Healthcare Analytics’ New World | Healthcare Informatics Magazine | Health IT | Information Technology Skip to content Skip to navigation

Healthcare Analytics’ New World

November 10, 2016
by Rajiv Leventhal
| Reprints
Recent federal mandates and developments around bundled payments, readmissions reduction and accountable care have again confirmed that data analytics and information technology will be crucial to healthcare’s value-based transformation

In back-to-back months this summer, announcements around new mandatory bundled payment programs from the Department of Health and Human Services (HHS) as well as the latest updates regarding Centers for Medicare & Medicaid Services (CMS) penalties on hospitals for failing to lower their rehospitalization rates, collectively signaled to healthcare leaders that payment reform is here to stay.

The July 25 announcement of the mandatory bundled payment program for heart attack care and for cardiac bypass surgery stated, “The hospital in which a Medicare patient is admitted for care for a heart attack or bypass surgery would be accountable for the cost and quality of care provided to Medicare fee-for-service beneficiaries during the inpatient stay and for 90 days after discharge. The proposed cardiac care policies would be phased in over a period of five years, but would begin July 1, 2017 for hospitals located in the 98 metro areas participating in the model (about one-quarter of all metro areas in the nation).” These new bundled payment models for cardiac care, in addition to the extension of the existing bundled payment model for hip replacements and other hip surgeries, are yet another major step in forcing reimbursement forward into value-based purchasing.

Meanwhile, on the hospital readmissions front, although the news didn’t come out of CMS directly, an August 2 Kaiser Health News report revealed that the federal government’s penalties on hospitals for failing to lower their rehospitalization rates will hit a new high as Medicare will withhold approximately $528 million—about $108 million more than last year. CMS will penalize more than half of the nation’s hospitals—a total of 2,597—for having more patients than expected return within a month, as mandated by the government’s Hospital Readmissions Reduction Program, which adjusts payments for hospitals with higher than expected 30-day readmission rates for six targeted clinical conditions.

These revelations point to a realization beyond payment reform that patient care leaders likely already knew, but is now confirmed: U.S. hospitals are under more pressure than ever before to produce optimal clinical and cost outcomes. Key to this transformation will be leveraging robust data analytics and information technology to help drive continuous performance improvement.

Payers and Providers Converge


Experience New Records for Speed & Scale: High Performance Genomics & Imaging

Through real use cases and live demo, Frank Lee, PhD, Global Industry Leader for Healthcare & Life Sciences, will illustrate the architecture and solution for high performance data and AI...

A critical element to providers planning for a value-based care future is aligning their needs and goals with those of payers. While this hasn’t always been easy to accomplish, most of the sources interviewed for this story agree that real strides are being made. Tim Moore, M.D., executive vice president of health affairs and chief medical officer of technology provider AxisPoint Health, a Westminster, Colo.-based spinoff of McKesson, which works primarily with payers, says there are plenty of new opportunities emerging around getting payers and providers on the same side of the table to sort out risk-based contracting challenges.

“With better integration and better relationships between payers and providers, through value-based reimbursement, there should be much better use of clinical data that is more timely and can provide interventions that are more appropriate to drive opportunities for savings,” Moore, previously chief medical officer at WebMD Health Services, says. Historically, he notes, payers would be straddled with only 60-day or 90-day-old claims-based administrative data, and by the time they did something with that, 30 more days would pass. “So there was a limit from a time perspective and also an accuracy perspective,” Moore says. He adds, “Providers have more timely claims be it through the electronic health record [EHR] or through the hospital with admission/discharge/transfer [ADT] information. If you have that timely information and you can leverage it, you can much better leverage algorithms and analytics to help predict who needs better support and guidance, from their own real data rather than administrative claims data that’s 90 days old.”

Tim Moore, M.D.

The thing that payers can bring to the table that providers sometimes cannot, continues Moore, is a higher level view of the population that the providers are delivering service to. “Providers sometimes don’t get a good view of the whole population they are serving, as they are only serving one patient at a time. But payers see a longitudinal view of patients over the past year or two,” he says.

Moore gives an example of how some hospitals throughout the country leverage health information exchanges (HIEs) that have good ADT data that hasn’t been shared or used by industry players such as the payer market. “With this ADT data, you can pull out other information including how many ER visits someone has had in the past six months, his or her diagnosis, and when he or she was in the hospital, so you have timely information that says here is a patient that has been in the hospital and because of this condition they have a higher risk of a readmission,” he says.

