• Promising Innovations for Complex Problems

    Oct 31, 2012

    By Maggie Van Dykefeature-art-innovations

    An easy-to-remember mnemonic and robots that can lift heavy boxes do not seem to have much in common. Yet these and other innovations hold potential for achieving the dual goals of improved quality and reduced costs.


    Before vaccines, countless died from diseases that most children today have never even heard of. Before MRIs, many patients had to undergo surgery to get an accurate diagnosis.

    Innovations such as these don't just improve a situation. They completely change the story line.

    As everyone knows, healthcare delivery is currently in need of a few major plot twists. Despite all the medical advances that have extended lives and reduced suffering, the healthcare system ails from numerous broken and inefficient delivery approaches. Many of the problems, while complex and over lapping, can be boiled down to two primary issues: reducing costs and improving quality.

    Motivated by their missions and declining reimbursements, progressive providers are stepping up to the plate to identify and test new approaches that stand to contribute to a less costly and safer healthcare system. The four innovations described in this article vary from technological marvels to group dynamics. What they have in common is the potential to radically alter clinical and operational processes, resulting in new paradigms and dramatic improvements in value.

    Innovation #1: Moving the Acute Care Model Home

    Centura Health at Home is among a handful of providers implementing a well-tested but still novel idea: Bringing the hospital to patients versus the other way around.

    First piloted at John Hopkins Bayview Medical Center in 1996-1998, the Hospital at Home® model involves providing comprehensive acute-level care-including advanced tests and treatments and frequent visits from physicians and nurses-to patients in the comfort of their homes. While not suitable for all patients, the Hospital at Home program has been successfully tested among elderly patients with diseases that can be safely treated at home, such as pneumonia and congestive heart failure. The model also works well as an early discharge option, allowing ICU patients to transfer to Hospital at Home versus a medical-surgical unit.

    Access related sidebar: Is Your Organization Ready for Hospital at Home?

    Meeting the Triple Aim. "There is no place that is more 'vulnerability making' than the hospital," says Erin Denholm, president & CEO, Centura Health at Home. "When you're in the hospital, you have no control at all. In contrast, at home, you are the king of your castle."

    Hospital at Home patients appear to need less pain medicine and have a lower risk of complications, such as delirium and falls. One 2009 analysis found that Hospital at Home patients had a 38 percent lower risk of dying at six months than traditional inpatients (Shepperd, S., et al., "Avoiding Hospital Admission Through Provision of Hospital Care at Home: A Systematic Review and Meta-Analysis of Individual Patient Data," CMAJ, 2009, vol. 180, no. 2).

    Costs also go down. A June 2012 study found a 19 percent cost difference between Hospital at Home patients and similar hospitalized patients at Albuquerque, N.M.'s Presbyterian Healthcare Services. The cost savings was primarily due to shortened lengths of stay and fewer lab and diagnostic tests, according to the researchers (Cryer, L., et al, "Costs for 'Hospital at Home' Patients Were 19 Percent Lower, With Equal or Better Outcomes Compared to Similar Inpatients," Health Affairs, June 2012).

    Not surprisingly, patients really like the program. Presbyterian's Hospital at Home program has a 90.7 overall patient satisfaction score-versus 83.9 percent for its traditional inpatient services.

    Making it even better. Centura Health at Home is about to join a small but growing cadre of providers that offer home-based acute care. A participant of CMS' Innovation Advisory Council, Denholm aims to combine the Hospital at Home concept with a telehealth program in hopes of achieving even better outcomes.

    "In the Hospital at Home program, the physicians and nurses cannot be by the bedside all the time," says Denholm. "Our telehealth program will allow us to monitor patients' vital signs and conduct virtual exams through the telephone lines and respond rapidly, if need be. We can see weights, blood sugars, oxygen saturation, anticoagulation, etc."

    photo-telehealthCentura's telehealth program already has a successful track record. The program started eight years ago with a pilot of congestive heart failure patients. "We decreased ED visits by 100 percent and readmissions by 90 percent. We followed that cohort for three years and none of those patients returned to the ED and only one returned to the hospital with a fractured hip."

