tries to assess whether they’re risk for pressure ulcers. If they see anything that’s going on, they’ll give you a bunch of recs. Right? And you can imagine a falls prevention team that comes around and sees everybody over a certain age or everybody who has a certain set of comorbidities and makes those same kind of detailed person by person recs about what you’re supposed to do that. You’ve got it. That’s your idea. Somehow some budding hospitalist researcher out there can take it away.

Eric: So, Adela, I got a question for you. You’re a hospitalist. You’re practicing out there. Are you doing anything different after completing this article and seeing the results?

Adela: No. Currently we’re not. We don’t have any alternatives to sitters here. And so sitters are what we order when there’s concern that we’re going to have a patient that falls. I hope that in light of what we’ve found, it might start some discussions here about the possibility of alternatives to sitters, because, as you said, there’s always that great debate. When do you stop the sitter? That we can transition the patient out of the hospital setting. There’s always resources. Because a lot of times there’s extra nursing has to be called in for the sitters. I would hope that maybe now it’ll be a talking point for what kind of alternatives might we consider here.

Alex: Yeah. That was going to be my similar last point here. Paul, at the beginning of the podcast, you mentioned that sitters have been in existence since at least 1982, when you started clinical practice…

Paul: The dawn of time.

Alex: The dawn of time. [laughter] That is a long history. And certainly you also, both of you, mentioned that this is usual care now. What is it going to take to move the needle here? Is it a randomized trial? What do you envision is the next step?

Paul: Yeah. My guess is there aren’t going to be randomized trials of sitters versus no sitters, because I think it’s just going to get… that patient accrual is going to be really hard to do that. I can imagine some cluster randomized things. So this ward continues to have sitters and this ward is going to have some alternative or whatever, but the thing that’s going to move the needle is money. And so it’s like if a hospital’s spending X on sitters, it’s just a matter of time until that comes into the cross hairs in terms of saying, “Oh gosh. Look. There’s this article that was in Annals that looked… That showed how people could save money and not change fall rates by doing X, Y, or Z.”” And every hospital has their own set of things that are cooking on the front burner. But at some point this is going to rise up the front burner status I would suspect pretty much everywhere.

Alex: And, Adela, any thoughts from you about next steps that you’d take?

Adela: No. I think it’s just dialoguing with administration, and finding out is this an important enough topic for them want to look into alternatives to sitters.

Eric: Because you said the VA central office was… They’re the ones that kind of helped initiate this. Hypothetically, I know you’re not speaking for the VA central office, but if you were sitting in that central office, let’s say of a different institution, I won’t even say the word, V something A. What would you do differently if you were now, not from a provider perspective, but from a institutional perspective, what would you do differently as a next steps after this article?

Adela: I think looking at fall rates, looking at how much additional nursing is needed for the sitters and whether or not it would be in the best interest and the safety of the patients to consider alternatives to sitters.

Paul: Yeah. I mean, falls are a JCAHO kind of thing. Right? And so, again, every hospital administrator probably has a list of this long of the things that he or she is having to deal with. And so there may be other things that’s higher up on their priority list, but if fall rates in the hospital are priority or are high enough for a priority, then I think that you could either look into the video, or if you wanted to go lower tech, you could look into the article that was [crosstalk 00:28:03] by Adams. About how they did it, about this sort of careful little, tiny [inaudible 00:28:09] cycle kind of thing.

Paul: And you probably have a pretty big impact within, again, it’s not overnight cures. Right? This could take 12, 18, 24 months before you finally got to where you were seeing the big effect.

Adela: And those… The video monitor studies, and it wasn’t one of those one size fits all that you wheel a video camera into the room and it was done. They did everything very systematically. And then they had escalation protocols and training for the people who are going to be looking at the video monitors. What kind of behaviors should they be looking for that could possibly lead to a fall? What would their next steps be? So it’s not just a matter of installing video cameras with two way microphones, and you’re done.

Paul: Yeah. Just like any other HIT intervention. There’s a lot of other stuff that has to happen.

Eric: Great. Anything else you guys want to discuss before we hear a little bit more REM?

Paul: No, man. Bring on REM. Michael, go for it.

Eric: Well, I want to thank both of you for joining us.

Alex: Thank you so much, Adela. Thank you so much, Paul. Really appreciate it.

Eric: I’m going to give Alex another try.

Paul: You bet.

Alex: (singing)

Adela: If I had a lighter, I’d be lighting it right now.

Eric: That was perfect.

Alex: That was slightly better. Well, thank you so much again.

Eric: And thank you to all of our listeners for joining us this week, and a big thank you to Archstone Foundation for their support.

Alex: And we’ll see you next time. Thanks so much. Bye.

Back To Top
Search