UPMC Physician Resources
Sorry CME is not currently available for this activity.
Pulmonary Aspiration: A Pathway to Prevention Throughout the Continuum of Care
Dr. Katie Nason discusses prevention of hospital associated aspiration through early identification and prevention.
Upon completion of this activity, participants should be able to:
- Improve diagnostic skills for identifying patients who may be at risk for an aspiration event
- Identify patient populations which may pose an increased risk for an aspiration event
- Improve the management of the at-risk patient population and implement safeguards to minimize the threat of an aspiration event for the high-risk patient population
- Barker, J., Martino,R., Reichardt, B., Hickey, E.J., Ralph-Edwards, A. “Incidence and impact of dysphagia in patients receiving prolonged endotracheal intubation after cardiac surgery.” Can J Surg. 2009 April; 52(2): 119–124.
- Danto J, DiCapua J, Nardi D, Pekmezaris R, Moise G, Lesser M, Dimarzio P., “Multiple cervical levels: increased risk of dysphagia and dysphonia during anterior cervical discectomy.” J Neurosurg Anesthesiol. 2012 Oct;24(4):350-5.
- Fine, M. J., T. E. Auble, et al. (1997). "A prediction rule to identify low-risk patients with community-acquired pneumonia." N Engl J Med 336(4): 243-50.
- Hinchey JA, Shephard T, Furie K, Smith D, Wang D, Tonn S; (2005). “Formal Dysphagia Screening Protocols Prevent Pneumonia.” Stroke Practice Improvement Network Investigators: Stroke. 2005; 36: 1972-1976.
- Johnson, J. L. and C. S. Hirsch (2003). "Aspiration pneumonia. Recognizing and managing a potentially growing disorder." Postgrad Med 113(3): 99-102, 105-6, 111-2.
- Marik, P. E. (2001). "Aspiration pneumonitis and aspiration pneumonia." N Engl J Med 344(9): 665-71.
- Pennsylvania Patient Safety Advisory, Vol. 6, No. 4—December 2009: 115-121.
- Skoretz SA, Rebeyka DM. “Dysphagia following cardiovascular surgery: a clinical overview.” Can J Cardiovasc Nurs. 2009;19(2):10-6.
Dr. Nason has no relevant relationships with proprietary entities producing health care goods or services.
The University of Pittsburgh School of Medicine is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians.
The University of Pittsburgh School of Medicine designates this enduring material for a maximum of 1.0 AMA PRA Category 1 CreditsTM. Each physician should only claim credit commensurate with the extent of their participation in the activity. Other health care professionals are awarded (0.1) continuing education units (CEU) which are equivalent to 1.0 contact hour.
For your credit transcript, please access our website 4 weeks post-completion at http://ccehs.upmc.edu and follow the link to the Credit Transcript page. If you do not provide the last 5 digits of your SSN on the next page you will not be able to access a CME credit transcript. Providing your SSN is voluntary.
Release Date: 6/27/2013 | Last Modified On: 2/13/2017 | Expires: 2/13/2017
It’s great to be here today. This is a little bit off of what my primary research interest is although it’s a topic that does affect our esophageal cancer patients pretty regularly and so it’s something that we as thoracic surgeons pay attention to. And at the end of this talk today I’d like I’m hoping that I will have convinced you that it’s something that you should also be paying attention to with every patient that gets admitted to the hospital. This should be something that crosses your mind as to whether or not the patient is at risk and whether you should be doing something different to modify and mitigate that risk so that the patient can safely leave the hospital without having suffered this devastating event.
I got interested in this topic as a beginning assistant professor here when we were approached by the Mosites family to address the issue of aspiration. They suffered a personal family tragedy and decided to take that tragedy and turn it into something good by attempting to impact and reduce the risk of aspiration for other patients. We’ve also had additional funding for the projects that we do here from the Shadyside Hospital Foundation and also the Shadyside Hospital Auxiliary are all supportive of these initiatives.
So aspiration is defined as the taking of foreign matter into the lungs with the respiratory current, this is a very extreme example of that. I’ve personally taken out strips of chicken from the airway of the gentleman who was eating his Chinese food too fast so I know that it can actually look like this and fortunately he did not die. Often it presents on chest x-ray as opacification usually on the right side but occasionally on the left side as well.
We looked at this when the initiative began to see what really is the impact of this and how often is this occurring in our patient population. We pulled data on over 100,000 admissions to all of the UPMC acute care facilities and we found that the incidents of aspiration was about 2.3%. We were very concerned when we looked at the discharge disposition for these patients and found that the mortality at discharge for patients who aspirated had a diagnostic code for aspiration, mortality was about 18%. This compares to about 1-2% for the overall general population. So a lot of factors go into risk for dying during the hospital stay but certainly the addition of the aspiration is likely to be contributing significantly to that.
