Today’s healthcare landscape is complex. Health systems are highly focused on efficiency, productivity, and optimal patient outcomes; however, being efficient while maintaining high level of quality and impeccable patient satisfaction is not simple. Having success in the healthcare industry involves commitment, effort, and intricate coordination from highly skilled teams and leaders to drive efficiency and reach goals. Meeting goals and being efficient will also support the financial stability and growth endeavors of health systems. Even in the non-profit healthcare sector, there needs to be a financial margin to continue to support the needs of the community and provide exceptional care to those who need it. One way to meet these demands is through data analytics. Data analytics can assist in driving the direction of work within a healthcare system by telling a story. Data can assist leaders in making informed decisions, identifying trends, optimizing processes, benchmarking success, and ultimately driving change when needed.

Identifying areas of inefficiency and implementing performance improvement initiatives is a standard process within healthcare to improve outcomes. To be efficient, organizations need to use available resources effectively, while minimizing waste. Resources can be people, supplies, and even time. Every year, there is an estimated health spending loss due to inefficiency. Reducing health system inefficiencies can improve the availability of quality healthcare to the communities that need it, which in turn will yield better health outcomes overall. Operational efficiency can be monitored through many different data analytics platforms, in many different areas of a health system. It is imperative that data is available to teams to improve workflows and minimize waste. There should be a standard approach to measuring, analyzing, and utilizing data to engage teams to close gaps when opportunities exist. Some areas within health care that might use data analytics to improve workflows could be the: operating room (OR), emergency department (ED), or even hospital medicine to improve quality metrics or length of stay (LOS) standards.

Prioritizing the opportunities from data analysis should be a strategic step to understand how to drive success.

The OR is an area of intricate skill and complex coordination. OR efficiency is crucial for patient safety, employee satisfaction, and overall healthcare effectiveness. Often, a significant amount of hospital revenue comes from the OR. Hospitals facing economic challenges from shrinking reimbursement should consider maximizing resources within the OR to allow for continued financial success and data analytics can be a means to guide and focus workstreams. Some of the data-driven outcomes that assist with OR efficiency include turn over time (TOT), first case on time starts (FCOTS), room utilization, cancellation rates, and scheduling accuracy.

Additionally, data from supporting departments such as sterile processing can assist with care coordination and efficiency. Benchmarking this data is also important to show success rates based on like organizations/departments in the region/country. Many benchmarks will show that OR utilization between 75%-80% is acceptable and appropriate; however, what happens if utilization is under or over the benchmark? In an organization where the utilization rate is over 85%, a cascade of challenges can occur as the volume outpaces the space available. Data is imperative in these situations to assist with short-term and long-term strategies for success.

OR utilization refers to the percentage of time an operating room is being occupied over a certain period. It’s a key metric used to assess the efficiency of OR operations. Through deep dives with data analytics, care delivery can be transformed to provide better care and profitability for a health system. In ORs, leadership typically focus on room and block utilization. Room utilization focuses on how efficiently a physical single operating room is used. Block utilization looks at how surgeons use their allocated room throughout a period of time. Focusing on these outcome metrics can assist with developing an action plan and process metrics to yield success. Many action plans or process metrics will involve workflow redesign.

When working on OR workflow redesign, data analytics can support several areas. Optimizing patient preparation, minimizing delays, and enhancing team collaboration can create safer and more efficient surgical environments. Standardizing physical and workflow designs will also assist in workflow efficiency. Assessing the previously discussed data can allow teams to focus on areas of opportunity to improve workflow. Additionally, data analytics can assist teams in developing simulation platforms to work through barriers or obstacles in workflow and design a collaborative approach to parallel processing or other methods to assist with efficiency. Data analytics in the OR can ultimately enhance decision-making, assist with resource management, allow for surveillance of processes, and reduce healthcare costs. Data must be collected and shared routinely, but more importantly, data must be acted upon.

There are a multitude of platforms to receive data in the OR. The primary data source is the electronic health record (EHR); however, there are other platforms, such as patient surveys, employee surveys, wearable devices, and financial data. All this data can be analyzed through multiple tools to explore opportunities as well as existing areas of operational excellence. Prioritizing the opportunities from data analysis should be a strategic step to understand how to drive success. Focusing on standard pillars such as quality, growth and finance, patient experience, and research or scientific advancement may all yield opportunities of focus to be key drivers of optimal outcomes. It is up to each healthcare organization to strategically analyze their data and identify priorities. Prioritizing initiatives will enable organizations to develop strategy, play books, or other action plans to optimize efficiency, improve patient outcomes, positively impact financial performance, and ultimately allow for continued success to support community needs and healthcare sustainability. Data is the future of medicine and trends are moving toward machine learning (ML) and artificial intelligence (AI). Systems that are poised with data will be well positioned for success in the rapidly changing healthcare landscape.

The post Driving Change Through Data Analytics: The Crucial Component to Operating Room Success appeared first on HealthTech Magazines.

No one in healthcare will forget March 2020, staring down the worst pandemic in living memory. Society shut down. People started dying at unheard-of rates in Italy, and shortly later NY City. MultiCare Health Systems, an 11-hospital healthcare system in the Pacific Northwest, near where the first US cases were reported, needed to know what to expect.

The data science team at MultiCare addressed the problem by modeling case increases in the communities we serve, first with exponential growth models, but within two weeks, we realized logistic (S-shaped) growth models fit the data better. We were not hit hard in that first wave, and we have continued to model Covid cases across our system through subsequent waves, giving advance notice of a surge and providing our leaders with data-driven insights.

The future of healthcare is big data. But big data by itself just sits in an enterprise data warehouse and runs up storage fees. Healthcare data scientists are essential in moving beyond reports and static dashboards. Data needs to be turned into actionable intelligence for a healthcare system to benefit.

