Leveraging Big Data in Healthcare


Post by Hamid Al-Azzawe


Vice President of Engineering, Caradigm

“Big Data” has become as prevalent a term as “EMR” and “HIE” in the healthcare industry. Is it the new technology that holds the promise to revolutionize healthcare? Should a CIO consider adopting it? What are the real benefits? What are some of the key factors to consider when reviewing the multitude of options available?

While carrying out the Monitoring and Diagnostics mission of the Microsoft Autopilot team, we had to process petabytes of data on a daily basis. Big Data was not science fiction; it was a fact of our daily lives. Our customers, and our own infrastructure team, demanded fast and efficient processing of huge amounts of data to facilitate operational, business, and engineering decisions critical to their success. Many of the insights gleaned from this data relied on Cosmos, Microsoft’s internal big data store.

In healthcare, as I outlined in my earlier blog and whitepaper, we are witnessing an explosion of digital data collected from cell phones, voice, images, notes, EMRs, HIEs, and even social media, coupled with an ever increasing number of medical devices that generate large amounts of healthcare tests and diagnostics data. Hospitals of all sizes are finding themselves overwhelmed with this growing data asset. As a result, this data asset may go unused or worse be purged periodically due to a lack of perceived value or the assumed complexity and cost of archival storage!

One way to tackle this data explosion problem is big data. So what is big data? Big data refers to the collection and management of very large data sets and storage facilities. Big data can offer solutions for data that is described as high in:

• Volume – the large amounts of data,
• Velocity – the speed of growth of data, and
• Variety – the mix of structured and unstructured data.

Some organizations also add Veracity, the quality of data, as a challenge because more often than not, data requires cleansing before use. Some of the key advantages of adopting big data stores in your IT infrastructure include:

Cost & Reliability – Low cost archival storage of historical data that can be retrieved as new applications are developed. Low cost does not equate to low availability or complexity as most big data stores offer triple storage redundancy and a host of management and monitoring tools to track your data.

Scalability & Elasticity – Big data is not just about storage; rather, it entails efficient data processing that can scale up or down based on your needs without requiring costly dedicated or upfront investment in expensive SAN (storage area network) or data processing servers in your data centers.

Performance – Processing large amounts of data—whether it is structured, semi-structured, or unstructured,—requires a massive amount of storage bandwidth and a considerable amount of processing power. Most big data engines support distributed redundant storage and some form of MapReduce distributed processing job execution engine, delivering unprecedented performance that can scale up or down based on an organization’s changing needs.

Flexibility – Multiple offerings exist today that emphasize one or more areas spanning performance, atomicity, read vs write, and data query flexibility based on an organization’s specific needs. These solutions range from publicly available open source libraries to more professional SaaS (software as a service) based commercial offerings.

Caradigm was able to recently leverage Microsoft’s HDInsight, a service that deploys and provisions Apache Hadoop clusters in the cloud, to establish a software framework designed to manage, analyze, and report on big data – see the illustrative diagram below. Caradigm’s population health and analytics solutions, with the power of HDInsight and other complementary big data solutions, can now be configured to allow for archival storage, unstructured data queries using your choice of NLP (natural language processing), and data analytics capabilities that allow care managers to glean actionable insights from volumes of data.

Caradigm Intelligence Platform Diagram

Make no mistake, big data alone will not be sufficient to address all of the data storage and data management needs of healthcare organizations. Caradigm uses a hybrid model that enables us to augment our near real time transactional and analytics data store with a big data store to deliver the most value to healthcare IT, providers, and patients.

Houston Methodist Addresses Operational Challenges


Post by Azam Husain


Senior Product Manager, Caradigm

Recently, Houston Methodist gave a great webinar on how the organization was able to address core operational challenges in their environment by using the Caradigm™ Provisioning solution. Methodist is a mid-sized provider serving the Houston area.  With 17,000 employees and 4,500 affiliated physicians, the Methodist IT organization had significant operational challenges in managing the lifecycle of user access to clinical systems. Kevin Conway, the Security Infrastructure Manager at Houston Methodist, explained that the operational benefit of Houston Methodist using the Provisioning solution comes down to four key tenants:

  • Lifecycle management of all types of clinician accounts
  • Account standardization and creation/removal consistency
  • IT oversight and data control
  • IT efficiency

Kevin articulated three key scenarios that enabled Houston Methodist to be successful with the implementation:

