Published in the October 2013 issue of Today’s Hospitalist
WHEN IT COMES TO ANCIENT CLINICAL PRACTICES backed by virtually no evidence, consider the routine taking of vitals in every hospital patient, every four hours around the clock. Given what we now know about the devastating effects of sleep deprivation on health and healing, a research team at the University of Chicago figured that it was time to scrutinize a practice that’s been in place since before the turn of the last century.
In a study posted online in July by JAMA Internal Medicine, researchers described extracting vital-sign data for ward patients from late 2008 through mid-2011. They first used those data to calculate patients’ modified early warning score (MEWS), then stratified patients into different risk categories based on their 11 p.m. MEWS.
Unsurprisingly, patients with a higher evening MEWS ran an increased risk of clinical deterioration, but patients whose score was 1 or 0 had less than a 1% chance of an adverse event. The analysis also found that nearly half of all patients (45%) in the sample had a MEWS of 1 or less.
Instead of taking a blanket approach to nighttime vitals, the researchers suggested an alternative: Why not tailor vitals to patient risk? Letting low-risk patients sleep would not only promote better health (and patient satisfaction), but it could save hospitals money. And quality might improve, authors noted, if nurses could spend more time monitoring high-risk patients more closely.
“We’re increasingly understanding the value of sleep in health and the detriment of interrupted sleep,” says senior author and hospitalist Dana Edelson, MD, MS, assistant professor of medicine at University of Chicago. Dr. Edelson spoke with Today’s Hospitalist.
Why has the understanding of the value of sleep been so long in coming?
All this sleep research is relatively new, so this study of ours follows all the great sleep research that’s been done only in the last decade. It’s amazing how something that we treat as so insignificant and easy to ignore can have such a big impact.
What findings surprised you most?
How frequently “and how consistently ” everyone was being woken at night. Regardless of risk category, everyone was woken up twice a night.
Is a MEWS of 1 or less the cutoff you’d suggest to stop waking everyone for vitals?
It’s an easy threshold, but it’s really a question of balancing risk and benefit. Each hospital would have to decide for itself.
There are always going to be cases where someone’s clinical status could change quickly. Even in our lowest risk group, the risk wasn’t zero: For five out of 1,000 patient days, there was an adverse event within the next 24 hours, so that’s not zero.
You could say, ‘I’m unwilling to accept any risk,’ which I think would probably be a mistake. Less than 1% is fairly small, given the risks involved in interrupting sleep.
So you’d accept a 0.5% risk?
I would. But that’s a decision to be made at the minimum by every hospital, if not every clinician.
Have these data convinced your administration to cut back on taking nighttime vitals?
This was not an interventional study, but a retrospective analysis. So we don’t actually know what would have happened if we had prospectively decided not to wake patients.
But these are enough data for us to move forward with our next study: an interventional trial of not taking nighttime vitals in low-risk patients. If that study, which we hope to publish next year, shows no significantly increased risk and better patient satisfaction, I think it would be a no-brainer to roll it out for everyone. Right now, it still requires a deviation from the standard of care.
How hard do you think it would be to change the practice of taking vitals every four hours at night?
It’s a totally different paradigm. But it’s so intuitive, I don’t think it would be that hard of a sell.
In putting together our next study, I’ve been surprised at the lack of political opposition. Even in oncology patients where people might be concerned that we’d miss a fever, there’s been a lot of support for not waking patients.
In the hospital, it’s always uncomfortable for people to do less because we tend to want to keep doing more. But we’re learning, and this isn’t just with sleep research, that more isn’t necessarily better. We tend to do more screening, which results in more false positives and more unintended consequences.
If you consider vitals to be a screening test, the key “like for prostate or breast cancer screening “is to find the sweet spot where you’re catching more cases but not doing unintended harm.
Still, tailoring nighttime vitals would entail some work before patients went to bed.
There would be some technical complexity and logistical challenges to pull the data in real time, calculate them and then communicate which patients are low risk. In our research protocol, we’ll design a sign to post on patients’ doors, saying something to the effect that, ‘This patient is low risk tonight.’
Other avenues to pursue are wireless technology and contactless monitors. In our next pilot study, we’re working with the manufacturer of a device that sits underneath the mattress and measures heart and respiratory rate without ever touching the patient directly.
For the study you published, did you decide to use the MEWS as the risk-scoring system because everyone knows it?
It’s published, but it’s not scientifically derived. So in the study we’re getting ready to run, we’ll use a much more refined algorithm for defining who is high or low risk. We’ll include not only vital signs, but some laboratory and demographic data.
White blood cell count and blood urea nitrogen are two labs that we’ll factor in. And one demographic that turned out to be important, aside from age, was whether patients had been in the ICU during an admission. That tends to make them higher risk during the rest of their hospital stay.
That said, hospital staff know MEWS because it’s used by many hospitals to call a rapid response team when a patient becomes high risk. What we did was turn that trigger system around to look at the other end of the spectrum for low-risk patients “and that’s new. I’ve not seen anyone do that before.
Vitals are only one reason why people parade through patients’ rooms at night. What else could hospitals change to protect sleep?
Another big frustration for patients is lab draws. Physicians routinely order our lab draws for 4 a.m. so the results are fresh and available for rounds at 7 a.m.
That’s really convenient for us but unfortunate for patients, and it isn’t really patient-centered care. The solution is either to draw routine labs at 10 p.m. the night before so results are older when we round, or draw them at 7 a.m., but then they’re not available for rounds.
I prefer the 10 p.m. blood draw time. That would be a small modification, but it would certainly be significant for patients.
Phyllis Maguire is Executive Editor of Today’s Hospitalist.