Blog

Blog

Samara Rosenfeld


Recurrent neural networks (RNN) provided significantly better accuracy levels than the clinical reference tool in predicting severe complications during critical care after cardiothoracic surgery, a new study found.
 
Alexander Meyer, M.D., department of cardiothoracic and vascular surgery at German Heart Center Berlin, and his team used deep learning methods to predict several severe complications — mortality, renal failure with a need for renal replacement therapy and postoperative bleeding leading to operative revision — in post-cardiosurgical care in real time.

“For all tasks, the RNN approach provided significantly better accuracy levels than the respective clinical reference tool,” the researchers wrote.

Mortality was the most accurately predicted, scoring a 90 percent positive predictive value (PPV) and an 85 percent sensitivity score. Renal failure had an 87 percent PPV and 94 percent sensitivity score.

The deep machine learning method also showed area under the curve scores that surpassed clinical reference tools, especially soon after admission.
 
Of the data studied, postoperative bleeding was the most difficult method to predict, due to how accurate the predictions were for mortality and renal failure. Postoperative bleeding had a PPV of 87 percent and sensitivity of 74 percent.
 
The team studied electronic health record (EHR) data from 11,492 adults over the age of 18 years old who had undergone major open-heart surgery from January 2000 through December 2016 in a German tertiary care center for cardiovascular diseases.
 
Patients’ data sets were studied for the 24 hours after the initial study, and if any complication occurred, patients were labeled accordingly.
 
Researchers measured the accuracy and timeliness of the deep learning model’s forecasts and compared predictive quality to established standard-of-care clinical reference tools.
 

Meyer told Healthcare Analytics News™ that one of the major findings of this study was that the system developed outperformed all three pre-existing benchmarks. He added that it is possible to work on a real-time uncurated clinical data stream.

With this information, physicians in emergency care units can perform interventions immediately if a patient is experience complications.
 
“Health systems should openly embrace this technology and ideally try to make use of it,” Meyer said.
 
At the very least, health systems can try to get regulations and developments so that this technology can be used.
 
In a clinical setting, technology like this is difficult to implement and generally demands a financial incentive.
 
Hospitals can work with researchers and companies to push this technology forward and gain support from politicians to help provide financial means and ways to attain these tools.  



Case Western Reserve University


Electronic health records (EHRs) produce savings for hospitals by reducing the average length of patient stays—but only in facilities meeting the highest federal standards for implementing the technology, according to new research from Case Western Reserve University.

The findings are significant for a health-care industry with growing levels of spending—now roughly 18 percent of the nation’s gross domestic product.

In hospitals meeting the federal government’s measure of “meaningful use” of electronic health records, patients are discharged nearly four hours earlier—approximately a 3 percent reduction of the average five-day hospital stay.

For sicker patients, the benefit was even greater: Those with complex or multiple chronic conditions see up to an additional 0.5 percent reduction in their hospital stays.

What’s more, researchers found that these shortened stays did not come with an increase in re-admissions. With prolonged patient stays costing hospitals an average of $600 a day, the use of electronic records could help contain growing costs, especially amid a trend of reduced reimbursements from insurance companies and entitlement programs.

“Longer hospitals stays cost more money for all involved,” said Manoj Malhotra, dean of the Weatherhead School of Management at Case Western Reserve and co-author of the research.

“Electronic health records, when meaningfully implemented help patients go home sooner, reducing their exposure to germs in the hospital and likelihood of having to come back," he said.

Hospitals that did not fully engage in meaningful use of electronic records showed no significant reductions in length of patient stays, according to the study, which was published in the  Journal of Operations Management.

“Any efficiencies, even small improvements, can produce significant savings when adopted in a large health-care system—and are certainly preferable to the alternative,” said Malhotra, who is also the Albert J. Weatherhead, III Professor of Management at the university.

Health-care savings, thanks to federal perks (and penalties)

While electronic records are touted for their potential to reduce hospital errors and inefficiencies, their adoption had been slow among U.S. hospitals.

In 2010, a $27 billion package included in the Health Information Technology for Economic and Clinical Health (HITECH) Act encouraged hospitals to adopt and meaningfully use the technology—and established penalties for failing to do so, such as negative adjustments to Medicare and Medicaid reimbursement.

The approach has been successful in pushing increased adoption of electronic records: By 2015, the level of adoption had reached 80 percent of hospitals nationally. But a more proactive approach that meaningfully uses the technology beyond mere adoption may be needed to see more progress, researchers conclude.

The researchers categorized hospitals into one of three categories—partial adoption of EHRs, full adoption of EHRs and “meaningful assimilation” of EHRs.

“Whereas partial or full adoption showed no benefits for reducing patient stays, meeting the government’s highest standard of meaningful use reduced length of stay without any adverse impact on readmissions,” said Malhotra. “Results from this study indicate that meaningful assimilation of technology is likely to help free-up clinicians and other valuable resources –this approach is preferable to making additional investments in facilities or hiring additional employees as more people seek hospital services.”

The study

The research—co-conducted with Deepa Wani, an assistant professor of management science and statistics at the University of Texas at San Antonio—used four years of detailed patient-level data from all acute-care hospitals in California, in addition to as data reported by the Centers for Medicare & Medicaid Services (part of the U.S. Department of Health and Human Services) on hospitals that successfully attested to meaningful use criteria stated in HITECH.

Austin Fitzgerald / U. of Missouri


Nursing homes that adopt more sophisticated information technologies are seeing specific improvements in the quality of care, a new study shows.


These improvements include significant decreases in urinary tract infections, patients reporting moderate to severe pain, and patients with new or worsened pressure ulcers.

Health care providers in hospitals and ambulatory care are currently incentivized with federal funds to adopt health information technology (IT). Nursing homes, however, have been largely left out of these incentive programs, although this health care sector is beginning to see some benefits. For example, IT systems support health information exchange and access to electronic health records by care providers across settings, enabling them to address patients’ needs better.

Now, a new study has linked more sophisticated nursing home information technology, including electronic medical records and other digital data systems in resident care, clinical support, and administrative activities with specific improvements in quality.

“We already knew that information technology can help create better care outcomes, but this study helped us see which technologies improve which elements of care,” says Gregory Alexander, a professor of clinical informatics at Missouri University. “As IT capabilities and extent of IT use improved in nursing homes, we saw an associated decline in urinary tract infections, among other correlations.”