Healthcare data scientists can help an organization get the greatest return on their data infrastructure investment.

Endless Opportunity. The problems that can be addressed with data science are essentially endless. Will a patient be readmitted after discharge? Which patients will have a prolonged hospital course? Can we identify those patients who will develop sepsis earlier and start lifesaving treatment sooner? These problems are readily amenable to predictive modeling and are already commonly deployed in hospitals across the country.

This is just the leading edge of what predictive modeling can do. Many systems are large enough to provide comprehensive datasets on a wide range of patients and conditions. Each disease can be approached using predictive modeling. For many common ailments, a patient’s journey can be mapped along a pathway, with each node in the pathway representing a decision point. It is not hard to imagine a future where dozens, even hundreds of pathways guide a patient’s care, with each downstream step in the journey modeled, and the optimal course of action presented for each individual.

Healthcare is undergoing a generational change as we move away from fee-for-service and towards shared savings and population-based care. Therefore, the need for accurate predictions will increase: who is most likely to be admitted to the hospital? Who will benefit from an intervention to keep them out of the emergency room? Should that intervention be an additional visit with their primary care doctor, transportation assistance to a specialist or another intervention, like a home health visit?

And, as great as the opportunities are in clinical care, as data becomes more and more available, improving a healthcare system’s operational processes may have just as much potential.

Business Analyst or Data Scientist? Do we really need data scientists? A healthcare system’s core competency will always be healthcare delivery, isn’t an analyst enough?

There are a number of steps involved in turning data into a true understanding of the problem. Querying data is often surprisingly challenging: the databases of some electronic medical records (EMR) are composed of more than 20,000 unique tables. Data needs to be aggregated, its quality assessed, and presented in a way that it can be understood by end users. Advanced analytics include machine learning (ML), forecasting, cluster analysis and the ability to hypothesis test. These skills usually require an advanced degree, although not necessarily these days given the availability of online training. But with these abilities, a data scientist can provide insight beyond what you can gather from a dashboard.

The Role of the Vendor. Can’t advanced analytics just be purchased from specialized companies? Certainly, the level of sophistication needed to develop deep learning algorithms is not something many healthcare systems will be able to support. Natural Language Processing (NLP) in particular has made a lot of progress in the last few years and will soon be pulling knowledge out of free text. Third party vendors will be helpful here, but fully realizing their potential will require people who understand both the algorithm and the use cases. 

In a larger sense, data science can help an organization develop insight long before it gets to the level of an RFP. Many people in healthcare have deep knowledge about esoteric fields. They may want to explore a hunch with genuine financial and clinical implications. Having access to a data science team, people who leaders know personally and can connect with to talk through a problem, can make all the difference.

Additionally, healthcare organizations need to develop the sophistication to evaluate a third party’s claims. Does a purchased model accurately predict what end users think it does? Even models published in peer-reviewed, academic journals cannot be assumed to be accurate on a specific organization’s population. A model needs to perform for an entire population, but also needs to be evaluated for bias on the vulnerable groups a system cares for.

The Clinician Data Scientist? Two decades ago, it was hard to imagine that we would have doctors who specialize not just in patient care but in optimizing the use of the electronic medical record. Now, over two thousand doctors are board certificated in clinical informatics, with hundreds more becoming certified every year. The value proposition for clinician data scientists may be even greater.

Big Data and Data Science are Essential to the Future of Healthcare. Data science is not an add-on, but a process to integrate into healthcare decision-making. As organizations make bigger and bigger investments in enterprise data warehouses and data aggregation platforms, healthcare data scientists are the best way to assure return on that investment.

The post The Case for the Healthcare Data Scientist appeared first on HealthTech Magazines.

It is simply not possible in today’s world to talk about technology and healthcare without playing the acronym Scrabble or buzzword Bingo – AI, ML, RPA, NLP, Digital Transformation, Cyber Security, Blockchain, Interoperability, Chat Bots, Predictive Models, and the list goes on. Since the rules that govern these games change daily, there is no clear winner. It can be overwhelming to think about these emerging technologies when deciding what to spend time and money to implement. Back when things were ‘normal’, analytics and IT executives had the luxury of mapping out a long-term strategy, planning for capital, and building a technology and talent portfolio to support that strategy. While a high-stakes endeavor to be sure, there was not an innovation imperative that required lightning-quick decisions and implementations. We are now entering a new normal where this paradigm has shifted. The pressures of a global pandemic, supply chain disruptions, staffing shortages and capacity constraints have provided a sense of urgency to innovate. Now more than ever, we need to leverage emerging technologies to augment human decision-making, automate manual tasks and connect data sources. Traditionally behind the innovation curve, healthcare is playing catch-up. While it can be daunting to deploy, I would ask what better time than now to implement innovative tools to help support our patients and caregivers? But how do you choose which technology? How do you create and sustain a culture of adoption? How do you ensure the valuable resources: money, time, and talent you spend on these technologies will ultimately provide the outcome you desire?

In our organization, we’ve been able to use the innovation imperative to advance new technology initiatives in record time.