We then went on to query about 650 patients here at Shadyside Hospital and looked at their long term outcomes. We selected a random sampling of these, 20% of them were dead at discharge and when we followed them out to 5 and ½ months after discharge, 75% of those patients had died. So obviously this is not your usual population of patients and given that they have such a high mortality we figured it’s likely that we would be able to develop a better understanding of who these patients are by further investigation.
So who aspirates, why does it happen? Well it’s probably a combination of patient factors and hospital factors. This is just an example of the various types of patients that we see here and any one of these patients could be at significant risk for aspiration. This is the first esophagectomy actually that was ever done and you can see he was reconstructed actually with a tube that came out attached to his esophagus came out through his skin and then went back into his abdomen so that he could eat. Elderly patients, patients who have been injured, patients who have had head and neck surgery, patients who’ve had tracheostomy these are all folks who are likely to be at increased risk for aspiration. But how do we know that from all the patients that we see in the hospital?
When you review the literature there’s a whole list of associated medical conditions and diagnoses that people think probably increase your risk of aspiration, I’m not going to read through all of these but you can see that a number of them are related to neurologic dysfunction, to head and neck abnormalities, to altered mental status from such things as sepsis and narcotics, and then just weakening of the overall health status, weakening of your swallowing mechanism, these are all things that can increase your risk.
Medications have been looked at and been found to be associated mostly with how they affect your swallowing mechanisms and your mental status.
The biggest risk factor for aspiration is dysphagia, difficulty swallowing. This is an impairment of the movement of food from the front of the mouth to the back of the mouth and into your esophagus. If there is a disconnect between the mental functioning and the swallowing mechanism you can aspirate. Patients who have had a stroke are a perfect example of that. If there’s weakening of the muscles that move the food bolus from the front of the mouth to the back of the mouth, you can have difficulties with aspiration and this is common in elderly patients. It’s been looked at prospectively and about 30% of all elderly patients admitted to hospitals will have dysphagia and those patients who are confined to living in long term care facilities the rate is approximately 50% and again the various causes of these are similar to the causes of aspiration.
Intubation is probably a risk factor for dysphagia. This has been looked at in patients after having cardiac surgery. We are actually looking at this with the folks over at Presby and hopefully the next time I talk to you we’ll have some definitive results from their work over there. But it does look like patients have dysphagia about half the time after they’ve been intubated for longer than 48 hours after cardiac surgery. This increases if they’ve had a stroke or if they are suffering from sepsis. And when they looked at preoperative risk and the index procedure they did not find any associations with those but as you’ll see later I’m going to argue that there are things that we can know about the patient at the time they come into the hospital or into the operating room that can help us determine whether or not they are at increased risk compared to other patients. They looked at it based on the distribution of the duration of intubation and they found that as the intubation time increased the risk increased, the proportion having the dysphagia increased such that after 48 hours of intubation, 68% of the patients had dysphagia.
Spine surgery has also been associated with increased risk of dysphagia, multiple spine fusions have a higher risk than single spine fusions. If you’ve had more than 4 or more levels you are 4 times more likely to have difficulties with your swallowing and also with your speech after spine surgery.
So again, who is at risk? Well there is a variety of physical factors within the patient and diagnoses that are associated with that, but how do you know when you see that patient sitting in front of you? Patients who have a persistent immediate or delayed or weak cough with oral intake, so you give them a glass of water, or their family describes that you know when they are eating they are often clearing their throat. They’ll take a swallow and then they’ll cough afterwards. Now that’s a patient who is likely having at least microaspiration if not frank aspiration. After they eat their voice is wet in quality or it’s weak or they can’t speak, they have to swallow several times when they take a bite, so they take a little sit and they swallow and then they swallow again and then they swallow again. They might have a loud swallow or a very deliberate swallow that is not normal, or they might be found to be gulping their food. Poor dentition is likely to contribute to that, they may complain that they are having trouble swallowing. If they do that please, please have them looked at because it’s – swallowing difficulties can be a whole lot of things and most of them are bad; patients who are drooling after they eat, that would be another area of concern.
So it’s not just the patients, we as hospitals and hospital providers are also at fault and probably have the, the most ability to impact on this disease process and reduce patient risk for aspiration. There is not a lot you can do to change the patient except to work with them on their swallowing and modify the way that we – that you feed them and the way that we deliver your medications. But the hospital needs to recognize that risk and needs to do something about it, and in general we do a very, very bad job.