Alexander and his colleagues collected surveys once per year for two years from nursing homes nationwide. The researchers compared the responses, which rated the sophistication of a given facility’s information technology, against federal data describing 18 quality measures in those same facilities, and technology had positive affects on quality of care. For example:

  • Facilities reporting greater IT capabilities in administrative activities were associated with a lower percentage of residents reporting moderate to severe pain.
  • Increased use of clinical support technologies, which include laboratory systems, was associated with lower percentages of residents with urinary incontinence.
  • Increases in technology use in resident care were associated with lower percentages of residents with new or worsened pressure ulcers.

The researchers also observed that while the overall trend was an increase in IT adoption, some nursing homes actually lost capabilities between years one and two. Though these facilities were outliers, Alexander says they reflect the challenges nursing homes face when adopting new technology.

“Federal incentive funds are going into hospitals and ambulatory care, not nursing homes,” Alexander says. “Many homes don’t have a trained expert to manage the technology, so even if they do decide to upgrade their IT capabilities, they may abandon certain ones because they are too difficult or expensive to manage. If they aren’t being reimbursed for investing in information technology, they may decide it isn’t worth the time and money.”

Alexander says that because the study detailed the impacts of a variety of specific IT factors on different aspects of quality of care, the data could help inform nursing home administrators about which features of an IT system are important to adopt to improve quality of care. This information could be very helpful to administrators and other leaders in making decisions about how to design and implement information systems.

The study appears in the Journal of Nursing Care Quality. A grant from the Department of Health and Human Services’ Agency for Healthcare Research and Quality supported this research.


Rosy Matheson


Artificial intelligence looks set to transform nursing over the coming years.


If you think the digitisation of nursing is just about nurses filling out scores on a mobile device, it’s time to think again because artificial intelligence (AI) could be about to revolutionise the way nurses do their jobs. Recent digital developments include bottles which automatically issue reminders to drink, diapers that sound an alert when wet and sensor-equipped stoma pouches.  

Heiko Mania, NursIT CEO, and a former nurse, believes AI will change the focus of nursing care: “Modern nursing expert software not only streamlines nursing documentation, it will automate it using AI, sensors and smart nursing aids. At the same time, professional nursing care will change from reactive to predictive, preventive nursing care.”

Mania said they had developed a nursing care expert system, CareIT Pro, which supports automation in nursing. He explained that smart algorithms and AI could reduce the need for information to be entered and could link content, so that further workflows and tasks could be automatically initiated at the right time. He added that the software automatically recognised patterns, evaluated the planned nursing goals and recommended necessary adaptations.

He said sensors, wearables and smart devices were also enabling increased automation: “Intelligent tools automatically deliver data on the patient to the nursing expert software and thus allow automated documentation. Alarms, nursing tasks and digital processes can be generated and started independently. Nursing staff not only receive digital to-do lists, but can also see the current status and quality of the nursing processes at all times and react to them at an early stage.

An intelligent drinking cup can automatically fill the drinking protocols and remind the patient to drink regularly or the stoma pouch sensor generates an automatic care task for changing the bag when it is almost full. We are currently developing an intelligent nursing mattress with a partner company that can detect not only the patient’s movement, breathing, position, pressure and sleep, but also incontinence.” 

Improved quality of care and patient safety

If the Internet of Things (IoT) is set to transform nursing, it is also starting to change the way nursing is taught. Widener University in Pennsylvania has introduced a range of simulation training from programming intravenous pumps and pumps for medication to updating electronic health records.  It also runs disaster simulation training as Widener, in line with other nursing schools, has recognised the need to prepare nurses for such incidents in the wake of 9/11.

Nancy Laplante, Associate Professor of Nursing at Widener University, has recently published a paper arguing the case for introducing IoT in disaster training. Laplante believes this would highlight the application of these technologies in a meaningful way and enhance the experience for nursing students. She would like to teach the students to use mobile apps to track patients, triage them and track them to different hospitals, rather than using, for example, old-school paper-tagging of victims.   

She said that downloading simple drawings, like Rich Pictures and Use Case Diagrams, which show all the participants in the disaster scene at a glance, can also improve understanding: “We were looking at what we call this rich picture for disaster scenarios and it was one way to help visualise all the interactions that would occur.  What we wanted to do was to give students an understanding of how complex communication is in a mass casualty disaster scenario. It is not just nurses talking to patients; they are going to have to deal with fire fighters, police officers, bystanders and health providers that are off site.” 

Laplante said students had to get to grips with new technology as it was a growth area. She said that nursing students had to understand, embrace and help develop new solutions as they could transform their practice. However, she pointed out that while IT was an important aspect of nursing, it could never take the place of nurses: “I don’t personally believe that nurses can ever be replaced because you always need that human touch. My hope and my feeling for technology is that it can help enhance our care.” 

It is likely that technology will fundamentally change nursing over the coming years and, provided it is used correctly, it seems it really could improve the quality of care and lead to increased patient safety.  



Jessica Kent 


With more than $3 million in funding, the Center will use big data analytics to advance precision medicine worldwide.

Duke University Pratt School of Engineering has established a new big data analytics center that will support global research to advance precision medicine.

Launched last month, the Sherry and John Woo Center for Big Data and Precision Health will receive more than $3 million in funding over the next three years from philanthropist and biotech industry executive John Woo. The Center will help Duke faculty and students develop innovative methods for turning big data into actionable clinical insights.

Investigators will have new opportunities to work with hospitals, government agencies, and biotech companies worldwide to advance data-driven health research.

“Big data, analytics, and machine learning are changing our world significantly, and nowhere will the change be more significant and meaningful than in healthcare,” said Ravi V. Bellamkonda, Vinik Dean of Engineering at Duke.

“Duke Engineering and Duke Health are collaboratively leading this change, and the Woo Center will help catalyze this further by coordinating new partnerships, expanding access to diverse, well-curated datasets and fueling transformative research ideas in this space.”

The center already has research efforts under way in China, where a team is developing a national network of health data parks to improve rural care delivery.

The new facility will also award annual pilot grants of up to $150,000 to Duke faculty so they can explore new ideas for collaborative projects.