Here are some recommendations for implementing new, innovative technologies during this challenging and disruptive time:

• Begin with the problem itself. What exactly is the problem you are trying to solve?
  • Understanding the problem, the cause, and how you would measure successful resolution is key
• Ask important questions 
  • How will technology solve this problem?
    • What specific technology is needed?
  • What barriers will there be to adoption?
  • Will the support needed for this technology outweigh the benefits?
• Stay true to organizational and divisional long-term vision
  • Ensure executive alignment with new projects
• Be open-minded and flexible
  • What worked in the past may not work now
  • Think outside of the box
• Tap into existing resources
  • Current vendors may be willing to partner to enhance their products with little cost
  • Team members may have creative ideas to innovate internally
• Take advantage of the urgency of the situation to implement quickly
  • Adopt rapid decision-making processes
  • Utilize existing culture and guiding principles to inform choices
• Partner with operational and strategic leaders to align on objectives
  • Create excitement with the why
  • Ensure alignment with budget constraints – have a solid ROI
• Communicate effectively
  • Target audiences for biggest impact
  • Consider alternate forms of communication such as presentations at meetings or drop-in networking in place of e-mails and newsletters
• Avoid the traps 
  • Don’t implement technology for technology’s sake
    • Many times, people hear of or see a new technology and want to implement it because everyone else has it
  • Don’t believe everything you hear from vendors
    • Ask for evidence – consider doing a pilot or proof of concept before you spend
    • Carefully vet references – use your network and not just vendor recommendations
  • Don’t throw technology at process problems
    • Technology can’t solve deeply rooted process problems
    • Ask the question, how did we get here?
  • Avoid analysis paralysis
    • Be willing to make calculated mistakes and learn from them

I believe the current landscape is the crucible of crisis. A crisis that provided the opportunity to leapfrog outdated practice and process and replace with advanced technologies that make a difference. In our organization, we’ve been able to use the innovation imperative to advance new technology initiatives in record time. We have been able to rapidly scale up our predictive analytics capability to support our clinicians with accurate census forecasting. We have been able to partner with our supply chain department to help predict PPE needs. We have supported our nursing and facilities divisions with real-time alerting for air-flow monitoring. We have transformed manual process to automate submission of data to the State. We are in process of deployment of NLP/AI technology to support registry submission. Ultimately, we were able to take advantage of the urgency of opportunity to deploy meaningful innovations to support our patients and caregivers. The ability to leverage these tools has enabled us to shift from a traditional transactional model to a partnership-based solution delivery model. 

Healthcare is now positioned to be at the leading edge of technology innovation. Much of this is driven by external forces. If we do not respond with a sense of urgency to innovate and implement, we put our patients and entire workforce at risk. It is my hope that we can continue to ride the wave of this innovation imperative to continue to drive business transformation, improve care, increase efficiency, make smart decisions and reduce cost. If now is the time, the question that remains is how? My answer is that we will do this with the same care and consideration we use when at the bedside. We will assess the situation, identify the problem, leverage a multidisciplinary approach and judiciously select the right tools to address critical needs. 

The post Innovation in the Crucible of Crisis appeared first on HealthTech Magazines.

The objective for any business is to run efficient operations to reduce costs. The healthcare industry can benefit by using the same mindset to add value to existing traditional data sharing processes to enhance patient care, reduce adverse medical events and medical procedure revision costs across the clinical industry. According to Kopel and White (2020), the shift in data sharing processes is essential as patients demand diagnosis based on evidence for physicians to make the best treatment decisions. Evidence-based medicine and spending depends on data provenance and receiving healthcare data timely using modern technology for to collect relevant patient data for making patient care decisions. Greenhalgh (2020) posited that evidence-based medicine depends on querying singular truth sources and applying linear logic to determine the best effects of physician intervention. According to Hasanpoor et al. (2020), evidence-based medicine reinforces physicians’ skills, which leads to quality health services and better cost management. The evidence-based approach augments the medicine decision process with data analytics to reduce the overuse of medical services to reduce costs (Hasanpoor et al., 2020). However, changing the traditional ways healthcare managers collect and share data to gain informed decision insights poses a competent resource effort and challenge. Healthcare leaders who are not aware or technically savvy to understand how to serve the next generation of healthcare consumers and use of the social media platform methods for real-time data sharing instinctively disjoin care responsibility between patient and physician leading to poor satisfaction scores.

Evidence-based medicine and spending potentially creates an environment that promotes greater knowledge and diagnosis insight to make informed decisions.

Healthcare managers can promote better healthcare and business sustainability by applying new technology approaches. The best way to provide guidance is to start with a small-scale proof of concept to gain data sharing trust and then scale iteratively for continuous process improvement. Many other factors such as overtreatment, technology development, insurer payment mechanisms that increase cost, regulations, patient values, and liability are part of the evidence-based medicine and spending equation. Zack (2020) proposed a pragmatic method to implement and promote a medium for user acceptance of data collection and sharing technology that involved redesigning the process for tracking medical devices and using Blockchain distributed ledger technology (DLT) to share the data ubiquitously to all providers involved in the care. The benefits of using DLT technology opposed to traditional Search Query Language (SQL) are the decentralized database management system, immutable audit trail, persistent availability, data provenance, and robust flexibility (Tsung-Ting et al., 2017). A DLT data-sharing platform in the healthcare system eliminates a single point of data connectivity failure for continuous health record availability (Tsung-Ting et al., 2017). Zack (2020) indicated that integrating medical device data in real-time is congruent to integrating complex electronic health record applications and a plethora of supporting medical application data, including physician contextual data, insurance claim, and device manufacturing data. Interfacing and integrating applicable healthcare data frameworks to capture contextual data and combining the relationship to measurement data adds semantic relevance and potentially improves treatment evidence. Eliminating interface and intermediary boundaries to collect and share data potentially allows the physician managers to obtain unique treatment insights by combining knowledge and prior performance to identify the best care value.

