Richard H. Fortinsky: Current Trends in Home Health Care

This is a series that we have been doing for about three and a half years and we are very lucky to be able to invite the most distinguished researchers in public health today from across the country, different parts of the world. And so our incoming students have a taste of what we typically do every other Monday. The continuing Masters in Public Health students have a wink in their eyes because we don't give this much food every other week. But this is a special occasion. And all public health faculty members and students are welcome to nominate speakers for this series. And I'm so delighted that Dr. Dana Mookamil took us up on that offer. She's a faculty member professor in the School of Medicine and also in public health. She's also a member of the Health Policy Research Institute, which is a campus-wide Institute based in the School of Medicine. So I am delighted that she was able to join us today and nominate the speaker. And she will introduce Dr. Fortinsky. Thank you. Thank you, Deli. And I would also like to mention that I'm also a member of the American Public Health Association, public health. And as Dr. Fortinsky, we are both members of the gerontological, well actually we changed our name. It's now called Aging and Public Health. So for those of you students, new students, existing students who are interested in public health and aging, or as you develop interest in those areas, I would welcome you to come and talk to me and get you involved in the section. But for today, it's a real pleasure and a real honor to welcome to UCI and to this program and to this lecture series, Dr. Rick Fortinsky. Richard Fortinsky is a professor at the University of Connecticut School of Medicine. He holds the Physician's Health Services Endowed Chair in Geriatrics and Gerontology and has a secondary academic appointment at the University of Connecticut School of Nursing. For more than 30 years, Dr. Fortinsky has conducted observational intervention studies with colleagues in medicine, nursing, physical therapy, social work, and social and behavioral sciences, with the goals of preserving and improving health and function of older adults and their families. Real interdisciplinary. His research has been funded by the NIH, other federal agencies, and national and local private foundations. And his major areas of investigation include health-related outcomes and resource use among older adults receiving home health care, family and physician care for older persons with memory and other cognitive disorders, interventions to maximize functional capacity in older adults following hip fracture, and evaluation of evidence-based programs to prevent falls. He also finds time to teach master's in public health students, like you, at the University of Connecticut, and he serves as academic and research mentor to numerous students and faculty members at the University of Connecticut. He received his doctoral degree in sociology at Brown University. Rick? Thank you, Dana and Deli. It's a real pleasure to be here on a special day for many of you here at UCI as well. This is my first visit to this part of California. I've been to LA, I've been to San Diego, but I have not been anywhere in between until today, so, or until last night. So it's a pleasure to be here coming from New England, about as far away in the United States as one can come to get here. I am teaching a course this semester on aging and public health at UConn in the MPH program, and we have some PhD students as well. We are trying to build up the area of intersection between aging and public health at UConn, but most of the courses we teach are probably very similar to the coursework offered here at UCI in the public health program. So the slides I prepared today are really to give an introduction to home health care for older people, because for some of you who may not have done that much work in the field of aging or know much about health care for older people, receiving care at home is often the most important thing for older people who have some kind of disability, because it really helps keep them living at home, as opposed to having to go to a nursing home or to assisted living. Are there any any students in the audience who have relatives who have, older relatives who have some kind of disability? You just raise your hand if you do, just so I have some idea. Okay, so some of you do. And do any of you have any personal experience working in, either as a volunteer or paid sometime during your lives, working with the elderly? Anyone? Oh, quite a few of you. Okay, good. Okay, well this is just acknowledgments. The research that I have done over the years in home health care is funded by a whole number of different foundations and I have numerous colleagues, both at UConn and at Case Western Reserve University, where I was a faculty member before UConn, who I've continued to work with in the field of home health care, and also at Yale University, which is just down the road in Connecticut from where I'm located. So this is an outline of what I plan to cover. I have a lot of slides and some of them are fairly detailed and I probably will go off message, quite honestly, because I think it's less important to go into the nitty-gritty details for students like yourselves, then it's probably more important just to talk about some of the broad brush trends in home health care and to give you more of an overview of the kind of work that I've done. So I'll probably skip over some of the slides as we go that contain some details of research results, but I'll give you at least enough of an overview so you know about the kinds of studies that we've done in this area. Hopefully we'll intrigue you a little bit about this area of research. So the topics that I'll cover will include just an overview of what is Medicare home health care, what are some of the trends over time and the use of home health care, and a couple of key topics that I wanted to focus on because they have particular importance for public health and public policy. I wanted to just talk to you a little bit about quality of care and how that is measured in home health care in the United States, talk about hospitalization and trying to prevent hospitalization in older people with home health care because that's also a very poor outcome in terms of older people. If they're getting care from home health nurses and physical occupational therapists, they want to try to avoid going to the hospital as well. So I'll talk a little bit about that and then I'll talk a little bit about the work we've done in trying to prevent falls in the home health care population mostly by working with the staff, the nurses and physical therapists to try to teach them about how to do a better