At the same time, payers can help by looking across different hospitals and pick out which ones are outliers in terms of high readmission rates. “Some hospitals are good at [avoidable readmissions], so you need to put resources towards the ones that are outliers,” Moore says. “Providers don’t have that full view like payers do. I think that leveraging the two sides can open up a whole new way of taking the data, and putting together and focusing the resources on where it will be most impactful,” he says.

To this end, Independence Blue Cross Blue Shield in southeastern Pennsylvania, serving two million members in five counties in and around Philadelphia, uses a predictive tool that calculates an individual’s likely future health state based on associated clinical conditions or diagnoses. The risk matrix, from San Mateo, Calif.-based healthcare analytics company Lumiata, helps the payer identify where members might be at risk for or might have certain conditions, and then helps alert their providers, explains Michael Vennera, senior vice president and CIO at Independence Blue Cross Blue Shield.

Vennera says that with the analytics tool, the payer can go to providers in its market and say that there is a chance patient X has a certain condition, even though it’s not diagnosed on his or her claims. All different types of data goes into that risk engine, says Vennera—medical claims data, prescription drug claims, lab results, and also basic demographics such as age, gender, and location. “Then what we get out of it is a prediction around diseases with different confidence levels for different members. And then you can use that to follow up,” he says.

Michael Vennera

Being right in the thick of the payer-provider relationship, Vennera notes how there are now lots of opportunities to combine the depth of information that a payer has with the depth of information a provider has, and put analytics on top of it. “What you typically see in most markets is that providers will have deep information, such as services rendered in an EHR,” he says.  “So if you go to a hospital, all the information about that stay and the procedures done make for a rich and deep data set. But then the challenge is when people go to multiple providers. We know interoperability is long way off. But payers do have a broad set of data which covers most of your healthcare since most of healthcare flows back to your insurers,” he says.

Rose Higgins, president of the West Hartford, Conn.-based analytics solution company SCIO Health Analytics, additionally notes the challenge of payers and providers being as transparent with the data as possible. Higgins says that while it begins with recognizing that the data has intrinsic value to both sides, there has to be a willingness to be open with respect to the information, and share it, for opportunities to be identified and acted on. “It’s challenging to mix different types of payer data sets. Providers have multiple contracts, so there are different approaches with each payer. This means that a payer may not want to share data with a provider when they know another payer’s data may be mingled with their own,” Higgins explains.

Drilling Down with Policy Implications

When CMS’ Bundled Payments for Care Improvement (BPCI) initiative first got off the ground five years ago, there were high expectations for investments in technology—to track performance on bundles, to make more predictions on performance, and to potentially price commercial bundles, notes Matthew Cinque, executive director, product management at The Advisory Board, a Washington, D.C.-based consulting and technology company. “But at that time, the market did not move as quickly as was expected on the analytics side,” Cinque says. “Folks signed up for the bundled payment programs, so there were lots of conversations around bundles in general, but when push came to shove, there was not a lot of movement.”

Cinque explains that one of the reasons for this was the upside in the CMS program was not big enough to justify freestanding investments in new analytics. “Folks would get by with what they had in Excel. On the commercial side, we saw a lot of interest but there was hesitance on the part of payers to try to adjudicate bundled payments,” he says, adding that with a commercial population under the age of 65, the numbers showed that there was not a lot of volume of any one thing.” Even with the [Comprehensive Care for Joint Replacement Model] announced last year, there has not been “a huge move around analytics investment,” Cinque says, noting that he expects that to “get more serious sooner than later.” He says, “I would say it is an immature market, but one that I expect to have more dedicated focus on bundled payment-specific analytics as CMS rolls out more mandatory programs related to this.”

Matthew Cinque

Cinque adds that “getting more serious” involves an investment in integrating different data sources. “One thing that makes bundled payments so challenging, especially if you look at the cardiac care model, is that so much of what you’re trying to manage happens outside of the four walls of the hospital. You need to be able to get data across inpatient metrics and get visibility into what happens in the physician office and skilled nursing facilities. Those are almost always entirely different data sets,” he says. Thus, data aggregation has to be a big point of investment, be it through a data warehouse or something else, he notes. “It’s about accumulating that data and then manipulating it. The data aggregation component of it is really what makes it cost prohibitive today,” he says.