Sharing evidence-based healthcare data is essential to improve care quality through better health information system access and communication. Sharing healthcare data supports evidence-based medicine and can help clinical leaders to understand trends and patterns of disease and treatment outcomes to ensure better care quality (Yue et al., 2016). As advances in medical technology improve the treatment of disease, health care managers should consider implementing innovative strategies to redesign the health care monitoring process on a real-time continuous basis and leverage integrated healthcare data into a standard accessible format that is unambiguous. For instance, a standard format can contain clinical and contextual data, known as metadata, to provide effective data collection for effective understanding and treatment.

Improving the quality of health care decisions influences the delivery of medical care. Evidence-based medicine and spending potentially creates an environment that promotes greater knowledge and diagnosis insight to make informed decisions. Data alone may be insufficient, as many patients and providers are ill-equipped to make sense of the volume of detailed information that populates their EHRs and insurance claim repositories. Organizing and interpreting the data, historical content, scientific literature, and the health care resources available to them in their own communities can help in the health care decision process. Perhaps using a social media health care platform that is patient data-driven could potentially serve as a ubiquitous tool to collect relevant resources to establish reliable real-time evidence to conduct qualitative and quantitative systematic analysis for clinical and managerial practice. The lack of data connectivity and contextual communication between health systems, physicians, patients, and insurers impacts medical decisions. Implementing an integrative DLT technology in a social media type framework is a novel methodology that can evolve into an evidence-based approach having the potential to become the next gateway to better health care. Application of the DLT healthcare digital agnostic data medium has the potential to unlock the interoperability gateway challenges of volume, velocity, variety, and veracity to produce the most patient value pragmatically. Future work regarding data sharing platform management tools extends the knowledge to resolve the unknown gateway lock challenges. Social media platform designs are the next healthcare data sharing frontier that potentially combines functional semantic integration, semantic axioms, and network configuration to bridge the gap between healthcare standards, data, physicians, and patients. The novelty of the social media approach opens interoperability gateways that move the data closer to the user for better decisions for evidence-based medicine and spending.

The post Applying Intelligent Data Share Gateways for Evidence-Based Medicine and Spending appeared first on HealthTech Magazines.

How do we achieve an organization’s data ambitions by bringing its data and analytics capabilities into alignment? Data and Analytics are often the top priority for nearly all CIOs that I speak with. What should we consider as we work toward meeting the vision of a data-enabled and information-driven enterprise? 

It is no longer enough to simply consolidate demand across the enterprise for data and analytics and create an execution plan for information services as a delivery center. For an organization to meet the ambitions of being digital and data-driven, integrating this into the business requires a shift from analytics that look at delivering data services as projects to a model that develops and enables enterprise-wide data use and human analytic competencies with information tools. . The movement from reporting to dynamic access and understanding of data as a catalyst for business decisions requires a journey and a change in mindset. This shift in culture needs to be woven into the thread of everyday activities This is more than a delivery of data tools. Success requires a shift in leader use and understanding of information.   

There are a series of essential elements to consider in achieving a data-driven vision. The recognition that it requires executive leadership and ownership should be considered a first step. The appointment of the CDO (Chief Data Officer) and a network of leaders, bring into focus the priority and recognition that this is an enterprise transformation. Create a vision and an understanding of the drivers for change. Leaders will require changes in day-to-day behaviors, use, and synthesis of information, while integrating information into making informed business decisions.

The CDO and team need to formalize enterprise strategic focus areas covering a wide gamut across digital business transformation and optimization with senior business executives. The identification of direct opportunities should align with the organization’s strategic mission and plan. This will vary by organization and focus on new revenue platforms, digital revenue opportunities or lines of services, better informed customer-connected experiences, enterprise efficiencies, and revenue growth/optimization.

This approach enables a clear articulation of the organizations’ priority and corresponding requirements of business leaders that will need to be part of the journey to data literacy and information-driven decisions. Establishing enterprise capabilities, ownership, responsibility, and cultural shift enables a data literate organization to drive the tools and adoption of practices required to bring information together and respond to the enterprise’s strategic demands. This shift will clarify the required data types, sources and level of quality to drive project-related work. This then aligns the information strategy with the enterprise strategy to create a journey for the enterprise to be a data and digital organization. This work must be orchestrated and deliberately driven with clarity. Today’s data and analytic tools, including AI, are built to be generic across industries. To truly garner the value, each organization needs to have clarity in the specific enterprise use cases to bring the expected and needed value.  

There is also the need to recognize the value derived when the program is defined and initiated. Establish a clear set of measures to begin to collect across the business domains highlighting the value and insights this change in culture, approach, and thinking brings.

In the smallest and most nimble of organizations, this shift will require fundamental change. Developing these competencies will require a holistic view with the alignment of leadership and skills within the enterprise aligned to a program of initiatives. This is a change from the legacy world of project-based reporting, analytics, and data governance to an organization whose data and analytics are entwined with its strategy and mission and built for purpose.

The post Achieving your Data and Analytic Ambitions appeared first on HealthTech Magazines.

On May of 2019, I was appointed New York City Health and Hospitals’ (NYC H+H) chief data officer. My first order of business was to lead the system to develop an enterprise data and analytics strategy. Our intent ultimately included driving behaviors that would lead to improvements in health care, quality and safety, care experience, equity, and better financial outcomes; merging our digital strategy with our data strategy seemed natural. Consequently, the enterprise application development team was placed under the data and analytics (DnA) department. While the link between data and digital has already proved to be a strong approach for connecting data insights to better engagement opportunities, it would take the pandemic to truly uncover the power afforded by having an application development team report to the data and analytics department.