  1. Clinician Onboarding – Student/resident onboarding events. Before using Provisioning, help desk analysts would struggle to build Active Directory accounts in a reasonable timeframe and application accounts would not be complete for weeks. After implementing Provisioning, students and residents would have the appropriate access on day one.
  2. Hospital Acquisition – Methodist acquired a new hospital. As part of the integration, Methodist needed to create 1,100 accounts for all acquired employees for all target systems. Caradigm Provisioning was used to create this access within one day.
  3. Terminations – Terminating employees from Methodist’s systems was previously inconsistent, and accounts often remained active in key clinical systems inappropriately, creating a compliance problem. After implementing Provisioning, HR would provide a daily feed of inactive users, and Provisioning would immediately de-provision these users from network access and all key clinical systems.

View the webinar to learn more about how Houston Methodist addressed their operational challenges with Caradigm Provisioning.

Houston Methodist Thumbnail

Simplicity – Too Often Missing in Health IT


Post by Ed Barthell, MD


Medical Director, Americas, Caradigm

(sim-plis-i-tee), noun  -  simplicity. The quality or condition of being easy to understand, deal with or use.

Simplicity relates to the burden which a thing puts on someone trying to explain or understand it.  We hear again and again from health care workers that the information systems they are asked to use are not simple and present them with a great burden. For example, according to a recent Johns Hopkins study that closely followed first-year residents at Baltimore’s two large academic medical centers, medical interns spend just 12 percent of their time examining and talking with patients, and more than 40 percent of their time behind a computer.1

We hear the phrase KISS (keep it simple stupid) as a worthy goal. One of my favorite books on effective messaging is from the Heath brothers entitled “Made to Stick”.They propose the acronym SUCCESS and the first S stands for “Simple”. In order for messages to stick they must be simple.

So why is simplicity in health IT so hard to achieve? The simple answer is lots of reasons. 

The healthcare domain is complex and rapidly changing, and many IT vendors try to solve many healthcare problems for many users at the same time. Committees that make purchase decisions tend to judge IT vendors on laundry lists of functionality and pretty colors rather than focusing on apps that solve common problems for specific groups of end users in a simple and elegant way.   Health care is not unique, I’m writing this blog with the word processor application, Microsoft Word, that dominates the market. It contains tons of functionality that neither I (nor probably 99% of end users) take advantage of in our day to day use.

The concept of simplicity is one of our value statements at Caradigm. It underlies our commitment to an open platform plus applications approach.  Ingest data from multiple source systems once, then allow that data to be used by multiple apps for multiple purposes. So apps can simply and elegantly solve problems for specific end user groups that are not well served by generic monolithic systems.

Simplicity in design allows apps to shield end users from complexity and enables providers to focus on what’s important. Specific examples? A surveillance app that automatically calculates a MEWS risk score in near real time on patients throughout a hospital, so that a simple numeric score triggers activation of the rapid response team. Another application that uses machine learning to generate a simple score to reflect likelihood of readmission, so a specific group of end users, in this case discharge planners, can prioritize the assignment of resources. An app that includes a rules engine to help guide care managers to focus on the most important of many tasks that need attention in a given day. The bottom line is health IT applications that are not a burden but instead make it simple for users to do their jobs well.

As our value statement explains, a simple solution is a beautiful thing. In a world of increasing complexity, we take joy in striving for and achieving simple excellence. This is a value we will continue to pursue with vigor.

  1. http://www.hopkinsmedicine.org/news/media/releases/doctors_in_training_spend_very_little_time_at_patient_bedside_study_finds
  2. Heath, Chip and Heath, Dan: Made to Stick: Why Some Ideas Survive and Others Die. Random House, 2007.

Healthcare in Developing Countries


Post by Wilson To, PHD


Product Manager, Caradigm

A small team from Caradigm, including myself, traveled to a number of small communities around Tena, Ecuador and volunteered our services and expertise to Timmy Global Health.  The organization provides direct medical assistance and healthcare services to low-resource communities in the developing world through short-term medical brigades. The goal is to facilitate long-term capacity development and strengthen local health systems.

Last year, volunteers from Caradigm and Microsoft implemented a lightweight electronic health record system across the villages in the Amazon jungle. With a gasoline generator powering a system of networked laptops and wireless router, the team was able to maintain a clinic-wide solution that was used by every volunteer to capture data in electronic formats. From capturing medical history, vital signs, laboratory reports, scribed physician notes, treatment orders, and dispensing instructions – we’re expecting to continue using this tool to provide a completely integrated system to help view longitudinal patient records and improve care quality in the region.