    The telehealth program has grown to include patients with all types of chronic diseases-and has an all-cause hospital readmission rate of just 7.8 percent, compared to a national average of around 15 percent.

    Getting the program off the ground. Centura's Hospital at Home pilot is scheduled to launch in October at one system hospital, Penrose-St. Francis Health Services in Colorado Springs. "We hope to see 40 to 50 patients this calendar year," says Denholm.

    Getting the program off the ground has involved intense planning. "Like any new project, we have a project team and many subgroups working on specific aspects of the rollout, such as clinical protocols and admission/eligibility criteria."

    One major challenge has been intake referral, she says. "We can get referrals from the ED, from patients ready to transfer out of the ICU, or from physicians who tell their patients that they need to go to the hospital. We need to figure out how to make that referral process fluid and seamless from each of those settings. We are going to start with one setting, such as the ED. Then, we'll apply what we learn to the other referral settings."

    Innovation #2: Taking an I-PASS Around Medical Errors

    Sometimes improving one serious issue magnifies problems elsewhere. Boston Children's Hospital ran into this situation when complying with new medical education standards that restrict resident working hours to no more than 16 hours per shift and 80 hours per week.

    That's a big change from the frequent 30+ hour shifts that Christopher P. Landrigan, MD, MPH, worked as a resident. Landrigan knows too well the dangers of physician exhaustion, having conducted experiments on this patient safety topic for more than a decade. "Sleep-deprived doctors are much more prone to make mistakes with their patients, and they are more at risk of harming themselves by, for example, crashing their cars on the drive home from work," he says.

    However, reducing resident work hours has amplified another serious patient safety issue: handoff communications. "As you cut down shift lengths, you have more resident transitions, and handoffs are a top cause of sentinel events," says Landrigan, who is the hospital's research and fellowship director of the inpatient pediatric service.

    In fact, miscommunication between caregivers during patient handoffs contributes to an estimated 70 percent of serious medical errors, according to the Joint Commission.

    To reduce handoff-related errors, Landrigan and his colleagues came up with an appealingly frugal solution. Boston Children's residents now use the mnemonic I-PASS for all patient handoffs. The use of the mnemonic-bundled with team training and computerized tools- has been exceptionally successful, with medical errors decreasing by 40 percent during a six-month pilot study at the pediatric teaching hospital in 2009-2010.

    Bundling solutions. Boston Children researchers are now collaborating with colleagues at 10 other training programs across North America on a full-scale test of I-PASS. A $3 million grant from the U.S. Department of Health and Human Services (HHS) is helping to pay for this multi-site evaluation.

    Deceptively simple, I-PASS is a studiously thought-out bundle of three strategies, which attack different aspects of the handoff problem.

    Training. Residents at the 10 participating pediatric hospitals have all taken part in a three-hour interactive workshop that uses true-to-life simulations and role- playing to drive home the importance of good communication and handoffs with other clinicians. This is reinforced with refreshers throughout the year and ongoing feedback. Attending physicians observe and critique the residents during handoffs.

    A lot of the I-PASS training is adapted from the TeamSTEPPS™ program developed by the Department of Defense and the Agency for Healthcare Research and Quality.

    "Our training really emphasizes the need for a more structured approach to handoffs," says Theodore Sectish, MD, program director, Pediatric Residency Program, Boston Children's. "Most doctors think they are doing a good job at handoffs until someone observes them and points out how they failed to communicate or miscommunicated some key details about a patient's case."

    The mnemonic. The I-PASS sequence helps residents communicate key information during shift changes, whether handing off one or 25 patients:

    I Illness severity: How sick the patient is.  
    P Patient summary: Why the patient is in the hospital and what the course of treatment has been.
    A Action list for the next team: What needs to happen for the patient during the next shift.
    S Situation awareness and contingency plans: What is likely to go wrong for the patient-and what potential response plans are in place.
    S Synthesis and read-back: What is going on with the patient-as summed up by the incoming resident

    Ideally, I-PASS handoffs are conducted face-to-face, or at least over the phone. "This gives the receiver of the handoff the opportunity to synthesize what the case is about and ask clarifying questions," says Sectish.

    While initially created for resident-to-resident communications, the handoff protocol is now being adopted by other groups at Boston Children's. For instance, nurses are adapting it for shift changes.