During the first and second COVID-19 surges, the DnA team at NYC H+H were repeatedly asked to develop dashboards and reports with data that our source systems are not configured, or in some cases designed, to capture. Under typical circumstances, our DnA team would have informed our health care system leaders that our line of business systems lacked the data necessary to create the required dashboards and reports. To better understand this, it is useful to recognize that most enterprise data and analytics organizations follow a similar data life cycle. Typically, source data from line of business systems are ingested into data lakes or an enterprise data warehouse where the data is reorganized into facts and dimensions necessary to present reusable data constructs. The data constructs are then leveraged by analytics cores, reporting developers, data scientists, power users, and others via a centralized or self-service model. The timely orchestration of processes and procedures to get even one dashboard to achieve the desired business goal can at times prove challenging. Often, inconsistent business glossary and data definitions, decentralized and asynchronous data repositories, and poorly entered data will be cited as reasons for poor data quality and dashboard publication delays. However, it is generally understood that the data and analytics team can only report on the data they have available to them. But during a pandemic, simply stating that the data is not currently being captured is unacceptable.

Our enterprise app dev team, led by Andrew Vegoda, deployed a number of rapid application solutions to address key data capture limitations in our line of business systems.

Health care institutions have overcome the data capture limitation by introducing ad hoc data capture tools such as excel, share point sites, surveys that function outside of typical workflows to collect the missing data. While these solutions help fill the data capture gaps, they introduce unintended consequences. Firstly, data generated from ad hoc capture solutions is often decentralized and difficult to aggregate for use with robust dashboard and reporting solutions. Secondly, ad hoc capture solutions typically introduce new processes that require training to assure compliance as well as uniform and timely data input. Finally, as more ad hoc capture solutions are introduced, staff report that inputting data becomes overburdening and begins to negatively impact their ability to keep up with normal operations.

This is where having an app dev team reporting to data and analytics is magic. Our enterprise app dev team, led by Andrew Vegoda, deployed a number of rapid application solutions to address key data capture limitations in our line of business systems. For example, early in the pandemic, we needed to maintain an inventory of personal protective equipment (PPE). Due to shortages, PPE distribution would need to be throttled to a level that could ensure our health care works and staff would remain safe for the duration of surge 1. Our app dev team designed and built an application tracking the PPE allocations and number of requests made by each health care worker at each of our 11 acute care facilities. Consequently, in less than three weeks, we captured the necessary data and created robust reports to keep our workforce providing care while keeping them safe.

Our team also developed an application, point of entry (POE), to facilitate the registration of family members seeking to visit loved ones hospitalized due to COVID-19. This was especially important for multiple reasons. We wanted to support our community in staying connected with loved ones. We needed to make sure anyone entering the facility would not get infected or possibly infect others; all during a pandemic and within visiting hours that vary by the facility to accommodate each community. Again, there existed no cost-effective and quick solution in the market to address this need. In less than a week, our team developed a solution that supported the operational needs and offered reports to various agencies interested in our progress in effectively supporting our patients, their families in a safe manner for them and our health care heroes.

Ultimately, our app dev team developed nearly 40 application solutions varying in size and complexity. These applications helped address operational needs for our health care system.  Moreover, they generated missing data necessary to offer critical situational awareness necessary to run the nation’s largest public health care system during the worst pandemic of our lifetime.

The post Adding App Dev Capabilities to Your Data & Analytics appeared first on HealthTech Magazines.

How do you use data to find long-term chronic health problems?

Healthcare practitioners function in an event-based system with some discreet data points, but most of the good information about patients is contained in unstructured notes.

How can we use these notes to piece together chronic conditions and their causes across multiple visits, multiple doctors, across a sea of time?

Fortunately, many other industries have been down this path before and found ways to transform how they used data and technology to differentiate themselves. Healthcare, meet Big Data, AI, and the Digital Twin!

The glut of data, data everywhere

Each visit to a healthcare provider generates a ton of data about who you are, why you are visiting, what the diagnoses are, additional information needed in labs or images, and next steps. After the visit, even more information is generated through back-end billing information to the insurance provider.

Over the course of many different visits, to different providers, different systems, and over many years, the amount of data can create a complex, messy picture.

The retail transformation brought on by Amazon offers a roadmap for how healthcare might handle this glut of data. Amazon users know that the company creates an efficient and tailored experience based on what the user has looked at and purchased before. It’s almost like they knew what you wanted, and it engaged you in a way that you have now come to expect. This did not happen without using many individual points of data over time to get patterns and actions.

In healthcare, unlike in retail, much of the important data must be teased out of free-text notes. In order to get a accurate picture of a patient’s health, all of the relevant historical notes and images would have to be read and assimilated.

Where is the good stuff?

In healthcare, unlike in retail, much of the important data must be teased out of free-text notes. In order to get a accurate picture of a patient’s health, all of the relevant historical notes and images would have to be read and assimilated. This type of investigation is laborious for clinicians but also necessary to truly understand chronic diseases.

Let’s take a typical MS patient, for example. Flare-ups may manifest themselves over years, coming in loss of hearing for a week or two, losing eyesight for a brief period of time, or numbness in the legs that seems to come and go. The patient might go to an ear doctor, an eye doctor, and a spinal specialist. The individual specialists might not put together a bigger picture because of the nature of the interfaced systems, the dispersed data, and the amount of time needed to review all of the notes. Those notes could include innocuous-seeming but insightful clues: headache, fatigue, the environment at the time of the flare-up.

This example shows that much of a patient’s data is not captured in regular columns but is contained in free-text fields that, unless read or pulled out, are the missing key to creating a full picture of health.

Unlocking it into something I can use

Using Natural Language Processing and Machine Learning, or ‘AI’, against this sea of data is the key to distilling out useful bits of data to be a part of the complete picture. By adding these key data points to the visits through NLP, the ML can start to recognize patterns that are more meaningful to understanding the overall patient health profile.