In addition, the center will hold a yearly symposium to highlight significant findings to further build a global community of researchers. Leaders plan to sponsor global internships and exchanges for Duke students, as well as business plan and pitch competitions.

“Big data and precision medicine have the potential to vastly improve human health, and Duke has a special role to play with its unique combination of strengths in data science and machine learning, biomedical engineering and medicine—our faculty are world leaders in each of these areas,” said Larry Carin, Vice Provost for Research at Duke.

“Through new partnerships in China and around the world, we hope to address pressing medical issues in emerging markets and reduce disparities to improve global health.”

Duke University expects that the new Woo Center will add to its existing research efforts and will help foster the study of healthcare big data analytics

“Duke is already at the forefront of bringing big data and precision medicine into clinical practice,” said Xiling Shen, the Hawkins Family Associate Professor of Biomedical Engineering, and director of the new center. 

“We’re excited about the opportunities this new center will open for our faculty and students to build productive new collaborations with clinicians and biotech companies to make an impact for patients.”

Other organizations have established similar facilities to improve care delivery.

In September 2018, the New Jersey Hospital Association launched a big data analytics center to identify and address gaps in care. Researchers plan to use predictive modeling and other analytics strategies to extract meaningful insights from big data.

“So many of the problems we see in healthcare today – racial and ethnic disparities, access to care barriers, variations in use of healthcare services, variables in access and funding of prevention and wellness – require a deeper dive intowhy,” said NJHA President and CEO Cathy Bennett.

“One of the ways we get closer to answering that question is to have solid data that shows us the root causes of these problems. We can then support design of solutions that address the foundation of the problem, rather than the symptoms.”

The University of California, Irvine (UCI), also recently launched an artificial intelligence center to help researchers develop deep learning tools and apply them to big data. The new center will allow researchers and faculty to collaborate and translate AI-based concepts into clinical tools that will improve patient health.

Additionally, Dell Medical School at the University of Texas at Austin has established a big data analytics center, called the Biomedical Data Science Hub. The facility will use big data analytics to enhance population health research, showing how both clinical and non-clinical factors affect health outcomes.

“To increase the pace of innovation in health, high-quality data needs to be ubiquitous and analysis much richer, and that’s what we’re trying to achieve with the data hub,” said Clay Johnston, MD, PhD, Dean of the Medical School.

“UT already has so much strength in this area, and now it’s about directing that toward the key questions in health including addressing health inequities in our community.”



Kate Monica


The regional health information exchange leveraged funding to engage more providers in health data exchange.

More than 60 healthcare providers throughout the greater Rochester area are contributing patient EHRs to Rochester RHIO after receiving grants from a New York grant program, according to the Monroe County Post.

The grant from the Data Exchange Incentive Program (DEIP) is being used to offset setup costs for enrolling additional providers and patient health records into the regional health information exchange (HIE).

The DEIP was established in 2017 by the New York State Department of Health (DOH), with support from CMS. The grant program was launched to increase HIE adoption across the state. The New York eHealth Collaborative coordinates the programs and awards incentive payments on behalf of DOH.

“Hundreds of health care organizations were already sharing patient information, but as extensive as that data was, records were still not complete in many instances,” said Rochester RHIO President and CEO Jill Eisenstein. “With the help of DEIP, we’ve expanded our data sources to include groups such as skilled nursing facilities and diagnostic treatment centers.”

Rochester RHIO is one of eight qualified entities (QEs) part of the Statewide Health Information Network for New York (SHIN-NY). Organizations can add patient health data such as medications, lab test results, care plans, procedures, and other health information to offer providers a more comprehensive view of patient health spanning multiple care settings, facilities, and care teams through RHIO’s Contribute service.

Contribute allows providers to share patient data in the form of C-CDAs through provider EHR systems. Utilizing the Contribute service to add new patient health information to the exchange helps to enable better-informed clinical decision-making and improve care coordination.

Hospitals, healthcare organizations, private practices, and ambulatory care sites can gain access to this data with patient consent.

“By having a more complete digital record of care for each patient, health care providers can make more informed decisions,” Eisenstein said. “We’re looking forward to helping even more health care organizations connect, especially with the financial assistance from New York state.“

Increasing the number of providers contributing patient EHRs to the regional HIE will allow Rochester RHIO to provide a more complete view of patient health to each patient’s treating physician and care team. The grant supports the HIE’s mission to support high quality patient care across the community through the use of clinical data.

Grants from the DEIP can be used to build EHR interfaces that connect with QEs to increase the quantity and quality of data in SHIN-NY. The grant program was designed to help offset the costs of connecting to QEs for healthcare organizations by offering incentivizes to healthcare organizations that share a pre-defined set of data elements with other providers.

Two other New York-based HIEs part of SHIN-NY recently entered into a strategic partnership to boost HIE use among area providers.

HealthlinkNY and HealtheConnections partnered earlier this month after months of collaboration. HealthlinkNY first announced its decision to seek a strategic partner in 2017. Together, the two QEs cover 43 percent of providers across New York.

Prior to announcing its strategic partnership, HealthLinkNY stated its disapproval for another SHIN-NY QE that planned to expand its HIE services into HealthlinkNY’s territory.

Hixny planned to extend into nine additional counties over an 18 month period, stating health data exchange in those areas “historically lags.”

Several medical centers and clinics in regions covered by HealthlinkNY signed participation agreements with Hixny.

According to HealthlinkNY Executive Director Staci Romeo, Hixny’s expansion into territories covered by HealthlinkNY were the result of a “case of sour grapes after being passed over during our search for a strategic partner.”

HealthlinkNY covers 13 counties in the Hudson Valley, Catskills, and the Southern Tier.



Jennifer Bresnick


Cutting costs doesn't mean sacrificing on quality of care if hospitals focus on reducing unnecessary care variations and trimming down on wasteful testing.


High quality and lower costs can indeed go hand-in-hand for hospitals, according to new data from Advisory Board, if healthcare organizations can successfully reduce unnecessary variations in care.

An analysis of more than 460 hospitals revealed that the highest quality facilities delivered lower-cost care for 82 percent of diagnoses included in the study, indicating that investments in patient safety, standardized care delivery methods, and enhanced health IT tools may be worth the effort.

