Today we speak with Dr. Juliessa Pavon, MD.

Dr. Pavon is a researcher and associate professor at Duke University and works with the Durham VA for geriatrics. During our conversation, we get into the importance of mobility measurement. Dr. Pavon’s research revolves around using accelerometer data to measure mobility in the hospital, and one of the key takeaways from our conversation stems around the best way to use objective data in order to improve clinical practice. Just like many things in life, it is often hard to improve on something until you measure it, and that's something that Dr. Pavon’s research has really highlighted.

“I think the main reason why I feel it's important to objectively measure mobility through something like wearable devices is that it gives more awareness to the provider about whether the patient is actually moving or not moving. The converse is also true - a provider might assume that the patient is moving more than they are.”

- Dr. Juliessa Pavon, MD

Listen to the full podcast here.

Note: Podcast transcript below is a condensed version of the conversation generated using a combination of speech recognition software and human transcribers, and may contain errors. Please check the corresponding audio before quoting in print.

Hi Dr. Pavon! As we do with all of our guests - let us know what your overall definition of mobility in the hospital is and then bring us back to your original research on this topic in 2018.

Well, thank you very much for the opportunity to be here, and I'm very excited to talk to you about mobility, especially mobility in the hospital because what we've seen is that patients spend a lot of their time in the bed while they're hospitalized.

We've seen other research support that over 90% of a patient’s hospital stay is spent in bed, and that was very striking to me. (For us) the majority of our rounds are spent talking to patients over their bed and very little seeing them up in the chair, or even up and walking around in the hallways.

That only became more evident as we (started) to see the impact immobility has on hospital outcomes. I would certainly define mobility in the hospital as an activity that takes the patient from a lying position, even up to a sitting position that would still count as mobility. Mobility from the bed to the chair certainly counts - but walking activity outside of the room into the hallway is really where we want to see mobility (eventually) happen.

We ask that question because you can talk to 10 people, and you might get 10 different answers on what mobility in the hospital should look like.

What led you to notice these gaps in how mobility was being addressed in your system, and how were you able to take that into your research to start finding these accelerometer-based ways of measuring mobility?

I think part of what happened is that in talking to patients we would certainly ask, “Have you gotten out of bed today?” And a very simple answer sometimes is yes, but other times it's no. Even with the yes, that still doesn't say enough.

We’d then follow up with the question, “Well, how much walking were you able to do?” for instance, or, “How much time did you spend in the chair?” And again, generally, answers are very vague. Some patients can't really recall the amount of time that they spend up in the chair or walking. And we thought knowing those details was still a very important part of their recovery process to actually know how much time someone is spending out of bed or walking.

We would then dig into the chart to try to find out from a PT note, for instance, or a nursing note. It became difficult to consistently find that (mobility) information in the chart, and so there was this awareness of the need for some objective measure of mobility. We found that one objective way of measuring mobility certainly is with the use of wearable devices.

And I guess the hospital patient is unique in the fact that their gait speed is going to be slower. They're typically not as quick to move. And we know that accelerometers in consumer-grade technology are not 100% accurate all the time. So how did you go about validating this tool for the hospital patients specifically?

You bring up a good point about how mobility in the hospital is very different from mobility in the outpatient setting. There's more of a shuffling gait in a lot of our inpatients too so we did want to look for monitors that had taken some of those features into account.

And the striking part we found is that the median number of steps that patients were taking during their hospitalization was a little bit under 1300 steps a day. And certainly, that speaks to the minimal amount of mobility that is occurring in the hospital with a range from 200 steps to 6000 steps a day. But that also was consistent with what other literature has shown around mobility in the hospital.

Hospital-acquired disability (HAD) refers to functional loss acquired during hospitalization. According to Dr. Pavon, a key contributor to HAD is patient immobility. Dr. Pavon’s research defined HAD as having one or more of the following: 1) A new ADL deficit from the Katz Activities of Daily Living Scale 2) A decline of 4 or more points from baseline to discharge in the functional component of the abbreviated Late-Life Function and Disability Instrument (LL-FDI) 3) Discharge to a skilled nursing facility (SNF) determined by chart review or telephone follow-upIn the following graphic - average refers to the median total daily step count, which was 1456 steps per day. — Hospital-acquired disability (HAD) refers to functional loss acquired during hospitalization. According to Dr. Pavon, a key contributor to HAD is patient immobility.
Dr. Pavon’s research defined HAD as having one or more of the following:
1) A new ADL deficit from the Katz Activities of Daily Living Scale
2) A decline of 4 or more points from baseline to discharge in the functional component of the abbreviated Late-Life Function and Disability Instrument (LL-FDI)
3) Discharge to a skilled nursing facility (SNF) determined by chart review or telephone follow-up
In the following graphic - average refers to the median total daily step count, which was 1456 steps per day.

Right! And going back to your 2018 article about pharmacologic VTE intervention and relating it back to mobility. Where in your research did you find this connection between VTE prevention and mobility?

That was a unique study that (compared) wearable mobility data with someone's EHR data. We placed monitors on patients’ ankles and wrists and then had that monitor in place during the duration of their hospital stay for up to seven days. Then, we looked at the patient record to see what type of VTE Prophylaxis orders they had been receiving during that time. So we were looking at the start date for the VTE prophylaxis and the end date.

And what we found was that even for the patients that were moving the most in the hospital - that 4,000-5,000 steps a day range - their length of time with pharmacological VTE prophylaxis was just as long as the person that was moving with the fewest number of steps a day.