job of assessing the risk that older people have for falling in their homes. And then I'll end by talking a little bit about the work that really I've gotten started on more recently on diabetes care in home health care. So what is Medicare home health care? As some of you may be aware, first of all Medicare is health insurance primarily for the older population. So anyone who becomes 65 years old in the United States, as long as they've worked a certain number of years during their lifetimes and paid into the Social Security and Medicare trust fund, they're automatically eligible for Medicare. If you're younger than 65, you have to be disabled to get a Medicare card, but there are quite a few Americans who do get Medicare under the age of 65 and you'll see that that makes up a small part of the home health care population as well in Medicare, the under 65 group. Patients receive home health care for up to 60 days at a time under home health care. Most patients are discharged though within 60 days. Looks like I skipped ahead a little bit here. And then they have to be recertified to receive additional 60-day episodes of care. So I'm just going to go back a little bit. I can see I pressed the button incorrectly there. The services that you could receive under the home health care benefit include nursing. And so you can think about a visiting nurse agency. Probably a lot of you know about the visiting nurses. That's the kind of an agency and there are more than 10,000 of them nationwide that provide home health care to older people. And besides skilled nursing, older people who are Medicare beneficiaries could get physical therapy, occupational therapy. SLT stands for speech and language therapy. So especially after a stroke, a home health agency could deliver rehab services into the home. A person could get home health aid care, more personal care, under the Medicare home health benefit as long as they're getting some kind of skilled service as well. And there's also medical social work that they can receive. And the most common reason that an older person would receive home health care is if they have some kind of an acute illness, especially if they've been in the hospital and they're just not quite stable enough when they go home to just keep going on their own without getting some more help. So an order will be written for home health care. And a doctor does have to make the request for a home health care benefit to be provided. And patients who have acute exacerbations such as diabetes, if they have a flare-up of diabetes or heart failure, they could also get home health care. So this is just a brief policy history of home health care. The past 15 years have been quite turbulent in the home health care industry. Anyone who's worked in this knows that. Medicare itself was passed into law long ago at this point in 1965. But from 1965 all the way until 1997, there was just one way of paying for Medicare home health care. And it was really pretty straightforward. If the agency provided a service, they submitted a bill to the Medicare program and they got paid for it. And there wasn't a lot of questioning. You know, there were some claims that were denied. But it basically became a very costly program. And the Medicare officials talked to Congress about this. And Congress decided to change the way that home health care was paid for in 1997. And so they implemented a payment system that was in effect for just a few years. And then they went to the payment system that they have today, which is prospective payment. And what that means is the home health agency gets a bundle of money for a patient based on the kind of health problem that the patient has. So if a patient comes into home health care with diabetes, for example, there's a certain amount of money that the agency will get for having diabetes. And it's up to the agency really more or less to live within that amount of money. And if they end up spending more, the agency loses money. If they spend less, the agency could make money. But it's all averaged. And that's what's called prospective payment system. And then there's been a lot of quality assurance that's gone on recently in home health care as well. So the federal government, the Medicare program really tries to keep an eye on the quality of those services being provided to older people. So starting in 2002, just about 10 years ago now, quality of care and home health care has been monitored based on the outcomes of the patients, how well they do as a result of getting their home health care. And then more recently, just a couple of years ago, a new assessment process was implemented as well as process measures, which have more to do with how well the care is delivered. Not just does the patient get better, but are those services being provided according to the best clinical practice. So the quality monitoring system in home health care has really come a long way in the last 10 years. And you can actually go online, if you wanted to, to something called Home Health Compare, if you went to the Medicare website, and you'd be able to compare five home health agencies in your geographic area and compare them to each other. It's really like a report card, where the home health agencies are compared to each other in terms of, are they providing A-level care, B-level care, that kind of a thing. And it's based on, these measures are based on, how well the patients do in the home health agency when they're getting care. So this has been a lot for the home health industry to get used to. A different payment system, and then all this quality monitoring. So it has really been a turbulent type of an industry. And this is a graph that just shows the number of Americans who are served by home health care, going starting in 1990 all the way to 2010. So a 20-year period. And you can see that there was an increase up here to about three and a half million Americans who were receiving Medicare home health care back in the 1990s. And that's when that first kind of payment system that I talked about earlier ended. And all of a sudden there were many fewer people who got home health care. And that was because the payment system got more stringent. And home health agencies weren't paid as much as they had been previously. A lot of agencies closed at that period of time. There was actually quite a contraction of the home health care industry in this late 1990s period to about 2000. And then this is when that new prospective payment system came into play. And home health agencies thought that that was an okay system, really. They're doing okay in terms of financially. And so more agencies began opening up again. And you began seeing a greater increase again in the number of people getting home health care. So that by 2010, almost three