“Care variation reduction (CVR) is one of the few avenues for generating the level of savings needed to withstand downward pressures on hospital revenues without negatively impacting care, and hopefully improving it,” said Steven Berkow, Executive Director, Research at Advisory Board, an Optum, Inc. business.

Hospitals that follow the lead of their highest-quality, lowest-cost peers could save up to $29 million each year, the report added.

Advisory Board researchers derived the potential savings goal from analyzing cost and quality data from more than 20 million patients across 468 hospitals.  They found that the average hospital spends up to 30 percent more to deliver the same care than a hospital in the highest-performing group.

“Our high-performer benchmark is based on high-quality care, not low cost,” explained Veena Lanka, MD, Senior Director, Research at Advisory Board. 

The team explored variations in common quality metrics, such as rates of complications, to assess hospital performance.

“Closing just a quarter of the cost gap for less than 10 percent of the conditions we analyzed could net over $4 million in annual savings for a typical hospital and over $40 million for 10-hospital system—without compromising quality,” Lanka stressed.

However, Berkow pointed out, “Achieving a realistic chunk of this savings opportunity…will require most health systems to rethink how they prioritize, set and embed care standards.”

Reducing variations in care requires a collaborative effort that involves standardizing provider training, carefully choosing the appropriate settings for care, and fostering a greater reliance on meaningful health IT tools.

Reducing emergency room use by redirecting non-emergency cases to urgent care facilities can help to conserve resources in more expensive settings – as long as the urgent care clinics adhere to best practices for antibiotic stewardship and maintain high quality in other areas of care.

To ensure less variation in how services are applied, organizations may wish to consider clinical decision support (CDS) technologies that can ensure that providers are aware of the latest clinical guidelines for treating specific conditions. 

CDS tools may help to reduce unnecessary testing or imaging, and can help providers react more quickly to high-risk conditions such as sepsis.

Trimming down on repetitious or low-value imaging and lab testing can help to prevent billions in wasteful spending that lead to high costs without producing better outcomes. 

In a 2017 study from Health Affairsresearchers found that low-value testing and imaging contributed to more than half a billion dollars in spending per month in Virginia alone. 

Nationally, wasteful spending accounts for nearly a third of all healthcare dollars each year.

At Methodist Le Bonheur Healthcare, tacking the problem of variation in care and high spending involved significant investment in data analytics and health IT tools, explained Arthur Townsend IV, MD, MBA, Chief Clinical Transformation Officer for Methodist Le Bonheur Healthcare.

“Embarking on a journey to reduce care variation can be challenging, but our success is due to dedicated teams of physicians, nurses and administrators, all working toward the common goal of improving every life touched at Methodist Le Bonheur Healthcare,” he said.

The Tennessee-based health system initially targeted unnecessary laboratory utilization and blood transfusions, using data analytics tools to identify opportunities for improvement that would not negatively affect patient care.

The health system then moved on to develop standards of care for stroke and sepsis, creating Clinical Consensus Groups packed with subject matter experts to define guidelines for treating patients with these conditions. 

The experts, including administrative and clinical champions, took a close look at how to improve clinical documentation and standardize care delivery and infuse new best practices into the daily routines of care providers.

As a result of both efforts, the health system saw more than $800,000 in cost savings and revenue enhancements in a single quarter.  Atrial fibrillation is next on the list, promising even more gains in quality and cost.

“We see care variation initiative as the next frontier in improving overall quality and significant cost reduction across the system through physician leadership,” said Michael Ugwueke, president for Methodist Le Bonheur Healthcare.

While Advisory Board’s Lanka noted that it is not likely that hospitals will be able to stamp out all care variation due to differences in patient demographics, clinical severity, and other underlying socioeconomic issues, most hospitals will have some opportunities to reexamine care delivery and the costs associated with unnecessary utilization or discrepancies in delivery.

The goal is a very high priority for hospitals and health system, according to an accompanying survey of C-suite executives, with “preparing the enterprise for sustainable cost control” taking the top spot on their checklists for the remainder of 2018.

Organizations that hope to achieve that objective will benefit from assessing their current clinical processes for high-cost conditions, considering new technologies to support adherence to clinical guidelines, and investing in innovative initiatives to engage providers in quality improvements that simultaneously lower costs.



Brita Hansen


The medical world has declared zero tolerance for healthcare-associated infections (HAIs), but it is a massive problem to address. 

Seven out of every 100 hospitalized patients at any time and about 30% of patients in intensive care units will acquire at least one HAI, according to the World Health Organization.

HAIs such as Clostridium difficile (C. diff) and catheter-associated urinary tract infections (CAUTIs) take a heavy toll on patient outcomes and length of hospital stay. They are also expensive. The most common infection is CAUTI, accounting for more than 30% of HAIs, costing health systems about $500 million annually in the direct cost of treating patients. 

Making matters worse, Medicare does not reimburse for certain HAIs, and a portion of reimbursements are withheld for the quartile of hospitals with the most HAIs. When Medicare penalties and lost revenues are included, the cost likely exceeds $1 billion annually. 

Thankfully, health systems already have a powerful weapon that can make a major dent in infection rates: electronic health records. For true progress to be made toward zero HAIs, healthcare needs a greater focus on using this tool along with key clinical processes to guide the delivery of care.

Here are five specific areas where health systems can and must improve:

  1. Streamline workflows. Hospitals are reducing infection rates by using EHR systems to streamline workflows by making the right thing to do the easiest thing for clinicians to do. The idea is to remove unnecessary variation in the way care is delivered to ensure clinicians follow the best clinical practices. For example, it was once routine to place catheters in many hospitalized patients, but now catheters should only be used when the patients meet specific clinical criteria. Once inserted, EHR systems can be leveraged to remind clinical staff that a patient has a catheter and suggest they remove it or document why it is still indicated. Up to 69% of CAUTI cases can be prevented by following these and other evidence-based strategies.
     
  2. Stop overtesting. Hospital staff too often overtest for infections when there are not strong clinical reasons to expect an infection. For example, hospitals overtest for C. diff, the most common infectious cause of healthcare-associated diarrhea and a significant factor leading to morbidity, mortality and increased length of stay among hospitalized patients. This inappropriate C. diff testing leads to false positives and overdiagnosis. Using EHR systems to ensure care processes are aligned with current testing guidelines and workflow best practices can produce even bigger improvements. One hospital following this approach had a 50% reduction in C. diff infections.
     