All of the work we all do in the healthcare industry – whether developing population health management systems or providing quality clinical care to patients – ultimately provides a foundation towards improving global health. Much of the same science and best practices that we have in the United States can be modified and applied to low resource communities – a strategy that has yielded great progress in advancing the global health agenda. We have reduced healthcare inequities, reduced child mortality, and eradicated diseases. However, there are still a number of challenges and opportunities to improve the efficiencies in our systems. This adventure helped me understand a different side of healthcare and a better perspective as to what sort of disparities continue to exist today.

I’ll be providing a recap of the trip in a future post, so stay tuned!

Tena

Emerging Themes at HIMSS14


Post by Scott McLeod


Director of Product Marketing, Caradigm

I attended the annual conference of the Health Information Management Systems Society (HIMSS) in Orlando, Florida. Boasting 38,828 attendees, HIMSS is the largest health IT show in the country.

Attending educational sessions and touring the mile-long exhibit floor, one could identify the concerns and priorities of providers and vendors. Key drivers discussed included accountable care, big data and business and clinical intelligence; many vendors made claims of “population health” solutions, if they could address even a small part of a customer’s needs.

I believe the industry will have a bit of a shakeout as the market identifies those vendors that deliver the data control, broad analytics and care solutions needed to succeed in population health.Two emerging themes, in particular, caught my interest—an emphasis on patient engagement and a gap separating population-health pioneers and vendors from the market majority.

In presentations and through Twitter (HIMSS reported 63,839 tweets during the show) attendees discussed various aspects of patient engagement. One thread discussed the provision of more information—presenting quality and cost delivered by organizations—to enable and prompt consumers to make better healthcare decisions. Another thread discussed the need to support family caregivers—the least expensive healthcare providers we have. What struck me is the emerging expectation that consumers must shoulder the twin burdens of making appropriate cost/quality decisions and an increased reliance on self-management or family care for their health. For a population that has looked to providers to make healthcare decisions and deliver care, this change may run into resistance—and, according to a HIMSS Leadership Survey, only one percent of organizations identify “Consumer Healthcare Solutions” as a top IT priority over the next two years.

While signage on the exhibit floor promoted solutions for population health and advanced analytics as the solutions de jour, the topics of many educational sessions communicated a different place on the path to value-based care. A few sessions presented the learning of accountable care organization (ACO) pioneers. Many others, however, discussed other priorities—Meaningful Use, quality measures, health information exchange, electronic health records, HIPAA compliance and ICD-10 transition. This is consistent with the HIMSS Leadership Survey which identified “Achieve Meaningful Use” (25%), “Optimize Use of Current Systems” (19%) and “Complete ICD-10 Conversion” (16%) as three of the top four IT priorities over the next two years. “IT Support for Risk-Based Contracting” placed in a distant sixth place with only 3% of organizations naming it a priority.  

What this shows is that healthcare delivery organizations have more on their plates than a transition to value-based care. When they have the capacity to make that transition, and make the IT investments necessary to support population health, a role will exist for analysts and vendors and pioneers to help guide that transition. As one presenter stated, “the success or failure of a nationwide transition to value- based care hinges on the ability of hundreds of ACOs to learn very quickly from others.”

HIMSS14 Convention Center

Clinical Surveillance from Three Perspectives: Nurse, Patient, Vendor


Post by Jennifer Crandall, RN, BSN


Clinical Analyst, Caradigm

I have been on the nursing and vendor side of patient surveillance for more years than I would like to admit in print. In providing clinical surveillance, we nurses have the potential to minimize negative outcomes by preventing adverse events or deterioration in the patient’s status. As a vendor, it is our job to enable health systems to take enormous amounts of data from multiple sources to gain insight that drive decisions and actions. These actions can ultimately improve patient outcomes on a much larger scale. This large scale effect is one of the reasons I can be so passionate about a job that isn’t at the bedside helping patients. I am frequently asked “do you miss direct patient care?” At this point, I honestly can say I do not. I can help exponentially more patients by helping facilities understand their own data and use it to improve outcomes across all of their patient populations. And if I can make the job of collecting and using data less “painful” and more beneficial for my fellow nurses, then even better.

When I refer to Clinical Surveillance, I am specifically referring to the collection and analysis of health data about a clinical syndrome/condition that has a significant impact on the health of the population. This data is then used to drive decisions and actions.