The key is providing that information within an easy-to-consume and actionable manner that removes the friction of too much data and not enough information. At Novant Health, we implemented these processes in the electronic medical recordMeeting the clinicians where they spend their working time in the EMR means that all the data is put through the Big Data and AI machine to assist clinicians with investigation and treatment.

Making big progress and the challenges ahead

While implementing the NLP and ML is technically challenging, the bigger hurdle in our heathcare system and in others is AI’s acceptance. We’ve met that challenge by holding to explainable AI and no black boxes principles, and the technical teams have worked side-by-side with those who will use the new tools. That clinical user input and partnership help zero in on the differentiating bits of information and provide insights into what is needed next.

The next iteration of healthcare AI will be the development of a “digital twin,” a digital model of the patient created using the full range of collected data. Clinicians can use the digital twin, with all new compute power, to simulate different treatment factors.

In this next frontier of digital medicine, the digital twin will also provide new insights about how diet affects chronic diseases and will help narrow down food triggers without trial and error.

In this future of personal health, data will be a more powerful tool for clinicians and patients in managing disease because important information will no longer be left behind in clinician notes.

The post How Big Data and AI are helping sift through data appeared first on HealthTech Magazines.

COVID-Ready Qlik

Qlik, a leading data and analytics company, responded to the urgency of the COVID-19 crisis with myriad use cases across the global digital care community, from government agencies, health systems, health insurers, and policymakers. The company’s agile analytics platform automates the last mile of analytics-ready data delivery to the right people at the right time.

Many decision-makers were armed with Qlik’s advanced technologies for data exploration to monitor the ongoing situation and make accurate estimates of the health system demands to support initiatives and relevant plans. Qlik later offered solutions that helped users to implement a dynamic approach of data collection, analysis, and prediction – to inform policy decisions in real-time, and iteratively optimize public health recommendations for reopening. “Most reopening plans require extensive testing, contact tracing and monitoring of population mobility, and almost none consider setting up such a dynamic feedback loop until they use Qlik Sense and Qlik’s data replication solution,” says Andy Shore, VP of Public Sector, Qlik. “Relevant, timely data for analyzing the status/impact of these efforts and understanding health system capacity, has helped relief organizations and policy makers effectively model and adjust their response strategies. Agility is one word that best describes Qlik’s products in data and analytics for healthcare through automation of transforming raw data sources to analytics-ready data sets.

Making an Impact with Data Literacy

The company is guided by top minds in the industry with their steadfast commitment to excellence and core operating principles to aid enterprises worldwide move faster, work smarter, and lead the way forward with end-to-end solutions for getting value out of data. Qlik has a global customer base of 3,000 healthcare companies that leverage the firm’s world-class customer experience to support their users’ data strategy, whether on-premise, hybrid, or in the cloud. The very first version of Qlik’s product, QUIK, was introduced in 1994, followed by a patent application in the next year. QUIK stood for Quality, Understanding, Interaction, and Knowledge. Since then, the company has stood steadfast to its mission of empowering customers to make meaningful discoveries that drive actual change. In the healthcare community, Qlik has been helping systems unearth variations in care to better serve the patient community. In other words, make an impact.

At the heart of Qlik’s offerings for data and analytics lies the philosophy of data literacy. Low data literacy inhibits team performance, quality of care, and digital transformation in the healthcare sector. In layman’s term, data literacy is the ability to read, work, analyze and communicate with data. A skill that makes teams and workers unrivaled for their ability to ask the right questions on data and systems to build knowledge that empowers the rest of the team to make the right decisions. “In healthcare, data literacy means that everyone involved with patient care is able to interpret specific data and use it to make decisions to deliver positive patient outcomes, cost-effectively,” adds Shore stressing how data deluge can remain an ordeal for most organizations and their resources only unless they build data literacy for their workforce. “Data literacy is a strategy that can transform an organization while also building loyalty with a workforce that is energized and empowered by their employers’ investment on their professional development.” And rightly so, Qlik’s platform based on the DataOps approach is what most businesses need in today’s digital age to accelerate the discovery and availability of real-time, analytics-ready data to the cloud of their choice by automating data streaming (CDC), refinement, cataloging, and publishing. The company’s agile platform is the industry’s only end-to-end platform that delivers a near real-time analytics data pipeline, data warehouse automation, and data provisioning to the choice of analytics tools. Qlik Sense, one of the flagship solutions, is a complete data analytics platform that sets the benchmark for a new generation of analytics. With its one-of-a-kind associative analytics engine, sophisticated AI, and high-performance cloud platform, users can empower everyone in their organization to make better decisions daily, creating a truly data-driven enterprise.

With Qlik, the customer could bring down the refresh time from 24 hours to 5 minutes. The end result was not in about a few thousand data points, but instead over hundreds of millions of rows of data at a time!

Andy Shore, VP of Public Sector, Qlik

Active Intelligence and What it Means to Customers

Delivering care to thousands of patients annually means accumulating data regarding admission, patient IDs, time stamps, surgery schedules, medical records, electronic correspondence, and more. The challenge is to manage all these without data leaks. Security of patient information and compliance are integral to patient care. One of the customers of Qlik, Children’s Hospital of Pittsburgh (CHP), earlier relied on Cerner consultants to analyze data. They later deployed Qlik products to address their inability to steer decision making due to the odd systems as the patient reports underwent daily refreshing. In addition, the users lacked familiarity to handle the data systems to leverage data for meaningful use. As a result, the client was dealing with over 5,000 tables of medical records that Cerner Millennium created for them. “This type of view does not support fast, real-time decisions, and certainly is not user-friendly for a large staff who would all benefit from being able to view and interact with the data. Only the administrators could view the data, having to navigate the complex data models to find the right information,” adds Shore. Christopher Meyers, Systems Analyst at CHP knew that the hospital had to make the shift. And therefore, the client integrated hundreds of millions of rows of data using Qlik to create simple, visually appealing, and properly governed dashboards. With Qlik, the customer could bring down the refresh time from 24 hours to 5 minutes. The end result was not in about a few thousand data points, but instead over hundreds of millions of rows of data at a time!