  3. Focus on best practices. Getting CAUTI rates to zero means clinicians should set strict rules for testing and focus on the correct insertion, maintenance and removal of catheters. A study published in Infection Control & Hospital Epidemiology found that when ICUs at the Cleveland Clinic improved their catheter protocols that CAUTIs decreased from three per 1,000 catheter days to 1.9. Similar improvements can be made by ensuring clinicians focus on best practices throughout the care-delivery continuum.
     
  4. Minimize patient risk of infection. Whether it’s overuse of catheters or antibiotics, more rigorous hospital workflows can minimize infection risk. That was among the topics discussed in June at the Association for Professionals in Infection Control and Epidemiology (APIC) conference, where more than 4,500 infection prevention professionals gathered to discuss strategies to lower infection rates. The conference revealed that many hospitals and health systems still have multidisciplinary teams performing retroactive chart reviews of HAI cases to look for things that could have been done differently to mitigate the risk of infection. However, such retroactive assessments can never find the type of consistent process improvements that can be identified by a well-organized EHR, where you can look at and sort large data sets to find patterns. Rather than reviewing past cases, EHRs can even be set up to proactively make recommendations for clinicians. 
     
  5. Improve infection controls. Once an infection is identified in a timely fashion, hospitals need consistent procedures to stop it from spreading. Those should include isolating the patient, clinicians wearing special gowns and gloves, special hand-washing procedures and, crucially, communicating those standards to staff.

It’s not uncommon in American hospitals for nurses to attempt to manage catheter hours by physically walking around wards, seeking out patients with catheters to assess whether proper care was performed and which catheters can be removed. 

In an age when we use technology for everything from better navigation to movie recommendations, hospitals should use modern information technology available to them to push their HAI rates to zero as quickly as possible.



Jessica Kent 


The data sharing policy will support big data analytics by allowing investigators to reuse information and speed up innovations in health research.

The Patient-Centered Outcomes Research Institute (PCORI) has approved a new policy that will encourage data sharing among researchers with the goal of accelerating big data analytics and secure health information exchange.

The new policy strengthens PCORI’s commitment to open science by allowing researchers to verify and build on past findings from PCORI-funded studies and generate new evidence for healthcare decision-makers.

Research teams that have received PCORI funding will place the data generated from their studies, as well as documentation for how that data was produced, into a repository designated by PCORI.

The data, which could include deidentified participant information, full protocols, metadata, and statistical analysis plans, can then be made available for other research teams for additional analysis. PCORI will also provide funding to researchers so that they can prepare the data and other materials for sharing.

“Through this data sharing policy, we’re taking a major step in advancing open science,” said PCORI Executive Director Joe Selby, MD, MPH.

“By supporting how others may use information generated by the studies we’ve funded, we’re helping to enhance the quality and increase the quantity of evidence for healthcare decision making. We’re also reducing redundancy in collecting clinical data sets, which can speed research and the production of more useful evidence.”

PCORI will also require that all personal health information is de-identified to protect the privacy of study participants.

Additionally, informed consent from study participants is required to permit the reuse of data.   PCORI will review requests before granting access.

The new data sharing policy is part of a series of initiatives from PCORI that aim to support research transparency and ensure broad availability of high-quality health data assets.

The organization’s policy on peer review and public release of research findings ensures that all results from PCORI-funded studies undergo a review and are made publicly available on PCORI’s website in a final research report.

In addition to the reports, PCORI offers brief summaries of studies and their findings that are posted as public and professional abstracts on the website.

The Institute also has a public access policy in place to cover the costs for journals to make papers presenting the results of PCORI studies freely available to the public.

By approving this new data sharing policy, PCORI expects to expand on these past initiatives and accelerate healthcare innovation.



Jennifer Bresnick 


The combination of artificial intelligence and the Internet of Things will support the development of smart hospitals and fuel the ongoing growth of big data analytics.

The Internet of Things (IoT) is expected to combine with the power of artificial intelligence, blockchain, and other emerging technologies to create the “smart hospitals” of the future, according to a new report by Frost & Sullivan.

The IoT – also commonly known in the healthcare industry as the Internet of Medical Things (IoMT) – consists of any and all medical devices, patient monitoring tools, wearables, and other sensors that can send signals to other devices via the internet.

These tools generate massive amounts of data that must be stored, integrated, and analyzed in order to generate actionable insights for chronic disease management and acute patient care needs.

IoT data is a valuable addition to other clinical data sources, such as the electronic health record (EHR), that allow providers to monitor patients on an ongoing basis or predict changes in an individual’s health status.

“Escalating demand for remote patient monitoring, along with the introduction of advanced smartphones, mobile applications, fitness devices, and advanced hospital infrastructure, are setting the stage for establishing smart hospitals all over the world,” says the report.

Predictive analytics strategies are beginning to rely on the availability of data from wearables and IoT devices both inside and outside of the hospital. 

Predicting patient deterioration or infection in the inpatient setting requires continuous feedback from bedside devices, while home monitoring tools such as Bluetooth-enabled blood pressure cuff, scales, and pill bottles can keep patients adherent to chronic disease management protocols outside of the clinic.

According to a recent analysis by Deloitte, more than two-thirds of medical devices will be connected to the internet by 2023, compared to just 48 percent of devices in 2018. 

The uptick in connected devices will lead to the availability of more data for analytics, which will in turn require novel methods of extracting meaning from raw datasets.

Artificial intelligence and machine learning strategies are ideally adapted to managing and analyzing continuous data streams in large amounts, says Frost & Sullivan, and will be critical for ensuring that actionable insights are presented to providers without overloading their workflows.

“Sensors, artificial intelligence, big data analytics, and blockchain are vital technologies for IoMT as they provide multiple benefits to patients and facilities alike,” said Varun Babu, Senior Research Analyst, TechVision.

“For instance, they help with the delivery of targeted and personalized medicine while simultaneously ensuring seamless communication and high productivity within smart hospitals.”

The potential to improve efficiency, engage patients continuously, and get ahead of adverse events has created a significant commercial opportunity for device manufacturers, software vendors, and analytics developers, adds a separate report by MarketersMedia.