This is a slide showing a report from the Pennsylvania Patient Safety Authority that looked at 133 non-anesthesia aspiration events that were reported to them. Of all of those patients only 55% of them had been assessed for their aspiration risk prior to the event occurring. This is a problem, this is we know a lot about who these patients are and if we behave proactively we can identify them early and assess them for their risk and then make modifications. One of the main causes of the aspiration events was the delivery of improper nutrition, so patients were being given nutrition when they were supposed to be NPO. Patients who were NPO were being fed by their families, food that they brought in or, or water or coffee that they asked for. Another 5% of the occurrences were due to miscommunication. Healthcare providers weren’t talking to each other, they weren’t talking to the ancillary services like dietary and speech. The NPO notification wasn’t making it down to the dietary department so you know you order NPO and an hour before lunchtime bu the lunch tray still comes. That’s a big problem. Medications were related to 4%. Delivery of the medications is the biggest problem and this – they found that this was a staff knowledge deficit. The staff didn’t understand how they were supposed to deliver the medications to the patient even though the instructions had been given as far as you know crush the pills, put it in applesauce, that sort of thing. Finally, a small subset of patients had aspiration related to misplacement of indwelling endotracheal or naso/orogastric tubes.
So rather than have our head in the sand about all of this, the most – at the impetus of the Mosites family but with a strong interest in our own group we have decided to pursue how can we as a hospital and how can we as a provider anticipate, modify and reduce this risk? So in 2009 the Mosites Aspiration Prevention Initiative was born. I have a degree in epidemiology so I immediately thought of this from a large population perspective and a risk stratification and prediction rule type of an approach, and I quickly was brought back down to Earth when I met with the quality improvement folks here at Shadyside and discussed other possibilities for what we should do for this and with this funding and it became very apparent to me that both the Mosites family and the hospital would get better results and benefit more by having a joint effort where we actually addressed the issues at the bedside while trying to determine ways to predict risk through prediction modeling.
So we moved forward with a research project to identify clinical decision rule that could be perhaps easily used, it’s still a work in progress. But also hired a quality improvement nurse, Christine de Champ who is not here today but really deserves 99.99% of all the credit for everything that I’m going to show you and everything that’s happened here at Shadyside Hospital, unfortunately she’s not here for me to congratulate in person but we wanted to look at this from an admission to postdischarge process. And so we’ve done an enormous amount of things in order to review this, and we really started with who are the patients.
The research project began by looking at data from a single month at UPMC, January of 2006. We found 6,419 patients admitted to Presbyterian and Shadyside Hospital and 2.2% of them had an aspiration. This is very consistent with what the larger number of patients, the 100,000 patients, the rate of aspiration, so it’s pretty constant. We gathered their baseline data, demographics, admission source, did they come from the ER, through transfer or were they a referral? That’s the best we can do with large administrative data sources, we can’t tell you they came from a skilled nursing facility versus assisted living versus ELTAC, we can just say well they came through the ER, they were transferred here from another facility or they came in through some other referral process.
The administrative database also has comorbid diseases so the Charlson Comorbidity Index if any of you are familiar with that defines things like a history of myocardial infarction, congestive heart failure, peripheral vascular disease, that all gets captured at the – through administrative coding by chart reviewers and kept in a very large database of patients. You can use those to look at risk for subsequent events during the hospital stay.
Our outcomes were their discharge status and their length of stay, and we wanted to know what were the patients with an aspiration diagnosis, which of those characteristics did they have compared to the patients who did not have an aspiration diagnosis? This was a tedious process, there were 20,000 ICD-9 codes in those 6,000 patients, so you can imagine just the data management was a huge task. We ended up having to consolidate ICD-9 codes into major categories rather than looking at individual codes because they are so far broken down by the coding system that we couldn’t get a good idea.
When you look at the discharge group, patients who had an aspiration are here on the bottom row and then all of the control patients in that group. We had a total of – I don’t have the total here, but you can see that discharge as dead happened in 25% of the aspiration patients compared to only 3% of our control patients. Almost 50% of the patients who survived to discharge required discharge to a facility other than home, and only 25% of them were able to go home compared to 75, 76% of the patients who did not have an aspiration during their hospital stay. So probably a combination of comorbid diseases, but certainly having an aspiration is not – you are not likely to get home and you may not even leave the hospital. Significant factors associated with aspiration were male sex, admission through the ER and the transfer was marginally significant, older age. So the aspiration patients were closer to 70, the non-aspiration patients were closer to 60.
Looking at the Charlson Comorbidity Indices we found that having had a history of a heart attack you were – 8% of those patients had an aspiration event compared to only 2% of those who did not have a history of a heart attack or significant coronary disease, congestive heart failure, peripheral vascular disease, renal failure, history of liver disease, so these are the sicker of the patients.
We also in that large 20,000 ICD-9 codes, we narrowed them down to major categories and found 3 major categories that had a significant association with an aspiration event. Patients with oropharyngeal or esophageal cancer had a significantly increased risk of having an aspiration event, I as a thoracic surgeon know that that’s true for our esophageal cancer patients and it makes sense, they’ve got obstruction, you’ve altered the anatomy of their esophagus but 25% of those patients aspirated as compared to only 2% of patients who didn’t have a head and neck or esophageal cancer. Disorders of the central nervous system, nearly 10% had an aspiration compared to about 1.4% of those without, and then patients who had cerebral vascular disease as a completely separate category from the CNS disorders 7% versus about 2%.