    Handoff document. "The training and the I-PASS mnemonic are reinforced with a documentation tool that is integrated into the hospital's electronic health record," says Landrigan. This tool encourages the residents to use the mnemonic and it helps make the documentation more efficient because the resident can electronically import important information, such as patient allergies, into the handoff document."

    "Previously, residents were retyping a lot of information that wasn't always updated as things changed, says Amy Starmer, MD, MPH, who led the I-PASS pilot at Boston Children's and is now serving as project leader of the I-PASS study at Oregon Health and Science University.

    Paying dividends. In addition to dramatically reducing rates of medical errors, the initial I-PASS pilot at Boston Children's also improved resident satisfaction and allowed residents to spend almost twice as much time with patients and less time at the computer. "Although the handoff protocol takes a little bit more time than the old unstructured method, it seems that the investment of time pays dividends in the course of the resident's shift-presumably because residents are spending less time running back and forth to the computer checking on things," says Landrigan.

    Landrigan, Starmer, and Sectish hope to see similar positive results for the 10-site I-PASS study. Preliminary results are expected next spring. "This has been one of the most important things that most of us have done in our professional lives," says Sectish. "If what happened in the pilot study is borne out in this multi-site study, then that will mean huge reductions in medical errors, major improvements in communication, and the creation of efficiencies that create more time at the bedside for the residents."

    Innovation #3: Letting Robots Take Over Grunt Work

    Also interested in using staff time most efficiently, Parkview Regional Medical Center has hired a new type of mechanical employee that walks on wheels. These vacuum-like creatures can lift up to 500 pounds and deliver everything from patient meals and medications to trash and linens. The robots roll around hospital halls, politely announce deliveries, summon elevators, avert obstacles, and assume behind-the-scene tasks that tend to overtax and injure employees. photo-hospital-robot

    "The robots are exceptionally good at very simple functions, as long as it involves pulling and pushing," says Stewart Graham, director of finance. "They are programmed with architectural maps of our entire building, and they can be assigned to go wherever we need them. They are extremely reliable and can work 24/7."

    Service robots like the ones at Parkview are just one type of robot being used with more frequency in healthcare settings. For instance, physician specialists are using robots to conduct virtual visits with hospital patients, using the robot's telemedicine capabilities to take vitals and talk to patients and hospital staff. Some futurists also anticipate a day when home-bound patients can use robots to help them with housekeeping and self-care tasks.

    Assuming transportation duties. The service robots at Parkview were brought in as additional manual labor after the health system moved all of its tertiary acute care services to a newly constructed hospital this past March. The modern facility is also fitted with patient lift devices and a wireless real-time locating system.

    From a logistical standpoint, the new facility in Fort Wayne, Ind., does have one challenge: A much bigger footprint than the original 50-year-old hospital. This means that routine transportation tasks, such as delivering supplies, take significantly more time and increase the potential for employee injury. That's how the robots have proved helpful. "With the robots, we avoided having to add a lot of new positions solely to support internal movement," says Graham.

    The robots are also helping to address another ubiquitous healthcare issue: how to free nurses and other staff from nonvalue-added work so that they can spend more time with patients or engaged in other mission-oriented work, says Graham. "For instance, if I'm a housekeeper for a unit, I'm on the unit a lot more because I don't have to transport the trash. So I have more hands-on time and I can help the unit make sure the patients' rooms are clean and needs are met."

    Identifying robotic potential. To date, Parkview has been using the robots extensively in four areas:

    • Transporting supplies to decentralized storerooms on the units
    • Carting dirty and clean linen
    • Removing trash
    • Delivering medication to unit nurses via special secure carts

    In the near future, Parkview also plans to use robots for meal service deliveries and is assessing them for use in the lab.

    The robots are highly autonomous, using advanced sensors, motion detectors, voice communication, and other technologies to navigate and carry out their tasks. However, a robot needs to have a Parkview employee on each end of its route. For instance, with medication deliveries, robots convey medications to the nursing units. Nurses then use thumb scanners to open the robots' secure cart and load the medications into the units' medication dispensing machines.