Currently, the global IoT market is valued at $20.59 billion, and is anticipated to grow at a 25.2 percent compound annual growth rate (CAGR) until 2023 to reach $63.43 billion.

The market includes implantable tools, such as cardiac devices, as well as internet-connected ventilators, imaging systems, vital signs monitors, respiratory devices, infusion pumps, and anesthesia machines, MarketersMedia says.

Frost & Sullivan also anticipates that emerging categories of IoT devices, including adhesive skin sensors, will contribute to the financial opportunity, while developing technologies, such as blockchain, will enhance the security, interoperability, and analytics potential of these tools.

In order to succeed, providers and developers will need to collaborate on creating and deploying data standards and shared protocols to ensure the seamless exchange of data across disparate systems.

“The main objective of IoMT is to eliminate unnecessary information within the medical system so that doctors can focus on diagnoses and treatment,” said Babu. 

“Since it is an emerging technology, technology developers need to offer standardized testing protocols so that they can convince hospitals of their safety and efficacy and make the most of their massive potential.”



Noel Nevshehir


If there was ever an industry in dire need of increased efficiency, cost containment and improved outcomes, health care tops the list. Despite consuming 18 percent of our nation’s GDP—equal to $3.4 trillion in annual expenditures—it is responsible for nearly 250,000 deaths due to medical errors, poor record keeping and a dismal lack of shared data among doctors about patients in their care.

From blockchain technology to surgical robots, medical experts worldwide agree that big data and artificial intelligence (AI) will play a key role in vastly improving health care quality and delivery. Aided by advances in sensor capabilities, computational power and algorithmic ingenuity, the pace of medical innovation is accelerating rapidly. 

To be sure, AI and big data are not the next best thing, they are here and now. Digital medicine is currently tracking down and destroying mutant cancer cells faster than ever before. It is also commonly used in operating rooms by doctors tapping into pools of data accumulated from previous surgeries to receive guidance from computers systems that have analyzed learned procedures that can be scaled up in order to make appropriate recommendations before, during and after treatment. So instead of depending on one or two local practitioners determining the course of lifesaving treatments, patients now have access to a knowledge base of thousands of doctors worldwide.

AI—versus natural intelligence used by humans to power up their brains—is akin to a jigsaw puzzle that deploys algorithms to draw together seemingly unrelated dots of information to paint a clear picture of the underlying data. It has changed dramatically since the concept was first introduced at Dartmouth College in 1956. Today’s man and machine AI is being aided by neural networks and deep machine learning methodologies powered by quantum computers and sophisticated algorithms that can crunch raw data into meaningful and actionable analyses.

A recent Wall Street Journal article titled “The Operating Room of the Future” is a case in point. Verb Surgical Inc., a recent startup formed by a partnership between Alphabet and Johnson & Johnson, is designing neural networks which enables robots to learn from one another by connecting each of them to the Internet to create machine-learning algorithms. Called “Surgery 4.0," it is the next logical step after traditional open procedures, minimally invasive surgery and the introduction of robotics. Using machine learning methodologies, computer programs study past procedures to identify best practices and potential errors. Verb’s technology has rendered the da Vinci surgical robot ancient by today’s standards despite the latter performing more than five million surgeries worldwide since 2000.

Another area ripe for AI is mental health. Researchers are developing new drugs and pharmaceutical combinations using machine learning to assess chemical reactions of anti-depressants among individual patients. They then tailor them to closely match an individual’s unique biochemical makeup. The results thus far are promising. In addition to detecting when a patient is veering off into a bipolar episode even better than a psychiatrist could ever imagine, these drugs are mitigating some of the wrenching side effects associated with traditional serotonin re-uptake inhibitors. Taken one step further, Stanford University has created chatbots to combat this debilitating disease. Patients feeling an aura can tell their chatbot how they are feeling that day. Using predictive analytics, the bot can quickly suggest coping strategies drawn from numerous cognitive behavioral therapies. Again, the results are impressive, reducing depressive symptoms by 20 percent.        

AI and big data can also predict patient falls resulting in head traumas, bone fractures and other injuries costing on average $30,000 per incident. Businessweek recently featured California-based Qventus, a company that developed a program to help nurses overcome alarm fatigue and sensory overload from the constant beeping sounds and alerts found in hospital environments. In many cases, this results in medical staff missing critical and life-threatening alarms altogether. Qventus’ software extracts and analyzes data to recognize patterns from call lights, bed alarms, electronic medical records, patients’ prescriptions and age and other fall indicators. In turn, this has reduced injuries by 13.5 percent.

Today there is no such thing as TMI when it comes to data capture now that we have the tools to make sense of it all. We are far from Star Trek’s tricorder ability to instantly detect what ails us, but we are moving in that direction. Even in its embryonic stage, AI outperforms dermatologists in spotting skin cancer, helps pharmacists predict more effective drug combinations, and spots nuances on x-rays far better than radiologists. 

Quantum computers uncovering newfound data have provided medical professionals with keen insights into disease mapping and prevention, rendered speedier diagnoses and treatments for patients, accelerated scientific discovery aimed at curing the leading causes of death in our country, and have also played a major role in predictive analyses and detection. 

Last month, Oakridge National Laboratory rolled out the world’s most powerful supercomputer, Summit, capable of 122.3 petaflops (or 200 quadrillion) calculations per second. Comparatively speaking, the human brain clocks in at 10-100 petaflops per second. However, computers do not yet match the human brain in areas like reasoning, perceiving and intuition. AI will never replace humans or lead us to the dreaded robopocclypse of lore. Considered by many as an idiot savant, AI is well versed in single, closely supervised tasks but out of its element performing wider, more complex calculations. Equally important, big data and AI are only as good as the data fed to it by their mere mortals (you and me) whom program neural networks and plug and play algorithms that run the risk of being inherently biased or, worse yet, a victim of groupthink.  

Yet, the opportunities that big data and AI present in vastly improving health care and the quality of life for ailing patients far outweigh the challenges. Together, man and machine are teaming up to exploit unprecedented amounts of medical information churned out by powerful computers and advances in integrated software technologies. According to Kevin Lasser, CEO of JEMS Telehealth, “we are at an inflection point now and will soon look back and realize that today was only the beginning of a major revolution in medicine.

Tina Reed


When physicians get the right kind of alert in an electronic health record—and actually follow its recommendation—it could result in fewer complications and lower costs among hospitalized patients, according to a new study.