And that is not what we would have wanted to see because the guidelines certainly recommend that for patients who have mobility - VTE prophylaxis can be shortened. We would have wanted to see that those with the highest number of step counts would have had the shortest days in prophylaxis use.

NOTE: VTE prophylaxis was anything related to the heparins or, the Lovenox medications, pharmacological VTE prophylaxis as well as mechanical VTE prophylaxis. The study mostly looked at pharmacological intervention but sequential compression device orders were also considered.

And that makes sense - as a patient starts to walk and their calf pumps the blood back up to their heart for them, they don't necessarily need to rely on that pharmacological intervention and blood-thinning drugs. So how does the typical patient get assessed for their pharmacological VTE intervention and when do they decide to wane them off that drug?

The guidelines do vary around whether a patient is a general medical patient versus a surgical patient - but overall there is a recommendation to use VTE prophylaxis especially when someone is medically ill and always and surgically ill as well. And the main recommendation is around pharmacological VTE prophylaxis.

There are recommendations to risk stratify using scores such as the Padua or the Caprini risk stratification score as well. And that's where mechanical VTE prophylaxis becomes very important because there are a good number of patients who have a risk for bleeding, especially post-surgery.

So what is the importance behind mobility measurement even for the most various of patient types? Why is it important to measure mobility?

I think the main reason why I feel it's important to objectively measure mobility through something like wearable devices is that it gives more awareness to the provider about whether the patient is actually moving or not moving. The converse is also true - a provider might assume that the patient is moving more than they are.

(During our research) we interviewed providers and patients in terms of what type of information they would want about their mobility. And there was a lot of interest in knowing certainly the step counts but also the amount of time that someone spent out of bed and moving as well.

That was my next question of what metrics were important to you guys. Finding a standardized way to measure mobility is so crucial, but then understanding what's going to be important for the caregiver is the other half, right?

The literature is consistent around thresholds for the number of steps that someone should take in order to prevent functional decline in the hospital, and that threshold is somewhere around 1100 steps a day. So I would say step count is still one of those metrics and then also a position and time spent in that position out of bed.

Some patients might not fully make it out of the bed and into the chair - but there is still a difference between laying down flat and at least spending some time sitting up in the chair. Making that distinction is a difficult piece for some of these wearable devices.

So, what's the follow-up with these findings through the research? Are you just going to strap accelerometers on every patient just to see what they're up to?

I think what we've learned the most through this work is that even if we were to strap an accelerometer on every patient and that data was available, there's still a lot that needs to be understood on how the end user would use that data.

So how would the physician change their clinical management, knowing how much the patient is moving? How would it change nursing workflow? How would it change PT/OT workflow? And when we did start to do a deeper dive into understanding that we found that there was a big impact on the culture of mobility.

There is still such a big drive to minimize mobility because of this ‘falls never event.’ That is a federal quality mandate in the hospital where the goal is to have no (patient) falls in the hospital. That mandate is driving a lot of the culture you see around mobility. So even if patients and providers would know how much they were walking, would it really change the nurse being willing to move the patient out of bed if there's still all this concern around falls as a never event.

What we're realizing is that we're focusing on implementing a redesign of mobility in the hospital with the focus of changing some of that culture around mobility. I would say that’s the main next step and then using objective mobility within those types of interventions that are being developed to promote mobility in the hospital.

It seems to me that patient mobility is like a lot of things in life where if you don't measure it it's hard to change. The underlying reasons why hospital patients are often immobile are known - they're very ill in a lot of cases. But truly how do you change something unless you have an idea of what is going on in the first place? Or am I missing something?

That's right. A big part of the culture change is around documenting mobility. So even without having a widespread objective measure of mobility, at least there is this idea that some standardization needs to occur.

That decision must be within the institution to make it (happen). For example – are we going to measure mobility through steps? Are we going to measure it through time out of bed? All of those discussions definitely need to occur. But for sure, even just documenting something that in a standardized way is really part of the culture change.

You mention finding only 52% of older patients had any type of walking documented in the EHR. So what does that tell you? Obviously, that's not to the level of which it should be. Is that correct?

The first three days of being in the hospital, very few patients have any data around mobility because it's usually around day 3 when PT and OT are consulted - and only then do you start to get documentation of mobility. So for the first three days, when it's a very critical period of a patient's hospital stay, there's very little information about mobility.

And once those three days pass, only around 50% of the older adult population receive a PT or OT consultation. This becomes very challenging for the providers to make clinical decisions around their hospital care because minimal mobility can contribute to delirium, it can contribute to a longer length of stay, to functional decline in the hospital, and then requiring a higher level of care after discharge. So that's why that becomes very critical that for three days essentially providers are kind of driving with blinders on.

Yeah, exactly. That is a scary situation considering those are some of the most critical days in the hospital that there is essentially no data that is currently collected.

That's why I definitely like a lot of the work that your group is doing around trying to capture mobility in an objective way throughout the hospital by using medical devices. Because certainly, one way of capturing objective mobility is through devices that patients are already having to wear.

And then the low-hanging fruit would be a culture change around documentation. So for instance, there's a big focus in our institution in having nursing participate in some of that documentation around mobility and being able to screen for mobility using a bedside mobility assessment tool. There are also other tools like an AMPAC that other institutions are using - but having that be part of the structured daily workup for a patient is incredibly valuable.