    Expecting an ROI. Parkview currently leases 16 service robots. The contract allows the hospital to flex the number up or down, depending on the hospital's need.

    The company that provides the hardware also provides an on-site consultant for one year plus all service maintenance. "The robots are monitored remotely 24/7 via self-diagnostic software," says Graham. "The robots have a self-correcting technology and can fix themselves. Just in case, the manufacturer also has a 24/7 help desk with tele-operating capabilities via a camera in the robot."

    Parkview expects to break even on the robots and eventually see a modest ROI. "At the end of August, the robots had only been in place for five months," says Graham. "But they had already run nearly 4,800 miles around the hospital and been actively working for more than 7,600 hours. If we annualize that, the robots are already doing the work of more than seven FTEs. We expect this to improve as we fully optimize the robots."

    For instance, other health systems that are using robots for medication delivery are also using a related software that tracks each prescription from beginning to end. In this way, UMMC has cut medication delivery time by more than half, saved nurses more than 6,100 hours in one year spent tracking and retrieving missing medications.

    Graham also anticipates that the hospital will eventually see a decline in employee injuries-along with the associated lost hours and worker's compensation payments-now that the robots are assuming a lot of the manual labor around the hospital. "The bulk of injuries are often in dietary and housekeeping, which are areas associated with a lot of lifting. We expect a real return will come from the robots once we reduce injuries. That will be the bonus."

    Innovation #4: Putting Competitiveness Aside to Improve Quality

    The idea of forming a coalition is by no means new or rare. Yet when competitors get together to collaborate, they still accomplish wonders with the power of their combined might.

    That is what happened in Ohio in 2009. It started with eight CEOs who agreed to put business aside and work together to improve patient safety in their Ohio children's hospitals. Three years after that group handshake, 7,700 Ohio children have been saved from preventable medical errors and $11.8 million in unnecessary healthcare costs has been avoided. photo-ceos

    "As peers, we all agreed that safety is so important and foundational that we would not compete on it," says Michael Fisher, president and CEO, Cincinnati Children's Hospital and Medical Center. "We may compete in other ways, but we are helping each other get as good as we can be in terms of safety."

    The local business community is also lending its support to the initiative-called the Ohio Children's Hospitals' Solutions for Patient Safety (OCHSPS)-including early backing from the Ohio Business Roundtable. One private organization, in particular, has provided $3 million in support.

    OCHSPS is one of numerous regional and national healthcare collaboratives that have reported successful results in recent years. Common to all these improvement coalitions-whether they include a handful or hundreds of providers-is an emphasis on learning through benchmarking and sharing of best practices.

    "In terms of substance, in terms of speed, the collaborative approach has been beneficial," says Steve Muething, MD, safety officer, Cincinnati Children's. "The key is to identify the right practices, standardize those practices, then collect and share the data that demonstrate improved outcomes."

    Agreeing to transparency. Since OCHSPS launched in January 2009, adverse drug events at the eight Ohio children's hospitals have declined by 34 percent. In addition, surgical site infections have been reduced by 60 percent for high-risk cardiac, orthopedic, and neurological procedures.

    Now the Ohio initiative is aiming for even loftier goals. Not satisfied with just reducing harm, the eight Ohio hospitals are working to eliminate all serious harm at children's hospitals statewide. The initial focus is on serious safety events, which are near fatal incidents or deaths that occur after clinicians do not follow best or expected practices. Over the last two years, all eight OCHSPS hospitals have seen a dramatic 50 percent decline in their rate of serious safety events-but they hope to get the rate down to zero.


    To achieve results of this magnitude, each of the eight Ohio hospitals has committed to complete transparency and data sharing around patient safety topics. "We need to fundamentally understand how much harm is occurring and which hospital is doing better than the others," says Muething. "This helps us identify the best practices that we are all going to adopt to reduce each type of hospital-acquired condition."

    Rather than rolling out a single best practice, OCHSPS identifies bundles of best practices to improve different types of adverse events. For example, the surgical infection bundle includes five to six practices, including giving patients antibiotics in the hour before surgery and monitoring a patient's temperature and electrolytes during surgery. "We are not sure if it's one specific practice in the bundle or a combination of practices, but we know if the entire bundle is implemented 85 to 90 percent of the time, the rate of surgical infections drops dramatically," says Muething.