and a half million people again. They're almost back to where they were back in the old payment system days. But it's been kind of this swing over the last 20 years in terms of the people who are getting home health care. So it's been, and this is all, this is really the most visual way of seeing the effect of a policy change on an industry. Because the different payment policies have resulted in agencies opening up their doors or closing their doors. And then the next slide, it looks very similar, but this is actual payments. So this shows, that's 20 billion dollars. And so by the year 2010, the Medicare program was paying out almost 20 billion dollars for the home health care benefit. But again, you can see that same swing where it was going up and up and up, the costs, the expenses to the Medicare program, up until the payment system changed. And then the cost really took a dive. And again, it's the same thing as on the last slide with people served. Here you're seeing the dollars that Medicare paid went from a peak up here of 16 billion dollars down to less than half of that, just in a few years. So the system really went through a change. But then it's been going up again, all the way till 2010. And this is really the trend now, that home health care is becoming very, very popular once again. And another new trend is that hospital care is trying, they're trying to avoid hospital care in the Medicare program, because that's very expensive. So the more that older people could be taken care of at home, that's preferable from a cost point of view, as well as from the patient's point of view. They'd rather get taken care of at home as well. So as a consequence, there's a lot more being done at home for home health care. So this just shows you the kind, the number of people who are getting different kinds of care. So nursing care, you can see that almost a hundred percent of everybody who got home health care had at least a nurse in their home providing care. And then physical therapy was the next most common service. So over 70% of the home health care patients got PT, and then occupational therapy only about 30%. So the message I want you to just know about home health care is that nursing care and physical therapy and occupational therapy is really the core part of Medicare home health care. So people you might know who might need a visiting nurse type of a service, they might get a little bit of nursing, they might get some PT, and they might get some OT as well. And that's really the core of this Medicare home health care service. And this just shows the distribution of home health care visits, and it really looks like the same kind of a trend where about half of all the visits that are provided during the course of the year, this is for 2010, are nursing visits, again followed by PT, and then home health aides provide visits, about 15% of the visits. And this just gives you an idea of the age distribution of home health care patients. So very old age, as you can see. So this is basically showing the number of people in the Medicare population per thousand who get home health care, who got home health care in 2010. So this is showing that in the entire US population, in the 85 and over age group, almost 250 out of every thousand. So one out of four Americans aged 85 and over got a home health care visit during the year 2010. So it's fairly common. The older you get, the more common it is to get home health care. And that's probably not surprising, because older people tend to have more disabilities and tend to have a need for care. But this is just to kind of give you an idea of really where the age group is concentrated. And so this is probably the age of some of your grandparents, a lot of you in the room. So it is really a geriatric service for the most part. And this is that under 65 population that I mentioned, that they only have Medicare because they're disabled already. So that's why the under 65 group gets some home health care as well. And so what kinds of health problems are they most likely to have? Diabetes is the most common, but only about 10% of the patients in home health care have diabetes, and about almost 10% of hypertension, and about 8% of heart failure. Everything else, you can think of any other health problem that is possible, is experienced but by a smaller percentage of people. So the point is the home health care population is very diverse in terms of the kinds of health problems that they present. So a visiting nurse agency really has to be prepared to take care of just about any health problem that you can imagine. But again, the most common would be some type of diabetes, followed by some kind of a circulatory or vascular problem. So I really wanted to just point out what's most common, because I'll be talking about diabetes toward the end of my talk, but also to let you know that it's not like 80% of the patients have one kind of health problem. They can have any kind of health problem under the sun really, and so the nurses and the PTs and the OTs really have to be prepared to provide just about any kind of a disease-oriented care that's out there. And talking about the assessment tool that's used, it's called the OASIS, and it stands, well, that's the acronym, so that's all you really have to know. It's called the OASIS tool, and when a patient comes into home health care, no matter what agency they are served by in the whole United States, the same assessment form is used. And so for a researcher like myself, that's very attractive, because I can get a hold of all the OASIS data in the whole country, which I've actually done in some of the work I've done, and I know that a patient in Irvine, California, or a patient in Hartford, Connecticut, got assessed at the start of their home health care using the exact same tool. And nurses are the ones who fill out nearly all the OASIS forms, and they all have to receive training on how to fill out that assessment form. But it's a really nice research tool, because if somebody has diabetes in Irvine, or has diabetes in Hartford, or if they have a certain level of shortness of breath, for example, that will be documented the same exact way throughout the country. So it's actually a really nice tool for any students out there who are interested in doing large database analysis, or working with faculty members to do large database analysis. And the OASIS includes demographic information, obviously all the diagnoses, and a whole bunch of symptoms, pain level, fatigue level, just a ton of information that's available on the OASIS form. The ability to do activities of daily living, and instrumental activities of daily living, that's what these acronyms stand for. So how well can a patient bathe themselves, dress themselves, I mean this is really