Dan Golder, principal at Naperville, Ill.-based healthcare consulting firm Impact Advisors, agrees that the data integration piece could be the toughest. Golder says there are three levels when looking at value-based purchasing: claims data, clinical data, and eligibility data, and they happen to live in three siloes. Most groups, from what Golder has seen, have been working with claims data since it’s most available and “although it’s not easy to integrate it, you can,” he says. But the other two siloes are extremely problematic, he adds. “Linking claims data to other claims data from other payers, as well as clinical data and eligibility data to get it to be actionable at the point of care is an issue. Third-party tools are doing this right now, meaning providers are accessing a second application and leaving the EHR if they want to look at aggregate data and population health data,” Golder says.

Meanwhile, on the readmissions front, SCIO Health Analytics’ Higgins says that the organizations that have tackled this challenge head on are predicting where the outliers are, how to identify those earlier, and then work with those providers and patients to reduce these trends around readmissions. “The penalties are real and meaningful, so the approach for most organizations we are talking to is figuring out how to look at the providers who aren’t as strong in the primary care practice relationships that need to be in place with these patients to make sure there is a good plan of care prior to and after admission,” Higgins says. 

AxisPoint Health's Moore adds that payers are becoming increasingly frustrated since they have put programs in place that intuitively, and in an academic research setting, prove that they have value in terms of outcomes for lowering readmission rates, but when these programs are moved into a community-based setting, the community doesn’t act like an academic setting. “So many clients are frustrated that they have put programs in place that don’t work,” Moore says. We need more community-based studies and interventions that can be leveraged across the U.S.,” he says. Moore further points to the “LACE” index—a tool that identifies patients that are at risk for readmission or death—which is used by many hospitals and academic centers. “But unless you have those data points, like the ADT data, you can’t really do anything of significance,” he notes.

One area around readmissions that Cinque has seen an increased utilization of analytics for is psychosocial factors and financial barriers to care. “These readmission rates are higher in lower income areas than elsewhere,” he says. “Organizations are beginning to incorporate either a low-fi method, so collecting information while patients are in the hospital, or through more progressive methods, such as data mining to try to flag patients for readmissions. The integration of those data types in trying to become more predictive about risk factors for readmissions is an emerging area,” Cinque says.

Nonetheless, for some patient care organizations, notably smaller provider groups, incorporating this level of analytics can prove too expensive and overwhelming. To this end, Scott Pillittere, vice president of Impact Advisors, says that these smaller groups will be willing to “take the hit” from CMS regarding penalties for these policy mandates, since they won’t be able to pay for the data analytics that are needed. “We are seeing more consolidation in the marketplace, and this will be another factor that will push standalone or small physician practices into a much larger organization so they have the financing to pay for the data analytics groups that can help them with this part of their care,” Pillittere predicts. “There are not a whole lot of doctors who want to play in this business side of healthcare, so they are looking for help,” he says.

Golder adds that when he reads the tea leaves of Medicare’s new rules with how much payment they want tied to value in the coming years, much of it is budget neutral, meaning for someone to earn an incentive payment, someone else will have to pay a penalty. This represents a difference from the Meaningful Use program in which everyone could earn incentives. “So the inability to pay for systems and the lack of capability to run analytics to do better, will likely shift small practices into larger groups that can be successful in the world of population health and accountable care,” Golder says. 

Nevertheless, the shifting healthcare landscape isn’t stopping senior leaders at SCL Health from getting in front of the analytics game. The patient care organization, a nine-hospital health system with three safety-net clinics, one children’s mental health center, and approximately 200 ambulatory sites in three states—Colorado, Kansas, and Montana —last year selected Fort Collins, Col.-based Total Benchmark Solution (TBS) as its vendor for benchmark data and advanced analytics. The platform enabled the health system to quickly and easily compare performance using historical trends, and/or performance targets, and peer group data. It was then able to identify areas of undesirable variation to target for improvement, its officials say.

“The platform allows us to filter and adjust an analysis based on various criteria, such as a certain type of patient or a particular payer,” says Chris Bliersbach, senior director of clinical outcomes at SCL Health. We can integrate data sources, such as our ADT feed, Epic, and Press Ganey to see the whole picture through volume, cost, charges, supplies, quality, patient experience, and many other metrics.”

