Dr. Sawad Thotathil, Senior Director of Healthcare at Persistent Systems

home

authors

dr-sawad-thotathil-senior-director-of-healthcare-at-persistent-systems

Medical knowledge spans the history of humankind. From the first herbal treatments and natural remedies to the nanobots and genetic therapies of today, there has always been a process by which new medical knowledge is evaluated, its worth defined, and its application spread to other healthcare practitioners.

Hundreds of years ago, this process was defined more by first-hand observation and intuition then by any kind of objective analysis. Today, objective analysis increases at an exponential rate through wide-ranging data-collection capabilities, now processed through artificial intelligence (AI).

But while medicine becomes more objective, the diffusion of new information still has its challenges. 

 shows that it takes up to 17 years for medical evidence to reach routine clinical practice. The fact is that there are some blind spots in our data-driven world, and there are some underlying weaknesses in healthcare delivery that can be traced to deficiencies in the adoption, distribution, and application of emerging knowledge.

How Knowledge Moves Within Systems of Care

Let’s look at how knowledge dissemination has evolved over the years. Knowledge, as it becomes more complex and more specialized, further distributes among more and more participants in the value chain. The creation of new medical knowledge results in the emergence of new types of professionals to take on the new work.

In centuries past, one care practitioner would meet with a patient, evaluate their condition, and prescribe or perform a treatment. Now, a patient may meet with general physicians, cardiologists, nurses, physiotherapists and healthcare navigators. Add to this a huge supply chain of healthcare service providers, including everything from patient-centered medical homes to hospitals.

In this specialized state, an effective care process happens because a patient’s unique needs were matched to the body of medical knowledge relating to the patient’s disease. This match only works if data about the patient’s status is easily accessible and actionable.

But this is a challenge today, as data tends to proliferate in 2 dimensions. The first dimension consists of data generated at all points of care within the system of care. This takes on broader perspectives such as that of population health. The second dimension is more introspective, with data generated by the patient’s own condition with a continuously proliferating array of lab results, EKG, blood pressure readings, genomic tests, etc.

But these two separate dimensions must align into 1 point-of-view that is easily accessible at the moment of care. Data generated anywhere in the system must be available when it is needed and in whichever workflow it is needed in.

So, what does this mean when designing the algorithms behind AI and the predictive modeling that it enables? The best answer actually comes from an examination of the success driving a different industry: transportation and logistics. A parallel can be drawn between the movement of knowledge and how appropriate packaging facilitates movement of a physical product.

A quick history lesson: Until the 1950s, the shipping industry was constrained by long ship loading/unloading times, cumbersome transfers and intensive labor at marine ports. 

, an entrepreneur, believed that the freight business could be made more efficient only if goods could be moved seamlessly from the source to the destination, and had his engineers design a container that could be filled up at source, carried by rail or road and then transferred directly onto ships. It enabled ships to be unloaded and loaded within a few hours, and was eventually universally adopted to make cargo movement less about ports and more about how fast the cargo can be moved from source to destination.

This design principle can be applied to both medical knowledge and data. Medical best practices, such as clinical guidelines, decision trees and checklists can be packaged as simple or complex algorithms and kept relatively independent of the contexts within which they will be used. This separation essentially allows knowledge to be distributed based on the needs of the market. This relative independence of the algorithm could result in different implementations—as a drop down in an EMR, a decision process in a population health app, or a patient portal&mdash;but the key point is that this availability enables managers to insert knowledge into the appropriate workflow where it can yield the best results for their patients.

Think about how a decision support algorithm in chronic wound care could evolve. Management decisions are typically based on visual examination by trained doctors or APRNs at wound clinics. With the availability of automated wound analysis tools to objectively assess wound healing, this decision making task could shift to home visit nurses or skilled nursing facilities. These objective tools could be backed up by decision tree algorithms at different points in the care process. The decision trees act like treatment recipes for non-physician care practitioners to follow.

The core principle at work here is the acceleration of the adoption, distribution, and application of medical knowledge by bringing data and algorithms together in the right workflow, at the right time and to the right user. If AI is built and operated according to this principle, then it will yield exponential benefits in both outcomes and efficiencies of care.