    Identifying the bundle is relatively easy, he says, compared to the day-to-day work of getting physicians, nurses, and other staff across eight hospitals to consistently adopt the bundles. The quality improvement and patient safety staff at all eight hospitals have been kept very busy over the last three years, and all 30,000+ employees of the children's hospitals in Ohio have gone through patient safety training.

    "Clinicians are affected by data," says Muething. "We've had the most luck by showing them the benchmark data and how applying the bundle helps improve outcomes. Plus, you need to find clinical leaders who are willing to stand up and push their colleagues to change behaviors. It's often their peers who will convince them."

    Going national. In March of this year, the Ohio initiative went national with an additional 25 children's hospitals from across the country joining the OCHSPS National Children's Network. Informed by Ohio's example, the participating hospitals will be working together to achieve specific goals by December 2013, including reducing readmissions by 20 percent and serious safety events by 25 percent. Specifically, the national network will be working to reduce harm in nine healthcare-acquired conditions, including adverse drug events, pressure ulcers, and readmissions.

    The national network is being funded through a multi-million dollar contract with HHS, with a goal of reducing pediatric Medicaid costs.

    In addition to improving safety, OCHSPS is making the workplace a more meaningful place to work. "This initiative is incredibly uplifting and gratifying to our caregivers," says Fisher. "By focusing on learning, improvement, and sharing best practices, we are providing safer care to kids."

    Addressing Challenges to Innovation

    The featured providers point to a number of challenges that healthcare leaders may come across when adopting an innovative approach.

    Nothing will spread without a payment source. The widespread adoption of Hospital at Home programs is currently unfeasible because the current fee-for-service Medicare payment system does not pay for providing acute care in the home. Hospitals that have successfully implemented the program are either integrated systems that include a health plan (e.g., Presbyterian) or government-run hospitals (e.g., several VA hospitals).

    This is one reason why Centura chose Penrose-St. Francis to test the model. The hospital has a capitated Medicare Advantage contract with an insurer. "When you have a global cap, you can determine what you do with the money that the insurer provides for the services you provide," says Denholm. "Bundled payment and similar arrangements can also work for this."

    If the Hospital at Home pilot is successful, Centura hopes to spread the program to other hospitals, perhaps through bundled payment arrangements with payers. "Before you can work with insurers on this, you have to know what goes into a diagnostic episode and what it costs. So we are pricing it out now."

    Solutions-even frugal ones-cost money. Innovations, like I-PASS, may not require expensive technology. But adopting a new approach like this, which changes workflows and entrenched attitudes, often requires upfront investments in training as well as additional staff to launch the program and collect performance data.

    Finding the funding often takes some creativity. For instance, the initial pilot of I-PASS at Boston Children's was funded in part by the hospital's medical malpractice carrier.

    Then, before the national 10-site I-PASS project got HHS funding, Sectish and Landrigan asked the participating hospitals to show their support financially before getting involved. "We asked the CEOs if they would be willing to put up an initial $50,000 for the study if the HHS funding did not come through," says Sectish. "They all recognized how important communication failures are to improving patient safety."

    Simple and concise tends to stick. When trying to get staff to adopt new practices or behaviors, less is often more. For instance, before the mnemonic I-PASS was created, Boston Children's tried a longer acronym- SIGN OUT?-that physicians had trouble remembering. "I-PASS is shorter and more catchy so physicians are using it more readily," says Sectish.

    Relevancy is also vital. Boston Children's considered using the popular mnemonic SBAR (Situation, Background, Assessment, Recommendation). SBAR is intended to quickly brief a physician or caregiver about an acute situation, typically with a single patient. But SBAR is not in-depth enough for residents who need to communicate critical issues about 10-20 patients at the end of a shift. "SBAR is very, very good for what it was designed for," says Landrigan. "But we found that physicians found it unwieldly. That's what made us realize that the information may not be organized in quite the right way for resident handoffs."

    Excitement slips away if the change isn't integrated. The success of the initial I-PASS pilot at Boston Children's, with a 40 percent reduction in medical errors, created quite a buzz among staff. However, the hospital had a hard time sustaining the program.