Prior to this technology implementation, two SCL Health care sites and a commercial payer already had been particularly interested in hip and knee surgery improvement within the Comprehensive Care for Joint Replacement bundled payment model. Both care sites had desirable performance with length of stay (LOS) as measured against Medicare and all-payer benchmarks in the TBS database, its officials attest.

But they realized that customized benchmarks would provide a stretch goal appropriate to best-practice hip and knee surgery outcomes at the care sites. To develop the benchmarks, SCL Health and TBS collected and analyzed data from Healthgrades on organizations that had five-star ratings for hip and knee surgery. Importantly, organizations that could offer stretch goals for the care sites also required appearance on the U.S. News and World Report “Best Hospitals” list, and a similar patient volume to the care sites. Indeed, 80 hospitals providing knee surgery and 56 providing hip surgery met the criteria for “best practice” organizations with both low LOS and low complication rates. Data from those organizations were used to establish the tailor-made benchmarks, SCL Health officials say.

Advice for the C-Suite

All of the healthcare leaders interviewed for this story agree with the notion that with all of the initiatives that federal healthcare officials are creating now—around readmissions reduction, value-based purchasing for both hospitals and physicians, bundled payments, and accountable care—the leveraging of data and healthcare IT will be critically important.

So what is the best plan of action for CIOs and CMIOs right now around leveraging robust data analytics to bend the healthcare cost curve? Well, there isn’t one single answer for organizations nationwide, says Impact Advisors’ Golder, who notes several options for health systems such as: aggregating data by building data warehouses; integrating data sources themselves; looking at their existing vendor partners to help them since their doctors don’t want to leave the EHR; and finding third-party vendors for data aggregation. “So, for the provider organization, what’s your appetite for risk?” Golder asks.

Independence Blue Cross Blue Shield’s Vennera says that payers should be talking to the providers they work with in their market, if they aren’t already, about how they can share data, particularly if there is no regional data exchange program or HIE in place. And on the analytics specific side, he adds, “The big thing for CIOs is to strike the right balance between insourcing and outsourcing. With the analytics arms race and the cost of analytics resources, you can’t build everything in-house. But also you can’t bend your way to the answers. You need a combination of vendor solutions and building the in-house talent to interpret results, challenge findings, and think through and develop proprietary analytics.”

Moore agrees with this advice. He says: “Put stakeholders in a room together and have them each bring the data they believe is most important and share it with each other, so they could understand the data that exists rather than create something new.” Moore believes that there is a tendency in healthcare to try to create something new and constantly look at something differently. “Right now, we have many different data points that are not necessarily used as well as they could be and integrated as well as they could be. Start with the data you have and figure out how to best leverage it,” he says.

The Advisory Board’s Cinque further brings up the point that when an organization begins collecting information and data, and synthesizes it across different sites of care, it forces interactions with other EHR systems, or in some cases, places that don’t even have EHRs. “You need to understand the IT landscape of your partners and other providers in your community. That will influence your success on these programs,” he says. “For CIOs and CMIOs, there is a belief in if they are not on our EHR platform, we shouldn’t work with them or it shouldn’t matter. That’s not a tenable strategy with these interconnected programs.”

On the risk-based side, Moore also notes the fact that there is still such a mix of fee-for-service and value-based reimbursement, which ultimately slows things down. He calls it “a schizophrenic way for a provider to try to practice.” He says that provider organizations need to figure out how to segregate, meaning having a fee-for-service group and value-based group that is at least 75 percent reimbursed from one side or the other. “Until the majority of a doctor’s compensation is tied to one side, they won’t behave in that certain way. That’s one of the biggest challenges for us in the next five years,” he says.

The Health IT Summits gather 250+ healthcare leaders in cities across the U.S. to present important new insights, collaborate on ideas, and to have a little fun - Find a Summit Near You!


Have CIOs’ Top Priorities for 2018 Become a Reality?

December 12, 2018
by Rajiv Leventhal, Managing Editor
| Reprints

In comparing healthcare CIOs’ priorities at the end of 2017 to this current moment, new analysis has found that core clinical IT goals have shifted from focusing on EHR (electronic health record) integration to data analytics.