Published in the American Journal of Managed Care, researchers from Cedars-Sinai Medical Center and Optum Advisory Services teamed up to examine alerts that popped up on physician computer screens inside their EHR system when their care instructions deviated from evidence-based guidelines.

Those alerts were based on the 'Choosing Wisely' initiative in which different specialties have identified common tests and procedures in their respective areas of expertise that may not benefit patients and should be avoided.

Looking at nearly 26,500 inpatient admissions at Cedars-Sinai Medical Center between October 2013 and July 2016, the researchers studied whether the treating physician followed either all or none of the Choosing Wisely guidance. In 6% of visits, physicians followed all of the triggered alerts and in 94% of visits, physicians followed none of the alerts. In particular, they examined data in which one more of the 18 most frequent alerts were triggered.

For patients whose physicians did not follow the alerts, the odds of complications increased by 29% and the risk of readmissions within 30 days of the patients' original visit was 14% higher. There was also a 6.2% increased length of stay and an additional 7.3% or more than $900 per patient increase in costs after adjusting in differences in patient complexity. 

To be clear, the study was examining a product created by Stanson Health, a spinout company founded by Cedars-Sinai executives and in which the health system is a major shareholder.

Optum is a licensed reseller of Stanson's Health's Choosing Wisely alert content and, for the study, married Stanson's data with its own analytics on cost and complications and length of stay for those same admissions.

“We said, ‘Hey, let’s look and see if we can demonstrate that we’re not only able to get docs to look at alerts but we can actually change the ultimate outcomes we’re trying to change like complications and readmissions," John Kontor, M.D., an executive vice president at Clinovations within Optum told FierceHealthcare in an interview. "For us, it was risk-free but it was very risky for Stanson."

It worked out by showing the association they were looking for, he said.

But there are limitations to the study, they acknowledge.

There was no way to measure the impact of specific alerts on outcomes to see if one was more significant than others. There is also no way to know if providers who were more likely to follow alerts may also be more engaged around quality and efficiency in how they practice day to day, Kontor said.

More study is needed to more closely examine the direct impact of the alerts, officials said. But, they said, the study shows using tools could have a real impact on costs and quality.

"The next step is to look at the characteristics of overall alerts and say 'We see a positive impact. Now let's look at what’s most effective about clinical decision support in order to maximize the positive impact on both for patients and providers," said Anne Wellington, a co-author of the study and managing director of the Cedars-Sinai accelerator.



Kate Monica 


The state’s emergency department care coordination program enables collaboration between hospitals via a statewide HIE.

Virginia recently launched its emergency department care coordination (EDCC) program to connect all emergency departments in the Commonwealth for streamlined communication and collaboration across healthcare facilities.

The program utilizes a connection to Virginia’s statewide health information exchange (HIE) —ConnectVirginia — to enable health data exchange between healthcare providers, health plans, and care teams for patients receiving emergency services.

Additionally, the program integrates directly into the state’s prescription drug monitoring program (PDMP) and advance healthcare directive registry.

Enabling a connection to the state’s PDMP will equip care teams with patients’ comprehensive medication histories to promote safer prescribing practices in Virginia’s emergency departments and reduce opioid-related deaths.

“Virginia continues to be at the forefront of health care innovation, and the ED Care Coordination Program marks an important step forward in making sure Virginians in every part of the Commonwealth have access to the highest quality of care,” said Virginia Governor Ralph Northam.

“With this secure technology, we can provide emergency medical personnel with access to a patient’s critical medical information in a timely way, which will increase effective and efficient care, avoid duplicative tests, reduce unnecessary costs, and improve health outcomes,” Northam continued.

The Virginia General Assembly first established the EDCC program in 2017 within the Virginia Department of Health (VDH). The program was made possible through collabroations between health systems, health plans, physicians, VDH, the Department of Medical Assistance Services, and the Department of Health Professions.

“The ED Care Coordination Program will help ensure appropriate care in the appropriate setting for patients, while also ensuring that their personal health information is secure,” said State Health Commissioner M. Norman Oliver, MD. 

The program is headed by VDH. Health IT services provider Collective Medical assists in facilitating health data exchange for ConnectVirginia to enable a connection between emergency departments.

“This program can offer peace of mind to patients and health care providers alike. I applaud the countless individuals who have worked collaboratively to make this program a reality, which helps safeguard the health and well-being of all people in Virginia and moves the state closer to becoming the healthiest state in the nation,” Oliver stated.

Looking ahead, the State Employee Health Plan and non-ERISA commercial and Medicare health plans operating within the Commonwealth intend to join the EDCC program by June 30, 2019.

The EDCC program will also expand to include other providers including primary care physicians, case managers, nursing homes, community service boards, private behavioral health providers, and Federally Qualified Health Centers (FQHCs).

These entities will have the opportunity to receive alerts and contribute patient health information to the exchange for improved care coordination.

States across the country are increasingly making efforts to improve health data access and exchange for emergency services providers for improved care coordination.

In May, the North Dakota Department of Health’s (NDHS) Division of Emergency Medical Systems launched an initiative to allow emergency medical services (EMS) personnel to engage in EHR use during patient transport.

The state health department partnered with EMS, fire department, and hospital health IT software company ESO to develop a clinical data repository designed to collect and analyze EMS patient care reports.

Enabling providers to collect and analyze patient health data helps to improve patient injury prevention efforts, performance improvement, and patient health outcomes.

 “Smart data and insightful analytics can be a real game changer for organizations across the healthcare spectrum when it comes to patient transport and treatment,” said ESO Healthcare Vice President Allen Johnson.

North Dakota joins California, Colorado, Indiana, Oklahoma, New York, and other states that have taken steps to improve EHR use and health data access for emergency services organizations.

Kate Monica 


Using EHR data, researchers were able to predict short-term mortality for chemotherapy patients through machine learning.

Combining the power of EHR data with machine learning may be the key to more accurately predicting mortality among cancer patients undergoing chemotherapy.

This finding comes from a recent JAMA study by Elfiky et al. that explored the effectiveness of applying machine learning to EHR data to predict patient’s short term risk of death when they start chemotherapy.