So you can take all those things that are independently or individually associated with aspiration and put them together into a multivariant model that adjusts for all of the other factors. So – and understand how much that each of those factors increases your risk for aspiration assuming that all the other factors are the same. What we found is that there is a – the odds ratio, so this is the odds of having an aspiration event if you are a male patient adjusting for all of these other things. So if you are male you are nearly 2 times more likely to have an aspiration event than if you are female assuming that the female that you are compared to has the same characteristics in all of these other categories. Does that make sense? Okay.
So not unexpectedly with a 25% aspiration rate cancers of the esophagus and oropharynx 9.4 times more likely to have an aspiration, cerebral nervous – or central nervous system 4.6 times more like to have an aspiration, cerebral vascular disease 2.2 times more likely to have an aspiration. So these are things that you can look for in your patients when they are admitted. If they have them your index of suspicion for that patient should go up. Ultimately it would be a tool that you would have on your I-Pad or your I-Phone that you just pickup and you punch in yes or no for all of the characteristics in this model and get a result that tells you how likely is it that your patient is going to suffer an aspiration and therefore should be assessed for their risk for aspiration.
It all depends on what the probability of them having an aspiration is, and so in order to understand this model better we used a variety of probability cut points and looked to see you know how sensitive and how specific is this model for having an aspiration? We think that having a very sensitive test is a better, a better model than to have a very specific test and the reason is the risk of having an assessment for aspiration is very low. Someone gives you a little sip of water, they talk to you, they watch how you swallow, maybe the speech therapist comes in and looks at you. There is a little bit of cost associated with it, there is very little risk to the patient so we like our model to be extremely sensitive.
The limitations of administrating coding are pretty obvious here. Even with this strong association and these big P values this model is only about 65% sensitive for aspiration, so we clearly have work to do but it gives us a starting point to go from. It tells us who are the patients we should be most concerned about? And so this is just a listing again of those patients. When you take all these things into consideration you can discriminate between an at risk patient and a not at risk patient about 78% of the time just by paying attention to what their sex is, how are they admitted to the hospital, do they have a prior history of any of these comorbid diseases, do they have head and neck cancer, do they have esophageal cancer, have they had a stroke, do they have other CNS disorders, do they have cerebral vascular disease? And then increasing age, and I’m going to talk more about the increasing age aspect as we move into the test of change at the bedside.
All right, so now what do we do at the bedside? That’s our – that’s the research side of things, that’s what I get all excited about. Now we are going to move into the hospital and see how can we help these patients in the hospital in real time while the thinking heads and the researchers continue to play with their data for months and years on end? Well this is a team effort, this is a whole bunch of people who got together in a large room and started talking about the problems of aspiration and we had nutrition, we had speech therapy, we had the frontline nursing staff, we had physicians, we had patient advocates, at times we had the Mosites family there with us and we talked and talked and talked about what does aspiration look like, and who are these patients and how do we help them? Where are the problems?
We then started looking at different aspects of the things that contribute to breakdowns in care, communication being a key one, knowledge level of the providers and the patients is another key one, and then systems processes that can make our jobs easier or make our jobs harder in doing the right thing for the patient. So after all that conversation was done we assessed what the current process is and currently what happens is that the patient gets admitted to the hospital and things happen. The admission orders get written, the H&P gets written, therapy for the presenting problem gets initiated but no one does anything unless you have had a stroke to determine whether or not you are at risk for aspiration.
If something happens, God forbid it’s an actual aspiration event but maybe someone notices that every time the patient is eating they desaturate, or every time they drink a cup of water they have a lot of coughing, so there is a knowledgeable provider who witnesses the patient exhibiting risk factors for aspiration. That then leads to a communication to the physician who then has the option of consulting a speech/language pathologist. There is a lot of lag time and a lot of time in that process for people to drop the ball. Someone notices it, they make a mental note, I’m going to tell somebody about that but then the day gets busy, something bad happens to the patient next door, doesn’t get passed on until maybe the next day or never at all. And the patient continues to engage in behaviors that ultimately could lead to an aspiration event.
Assuming though that things fall relatively into place and a speech/language pathologist is then consulted they are required to see the patient within 24 hours, and they determine whether the patient is at risk for aspiration. If it’s no, they document that and let the patient go about their business and the providers order whatever nutrition they want to order and medication delivery however they want to deliver it. If the answer to that is yes they currently put their recommendations into the electronic chart and they also write a note in the paper chart. I say currently, things have shifted a lot because of our work here, so this is how things were in 2009, 2010 when we started. When we asked people where the speech/language pathologist recommendations were in the electronic record the nurses didn’t know, the residents didn’t know, the physicians didn’t know. So writing it in the EMR made it readily available to everyone except no one knew how to find it, so it was a very complex, very dense process and that’s why the speech/language pathologists actually wrote a note in the handwritten chart as well.
