For a place where the most obvious and necessary piece of furniture is a bed, a hospital room is a surprisingly difficult place to get a good night’s sleep. From the basic discomfort of sleeping in an unfamiliar place, to hospital-specific interruptions like beeping machinery and nurses coming in at night to check vital signs or administer medications, there are a lot of reasons why hospital patients might not get a full, quality night of rest.
But what if there are ways patients could get better sleep while they are still in the hospital? To try to find out, the Penn Medicine Center for Health Care Innovation is partnering with hospital clinicians to find simple interventions. They call it The Rest Project.
The Health Care Innovation team was not originally thinking about sleep. Instead, a few years ago when Penn Medicine was in the early stages of planning the Pavilion, the new inpatient facility at the Hospital of the University of Pennsylvania (HUP), they sought ways to ensure patients would find the new hospital to be quiet at night.
Initially, they took that metric—quietness—at face value. “We did a literature search to look at what other organizations had been doing, and we saw they were trying to make the hospital environment less noisy, focusing on physical changes to the hospital, earplugs, and so on,” says Megan Mariotti, NP, MPH, the assistant director of operations for the Acceleration Lab at the Center for Health Care Innovation. All of those things made sense as strategies, but Mariotti and her team dove in deeper. They interviewed patients at many of Penn Medicine’s hospitals—including HUP, Penn Presbyterian Medical Center, Pennsylvania Hospital, and Chester County Hospital—and they did observations to collect their first-hand impressions.
“We learned that the quietness score is really a proxy for ‘how did you sleep last night?’” she says. “Noise matters, but it’s more than auditory noise. Being interrupted to have blood pressure checked is a disruption at night, and being on IV fluids causes you to wake up to go to the restroom.”
They found that interventions to help patients get better sleep didn’t have to wait for a new building, and they didn’t need to cost much, either.
Sleep and Health
Getting a good night’s sleep is an important part of health even for people who don’t have underlying medical conditions that require a hospital stay. Charles Bae, MD, a Penn Sleep Medicine physician and associate professor of Medicine and Neurology, who wasn’t involved in the innovation center’s pilot projects, points out that the symptoms of sleep deprivation from interrupted sleep begin with common problems like irritability, lack of focus, ability to concentrate, and feeling down; the symptoms sometimes overlap with those of depression. Chronic sleep deprivation over time can be associated with either weight gain or difficulty losing weight, he notes. And large population studies have shown that there are associations between sleep deprivation and an increased risk of heart attack, stroke, and diabetes. “There is a strong connection, and more work is being done to establish links between sleep deprivation and these metabolic and other health issues,” Bae says.
As a sleep medicine physician who rounds on inpatient units, Bae notes that, for neurologic patients, sleep loss can sometimes contribute to delirium, a symptom that can increase a patient’s length of stay and require additional medications—raising the chances of drug interactions.
While that research is still being done, clinicians generally already know that sleep is important for hospital patients’ health. But supporting sleep isn’t generally clinicians’ top priority when their patients need serious medical treatment for whatever condition brought them into the hospital in the first place. Bae has observed a common adage that “patients don’t come to the hospital to sleep,” with a corollary that, “the sooner you get someone out of the hospital, the sooner they can get back to sleep.”
But, Bae says, “it does not have to be that way.”
The innovation team didn’t think so, either. After their interviews and observations on inpatient units, the team came up with three ideas for relatively simple, low-cost interventions and worked with clinical teams to run pilot tests of all of them.
The Clinician Intervention: Tracking Sleep Like a Vital Sign
One intervention the team tried focused on the behavior of clinicians.
“I’m a nurse by background and in working on this project I found myself remembering my own behaviors when I worked as a nurse in an inpatient setting,” Mariotti says. “One of the things we saw was that nurses in particular do pay attention to other aspects of patients’ health that are measured, but sleep isn’t measured. What if we tracked patients’ sleep and reported it back to nurses, especially night shift nurses, and see if it changed their behaviors?”
Mariotti led the resulting pilot effort to make information about patients’ sleep more evident to nurses. In that pilot, Mariotti’s team worked with an intern on the unit who wrote down the number of hours certain patients had slept the night before and how many times they had awakened, and posted those numbers somewhere prominent in the room so that nurses and other clinicians would see it. Another group of patients served as a control group for comparison, with their hours of sleep and number of awakenings recorded but not shared with their clinical team.