3 Keys to Cost-Conscious Population Health Management

What Tesla's Electric Truck Can Teach Us About Healthcare's Digital Transformation&#8203;

Personalizing Healthcare: The Netflix Approach

Dr. Sawad Thotathil, clinician and senior director, Persistent Systems, directs product strategy and management for the company’s Digital Healthcare group, helping health systems achieve their digital transformation goals.

Articles

There are blind spots in our data-driven world, and there are some underlying weaknesses in healthcare delivery that can be traced to deficiencies in the adoption and application of new knowledge.
 

Perfecting the Knowledge Flow for Medical AI

In the long and ongoing march toward electronic medical records (EMR) adoption, clinicians themselves have often been cited as a major impediment. It has been said that some physicians, especially older ones, were resistant to change because they are technological laggards. A closer examination calls this idea into question.

Physicians are innovators when it comes to advanced concepts, like care personalization. They have the clinical desire, necessary leadership, and enthusiasm to push adoption of personalized care forward: It is now the technology itself that is struggling to keep pace.

Clinicians already practice a type of personalization from the bedside, so they understand its potential value in the wider coordination of the care environment and ultimately in achieving better population health. But why does healthcare still struggle to tailor care to the patient at a personal level?

The reason is because the healthcare industry hasn’t moved far past the stage of getting EMRs up and running. Reams of data collected within the EMR, as well as the new data streams coming online, could be used to create precisely targeted therapies.

Think about data from emerging areas such as genomics, digital exhausts generated from doctor/patient encounters, a patient’s activities on social media, publicly available information about the environment where a patient lives, and the continually evolving range of wearables. The problem is that there is no strategic framework to translate this data into a better care experience for the patient. There is no functioning operating system for clinical care that is entirely geared toward personalized care.

However, there is another industry that has built a roadmap to follow in building such a system.

In 2006, Netflix wanted to better target video content to each member of its audience. The company established a 

 for a better algorithm that could predict what customers wanted to watch. As the contest moved forward, so did Netflix’s business model, with streaming video dramatically increasing the amount of data available on consumers' viewing preferences. Data points—like how long someone would play a video before stopping or what content was browsed before making a movie selection&mdash;all could provide a window into how customers were interacting with the streaming service.

What Netflix did with this data was to pair it with an immense number of sub-genre categorizations of movies (

 to be exact) that they developed by analyzing and breaking movies into attributes of interest that exposed the likes and dislikes of the customers. Here are just a few 

 to show how specific these categories became:

Sentimental Set-in-Europe Dramas from the 1970s

Just like biomedical research is trying to find out what type of gene or protein is related to a particular manifestation of disease in a patient, Netflix has invented a mechanism to squeeze out the most minute details about customers from their interactions with the service and map those actions in great depth.

Now, let’s apply this same level of specificity and personalization to a healthcare setting.

Imagine a platform that uses genomics and genetic profile testing to guide cancer treatment tailored to an individual patient’s unique life. It looks further than common population markers, such as age and lifestyle, but still considers those factors as part of a more holistic picture.

The difference is that now missing pieces of the puzzle are being filled in. The 

Cleveland Clinic considers personalized healthcare

 as any biologic information that helps predict risk for disease or how a patient will respond to treatments. This type of system can collect all of that data and put it into one comprehensive, on-demand view accessible to both clinicians and patients.

Healthcare needs a Netflix-like framework for personalizing care, and there is no reason to put off its development. The airline industry did not wait for the internet or the mobile phone to arrive before they started to personalize their customer engagement with frequent flier programs. They simply made use of all the data already available. And when Netflix launched its new platform, some of its content categories were still empty.

Similarly, with the data that’s available right at this moment, a functioning operating system for clinical care can be developed that is entirely geared toward personalized care. Such flexibility towards evolving data types in a clinical operating system will allow new capabilities that emerge from precision medicine research to move easily into clinical practice.