Unfortunately that required someone to actually pickup the handwritten chart and go to look for and find the speech/language pathology note and then do something about it. And the end result was a time delay in the writing of the orders to address the speech/language pathology recommendations, no one told the patient or their family that they were at risk for having an aspiration, communication to dietary wasn’t made and patients continued to engage in behaviors that put them at high risk.
So the other aspect of this is a knowledge aspect and not knowing where to find the recommendations was one part of that, but when we asked nurses what do you do with a patient who is at risk for aspiration or exhibiting signs of aspiration by and large they had about 150 different answers and most of them didn’t answer the things that we would think are relatively common sense. So we engaged in some efforts to improve communication and education and I’m not going to go through all the tests of change for these, but I’ve listed a number of them here. The first thing to do was to actually just have a reference tech, if you wrote aspiration precaution orders in 2009 what did that mean? It didn’t mean anything because there was no definition for it, there was no basic set of things that the nurse was supposed to do in response to that order, it was whatever the nurse thought was the right thing to do for aspiration precautions. So that was one of the first things that we addressed.
We also implemented a head of bed sign, I’m going to show you an example of that, you’ve probably seen it with your patients. We went to the volunteers, the volunteers were doing things like being nice to the patient and getting them a cup of water because they didn’t understand or know what aspiration was or when a patient was on aspiration precautions. We worked on the communication between providers and dietary as far as diet ordering, we actually sat down with the residents and taught them how to properly order the liquid thickening agents because that wasn’t being done right, and there was unnecessary opportunities for error introduced throughout that process. We are working on implementation of the speech/language pathology recommendations and we’ve developed modules for nurses and ultimately the physician assistants, probably not the providers, but education tools that can help people to understand better aspiration risk and reduction measures.
So here is just a few things that you might be seeing about the hospital, this is a sticker that goes on the front of the chart with a corresponding sign that is supposed to go on the doors and I say supposed to because we are still not 100% at this. And Christine gets an email from me every time I walk into one of my patient’s room who is on aspiration precautions and they don’t have the right stuff. Hopefully you guys will start holding everyone accountable as you leave this lecture.
So this is just a example of an effort to improve communications, this is something that we borrowed actually from the Overland Park Rehab Facility where there is a very clear and very extensive description of how the food should be delivered. So you know it says reduce concentrated sweets with chopped, put the pasta in a separate bowl, make sure that the fruits – that the liquids are the appropriate consistency. Let’s see – I had my mute on in my computer at home so I didn’t know there was an over text there.
This is another – this is a screensaver that we use that just sort of floats through on the list of screensavers to help people remember what proper consistency is and now all the different items that we have available to increase the consistency to what is recommended for that particular patient.
We also found out that the thickening agents were not available on the hospital wards, you actually had to call down to dietary and get those thickening agents sent up for your patient, and what that was leading to was nurses delivering the medication anyway, arguing well it’s just a one time, it’s just one pill, I’ll just given them a little sip, it won’t be a big deal. But as we all know it only takes one time for that to go down the airway and suddenly have a pretty big problem.
This is another screensaver, this was actually directed towards helping providers understand where in the electronic medical record the orders were supposed to go, this – I think this helped a little sitting down with them and sitting down the residents helped an awful lot. Another screensaver alerting people to suggest that a patient might at risk so you know a little image pulled off the internet and you know are you coughing when you are eating, if so have speech/language see your patient.
This is the reference text that we developed. This was in communication amongst all the UPMC hospitals. One thing we’ve learned in this process is that if you want to do something systemwide you actually have to talk to all of the other hospitals and get a consensus in order to make that happen, so we recognized in September of 2009 that this reference text was necessary and I believe that it was – oh it says right there, February of 2011 before we could actually get reference text there for the nurses to understand what we mean, we as doctors mean when we write aspiration precaution orders, so about a year and a half in order to make this happen.
So we also developed some ordering guidelines to help minimize aspiration risk, things that you can do as providers. You can order the RN to perform a bedside dysphagia screening test. So if you think that your patient is having difficulty swallowing the nurses here are all trained to do bedside dysphagia screening, don’t let them tell you otherwise. If they don’t – if they tell you they don’t know tell them that they need to get a hold of their supervisor because they are all supposed to be trained, because they are required to do this with the stroke patients. So if they say they don’t know tell them to take it up with their supervisor and make it happen. But you can order that to be done and get a bedside assessment of whether or not what you think is going on with the patient is actually difficulties with their swallowing. It’s important to make sure these patients are NPO, if you are worried about it go write the order, make them NPO, tell the nurse, tell the patient not to eat or drink anything until they are assessed because a long time can go between when you are worried about the problem and when they actually get the formal assessment, usually within 24 hours but you know if you order it in the morning and they eat all day and speech/language doesn’t get there until late afternoon, then they are franking aspirating, they’ve had maybe 2, sometimes 3 meals between when you decided that they were at risk and when they actually had a formal assessment.