In December 2017, hospitals CIOs said they planned to mostly focus on EHR integration and mobile adoption and physician buy-in, according to a survey then-conducted by Springfield, Va.-based Spok, a clinical communications solutions company, of College of Healthcare Information Management Executives (CHIME) member CIOs.

The survey from one year ago found that across hospitals, 40 percent of CIO respondents said deploying an enterprise analytics platform is a top priority in 2018. Seventy-one percent of respondents cited integrating with the EHR is a top priority, and 62 percent said physician adoption and buy-in for securing messaging was a top priority in the next 18 months. What’s more, 38 percent said optimizing EHR integration with other hospital systems with a key focus for 2018.

Spok researchers were curious whether their predictions became reality, so they analyzed several industry reports and asked a handful of CIOs to recap their experiences from 2018. The most up-to-date responses revealed that compared to last year when just 40 percent of CIOs said they were deploying an enterprise analytics platform in 2018, harnessing data analytics looks to be a huge priority in 2019: 100 percent of the CIOs reported this as top of mind.

Further comparisons on 2018 predictions to realities included:

  • 62 percent of CIOs predicted 2018 as the year of EHR integration; 75 percent reported they are now integrating patient monitoring data
  • 79 percent said they were selecting and deploying technology primarily for secure messaging; now, 90 percent of hospitals have adopted mobile technology and report that it’s helping improve patient safety and outcomes
  • 54 percent said the top secure messaging challenge was adoption/buy in; now, 51 percent said they now involve clinicians in mobile policy and adoption

What’s more, regarding future predictions, 87 percent of CIOs said they expect to increase spending on cybersecurity in 2019, and in three years from now, 60 percent of respondents expect data to be stored in a hybrid/private cloud.

CIOs also expressed concern regarding big tech companies such as Apple, Amazon and Google disrupting the healthcare market; 70 percent said they were somewhat concerned.

More From Healthcare Informatics


How One Community Hospital is Leveraging AI to Bolster Its Care Pathways Process

December 6, 2018
by Heather Landi, Associate Editor
| Reprints
Click To View Gallery

Managing clinical variation continues to be a significant challenge facing most hospitals and health systems today as unwarranted clinical variation often results in higher costs without improvements to patient experience or outcomes.

Like many other hospitals and health systems, Flagler Hospital, a 335-bed community hospital in St. Augustine, Florida, had a board-level mandate to address its unwarranted clinical variation with the goal of improving outcomes and lowering costs, says Michael Sanders, M.D., Flagler Hospital’s chief medical information officer (CMIO).

“Every hospital has been struggling with this for decades, managing clinical variation,” he says, noting that traditional methods of addressing clinical variation management have been inefficient, as developing care pathways, which involves identifying best practices for high-cost procedures, often takes up to six months or even years to develop and implement. “By the time you finish, it’s out of date,” Sanders says. “There wasn’t a good way of doing this, other than picking your spots periodically, doing analysis and trying to make sense of the data.”

What’s more, available analytics software is incapable of correlating all the variables within the clinical, billing, analytics and electronic health record (EHR) databases, he notes.

Another limitation is that care pathways are vulnerable to the biases of the clinicians involved, Sanders says. “In medicine, what we typically do is we’ll have an idea of what we want to study, design a protocol, and then run the trial and collect the data that we think is important and then we try to disprove or prove our hypothesis,” he says.


Experience New Records for Speed & Scale: High Performance Genomics & Imaging

Through real use cases and live demo, Frank Lee, PhD, Global Industry Leader for Healthcare & Life Sciences, will illustrate the architecture and solution for high performance data and AI...

Sanders says he was intrigued by advances in machine learning tools and artificial intelligence (AI) platforms capable of applying advanced analytics to identify hidden patterns in data.

Working with Palo Alto, Calif.-based machine intelligence software company Ayasdi, Flagler Hospital initiated a pilot project to use Ayasdi’s clinical variation management application to develop care pathways for both acute and non-acute conditions and then measure adherence to those pathways.

Michael Sanders, M.D.

Flagler targeted their treatment protocols for pneumonia as an initial care process model. “We kicked around the idea of doing sepsis first, because it’s a huge problem throughout the country. We decided to use pneumonia first to get our feet wet and figure out how to use the tool correctly,” he says.