nitty-gritty kind of stuff, but this is really what the home health agencies need to know when they're going into a home, is how dependent is a person on other people's help to do these kinds of things. And that's what these ADLs and IADLs are. IADLs are more like money management, and doing laundry, so I mean we get down a real, real detail here about a patient's clinical and functional profile with this OASIS form. There's also information on the OASIS form about depressive symptoms, about cognitive functioning, so you could get a really good portrait of this patient using this OASIS form. And then the OASIS-C, as it's called, is just the newest version of the OASIS that was just implemented a couple of years ago, and that includes some additional information that gets more at the kinds of processes of care that home health agency staff provide to their patients. So quality of care monitoring I talked about a little bit already. This home health compare, if you just googled home health compare, you'd be able to get on the web and, like I said, compare agencies as far as how they're doing on process measures of care, outcomes of care, and then what are called potentially avoidable events. And so process measures have to do with things like how frequently are agencies screening their home health care patients for things like fall risk or for the presence of depressive symptoms. And these are really important things to screen for, because if you pick something up at the beginning of a home health episode, even if that's not the reason a patient came in for home health care, it's really important to try to pick out if there are some of these underlying problems. If somebody's really at high risk for falling or has some depressive symptoms that really maybe weren't on the original note from the doctor to start the episode of care, the OASIS form picks this up, and now it's incumbent upon the home health agencies to document that all this kind of work is being done to assess patients really carefully. So that really gets into the quality monitoring system. The outcome measures that are of most interest in this home health compare have to do with hospitalization, because if agencies have a lot of patients who are being hospitalized and their percentage of patients who have to be hospitalized is really out of range compared to other agencies, well that might be a clue that there's something about that agency that has to be investigated. And these measures, when they're compared from one agency to another, are adjusted for how severely ill the whole patient population is. That's called case mix adjustment. That's probably enough detail, but you might wonder, well, gee, what if one agency has a lot of really sick patients and the other agency it's not quite so sick? Is it fair to compare them on hospitalizations? Well, the fact is that a lot of statistical work is done to try to make sure that when you're comparing hospitalization rates in agency A to B, that they've taken into account the difference in severity of illness of the patients. So that's part of the home health compare system. And there are some potentially avoidable events that have to do with emergent care, because there are some things that patients, if they're getting really good home health care, should really not have to go to the emergency room for if the home health agency is doing a really good job. So that's what these potentially avoidable events have to do with. So again, if there's an agency that's really way out of line in terms of its patients going to the emergency room a lot, well then that might be a reason to really investigate a little further about the quality of care being provided by the agency. So main point here is there's a lot of quality monitoring, but it's available to the public. And that's another really nice thing about this system, is that the reports that are available online are in really easy to understand comparative type language on this website. So in terms of hospitalization, as I already mentioned, it's a poor outcome for home health care patients. And it turns out that when we started doing our research several years ago, we realized that, well, what are the things that really put an older person at risk for being hospitalized? And that's what we really wanted to study based on that oasis at admission. That's really what we wanted to look at, the characteristic that they have that might then lead them to be hospitalized. And at the time we did our work, there were really no known studies of risk factors that were used nationally representative samples. So this is where it was an advantage to be able to get oasis data from everybody across the country, which tells us not only what these patients look like at when they were admitted to home health care, but then whether they were hospitalized sometime after they started their home health episode. And so we did a couple of studies, which I'm just going to run through quickly. This was our initial study, which was published, the results of this study were published now several years ago, where we looked in Ohio. This is when I was at Case Western Reserve. So we were working with the home care industry in Ohio, and we worked with 22 different visiting nurse agencies in Ohio, and they collected data on that number of patients, over 900 patients. And we took a look at what their characteristics were at admission to home health care, and then whether or not they were hospitalized. And this just stands for adjusted odds ratios, and this just has to do with the kind of an analysis that was done. So this pretty much tells you that if a patient had some kind of a skin or a wound problem at the time they were admitted, their chances of being hospitalized were more than three times greater, 3.4 times greater than someone who did not have a skin or wound problem at the start of care. That's what these adjusted odds ratios are all about. If somebody had diabetes in this particular sample, they were three times more likely than somebody without diabetes to then go on to be hospitalized. And if they had a stroke, they were about two times as likely. And then the more severe, dyspnea refers to shortness of breath, so the more short of breath they were, and that's a very common symptom among older people, particularly if they have heart failure or some kind of a cardiovascular problem. For every step up in the severity of shortness of breath on the OASIS form that a patient exhibited, they were one and a half times or 50% more likely to be hospitalized than somebody who was lower on the dyspnea severity scale. So this just kind of gave us a bit of a risk profile of the patients who were most likely to be hospitalized based on this single study. And then for the national study that we did as a follow-up