It’s also important to make sure that their oral medications have been addressed. If you think that they are at risk for aspiration or as aspirating don’t let them continue to give them pills. Make sure that you switch everything to IV or hold the things that aren’t absolutely necessary until you know whether then can be swallowed safely. Then once you have the speech/language pathology consultation and recommendations write the orders for them. Currently speech/language cannot write orders, they cannot make any changes to the care that’s delivered to the patient other than therapy and telling us what they think should happen. But if your patient is on a regular diet and they are recommending NPO they can’t go change that to NPO and the patient is going to continue to get their trays until you as the provider do something about it.
We are working on developing a way for the nurses to actually downgrade patients dietary status so if speech/language says this patient should be NPO the nurses would be – would have the authority to be able to do that. That’s something again that has to be agreed upon at the UPMC level. We would not recommend that the nurses be able to upgrade a patient’s diet. So if a patient is on clear liquids I don’t want a nurse changing my patient from clear liquids to a regular diet just because speech/language said so because I generally have them on clear liquids for other reasons other than their difficulty swallowing. But that’s something that we are working on, whether that’s feasible or not, whether that will be acceptable to the attending physicians or not is something that we are having to work out.
This is the head of bed sign that we have been using. We have new speech/language pathology here and so we are continuing to modify and discuss how to improve this and make it more useful for them but continue – but still provide the basic information to everyone else who walks in the room. It’s hard to miss this sign, so if your patient is on aspiration precautions it tells you what you have written for their diet and what the speech/language pathologists are recommending. So if you walk in and you see they are written up for a regular diet and speech/language is recommending nectar thickened liquids it’s a way to prompt you to go oh, I should probably make a change to this. Without ever having to go to the electronic record and find the note from them or have someone pass the information on to you, you can just see it when you are at the patient’s bedside. And then these are also very useful for the families. The families can look up there and you know I’m going to give you your pill, Mr. So and So and the daughter who is sitting next to them says only if you crush it and put it in applesauce and points at the sign. This really happened with one of our patients and it’s a way to empower the families and empower the patients to be advocates and help to minimize the risk.
All right, so how do you do this? This is just a schematic of how transforming care at the bedside works and it’s meant to – it’s meant to communicate that really this is a continuous process. There is no one point in here where one element can be the make or break, every single one of these things has to be addressed, every single one of these things has to be cooperative and worked together with the ultimate goal of being aspiration free. I’d like to add that – I couldn’t figure out how to change this – but it’s not just up to the point of discharge, this is a process that continues on after discharge. The recidivism rate for patients who aspirate is at 30 days is about 25%, so if you’ve aspirated in your hospital stay 25% of those patients are dead at discharge, another 25% of the patients who are discharged are back again within 30 days. So this is a process that we need to extend beyond, beyond the hospital stay and we have all kinds of issues that we are addressing with how to communicate that aspiration risk as the patient leaves the hospital.
So we worked with the Lean Six Sigma group to sort of look at this from a global perspective and analyze all of the factors that contribute to this risk of aspiration. Again I’m not going to read through this, but it just illustrates the point on the previous slide of all of the different areas that could be addressed that are contributing to this overall problem of aspiration and there is lots of areas that are patient specific, we are not going to make you any younger despite what some people might say, but we can improve communication. We can work with the hospitals to reduce staff shortages, we can change the attitude about this. This is not a casual problem. I’ve had a patient aspirate and die right in front of me, it’s a dramatic problem, you don’t ever want that to happen to you. But it’s real and until you’ve seen it happen it’s a little bit hard to understand how big a problem it is. But we can address that by doing things like this, by making sure that we as the providers are discussing patient’s risk for aspiration when they get admitted to the hospital and communicating that to the nursing staff and then educating one another.
So how well does that current process work? How well did that work? It didn’t work very well and so we decided to look and see how we could improve that by targeting what we consider to be a high risk population. And we took some of this from the research study that we are doing to develop the prediction rule but also extended that to say well you know who are the patients who are most likely to be aspirating? What does the literature say about that? And so we, we decided to look at patients who are over the age of 70 or who had been admitted from a long term care facility as a very target rich group of people likely to be at risk for aspiration. We also looked for patients who had a history of stroke or TIA, patients who had previously been found to have dysphagia or had a previous aspiration event, people who are demented or who had acute mental status changes at the time of their admission.
We looked initially at 89 patients, of those 39 patients met the high risk criteria. Only 36% of those had received a speech/language pathology consult. We are observing your actions at this point with this study, we are just making note of patients as they are admitted and then we are following them through their hospital stay and we are making notes as you go, okay. So in this high risk category 36% ultimately got a speech/language pathology consult, 33% were ordered a dysphagia diet and only 3%, sorry only 3 of the 39 actually had an order for aspiration precautions. It took over a day from the admission to the time of speech/language pathology consult and 27 hours for them to have a change in their diet to one that was specific for their swallowing risk.