The AI tools from Ayasdi revealed new, improved care pathways for pneumonia after analyzing thousands of patient records from the hospital and identifying the commonalities between those with the best outcomes. The application uses unsupervised machine learning and supervised prediction to optimally align the sequence and timing of care with the goal of optimizing for patient outcomes, cost, readmissions, mortality rate, provider adherence, and other variables.

The hospital quickly implemented the new pneumonia pathway by changing the order set in its Allscripts EHR system. As a result, for the pneumonia care path, Flagler Hospital saved $1,350 per patient and reduced the length of stay (LOS) for these patients by two days, on average. What’s more, the hospital reduced readmission by 7 times—the readmission rate dropped from 2.9 percent to 0.4 percent, hospital officials report. The initial work saved nearly $850,000 in unnecessary costs—the costs were trimmed by eliminating labs, X-rays and other processes that did not add value or resulted in a reduction in the lengths of stay or readmissions.

“Those results are pretty amazing,” Sanders says. “It’s taking our data and showing us what we need to pursue. That’s powerful.”

With the success of the pneumonia care pathway, Flagler Hospital leaders also deployed a new sepsis pathway. The hospital has expanded its plans for using Ayasdi to develop new care pathways, from the original plan of tackling 12 conditions over three years, to now tackling one condition per month. Future plans are to tackle heart failure, total hip replacement, chronic obstructive pulmonary disease (COPD), coronary artery bypass grafting (CABG), hysterectomy and diabetes, among other conditions. Flagler Hospital expects to save at least $20 million from this program in the next three years, according to officials.

Finding the “Goldilocks” group

Strong collaboration between IT and physician teams has been a critical factor in deploying the AI tool and to continue to successfully implement new care pathways, Sanders notes.

The effort to create the first pathway began with the IT staff writing structured query language (SQL) code to extract the necessary data from the hospital’s Allscripts EHR, enterprise data warehouse, surgical, financial and corporate performance systems. This data was brought into the clinical variation management application using the FHIR (Fast Healthcare Interoperability Resources) standard.

“That was a major effort, but some of us had been data scientists before we were physicians, and so we parameterized all these calls. The first pneumonia care path was completed in about nine weeks. We’ve turned around and did a second care path, for sepsis, which is much harder, and we’ve done that in two weeks. We’ve finished sepsis and have moved on to total hip and total knee replacements. We have about 18 or 19 care paths that we’re going to be doing over the next 18 months,” he says.

After being fed data of past pneumonia treatments, the software automatically created cohorts of patients who had similar outcomes accompanied by the treatments they received at particular times and in what sequence. The program also calculated the direct variable costs, average lengths of stay, readmission and mortality rates for each of those cohorts, along with the statistical significance of its conclusions. Each group had different comorbidities, such as diabetes, COPD and heart failure, which was factored into the application's calculations. At the push of a button, the application created a care path based on the treatment given to the patients in each cohort.

The findings were then reviewed with the physician IT group, or what Sanders calls the PIT crew, to select what they refer to as the “Goldilocks” cohort. “This is a group of patients that had the combination of low cost, short length of stay, low readmissions and almost zero mortality rate. We then can publish the care path and then monitor adherence to that care path across our physicians,” Sanders says.

The AI application uncovered relationships and patterns that physicians either would not have identified or would have taken much longer to identify, Sanders says. For instance, the analysis revealed that for patients with pneumonia and COPD, beginning nebulizer treatments early in their hospital stays improved outcomes tremendously, hospital leaders report.

The optimal events, sequence, and timing of care were presented to the physician team using an intuitive interface that allowed them to understand exactly why each step, and the timing of the action, was recommended. Upon approval, the team operationalized the new care path by revising the emergency-department and inpatient order sets in the hospital EHR.

Sanders says having the data generated by the AI software is critical to getting physicians on board with the project. “When we deployed the tool for the pneumonia care pathway, our physicians were saying, ‘Oh no, not another tool’,” Sanders says. “I brought in a PIT Crew (physician IT crew) and we went through our data with them. I had physicians in the group going through the analysis and they saw that the data was real. We went into the EMR to make sure the data was in fact valid, and after they realized that, then they began to look at the outcomes, the length of stay, the drop in readmissions and how the costs dropped, and they were on board right away.”

The majority of Flagler physicians are adhering to the new care path, according to reports generated by the AI software's adherence application. The care paths effectively sourced the best practices from the hospital’s best doctors using the hospital’s own patient groups, and that is key, Sanders notes.