    "Four or five months after our study ended, many of the aspects of our program had faded away," says Landrigan. "I don't think we had done an adequate job of integrating I-PASS fully into the culture and workflows here."

    Before launching the 10-site I-PASS pilot, which includes Boston Children's, Landrigan and his colleagues redesigned the curriculum to make it more engaging and meaningful. The goal: To make I-PASS a routine part of practice. "Across the sites, we engaged a large number of faculty and residents in weighing in on how to improve our curriculum and to give residents some realistic practice with using the I-PASS approach." The I-PASS curriculum is now available to anyone who requests a copy at ipasshandoffstudy.com

    Another challenge was getting the faculty, or attending physicians, to agree to give up their time to observe and critique residents during handoffs. Boston Children's is exploring two possible ways to repay faculty. One is offering them continuing medical education credits. A second possibility is helping them maintain their board certifications by earning credits for a quality improvement project.

    "These incentives are helpful," says Sectish. "But I think the key to success will be the I-PASS intervention itself, which is really evidence-based and consensus driven."

    Value optimization. Just as I-PASS improved during its rollout, the robots at Parkview are still being tweaked so they can work to their full potential. The robot company's on-site consultant has been following the robots around and reprogramming them, as needed, to maximize travel times. "There's a lot of complexity to determining the best route for the robots, such as figuring out where there are frequent hallway log jams," says Graham.

    Parkview employees who interact with robots are also having to change their routines and roles, he says. "This is not a technology that you can just drop into an organization and expect it immediately achieve anticipated results. You have to change your processes so they efficiently incorporate the robots."

    Figuring out the best process often involves trial and error. For instance, the hospital is currently piloting the use of robots in food service. "We have a room service model that allows patients to order meals whenever they want. We've been doing some focused Lean-type work to figure out how to best use the robots in delivering meals. We use a process flow diagram to find those pieces that the robot is going to do and then we determine how to optimize the work on either side of the robot's journey. We're trying to get to the point where the robots can deliver the food up fast enough so it is still hot."

    Another major challenge, says Graham, is to not waste the extra time that employees find on their hands. "Once employees are freed from transportation tasks, thanks to the robots, hospitals need to figure out how to best divert that time to a value-added activity. Otherwise, you are not gaining anything."

    Spreading Innovation

    One of the dangers of innovation is that it sometimes spreads too quickly before people are truly ready for it, warns Lindsay Martin, executive director of research and development, Institute for Healthcare Improvement (IHI). "It's natural to get excited when you hear about an innovation. A unit nurse will say, 'I'm ready. Give it to me.' But the individuals in the spread site may not be ready, or the change/innovation may not be ready if it has not been sufficiently tested under a variety of conditions to help ensure consistent results."

    In other words, do not try the innovations in this article at home ... without the proper front work. (Access two papers about spreading improvements at ihi.org)

    "There's a whole science to spread," says Martin. "When spreading an innovation, you have to do your own rapid-cycle testing until you find the approach that works best for you, while still keeping pertinent pieces from the original."

    Maggie Van Dyke is HFMA's managing editor of Leadership (mvandyke@hfma.org).

    Interviewed for this article (in order of appearance): Erin Denholm, president & CEO, Centura Health at Home, Denver. Christopher P. Landrigan, MD, MPH, is research and fellowship director of the Inpatient Pediatric Service, Boston Children's Hospital, Boston (clandrigan@partners.org). Theodore Sectish, MD, is program director, Pediatric Residency Program, Boston Children's Hospital (Theodore.Sectish@childrens.harvard.edu). Amy Starmer, MD, MPH, is assistant professor of pediatrics, Division of General Pediatrics, Oregon Health and Science University, Portland (starmer@ohsu.edu). Stewart Graham, is director of finance, Parkview Regional Medical Center, Fort Wayne, Ind. (stewart.graham@parkview.com). Michael Fisher, is president and CEO, Cincinnati Children's Hospital and Medical Center, Cincinnati. Steve Muething, MD, is safety officer, Cincinnati Children's. Lindsay Martin, is executive director of research and development, Institute for Healthcare Improvement, Boston (lmartin@ihi.org).