to that single-state study, this is where we got OASIS data from all over the country. And this is just a long way of saying that we took all the OASIS data and linked patient data together so that we knew who the patients were at the start of their care and then who they were at the end of their care. So we were able to get a complete picture of what these patients look like during the course of an episode of care. And the average length of an episode in this study was about just really about six weeks. So again, patients aren't on home health care usually for a really long time, so the average is about six weeks. And we were interested in looking at whether they had to be hospitalized either sometime before they were formally discharged from care or within three days of discharge from home health care. That's really what we looked at for hospitalization. So as you can see, we had a very large sample size and this represented about 20% of all patients who completed one of those roughly six week episodes of care in the whole United States back in 2002. This is the linkage that we did. We actually got hospital claims data, which is also available from the Medicare program, and that's the real proof that somebody was hospitalized. If there was a Medicare claim from a hospitalization, we were able to link that data at the person level with their OASIS data. So for people who like to do this kind of stuff, it was kind of cool to be able to do that for a whole national sample of patients. And what we found was in this large sample, about 17% of all those 374,000 patients were hospitalized as a result of either during or immediately after their home health episode. And this just gives you an idea of what the patients looked like in this sample. And this gives you a good idea of a real profile of what are these home health care patients looking like anyway nationally. So not surprisingly, about two-thirds were women and that's because, as you probably know, in the older population there are a lot more women than there are men, because women tend to live longer than men. So it's not surprising that you had more females than males in home health care. And this shows you the age group distribution. And as I said earlier, you know, about 10% even though these are Medicare patients, these are all just Medicare patients, I'd make that clear again, were under the age of 65. And then you can see that they were really clustered in the 75 and over age group. And this is the race and ethnic distribution. So again, not too different than the race and ethnic distribution of the whole US older population if you looked in the census. About 10 or 11 percent of all older Americans in the US are black or african-american and about 4 percent, this is by today's figures, are Hispanic and Asian. You can see what they are. And then these are patients who also had Medicaid insurance, which meant that they were poor. They tended to be poor as well as old or poor as well as disabled. And they were called dually eligible. And then this next slide is kind of busy and I just want to point out a few things on this slide. Again, just to give you an idea of who are these people in home health care. The first thing I'll point out is that the average number of medical conditions that were listed on the OASIS form was between two and three different diagnoses that they had. So you can think back to the diabetes and hypertension or diabetes and heart failure, you know, those are the kinds of problems. Or they might have had some kind of a hip fracture and they had a stroke or something like that. But the average number of medical conditions is about two and a half. And this is cognitive functioning. And these are the categories actually that are on the OASIS form that have to be filled out by the home health nurses. So in terms of cognitive functioning, about two-thirds of the patients had no problem with cognition, but about a third had some kind of cognitive problem. They required some kind of prompting in order to do activities. And so this basically shows that about a third, one-third of the home health care population has some kind of cognitive impairment. And again, probably that's not so surprising because Alzheimer's disease and other kinds of dementia that you've probably read about, at least, are common in the older population. They're not universal. And I don't want you to think that every older person has dementia because that's not true. But this just gives you an idea that about a third of the home health care population has dementia of some sort, some kind of cognitive impairment. About 20% have depressive symptoms. And then this is that dyspnea that I wanted to show you that about, if you add all these together, that gives you almost 40% have shortness of breath just with moderate or minimal exertion. So some of them are really fairly impaired. And this is what we found in this study. We found that a lot, as in the single health, the single-state study that we did, having a skin or a wound problem as the primary diagnosis increased your chances of getting hospitalized. This is congestive heart failure. This is depression. Again, more severe dyspnea. And it turned out that we found in this, it turns out in the Ohio study, that single-state study, we had very few African-American, we had very few people of color in that study. So we really couldn't look at racial and ethnic differences in that single-state study. But because this was the national study, we had, you know, plenty of racial and ethnically diverse patients. So this, this 1.2 showed that after you control statistically for all the health conditions that we control for, blacks compared to whites, we're about 20% more likely to be hospitalized. And we don't know why that is. That's just, that's what we found. Yes? Both. It's both. We, we, we put those together. That's a good question. So this could have been people who had pressure ulcers because they were very sedentary for a long time, even before home health came on the picture, or because they had some kind of surgery in the hospital, and then they went home and they had a wound that was just really fresh, and a nurse came in to take care of it. But if they had any of those kinds of problems, we, we put them all together in one category. And you know, there's actually been a lot of research, our group hasn't done that work, on, on trying to improve outcomes in home health care patients who have skin problems. That is a really big deal, but not what I'm talking about today. It's another one of those really big risk factors for hospitalization. But you can see that diabetes, as in the other, as in the Ohio study, showed that if you had diabetes, you were more likely to be hospitalized, and the more medical conditions you had as well. But