“When we had conversations with physicians about the data, some would say, ‘My patient is sicker than yours,’ or ‘I have a different patient population.’ However, we can drill down to the physician’s patients and show the physician where things are. It’s not based on an ivory tower analysis, it’s based on our own data. And, yes, our patients, and our community, are unique—a little older than most, and we have a lot of Europeans here visiting. We have some challenges, but this tool is taking our data and showing us what we need to pursue. That’s pretty powerful.”

He adds, “It’s been amazing to see physicians rally around this. We just never had the tool before that could do this.”

While Flagler Hospital is a small community hospital with fewer resources than academic medical centers or larger health systems—for example, the hospital doesn’t have a dedicated data scientist but rather uses its in-house informatics staff for this project—the hospital is progressive in its use of advanced analytics, according to Sanders.

“We’ve been able to do a lot of querying ourselves, and we have some sepsis predictive models that we’ve created and put into place. We do a lot of real-time monitoring for sepsis and central line-associated bloodstream infections,” he says. “Central line-associated bloodstream infections are a bane for all hospitals. In the past year and a half, since we’ve put in our predictive model, we’ve had zero bloodstream infections, and that’s just unheard of.”

Sanders and his team plan to continue to use the AI tool to analyze new data and adjust the care paths according to new discoveries. As the algorithms find more effective and efficient ways to deliver care that result in better outcomes, Flagler will continue to improve its care paths and measure the adherence of its providers.

There continues to be growing interest, and also some hype, around AI tools, but Sanders notes that AI and machine learning are simply another tool. “Historically, what we’ve done is that we had an idea of what we wanted to do, conducted a clinical trial and then proved or disproved the hypothesis, based on the data that we collected. We have a tool with AI which can basically show us relationships that we didn’t know even existed and answer questions that we didn’t know to ask. I think it’s going to open up a tremendous pathway in medicine for us to both reduce cost, improve care and really take better care of our patients,” he says, adding, “When you can say that to physicians, they are on board. They respond to the data.”


Related Insights For: Analytics


At RSNA 2018, An Intense Focus on Artificial Intelligence

November 29, 2018
by Mark Hagland, Editor-in-Chief
| Reprints
Artificial intelligence solutions—and discussions—were everywhere at RSNA 2018 this week

Artificial intelligence solutions—and certainly, the promotion of such solutions—were everywhere this year at the RSNA Conference, held this week at Chicago’s vast McCormick Place, where nearly 49,000 attendees attended clinical education sessions, viewed nearly 700 vendor exhibits. And AI and machine learning promotions, and discussions were everywhere.

Scanning the exhibit floor on Monday, Glenn Galloway, CIO of the Center for Diagnostic Imaging, an ambulatory imaging center in the Minneapolis suburb of St. Louis Park, Minn., noted that “There’s a lot of focus on AI this year. We’re still trying to figure out exactly what it is; I think a lot of people are doing the same, with AI.” In terms of whether what’s being pitched is authentic solutions, vaporware, or something in between, Galloway said, “I think it’s all that. I think there will be some solutions that live and survive. There are some interesting concepts of how to deliver it. We’ve been talking to a few folks. But the successful solutions are going to be very focused; not just AI for a lung, but for a lung and some very specific diagnoses, for example.” And what will be most useful? According to Galloway, “Two things: AI for the workflow and the quality. And there’ll be some interesting things for what it will do for the quality and the workflow.”

“Certainly, this is another year where machine learning is absolutely dominating the conversation,” said James Whitfill, M.D., CMO at Innovation Care Partners in Scottsdale, Ariz., on Monday. “In radiology, we continue to be aware of how the hype of machine learning is giving way to the reality; that it’s not a wholesale replacement of physicians. There have already been tremendous advances in, for example, interpreting chest x-rays; some of the work that Stanford’s done. They’ve got algorithms that can diagnose 15 different pathological findings. So there is true material advancement taking place.”

