Florence Nightingale, in her seminal 1859 Notes on Nursing textbook, reminded nurses about the importance of not waking hospitalized patients up. It is not surprising that years later, the American Academy of Nursing, through the Choosing Wisely™ campaign, recommended that hospitalized patients not be woken for routine medical care to protect their sleep, citing numerous harmful health effects of hospital sleep loss.
Unfortunately, we still have a long way to go. Efforts to improve sleep have been undertaken but have not been adopted widely – likely due to the over-reliance on staff changing their routines. A “work harder, do better” approach rarely will stick in healthcare, when everyone is working hard and trying to do their best. Often for behavior change, a ‘work smarter’ approach is needed. In this case, we were inspired by a book that I read from my then upstairs neighbor Cass Sunstein on “nudges.” While he and Nobel Prize winner Richard Thaler described nudges to encourage people to make better health choices, we wondered if we could repurpose this idea to nudge staff to remind patients to sleep. This idea made good sense since one of the top problems noted by the staff was that the electronic health record was “hardwired” for routine vitals every 4 hours and included certain non-sleep friendly options for medications; it was hard to change “even for a VIP patient,” as one intern said to me.
We all know the electronic health record can be our nemesis, but can we actually use it to provide friendly reminders that are not obtrusive but nudge us to do the right thing? That is exactly what we did. But we did not rely only on the electronic health record (EHR) to change this culture. We also worked to include training for both physicians and nurses about the importance of these measures. It worked! Sleep for Inpatients: Empowering Staff to Act (SIESTA) was associated with more sleep-friendly orders for vitals and medications, fewer nighttime disruptions and room entries, as well as improved patient experience.
While I encourage you to read the paper in the Journal of Hospital Medicine and also join me and Charlie Wray on #JHMChat later this month, I wanted to take author prerogative and describe things you may not always see in a research paper that I learned as the author.
- Engaging nurses and physicians together in SIESTA was key. It was nurses who suggested adding SIESTA to their own nursing huddles. It was the residents and hospitalists who mentioned that it was an EHR problem, meaning they did not even know how to change ordering of vital signs in the electronic health record.
- Changing default options in the electronic health record is a powerful way to change behavior, but it is not enough. I will never forget the moment that I reached one of our EHR programmers who said, “I can easily change the setting so everyone is asked whether they want vitals to be continued through the night. Right now, it is defaulted to no and hidden so no one would even know how to change vital signs.” The EHR was respecting our workflow, but it was automating a bad process. Since everyone ordered q4h vitals, that became our “default.” Changing the default was a powerful way to motivate behavior, but it was not enough. Change is often not sustained unless it is owned by the frontline – in this case, that was when nurses added SIESTA to their daily nursing huddles.
- Data can be repurposed from strange places to monitor improvements. One of the most interesting things about this study to me is the use of hand hygiene data to monitor room entries. I was listening to our hospital epidemiologist, Emily Landon, who had just given a presentation on how the GoJo™ hand hygiene system monitors hand hygiene compliance by unit, and I realized the denominator for the unit-level compliance is measured through heat sensors wired in EVERY hospital door. Not only did she confirm that this was the case, we were able to work with her team to help us pull this data and monitor SIESTA. This data was crucial for our team since it showed that patients rooms are entered a LOT (14 times a night!) and that SIESTA made a big difference (44% reduction in room entries). Think of all the other ambient data that we may be able to use to monitor patient progress and recovery in the hospital and beyond.
- Changing lab ordering culture is very hard. One of the things we tried to change but could not with SIESTA was timing of daily labs. Most daily lab tests are ordered at 4AM in our hospital. Although we offered sleep-friendly times, no one wanted to inconvenience night float or nocturnists with night time checks on routine labs. The thought of ordering labs and having them go unchecked was heresy. 6AM was “way too late” because having no labs at attending rounds was an “intern fail.” Fortunately, over the past year, this is something we have been able to tackle separately using the electronic health record and have made progress here, too. So, stay tuned…
We are now trying out new nudges in the electronic health record and expanding components of SIESTA to our pediatric hospital as well as exploring partnerships in post-acute care settings. We are also working to patient coaching interventions and ways to automate or flag patients who may be appropriate for SIESTA. As we embark on our new work, we look forward to hearing from hospitalists, nurses, residents, patients, or hospital leaders and administrators about ways they can improve sleep in the hospital. All voices are welcome to join us to learn more on #JHMChat on January 28 at 9 p.m. ET when we honor Florence Nightingale’s vision… 160 years later. Better late than never.