what was interesting was, again, statistically, even knowing about all of these risk factors, they only accounted for about 10% of the total risk of hospitalization. So there's still a lot of things that cause hospitalization that we couldn't measure, because we didn't have the data on the OASIS. So there's still a lot of unmeasured reasons for hospitalization. Yes? In the Ohio study, skin and wound problems were 3.7. That's right. So nationally it's 1.5. Well, that's why I was talking with Dana about that this morning. Part of it could be due to regional differences and geographic differences in care. It could also be, what might be called a statistical artifact of the data, because we didn't have as large a sample size in Ohio as we had nationally. And what tends to happen when you get huge sample sizes like this is, on average, these risk ratios generally tend to be smaller, because you have just so many more things that are being taken into account with a large sample size. It's a good question. Yes? I was also kind of curious, did you discover any changes in the risk ratios or anything in, say, urban populations versus suburban or rural? For hospitalization, we actually did find that those who were in non-metro areas in rural areas were slightly more likely to be hospitalized. Yeah, we did. I didn't put that on there because it wasn't a patient characteristic. That had more to do—the way we measured that was the location of the agency. So patients being taken care of by agencies in rural areas did have a higher risk of hospitalization compared to those taken care of by agencies in rural areas. And it could be because there weren't a lot of additional supportive services besides that home health agency out in rural areas to kind of help bolster the care, and that could have led to hospitalizations. It's a great question. So these are just conclusions based on the hospitalization part of my talk. Based on both of these studies, Medicare patients who have skin and wound problems or diabetes or these other kinds of symptoms really deserve more aggressive care to prevent hospitalizations. And we are really interested in further investigating why the black home health care patients had a higher hospitalization rate. And then falls—it turns out that anecdotally there's a tremendous amount of information that implicates falls, just falling in the home that might lead to hospitalization. But because the OASIS form doesn't include information—or didn't at the time that we did our research—on whether a patient, let's say, had frequent falls in the home at the time that they started home health, we weren't able to measure falls in our study. But in the future, we think that there really has to be more focus on looking at patients who are at high risk for falls at the start of home health care to try to prevent those falls and therefore prevent hospitalization. And it turns out that falls—I'm going to now move to the next part of the talk, which has to do with fall prevention, and I'm mindful of the time here, so I'll kind of go through these slides fast. So what I'd like to do is just kind of give you more of a narrative of what we did in Connecticut as far as fall prevention goes. So falls are common in the older population. About a third of the whole U.S. population, 65 and over, experience some kind of falls each year—one-third. And if you look at the 80 and older population, about a half have some kind of fall. That doesn't mean that they're really injured badly, but they fall enough to report it in national studies or in local studies. But if they do fall and they break a hip or they break a wrist, that leads to injury, disability, and a lot of health care costs. And there have been a lot of randomized trials that have shown the benefits of interventions that really focus on different fall risk factors, which I'll just explain what they are on the next slide. And what we wanted to do in this study was find out how some of the results of these risk factor studies and fall prevention could be incorporated into everyday practice in home health agencies. And so that's what we did in Connecticut. And this is where we work with the group in Yale. Some of you may know the work of Mary Tenetti. She's actually world-famous for doing work in fall prevention. She's a geriatrician at Yale. And so her group went in and tried to do a better job of training nurses and physical and occupational therapists about how to do a better job of fall risk assessment. And this was before, several years before, the OASIS form developed some of its process measures that now start to look more in detail at fall risk assessment. So the main thing I wanted to focus on here was, well, what are some of the risk factors for falls? And they're right here. So if a patient is already known to have a mobility impairment, there should be some effort to get them to exercise or to do some kind of range of motion to try to strengthen, especially their lower body. Because there's actually been a lot of research that has shown that in older populations, the poorer your lower body function, so the slower you walk or the poorer your walk gait, the more likely you are to have all kinds of bad things happen to you. You're more likely to be hospitalized, go into a nursing home, have higher mortality. There's like a ton of research out there on this. But in the home health care world, there hadn't been a lot of focus on, well, how do we train home health care staff in a very systematic way, based on risk factors like mobility impairments, to do a better job of identifying them and then trying to implement some kinds of interventions to try to minimize their impact. So mobility impairments were a risk factor that were focused on in this Connecticut study. Balance disturbances, which is really often just an extension of mobility impairments, just, you know, you just tip over a lot or you just have a hard time maintaining your balance. If you're on multiple medications as an older person, that's a real high risk factor for falls. And the threshold tends to be if you're on at least four medications, and that could be prescription or over-the-counter medications as an older person, that's like the tipping point, literally and figuratively speaking, that sometimes leads to a greater risk for falls. So a lot of focus was put on in this study on reviewing the medication list that older people had at the start of home health care and encouraging patients to go back and talk to their doctors about, do I really need to be on all these medications? So there was really a focus in this study on trying to get the home health care nurses, particularly, to review the meds with the patients. And if the nurses were suspicious about, well, maybe there are two meds that are almost prescribed