Meanwhile, Dr. Whitfill said, “At the same time, people are realizing that coming up with the algorithm is one piece, but that there are surprising complications. So you develop an algorithm on Siemens equipment, but when you to Fuji, the algorithm fails—it no longer reliably identifies pathology, because it turns out you have to train the algorithm not just on examples form just one manufacturer, but form lots of manufacturers. We continue to find that these algorithms are not as consistent as identifying yourself on Facebook, for example. It’s turning out that radiology is way more complex. We take images on lots of different machines. So huge strides are being made,” he said. “But it’s very clear that human and machine learning together will create the breakthroughs. We talk about physician burnout, and even physicians leaving. I think that machine learning offers a good chance of removing a lot of the drudgery in healthcare. If we can automate some processes, then it will free up our time for quality judgment, and also to spend time talking to patients, not just staring at the screen.”


Experience New Records for Speed & Scale: High Performance Genomics & Imaging

Through real use cases and live demo, Frank Lee, PhD, Global Industry Leader for Healthcare & Life Sciences, will illustrate the architecture and solution for high performance data and AI...

Looking at the hype cycle around AI

Of course, inevitably, there was talk around the talk of the hype cycle involving artificial intelligence. One of those engaging in that discussion was Paul Chang, M.D.., a practicing radiologist and medical director of enterprise imaging at the University of Chicago. Dr. Chang gave a presentation on Tuesday about AI. According a report by Michael Walter in Radiology Business, Dr. Chang said, “AI is not new or spooky. It’s been around for decades. So why the hype?” He described computer-aided detection (CAD) as a form of artificial intelligence, one that radiologists have been making use of for years.

Meanwhile, with regard to the new form of AI, and the inevitable hype cycle around emerging technologies, Dr. Chang said during his presentation that “When you’re going up the ride, you get excited. But then right at the top, before you are about to go down, you have that moment of clarity—‘What am I getting myself into?’—and that’s where we are now. We are upon that crest of magical hype and we are about to get the trench of disillusionment.” Still, he told his audience, “It is worth the rollercoaster of hype. But I’m here to tell you that it’s going to take longer than you think.”

So, which artificial intelligence-based solutions will end up going the distance? On a certain level, the answer to that question is simple, said Joe Marion, a principal in the Waukesha, Wis.-based Healthcare Integration Strategies LLC, and one of the imaging informatics industry’s most respected observers. “I think it’s going to be the value of the product,” said Marion, who has participated in 42 RSNA conferences; “and also the extent to which the vendors will make their products flexible in terms of being interfaced with others, so there’s this integration aspect, folding into vendor A, vendor B, vendor C, etc. So for a third party, the more they reach out and create relationships, the more successful they’ll be. A lot of it will come down to clinical value, though. Watson has had problems in that people have said, it’s great, but where’s the clinical value? So the ones that succeed will be the ones that find the most clinical value.”

Still, Marion noted, even the concept of AI, as applied to imaging informatics, remains an area with some areas lacking in clarity. “The reality, he said, “is that I think it means different things to different people. The difference between last year and this year is that some things are coming to fruition; it’s more real. And so some vendors are offering viable solutions. The message I’m hearing from vendors this year is, I have this platform, and if a third party wants to develop an application or I develop an application, or even an academic institution develops a solution, I can run it on my platform. They’re trying to become as vendor-agnostic as possible.”

Marion expressed surprise at the seemingly all-encompassing focus on artificial intelligence this year, given the steady march towards value-based healthcare-driven mandates. “Outside of one vendor, I’m not really seeing a whole lot of emphasis this year on value-based care; that’s disappointing,” Marion said. “I don’t know whether people don’t get it or not about value-based care, but the vendors are clearly more focused on AI right now.”

Might next year prove to be different? Yes, absolutely, especially given the coming mandates coming out of the Protecting Access to Medicare Act (PAMA), which will require referring providers to consult appropriate use criteria (AUC) prior to ordering advanced diagnostic imaging services—CT, MR, nuclear medicine and PET—for Medicare patients. The federal Centers for Medicare and Medicaid Services (CMS) will progress with a phased rollout of the CDS mandate, as the American College of Radiology (ACR) explains on its website, with voluntary reporting of the use of AUC taking place until December 2019, and mandatory reporting beginning in January 2020.

But for now, this certainly was the year of the artificial intelligence focus at the RSNA Conference. Only time will tell how that focus plays out in the imaging and imaging informatics vendor space within the coming 12 months, before RSNA 2019 kicks off one year from now, at the conference’s perennial location, McCormick Place.



See more on Analytics

betebet sohbet hattı betebet bahis siteleringsbahis