This intervention wasn’t quite a runaway success. The patients perceived that they slept worse. “I suspect that may be because they’re paying a lot more attention to sleep,” Mariotti says. “They would ask each morning, ‘How did I do?’ almost like how did I do on a test.”
But keeping track did seem to have an impact on the nurses. When nurses saw that their patients were only sleeping an average of about 5.5 hours over the course of an entire night and day, they adjusted their own behavior to the extent they could within the needs of providing necessary care.
“The nurses’ response to seeing the data was one of trying very much to reduce interruptions in the hospital at night,” Mariotti says. “They tried to cluster care, so rather than go in three times to do three tasks, can I go in one time and do all three things?”
Nurses also said that if they saw a certain patient slept poorly on a particular night, they made extra effort the next night to try to help them sleep.
Encouraged by this change in the nurses’ behavior, Mariotti and her team wanted to know if that might really impact patients’ sleep, even if the patients themselves didn’t perceive an improvement. So they repeated the pilot but this time gave sleep trackers to an intervention group whose data was displayed for nurses, and to the control group, while a second intervention group had self-reported sleep data recorded and displayed. (Another reason for trying the sleep trackers: the team wanted to know whether sleep-tracking technology might be a valuable addition to the patient beds in the Pavilion.) When they actually measured the amount of sleep patients got, they found patients really did sleep more if their sleep data was shown to nurses, in comparison to control patients on the unit. They slept an average of an hour longer, with two fewer awakenings per night.
“Part of the problem is we don’t track sleep like we’d track another vital sign,” Carolina Garzon Mrad, MSE, a design strategist in the Center for Health Care Innovation said.
The clinical intervention pilot changed staff members’ behavior even almost two years after the intervention, according to Garzon Mrad. “We met with a nurse on the team recently and she mentioned that certain things like weighing a patient before 6 a.m. are ‘no-gos’ now and staff are very adamant about it,” she says. “That is just not how they work anymore.”
The Patient Interventions: Room Concierge and Comfy Cart
Another pair of pilot interventions focused on enhancing patients’ experiences so they could get better rest.
One intervention, a “room concierge” service, helped patients set up their hospital room according to their personal preferences for optimal sleep—the right number of blankets, night lights off or on, window blinds open or closed, turning on a fan, and so on. In the unit running this pilot intervention, the team asked patients about those preferences and furnished the unit clerk with that information. The unit clerk then took care of those needs for each patient every night. “It’s hard for our patients to go and change things around the room, especially if they are at risk of falling,” Garzon Mrad notes. (In thinking ahead to the new hospital Pavilion, the team is evaluating whether room automation technology is a cost-effective way to provide a similar service to patients at the press of a button.)
A related intervention effort extended the concierge concept to offering items patients might find useful to get into a sleepy state of mind. The unit clerk came by patient rooms each night with a “comfy cart” stocked with items like hot tea, extra blankets, and bedtime snacks, that patients had said in interviews were things they liked to use at night before they sleep.
After piloting these concierge services for 12 days, the innovation team collated data on patients’ self-reported perception of their sleep before and after. “Those very small interventions made patients feel like they were sleeping better,” Garzon Mrad says.
And the intervention was very low in cost, about $23 per patient stay, including the retail cost of items used from the comfy cart and the hourly rate of staff time. The unit clerk spent about 4 minutes per patient per night offering the concierge services. Based on the pilot experience, the Center for Health Care Innovation team recommended that the “comfy cart” service be offered to units on an opt-in basis at HUP. Interested units can consult with the innovation team about how to implement the idea in their area.
Because the patient-experience interventions helped patients to qualitatively perceive they were sleeping better, while the clinician intervention helped patients to quantitatively sleep more—even if they didn’t perceive it—the innovation team believes that a combination of the two interventions might offer the best of both worlds for patients’ health outcomes and their feelings about their hospital stay.
“After talking about the project in different venues, we have been approached by three different teams regarding implementation of what we did in combination with another intervention they are planning,” Garzon Mrad said. “We are advising them on how to implement our interventions and merge them with their own plans. It looks like sleep is having a comeback at HUP and we hope once three floors do it, others will join.”