for the same thing, and there's some drug interactions that might be going on. It's ultimately up to the doctor to make a change in the medications, but the patient has to, needs to know that. And so that was what they were trying to teach the home health nurses, trying to think of home health care as like a teachable moment, because the agency nurses were in the home, they were reviewing their medications, and they used that as an opportunity to tell patients, we think you might be on one too many meds here, or a couple of them, and so go talk to your doctor about that. Postural hypotension is when your blood pressure, when you go from sitting, like if you were just sitting there, and you stood up, and you got real dizzy, and this is not drinking alcohol or anything, this is kind of in your current state. That could be because your blood pressure really plunges, and that's actually very common in advanced age. And sometimes it's secondary, it's often secondary to some kind of a health problem, but again, you can think of an 85 year old person sitting in their chair for a couple of hours, and they get up real fast, their blood pressure takes a dive, and they can fall over, and that happens a lot. It also happens in the middle of the night a lot. An older person has to get up to go to the bathroom, they've been lying in bed for a few hours, they get up real fast, and all of a sudden they're on the floor. So trying to identify patients who have postural hypotension, and to teach patients to get up slowly. Take your time getting up, there's no hurry to get anywhere, and that is a really good little tip to try to prevent falls in older people. And then finally, checking out the home itself for environmental hazards. Throw rugs that really shouldn't be there, wires that are just on the floor, and you know, older people who might have been living in their homes for decades and decades, they kind of get, you know, this would be, this would be a really fall hazard right here, you know, throw rugs like this, just exactly like this. Or pets, small pets. Older people who have cats or dogs, and they're right under their feet, because that's what cats and dogs like to do. They like to hang out with the people who own them. Well if you're an older person, and you kind of forget that the cat or the dog is there, easy tripping. So it's not like they're saying get rid of pets, but again, to raise the awareness of patients that your pets could really be a fall hazard, and you just don't want to find yourself on the floor calling 9-1-1, having to go to the hospital. So all these different risk factors for falls were what the Yale group trained the Home Health Agency people to look out for and teach their patients about how to try to minimize those risks for falls. So what we basically did in the study, and I'm not going to go into any detail in any of this, was we, at University of Connecticut, we evaluated this training program that the Yale people did for the Home Health Agencies, and I'm just going to kind of quickly go through the slides. What we did was we had the Home Health Agency personnel fill out a questionnaire several months after they got the training, to find out what their routine risk factor assessment was for their older patients, now that they'd had this training. And what we knew was that at all these agencies, they had received no formal training ever before the Yale people came in. I mean, if a home health nurse was resourceful enough to go to some kind of a workshop, maybe they might have gotten a little training. But this Yale training was really the intervention that we were really interested in, and then we basically asked them, an average of 21 months after their training, whether they were doing fall risk assessment. And we had almost 200 home health personnel, which represented about half of all the clinical staff who got that training from the Yale people, took part in filling out the questionnaires. And the main thing that I just wanted to point out in this slide was that the risk factor that was most likely to be correctly assessed was that one about postural hypotension. That was the one that the agency personnel really tended to zone in on, because that came very naturally, particularly to the nurses, because they're really, nurses are really used to taking blood pressure of older people in sitting positions and standing positions, so that one really rang true to them. And so they really did a good job. Balanced impairments and medications were less likely to be correctly assessed across all agencies, and by correctly, again, we had a certain way of matching the training that they got with the correct answers that they were supposed to provide when we asked them, what are you doing now for these different risk factors? Well, it turned out that that multiple medications risk factor was less likely to be assessed, and when we went back and asked them later, they, the nurses often told us, well, we can tell the patients everything in the world, but if they don't go to the doctor, what's the use? So the nurses were really concerned about compliance and things like that with the patients, so they were a little less likely to do a full-blown multiple medication assessment. And what we found was that across the agencies, there were some agencies where the personnel did a really good job of these fall risk assessments, other agencies not so much. So we found that the impact of the intervention for fall risk assessment really varied a lot across the agencies, and so we went back and kind of did a second teaching to try to give them the results of this, and in Connecticut, it turns out, there are a lot of home health agencies, particularly because of the Yale group, that have really incorporated a lot of fall prevention assessment into their routine home health care, and now with the Home Health Compare, there's more of a universal need to do fall prevention training for older people, because that's now part of the Home Health Compare system, but this study was done before that. So I'm going to finish up, I think, let's see, so these are the conclusions from the fall prevention study. It predated efforts, this is the Medicare program, Centers for Medicare and Medicaid Services is doing their risk assessment now for falls, and this OASIS-C data will enable better monitoring, but there's still no requirement that home health agencies train their staff. I mean, there's no systematic training across the country for fall prevention, so it just so happens a state like Connecticut is where Yale was located and Mary Tinetti, and so there was really a lot of focus on fall prevention in Connecticut. I know in California, there have been a lot of efforts to develop fall prevention coalitions among agencies, but I don't know how