It wasn’t until a recent experience from the patient side did I really take a look at what all of this surveillance data meant to me personally. In the days leading up to my surgery, in typical nurse fashion, I began to run through every worst case scenario in my mind. Suddenly hospital acquired conditions and infections (HACs) were no longer just statistics and numbers. Drug resistant wound infections, blood clots, air embolisms or a catheter associated infection, were all conditions that could now happen to me. I thought of the phrase “ignorance is bliss.” Wouldn’t it be better if I didn’t know all of the things that could happen to me?

I was determined that I would dictate a great deal of my care to help ensure I escaped the inpatient experience unscathed by any HAC or other negative outcome. While my intentions were strong willed, once I was post op and medicated, preventing a negative event never again crossed my mind during my stay. Most of my cognitive exercise was spent trying to locate the button for the wonderful machine that delivered the elixir of pain relief or making the long eight foot trek across the floor to the rest room.

It wasn’t until after I was home that I realized that despite my good intentions, we must rely heavily upon the care teams and facilities to keep us as safe from these conditions. It was clear that they were using past surveillance data and studies to keep me safe. Because my surgery (per historic data) did contribute a risk factor for Deep Vein Thrombosis (blood clots), they put the SCD (Sequential Compression Device) on my lower legs from post op to discharge. I was given the proper pre-op and post-op antibiotics to ensure my risk of surgical site infection was minimal. They also kept all devices that came in contact with me – such as blood pressure cuffs and stethoscopes – in my room, thus cutting down on my chance of getting a drug resistant infection from other patients.

I have no doubt that my inpatient experience was influenced by clinical surveillance data collected and studied. While it is always very important to be an informed patient engaged in our own care, I do now realize that clinicians and vendors who collect and study this data are the ones who are able to get insight and drive actions for our positive outcomes.

The Buzz of Big Data and Analytics


Post by Dana Alexander, RN, FAAN


Chief Nursing Officer, Caradigm

Big Data can drive improvements in care processes, delivery and management. This is particularly important in today’s growing environment of risk sharing, fixed reimbursement and penalties incurred for not achieving expected quality and outcome benchmarks. We are in a time of value-based business models and value-based care. . . organizations need analytics to manage the higher level of risk and reward. Data that is transformed into meaningful information is essential, but the challenge for organizations is how to harness this data for purposeful use. Data analytics is necessary from the individual level to population health, with the ability to include not only acute care, but settings across the care continuum. The post-acute care settings are of particular focus as a way to manage care in a less costly but quality way, the latter becoming an important consideration in risk sharing and capitated models. The capability to blend and align clinical and financial data into analytics is particularly crucial with the risk-reward models that are in play. What may appear on the surface as reasons for increased costs and utilization of services may not be the true root cause.

As an example, Geisinger Health Plan had long-used clinical and financial data for population health management to identify opportunities for cost and quality improvements. Geisinger has learned that underlying factors such as “behavioral health” issues may be significant drivers in utilization of health care services. This is a proof point that data must be analyzed with an open and creative lens, as never before, to better manage care.

So what types of analytics are necessary? While retrospective reporting and comparative analytics are important, explorative analytics to investigate for root cause analysis can give insight to improve processes and care management. Of great value and of growing importance are analytics that are more near-real time, providing guided analytics translating insight into action. Guided analytics identifies a problem or risk situation such as a predictive risk score that alerts clinicians for awareness and decision making to take action Even more advanced are predictive analytics that allows for manipulation of variables and assumptions based on algorithms and “if and then” situations to forecast a future event and impact on outcomes. Last, prescriptive analytics based upon evidence can guide specific actions in managing individuals and population health. Evidence-based protocols continue to demonstrate the value to achieve quality and cost management goals, and with the right analytics, standards-based protocols can be customized to individuals to effectively achieve desired quality and performance goals.

It’s an exciting time in the world of Big Data. The possibilities for data analytics in healthcare to drive insights and take action are burgeoning. Unlike other industries that have been using analytics in a variety of ways with proven value, healthcare is relatively young in optimizing analytics for clinical and business intelligence to drive performance and forecast new business models. As healthcare systems take on the challenge of managing higher risk, Caradigm solutions are well positioned to support the emerging frontier of Big Data and analytics for today and the future.

Synthesis in Healthcare


Post by Ed Barthell, MD


Medical Director, Americas, Caradigm

(sin-thuh-sis), noun, plural – syntheses.   A complex whole formed by combining.