While chemotherapy significantly lowers the risk of recurrence in early-stage cancers and can improve survival rates and symptoms in later-stage disease, the treatment is challenging and costly for patients.

“These patients experience burdensome symptoms without many of the potential benefits of chemotherapy,” wrote researchers in the report.

Researchers set out to find a way to more accurately predict mortality risk before administering chemotherapy treatment to ensure patients that undergo the stress and burden of treatment will also reap its benefits.

In the cohort study, researchers analyzed the EHR data of 26,946 patients starting 51,774 chemotherapy regimens at Dana Farber/Brigham and Women’s Cancer Center from January 1, 2004 to December 31, 2014. Researchers identified the date of death for patients by linking their health records to their Social Security data.

The team classified patients by primary cancer and presence of distant-stage disease using registry data codes for metastases. With this information, researchers attempted to accurately predict death within 30 days of starting chemotherapy with a machine learning model based on single-center EHR data.

Ultimately, the machine learning model was able to accurately predict mortality rates despite lacking genetic sequencing data, cancer-specific biomarkers, or any detailed information beyond EHR data. Specifically, patient EHR data used in the machine learning model including symptoms, comorbidities, prescribed medications, and diagnostic tests.

“Mortality estimates were accurate for chemotherapy regimens with palliative and curative intent, for patients with early- and distant-stage cancer, and for patients treated with clinical trial regimens introduced in years after the model was trained,” stated researchers.

Researchers emphasized that EHR data contains “surprising amounts of signal for predicting key outcomes in patients with cancer.”

In addition to proving accurate, the machine learning model developed by researchers would also only minimally increase administrative burden on clinicians. The machine learning algorithm would not require manual data input from clinicians.

Instead, the algorithm could pull directly from existing patient EHRs.

“Although our algorithm was developed using a single institution’s data, its inputs are available nearly everywhere with an EHR,” wrote researchers.

“In addition, no special infrastructure is required to pull these data from an institution’s data warehouse; in the same way that today’s EHR systems pull a rich set of data from a database to present it to clinicians, an algorithm could pull and process the same data in real time using the processing power on a desktop computer,” the team continued.

The team also suggested the machine learning algorithm could potentially be designed to support EHR integration. Healthcare organizations could integrate the algorithm directly into existing health IT systems. 

“Algorithmic predictions such as ours could be useful at several points along the care continuum,” wrote researchers. “They could provide accurate predictions of mortality risk to a clinician or foster shared decision making between the patient and clinician.”

By predicting short-term mortality for cancer patients, clinicians can identify patients who are unlikely to benefit from chemotherapy and instead may be better suited for early palliative care referral and advance care planning.

“For patients receiving systemic chemotherapy, an estimate of 30-day mortality risk may be a useful quality indicator of avoidable treatment-associated harm,” researchers concluded.

Leveraging EHR data to predict patient health outcomes may help providers to avoid clinical decisions that add unnecessary strain on patients for minimal benefit.

Kate Monica 


ECRI recently outlined three ways healthcare organizations can improve health IT use to reduce delayed, missed, and incorrect diagnoses.

Improving communication between providers, diagnostic testing and medication tracking, and documentation through health IT use can help to reduce delayed, missed, and incorrect diagnoses.

This set of safe health IT use recommendations was released as part of ECRI’s Partnership for Health IT Patient Safety collaborative, which was established in 2014. The multi-stakeholder partnership is open to participation from providers, health IT companies, professional organizations, and other industry insiders.

The partnership’s most recent workgroup focused on improving patient safety during diagnostic testing, test tracking, and medication changes. Chaired by Vanderbilt University professor of pediatrics and biomedical informatics professor Christoph Lehmann, MD, the group identified three strategies healthcare organizations should keep in mind when implementing health IT solutions.

In combination, these three strategies can be effective in closing the loop of patient diagnoses.

“The goal of this Partnership workgroup was to look for technology solutions that all stakeholders could implement to close the loop — the tools provided here will help to do just that,” said ECRI Institute Program Director Lorraine Possanza.

First, the workgroup emphasized the importance of developing and applying health IT solutions that will communicate the necessary information to the correct members of a patient’s care team at the right time.

“Improve the transmission of information using standards for the formatting of normal, critical, abnormal-noncritical, and abnormal results,” wrote members of the workgroup in the report.

Efficient and effective communication between testing facilities, pharmacies, providers, and patients can enhance patient care across care settings, ECRI suggested.

“Designing, testing, deploying, and implementing health IT solutions to improve these communication pathways has the potential to make closing the loop a seamless and elegant process, with all diagnostic results and medications communicated to the provider, the pharmacy, and the patient,” wrote the workgroup.

ECRI also emphasized the importance of tracking diagnostic results and medication changes.

“Tracking of diagnostic results and medication changes is a time-consuming, burdensome task, but necessary to ensure a closed loop,” noted report authors. “Identification of interruptions and potential failure points in the process is critical to find and react to failures to close the loop.”

The workgroup recommended healthcare organizations identify where health IT can be useful for resolving deficiencies and improving medication and test result tracking.

Using EHR functionality to track diagnostic testing and medication changes may also be helpful. Furthermore, implementing lab standards — such as LOINC — may help to automate accurate matching of results and ordered tests to close loops.

Finally, ECRI suggested healthcare organizations use health IT to link, acknowledge, and document the review of information and action taken.

“This step includes the actor reviewing and acknowledging or acting upon information,” clarified report authors.

Toward this end, the workgroup recommended healthcare organizations help to improve interoperability by integrating EHR systems and other health IT modules across the care continuum to facilitate communication and documentation across care settings.

“This is to facilitate communication and acknowledgment, including the use of application programming interfaces (APIs) to allow laboratory systems and hospitals to communicate, as well as the use of HL7 and fast healthcare interoperability resources (FHIR) to aggregate and merge patient data from separate data sources,” wrote the workgroup.

Developing EHR functionality such as diagnostic results notifications may help to promote communication, acknowledgement, and documentation of diagnostic testing, medication changes, and actions taken.

Implementing these health IT use practices and closing the loop will help to avoid delayed or missed diagnoses, which may lead to patient harm.

“These safe practice recommendations are a call to action,” maintained the workgroup. “Although the EHR and its technology components have the potential to facilitate timely follow-up across all healthcare settings, it may take regulatory efforts to make this possible.”