Within a year, the client incurred significant cost savings by reducing previous inefficiencies.  CHP reduced the cost of care for patients with appendicitis by 11 percent and data errors were found and corrected, such as duplicated records, and missing or incorrect timestamps. Readmissions were reduced due to better management of health records, and the overall registration workflow was significantly improved through the standardization of canceled or duplicate registrations.

Qlik apps provide analytics to more than 500 personnel in CHP and focus on areas such as ED, Surgical Analysis, and monitoring adherence and impact of clinical pathways for Acute Appendicitis, Isolated Hyperbilirubinemia (jaundice in newborns), Cellulitis and Simple Abscess, and Bronchiolitis. 

As the demands in healthcare surge driving a change in modern analytics, a need has emerged in the industry to overcome the shortfalls of legacy analytics approaches and Qlik is set itself up to the task of delivering a state of continuous intelligence from real-time, up-to-date information designed to trigger immediate actions. “We call it Active Intelligence,” says Shore. “Active Intelligence closes the gaps among those components, creating a multidirectional conduit for the continuous flow of data and information across the organization. This intelligent data analytics pipeline enables fresh data to reach users almost as soon as it comes in the door.”

The post Qlik-Driving Data Literacy through Agile Analytics appeared first on HealthTech Magazines.

Many organizations aspire to be “data-driven’ or “analytics-focused,” but what do those terms really mean? Community Health Network was no different, embarking on a data and analytics improvement journey which resulted in a unique organizational structure: forcing form to follow function. “Form follows function” is a principle typically associated with industrial design, meaning the shape of an object should relate to its intended function or purpose. To extrapolate, how an organization is shaped ideally is directly related to the intended outcome.

After years of frustration with the lack of data or insights regarding how the business was performing, Community Health Network (CHNw), in the summer of 2018, engaged with Gartner to undergo an enterprise-wide evaluation of the function and future state vision regarding analytics as a strategic asset. Prior to this assessment, information was scarce despite data being abundant. The Analytics department, led by a Business Intelligence Director, was a subset of the Information Technology Department. There was a limited connection to the overall strategy of the business. Prioritization of projects and resources was either based on “first-come, first served” or by whichever Executive could strong-arm reports to be generated. Ultimately, this prioritization method led to operational units hiring their own data analysts which resulted in more silos as they did not have access to the central data sets.

As a result of the Gartner assessment, recommendations were made to create a freestanding Department of Network Analytics, headed by a new position; the Chief Analytics Officer (CAO).  Additionally, as part of the new Analytics department, a Center of Excellence was formed. With the CAO reporting to the CEO, it ensured data and analytics had representation at the organization’s highest levels. The Analytics Center of Excellence was charged with reimagining how data was utilized within the organization and moving from mostly descriptive-analytical reporting to more advanced analytics, including predictive models powered by machine learning. Ultimately, the goal was to tie the network strategy to analytic output to inform business decisions with data.

After approximately nine months under this structure, additional organizational structures were put in place. Previously, all Information Technology functions reported under the Chief Operations Officer (COO) and Clinical Informatics activities fell under the Network Chief Physician Executive (CPE). In order to continue in advancing the desires of a data-driven organization, Information Technology, led by the Chief Information Officer, moved under the CAO. Additionally, Clinical Informatics, led by the Chief Medical Information Officer (CMIO), also moved under the CAO, along with Regulatory Reporting and all Enterprise Services including Business Process Management and Continuous Process Improvement.

Ultimately, after said changes, the Chief Analytics Officer has operational oversight of all network data and analytics (excluding finance), all technology and digital transformation, clinical informatics, process improvement activities, and regulatory reporting. This allows for direct connections between network strategy with analytics, technology, and business processes. In the truest sense of form following function, the organizational structure at Community Health Network is shaped to allow data to be a driver of network strategy by driving business outcomes.

After the pandemic passes, technologies such as electronic check-in, early hospital discharge programs, and remote chronic disease monitoring will quickly become the expectation for many patients and health systems.

In a year such as 2020, this structure has been incredibly beneficial. Early in the pandemic, data was scarce and technology needs were changing rapidly. For example, the conversation from a purely in-person care delivery model to a near 100% virtual model required expedient implementation of technology to support. The needs of measuring this new business delivery were no less important. Being able to predict volumes of covid infections and align with inpatient operational changes was critical. The strategy focused on preparation and taking care of infected patients while protecting our healthcare workforce. 

As the pandemic continues to alter healthcare delivery, thus requiring continued use of virtual care and digital tools, the alignment of the support services with enterprise strategy will continue to be of utmost importance. After the pandemic passes, technologies such as electronic check-in, early hospital discharge programs, and remote chronic disease monitoring will quickly become the expectation for many patients and health systems.

Silos have long existed in many industries; healthcare is no different. Often, organizational structure contributes to the continued difficulty in communication and collaboration. As a result, three interventions will assist in the journey.

First, look at the organizational structure and understand if it is truly working. Historical and legacy structures, if in place for long periods should be examined. This is not to advocate that reorganization for the sake of change is the answer; however, taking a deeper examination into current structures is well worth the exercise.

Second, explore if non-traditional reporting structures might benefit the organization. Certainly, having the CIO report to the CAO is not typical. However, it is effective in our organization.