much that's filtered down to home health agencies, and many, many states have really gotten on the bandwagon of fall prevention way outside of home health care, but it's one of those things that just never ends. You can never stop doing this kind of training, because you get new home health nurses that come in, and if they hadn't gotten the training earlier, well then that agency's not going to benefit from them going into the home if they don't know how to do fall prevention. So fall prevention in home health care is really, really key, and the trick still is trying to figure out how to allow ongoing training, or just have at least a training module. Maybe it could be something they go online and check out, you know, could be a training, but exactly the best way to do that isn't well known yet. Honestly, any of you interested in projects, not just with home health care, but with older people, preventing falls is really, it's a public health issue. I mean, there's just no doubt about it, and that's really why I wanted to spend a little bit of time on that one today. I'm not going to have time to get into the work we're doing with patients with diabetes, because I want to leave time for a few questions at the end, but I'll just kind of mention why we did this study, which really had to do with racial and ethnic disparities, and how much and what kind of home health care patients get. So the reason we did this was because, as I showed you earlier in my talk, diabetes is the most common home health care diagnosis, and we already found out that it's a risk factor for hospitalization. We talked about that earlier, and it's known that racial and ethnic differences exist in the end-stage health outcomes of diabetes. So, for example, African-Americans are much more likely to develop the really bad complications of diabetes, and basically what diabetes is, is blood sugar is not really getting into your system well enough, and so parts of your body that really depend on blood sugar kind of getting into your bloodstream, they really suffer, and they sometimes get numbness in their extremities, or the bottom parts of their legs or their feet really get numb, and they really have a hard time being mobile. Eye problems, retina problems develop from diabetes, so a lot of really bad things happen if your diabetes is untreated, and it turns out that African-Americans and Hispanics have a higher prevalence of dementia than whites do of diabetes. That's just known in later life, in middle life, as you know, middle age as well, but that carries over in the middle into later life as well. And racial disparities in health care, you've probably heard about that. That's a theme I'm sure that you'd be exposed to in any public health program and any kind of health service, but there hasn't been a lot of work done on racial disparities or the extent to which they exist in home health care, or particularly around diabetes. So that was why we embarked on this study, and I'm just going to kind of go to the end of the slide. These were the results. This is what we found. After controlling for a bunch of characteristics, we really wanted to compare whites, blacks, Hispanics, and Asians in terms of how much of different kinds of home health care they get if diabetes is their primary problem. And this is what we found. We found that African-Americans were less likely than whites to receive either skilled nursing services or physical or occupational therapy. Hispanics were less likely to receive physical therapy or home health aid visits. And then among those who did get nursing visits, African-Americans received fewer nursing visits per week from nurses, as well as from all home health care disciplines combined. And then Asians and Hispanics actually received more nurse visits per week than whites. But it was the African-American and white disparities that really stunned us after controlling for so many different things. So what that really means is that patients have diabetes. If you're an African-American patient with diabetes getting home health care, you're simply less likely to get skilled care. And again, we don't know why that is. We just found this trend. That was what we were looking for. And so we're kind of left trying to figure out what's going on there. And so one of the reasons is, I mean, it might be that African-American patients are less assertive as far as asking for skilled services, because it's actually known that in other areas of health care, particularly in terms of getting cardiac procedures, there's actually been a lot of research on this, that African-American patients are less likely than white patients to request procedures to be done if heart problems are shown. This has been shown in the VA health care system, the Veterans Administration, as well as in the non-VA system. And so we suspect that African-American patients, again, on average, and this is based on large population samples who have diabetes, may not be prescribed nursing care or rehab care associated with their diabetes, but they may not be requesting it. They may just think, okay, well, that's what we're supposed to get, and so we'll accept it. And again, this is in today's older population of 70 and 80 year olds. Oh, and it also may be that African-Americans with diabetes are getting more, they tend to get as much home health aid care as whites do, but not the skilled care. So we think that more emphasis has to be placed on determining the rehabilitation potential in patients of color who have diabetes in Medicare home health care. And so again, we're just starting to investigate this, and we're going to be looking at some more recent data, because these are all based on data that are now 10 years old. But the study that we published on this was really the first one of its kind that appeared in the literature. We were really able to find virtually nothing else that's been done in this area. And because diabetes is such a big health problem, we really think that there's a lot more to be investigated in the whole world of racial and ethnic disparities. So just the summary points, and then I'm pretty much done. So Medicare home health care is a rapidly growing program within all of the Medicare services, but it's really understudied compared to all the other Medicare covered services. Secondly, these risk factors for hospitalization really require a lot of further study in the home health care population. Third, fall prevention is really important, and Medicare home health care really represents this teachable moment that I mentioned earlier for patients and their families because of the home-based care setting. And then finally, racial and ethnic variations in diabetes care really deserves further investigation. And I think that's it. So thanks for your attention.