So that - what we concluded from that is that we weren’t doing very well with getting high risk patients the care that they needed in a timely manner. So we developed a new process that involved assessing patients at the time of admission and then contacting the physician, letting them know that their patient was at risk and then initiating the pathway through the speech/language pathologist. This target condition is designed in order to immediately identify those patients so that within hours of their admission the correct orders have been written, the speech/language pathologists are seeing the patient, they have a sign above the head of bed saying that they are at risk and there is no opportunity for incorrect nutrition, incorrect medications or even patients’ families to contribute to an increased risk for aspiration.
So when we looked at the outcomes from the tests of change at the bedside our short term outcomes were looking at the time to obtaining a speech/language pathologist and a consult and the percentage of patients who were – who we deemed to be at risk who actually received that, the use of the dysphagia diet, the use of aspiration precaution orders and then looked at the discharge diagnosis for aspiration or evidence for aspiration as documented in the chest radiographs in the discharge summaries. This particular study we can thank Dr. Gennari and Dr. Rubin and Dr. Hassan for assisting with this, we talked with them upfront and asked them if they would be willing to be our test doctors where we would do the assessment, notify the physician and then have the appropriate orders written in a more timely fashion, and those are the physicians who admitted to these same units, 4M, 5M and 6M who are not in that group, we just watched your patients and watched to see how they did. So again here is the current process we are inserting into the top of this an immediate assessment, Christine did most of these assessments within the first day of admission and then leading to the speech/language pathology consult.
So what happened in the control group? These were the patients who were admitted within the same 24 hour period as our test patient and admitted to a different service, okay. They did not receive a screen, they did not get the bedside screening test, and their risk factors were a median age of 81, again targeting patients over the age of 70 or who came in from a long term care facility. 48% of then had none of the other pre-described risk factors, 26% had a TIA or stroke, in the past 17% had a history of dementia, 5%, 2 patients out of 42 had a prior history of dysphagia or aspiration, you’ll see why that’s important on the next slide, and 12% had been admitted for acute mental status changes.
Actions taken by the medical team caring for that patient, only 1 patient had aspiration precaution orders despite 2 patients having had a history of dysphagia and/or aspiration. So already we’ve dropped the ball on one patient who we know has had an aspiration event and is definitely at risk for aspiration because they came in with that diagnosis. Only 3 of the 42 patients ever had a speech/language pathology consult and after that speech/language pathology consult was one only 1 of the 3 patients actually had any change in their orders based on the speech/language pathology recommendations so only 1 of 42 patients who were considered to be high risk had anything done to address that risk. Two had a dysphagia diet ordered and none of the patients had adjustments to their medications based on their risk for aspiration.
In this group 9.5% of the patients had an aspiration event during their hospital stay. If you remember overall in the entire UPMC population 2.2% of patients have an aspiration, so using these screening criteria just you know a small list of diagnoses and patients age and where they came from we took a big group where the risk was 2.2 and we narrowed it down to a group that’s not very big, half of our population, but it’s something that you can narrow down using these criteria and suddenly you are finding a group that has a 9.5% risk of having an aspiration event during that stay. That’s not being at risk for aspiration, that’s not having dysphagia or needing to be on precautions those are actual events happening to the patient, okay. So that’s a big change.
What happens if you engage in the process change where you do this early risk assessment, you have the nurse do the bedside dysphagia screening and then you start the process based on that? So this is our test group, these are the patients that we evaluated and then we called Dr. Rubin and Dr. Gennari’s groups and we said hey, the patient is at risk based on our bedside assessment. Again these are all 70 and older and/or came from a long term care facility, they had a formal bedside dysphagia screening test done and they had similar risk factors to the control group. Median age in this group was 81%, 48% of them had none of the other pre-described risk factors, 16% TIA or stroke, dementia, prior history of dysphagia or aspiration, slightly fewer patients were admitted with acute mental status changes although with the numbers the difference was not statistically significant. So when you screen these patients at the time of admission 84% of them were positive for the bedside dysphagia screening, so 21 of the 25 were positive. Of those after putting in a call to the physicians about half of them were ordered for aspiration precautions, a little over half of them had a speech/language pathology consult but 100% of the speech/language pathology recommendations for diet changes were ordered after the consult. So that’s a huge improvement from what we saw in the control group which was only 1 of the 3 patients.
Dysphagia diet was written for 33%, and the physician followed all of the aspiration precautions recommended by the speech/language pathologist in 24% of the patients. Aspiration events were only 4% in this group. So now we are getting much closer to that overall risk of 2% in the entire population as opposed to that 9.5% rate of aspiration that we saw in the group that we did nothing for. Again the numbers are small, this needs to be looked at on a much larger scale and we need to continue to work on this because even though you know we’ve identified a group that’s nearly 10% risk of having aspiration and we have a group of physicians who are very willing to work with us we still only got half of the patients to have aspiration precaution orders. We only got 60% to get a speech/language pathologist to see the patient and so we haven’t convinced everyone and even with people who are convinced the actual execution of this is not happening as well as we’d like. So we are going to continue to work on this and hopefully educate people.