The world of medicine is full of activities that involve synthesis of data.  Every day doctors and nurses synthesize data from multiple sources as they care for patients.  They aggregate verbal data from the patient, the patient’s family, emergency medical technicians, care managers and other caregivers.  They combine this information with such things as data reports from laboratories, medication lists from pharmacies, notices from insurance companies, and imaging reports from free standing centers.  They synthesize this data with other information such as trends reported in medical studies involving similar patients, news reports and weather information.  And notice all of this data synthesis needs to occur in addition to reviewing the data available from various electronic medical record systems used by multiple providers involved in a given patient’s care.

Of course all of this data, even when available in electronic form, can be very difficult to aggregate, not to mention synthesize in an effective way to drive clinical and administrative decision making.  It is rare to find a health care worker who will not admit wasting time every day searching for data, even those who work with a supposedly comprehensive electronic health record system.  And I’m not proud to admit that in my own clinical practice I saw directly the impact of having to make decisions without knowing what I didn’t know, and subsequently seeing less than optimal outcomes.

I strongly believe the health care industry, like other big complex industries, is best served by a diverse range of systems that support ongoing rapid innovation, rather than dependence on a monolithic software system and a monopolistic business model.  But in a world of diverse systems, interoperability of those systems is needed, and aggregation of data from diverse systems is essential to enable data synthesis for various business purposes.

The Caradigm™ Intelligence Platform (CIP) was created as a robust solution to address this issue.  Now in its third generation as a commercial product, the platform includes state of the art tools that enable efficient ingestion and aggregation of data from multiple diverse source systems.  These tools provide a visual environment for mapping, interpreting and validating data, correctly matching data to patients, deriving calculated fields, applying predictive analytic algorithms, tagging to standardized terminology services, and grouping data into intelligent conceptual entities. 

In turn, CIP surfaces that data through an open application framework for multiple uses, whether end user applications are built by Caradigm, customers or third-party partners.  All of these mechanisms allow CIP to “pre-position” data for further data synthesis by healthcare knowledge workers, so they can do their jobs with high efficiency and high quality.  Diverse data systems supporting innovation, and at the same time aggregating data to support data synthesis tasks by smart people, result in improved outcomes – pretty cool stuff!

Healthcare Analytics


Post by Hamid Al-Azzawe


Vice President of Engineering, Caradigm

The world of healthcare has witnessed a major paradigm shift this past decade instigated by the now widespread use of electronic medical records and medical informatics.  This new world of electronically entered, stored, and exchanged medical information has all but eliminated the numerous shortfalls of the departed world of folder and paper medical records.

This shift is ongoing with new focus on accountable, affordable, high quality and patient-centric healthcare.  Technological advancements in analytics and big data also hold great promise for healthcare organizations as they embark upon this new era of medical services.

Although relatively new to healthcare IT, analytics and big data are well established technologies.  While working in the Bing group at Microsoft, I was in charge of the monitoring team where we built fairly complex subsystems for capturing, collecting and aggregating multiple petabytes of data on a daily basis from upward of a million servers hosted in Microsoft data centers worldwide.  The solutions we built provided near real-time insight from this data, which enabled us to support operational and strategic business decisions and actions.

Realizing the benefits and promise of these analytics and big data in other industries, the media is beginning to create a lot of hype about how these technologies can help transform healthcare.  Established companies, such as IBM, and new startups, have entered the race to turn this hype into reality; however, the intricacies of healthcare make it quite challenging to use these technologies to their full potential.

Caradigm’s analytics vision, described in this short whitepaper, is focused on enabling clinical analysts to derive insight from data and drive actions through decision support.

CIPAnalytics (2)

I firmly believe that by leveraging both analytics and big data in our intelligence platform, and delivering a growing portfolio of population health management and analytics solutions, Caradigm will enable healthcare providers to flourish in the new era of accountable quality healthcare and services.  The race is definitely heating up!

Payer Perspectives from AHIP


Post by Scott McLeod


Director of Product Marketing, Caradigm

I attended the Operations and Technology Forum, sponsored by America’s Health Insurance Plans (AHIP), in Chicago last month.

While targeted at health plans and insurers, the content may hold some lessons for healthcare delivery organizations as they take on more financial risk, operating under reimbursement methodologies that necessitate they think more like payers.

Three themes emerged from the slate of speakers featuring health plans, IT vendors and industry experts:

1. The challenges in launching health insurance exchanges (HIX),

2. The importance of analytics to understand the risk within the populations for which organizations are accountable, and

3. The role of cost and quality transparency in driving patient/consumer behavior. Continue reading