Third, ensure team members have a proper understanding of the strategy and vision of the organization. This is often an overlooked function with services typically considered supporting such as IT and Analytics. Additionally, helping them recognize the value in their work will also improve employee engagement. 

Fourth, these functions need a consistent voice at the highest level of the organization. In order to be data-driven and truly digitally transformed, the employees responsible for this change must be earnestly represented throughout the process, not merely in appearance.

Finally, organizational structures are typically complex with legacy personalities. Finding the right structure to be data-driven can be challenging and demanding. It requires crucial but honest conversations. When performed with the organizational strategy and best interest at the center, the outcomes will be incredibly rewarding for all involved.

The post Being data-driven leads to a new organizational structure appeared first on HealthTech Magazines.

The COVID-19 pandemic forced so many aspects of life to change and for health care to adapt in completely new ways. At St. Luke’s Health System, the need to adapt sparked innovations, many of which were driven by our Digital and Analytics organization. The St. Luke’s Advanced Analytics team, led by Dr. Justin Smith, developed COVID-19 modeling to help predict the virus’ path through our footprint and to optimize staffing. Our suite of Machine Learning and Artificial Intelligence (AI) models have helped shape our enterprise, providing insights towards safely and responsibly providing healthcare services during the COVID-19 pandemic.

Because the Rocky Mountains divide our coverage area, traditional susceptible, infected, recovered (SIR) models were not supplying meaningful results when forecasting the future surge of COVID-19. Our team developed a new SIR model that allowed us to adjust for the probability of social mixing between geographically distant populations – enabling us to accurately forecast our summer COVID wave within a week of reality. We also created simple but effective COVID regression models for forecasting hospital census up to two weeks into the future.

Perhaps our greatest advancement, though, came in a novel approach using Machine Learning and AI to project COVID patient hospitalizations – the XGBoost Model. This new model provides 30-day hospital and Intensive care forecasts that are accurate within +/- 4-5 patients across four hospitals and eight separate Intensive Care units, allowing operational decision-makers and executives to plan for future staffing and patient decisions. XGBoost was successful from early on despite such little data and more accurate than statewide models after accounting for Idaho’s unique geography. 

The XGBoost Department Forecasting model is determined accurately by a validation method that retrospectively withholds actual results and creates predictions for those data points. The metric used to analyze the accuracy is Mean Absolute Error (MAE). MAE measures the average magnitude of the errors in a set of predictions without considering their direction. This model’s validation is designed to withhold 7-days of historical data and then project those same 7-days of historical data. The projections from the 7-days are then compared to the actual result and the MAE is then determined.

XGBoost is a highly sophisticated ensemble machine learning algorithm used to determine the individual projection for each day. This model’s XGBoost algorithm has been slightly shifted from the original XGBoost model and finds a continuous outcome versus a binary outcome. XGBoost follows a gradient boosted tree format that creates a hierarchy of predictive features that will filter down to create a projection. The XGBoost algorithm will create a gradient boosted tree algorithm specific for each projection day, and the output generated will be used and inputted into the three and ten day horizons. This means that an XGBoost model can be constructed to project every single projection directly across a 3-day horizon and 10-day horizon.

Regression Forecasting 

The purpose of these forecasts is to supply a short-term projection for ICU and Adult Unit Hospital utilization, specifically for COVID-19 positive patients. The current projections are based on St. Luke’s data and are created using linear and polynomial regression. St. Luke’s Advance Analytics provides the two projections in an attempt to supply an outlook on short-term surges. 

Rolling 21-Day Forecasts 

Figure 1: Rolling 21-Day ICU: Lines illustrate actual (jagged), linear (straight), and polynomial (curved) projections. Right y-axis value represents the most recent midnight census in respect to the ending jagged line.

Direct Multi-Step Forecast with Multiple Times Series using XGBoost 

The purpose of these forecasts is to supply a one-month projection for ICU and Hospital utilization within relevant Hospitals, specifically for COVID-19 positive patients. The current projections are based on St. Luke’s data and are created using a novel, sophisticated machine learning algorithm. St. Luke’s Advanced Analytics now provides the novel projection for a 30-day horizon.

Direct Multi-Step Forecasting with Multiple Time Series (Direct Forecast) is a methodology that trains on historical data (data already observed and collected) and creates a projection for, in this case, a future date.

Three separate model horizons (1-day, 1-14 day, and 1-30 day) are then used to determine the timespan parameters of the model, which is then fed into the XGBoost portion of the model.

The model currently uses COVID-19 Positive tests per region (Magic Valley and Treasure Valley) as reported by the State of Idaho, Admissions, and Discharges from each Hospital, Idaho State Reopening Phases, and Holidays as predictive features. The addition of policies and discoveries such as the reopening of schools, potential masking policies, or the discovery of a vaccine can be used within the model once the policy or discovery has been enacted and given at least two weeks of training data.

One-Month Forecast: Direct Multi-Step Forecast with Multiple Times Series using XGBoost 

Figure 2: Direct Forecast Projection for ICU Hospitalization Census: Red line indicates the current date and the jagged line before the red line denotes the historical hospitalization census. Purple line after red line is the model’s projection per day with a confidence interval of +/- 2. 

What’s Next? 

The Advanced Analytics team that developed XGBoost is a finalist for an Aegis Graham Bell Award, an annual award sponsored by the government of India’s Ministry of Electronics and Information Technology. The category St. Luke’s is in is “Combat COVID-19 With Artificial Intelligence.”

St. Luke’s Health System is one of a very select few organizations in healthcare both has the capability and experience in utilizing advanced Machine Learning. We are very proud and excited to serve our communities with the support of cutting-edge technologies. 

The post Utilizing Advanced Analytics Capabilities to Respond to COVID-19 appeared first on HealthTech Magazines.