So this what we would recommend currently as the plan. I keep showing you this, and the reason is I’m hoping it will just be burned in your brain, but the key to this is early recognition, okay, early recognition, early recognition, understand the risk factors, understand whether those risk factors exist in your patient at the time you are seeing them, writing their H&P, talk about it as part of your routine presentation of the patient. This should be one of the things that goes in their past medical history. Is this patient having dysphagia, is this patient at risk for aspiration? Just like have they had a heart attack, have they had congestive heart failure, have they had pneumonia, have they been admitted? It has the same impact on the patient outcome, the same impact on our healthcare dollars as those other diagnoses and it should just be part of your routine assessment especially in the elderly and institution patients.
This is just to summarize everything that I just showed you between the test group and the current process group. The absolutely change in the rates of these, we increased orders for precautions by 46%, we increased the speech/language pathology consults by 55%, 77% improvement in actually implementing the recommendations, 48% medications. We want these all to be at 100%, okay, we want every patient who is at risk to have the right things done for them, so there is still room to work but you know when you do something as simple as this and you reduce aspiration events by 58% it’s easy to do and it’s worth having, the central part of how you assess them.
Other things that we’ve done, I didn’t want to get too far into this just in the interests of time but all of your patients who have been assessed by speech/language pathology and found to be at risk for aspiration now leave the hospital if they are going home with a discharge to home packet that has been developed through the generous support of the hospital auxiliary fund that has instructions for how to thicken their liquids, it has diagrams that show what dysphagia is, what aspiration is, how the swallowing mechanism works, where they can go to get the thickening agents that they need, resources that they can access on the internet or through local services. And it’s extremely helpful to the patients, they report that they have a much better understanding of what they are and are not supposed to eat, and they found it very useful. With additional funding from the auxiliary fund we are working on a discharge to other than home to address that poor communication between the hospital and the accepting facilities. Patients leave here, they go to another facility and it says oh we have to start all over at that other facility, the restrictions on their diet are not communicated, their risk for aspiration are not communicated. Even worse they come back here, it hasn’t carried through from this hospital stay to the next one. So you open that patient’s chart, there is no big banner that pops up and says you know this patient just 3 days ago was on aspiration precautions and had specific dietary restrictions. You are starting all over again, and if you don’t think about it, you don’t go back and look for it, you write for regular diets and then suddenly we have this cycle all over again where the first 24 to 48 hours the patient is at risk for having an aspiration event.
We are working also on accessing the worldwide web, we have a Facebook page that’s under development and we are working on other efforts to improve education for patients so that they at home have a place to go and see you know, I’m worried about my dad he seems to be having trouble eating, he’s avoiding some foods, he’s coughing a lot, he just doesn’t look comfortable, there’s not much on the internet to help people when they go searching for this and so we are trying to address that through social media and other worldwide web activities.
Overall, I had a picture of Hillary Clinton here with a baby about to punch her but I took it out because we are being filmed for this, so I didn’t think that was appropriate. Basically that’s where the it takes a village reference comes from, but the ideal is that we would have zero hospital associated aspirations. That’s the ideal, it’s going to be impossible for that to actually be achieved. We all know there are patients who have you know massive emesis and they’ve been sedated at the time of anesthesia or something like that, those are things that we can minimize but we are never going to get to zero, but we can get very close to zero in many of these patients who are not even coming up on our radar yet because we are not thinking about it. We have to have a high index of suspicion, we need a team oriented approach, the nurses, the doctors, the speech/language pathologists, dietary, transport. Transport has been a huge part of our initiative, we’ve actually trained all of them to recognize aspiration risk and in one month they brought 3 patients back to the hospital ward because the patient was exhibiting signs that we had informed them were risks for aspiration and about 2 hours later one of those 3 patients has a massive aspiration and ended up intubated in the ICU and ultimately died. So these are things that everyone who interfaces with the patient should be aware of and can help to intervene in a way that should improve patient outcomes. Communication, communication, communication, nurses handing patients off for transport, transport hands them off to the radiologist. If it’s not communicated that this patient is an aspiration risk and they are going to do a study that requires oral contrast you don’t want them giving the patient barium as the first or Gastrografin as the first contrast because if they aspirate Gastrografin they are going to end up intubated in the ICU. And then finally patient provider education has been a central part of our efforts and it really continues to be one of the key components of this initiative.
I’m happy to take any questions. I do want to point out they are videotaping this and if you have other people who want to see the presentation or know what we are doing or if you just want to go back and look at this again to review those risk factors or see what the process should be you can go to the UPMC Physician Resources page, look at specialties and then I believe it’s going to be under both Thoracic and the Pulmonology sections and you’ll be able to find the entire talk plus the slides for your reference.