This year’s RIV innovation winners reflect a nascent trend of applying informatics to quality improvement and patient safety initiatives.
“One striking thing is that all three winners used either EHR or Big Data and large collaboratives to achieve their goals,” Margaret Fang, MD, MPH, FHM, program chair for HM17’s scientific abstracts competition, and moderator of the winners’ panel, said in an interview. This year’s winners included a sleep-promoting “nudge” system that Dr. Fang said she expects will help improve sleep and lower rates of delirium and a source code that connects disparate data systems for daily updates on where quality can be improved. The third winner used what Dr. Fang, a hospitalist and the medical director of the anticoagulation clinic at the University of California, San Francisco, called a “classic quality improvement collaborative,” which simplifies the decision tree around venous thromboembolism (VTE) prophylaxis for better patient outcomes.
Calling uninterrupted sleep the “sine qua non” of patient care, RIV award recipient Vineet Arora, MD, an associate professor of medicine at the University of Chicago, described the rationale for her RIV award-winning SIESTA (Sleep for Inpatients: Empowering Staff to Act) program. She and her colleagues surveyed hospitalists, nurses, residents, and patients to determine the most common sleep disrupters in their institution and devised “nudges” to alter how staff performed various tasks that otherwise might interfere with patient sleep. Rather than use overt incentives, nudges are changes in what Dr. Arora called the “choice architecture” of people’s behavior.
Based on survey feedback, Dr. Arora and her colleagues worked with their electronic health record (EHR) vendor to consolidate the performance of certain tasks that were affecting patient sleep. Reminders were added to daily nursing huddles to prompt them to look for ways they could decrease patient interruptions, and empowerment coaching was offered to nurses to encourage patient advocacy when physicians had given orders that would interfere with patients’ sleep.
When tested and measured over the course of a year, SIESTA’s EHR innovations resulted in six fewer nighttime disruptions than before the intervention, compared with controls, a statistically significant difference. The nursing-based interventions resulted in one less nocturnal interruption on average, also a significant change.
“If every patient were admitted into a SIESTA unit, 84% would say they were not disrupted by medications, compared to 57%. For interruptions for vitals, it would be 17% vs. 41%,” Dr. Arora said. In terms of the Hospital Consumer Assessment of Healthcare Providers and Systems) data, this translates into as much as a 25th-percentile performance improvement for hospitals in related domains, according to Dr. Arora.
Nader Najafi, MD, an assistant clinical professor of medicine at UCSF, and his colleagues created Murmur, an open-source code data aggregator, which can be customized to solve a variety of quality improvement issues. RIV award winner Dr. Najafi applied the code to determine how systems failures in their institution were contributing to avoidable inpatient days, for example. At a daily appointed time, Murmur would determine which staff members were scheduled to work that day. Each provider would then receive a brief, customized survey about patients for that day on their cell phone. The data were then collected to create instant reports of where the delays in discharge were occurring.
Testing by gastroenterologists was pinpointed as a “huge source of delays, something we had never been able to quantify before, “ Dr. Najafi said. This led to brainstorming sessions with the department for solutions.
To reduce rates of hospital-associated VTE, 35 California hospitals with varying numbers of beds and locations collaborated on a project led by RIV award recipient Ian Jenkins, MD, SFHM, a health sciences clinical professor at the University of California, San Diego. Key components of the intervention were mentoring at the sites by VTE prophylaxis experts, group webinars in best practices, and a “measure-vention.” Teams were taught how to rate patient risk for VTE and apply specific protocols according to risk rating using the SHM-mentored implementation model. Real-time monitoring of the intervention was used to make any necessary adjustments. When before-and-after data were compared, following the 18-month period during which the intervention was measured, Dr. Jenkins said an average of 330 VTEs were averted annually. “We found the results very gratifying,” said Dr. Jenkins.
“These projects all reflect a broader trend in hospital medicine where we are using the wealth of data we have now for quality improvement and for outcomes research,” Dr. Fang said in the interview.
There were no relevant disclosures.