Health system leaders should support clinicians’ intent for personalization by providing resources and technology platforms, where clinicians and data scientists can work together in crafting algorithms that segment patients and their needs. These algorithms, when properly tested and proven, should be embedded into workflows where patient interactions, response, and compliance to treatment is generating easily utilizable data that can be captured for continuous improvement to that patient’s unique treatment plan. Just like the idea of a physical Blockbuster store now seems antiquated, our current models of care may seem patently obsolete just a few years from now.

Dr. Sawad Thotathil, clinician and senior director, 

, directs product strategy and management for the company’s Digital Healthcare group, helping health systems achieve their digital transformation goals.

Lessons that healthcare can take from the streaming video juggernaut as it tries to personalize care for every patient.

In the new age of analytics and algorithms, clinicians could be forgiven for wondering if a few decades down the road they might find themselves out of a job. One can imagine buggy manufacturers feeling the same way as the horseless carriage started to gain momentum. But there is one key piece missing from this comparison. And it’s critical to understanding how the digital transformation revolution will energize, not demobilize, a new generation of clinicians with capabilities that were only dreamed of a few short years ago.

To understand this transformation, it’s helpful to look at the new electric truck design just released by 

. The idea of automated, self-driving trucks has been understandably intimidating, with 

 that run the gamut from roadway safety to the elimination of trucking jobs. But this new approach, however, doesn't mean we replace human drivers with total AI control. The design of the new truck is focused on empowering the driver, with features including automatic braking, lane-keeping, and lane departure warnings.

 from Tesla and other companies include “platooning,” where a driver in the lead truck controls a line of vehicles augmented with AI, or a “

” concept where a driver sits comfortably in a simulator to control the truck at a distance.

Within all of these ideas, you can find one common denominator: A driver spending less on a monotonous drive, allowing their human experiences, intuition, and decision-making capabilities to be maximized when they are needed most in ways where algorithms might not be able to predict.

From Trucking to the Exam Room: How the Core Value of Digital Transformation Stays the Same

How does all of this apply to the analytics revolution in healthcare? There’s no denying that a part of the clinician’s profession is slowly, but precisely, being replaced by algorithms. No medical professional is ‘safe’ from some level of automation. Robotics will most likely transform even surgical applications that require a high level of manual dexterity at some point.

But that doesn't mean that clinicians’ knowledge, experience, intuition, and expertise are no longer needed. Analytics will augment their ability to spread their wisdom with far greater reach to far more patients than was ever before possible. Medical professionals engage in continuous exploration and research focused on improving outcomes, with the hope of transforming their research findings into routine practice. The analytics-enhanced future of clinicians may become centered on managing and improving the algorithms that ultimately take over the more repetitive and mundane work in healthcare.

What this affords is the ability for clinicians to provide direct care to the most complex patient cases and let cognitive algorithms handle the more routine cases.

Digital Transformation Means Healthcare Delivery at a Transformative Scale

Just as automated trucks will redefine the scale of the reach and ability of drivers, a reallocation of cognitive resources in healthcare will help scale the impact and performance of every clinician. The result is reduced cost of care delivery and the ability to fill in some of the emerging healthcare 

In this perspective, the digital transformation of healthcare occurs at two levels. The first is the standardized execution of clinical knowledge through algorithms; and the second is the way algorithms become the tool for efficient redistribution of work across the system.

Advancements in healthcare analytics can keep healthcare organizations one step ahead of economic pressures from payers, regulations and the ever-evolving demands of the healthcare consumer. Yet, healthcare AI should never be viewed as a Darwinian replacement for the nuanced wisdom of the healthcare professional. In reality, analytics is both a survival and a growth strategy as healthcare moves into a newer and stronger digital age.

Dr. Sawad Thotathil, Clinician and Senior Director, 

Fear not the rise of the algorithms, Sawad Thotathil of Persistent Systems writes.

The US is facing a steep rise in healthcare costs. Despite having one of the 

highest per capita healthcare spending expenditure

 in the world, the quality of care is not on par with some other developed nations.

Rising healthcare costs have prompted the government to enforce policies that require healthcare providers to deliver value-based care, including personalized care that can help reduce readmission rates and improve overall health. However, to do that, providers need to better understand their patients and their conditions, as well as the likelihood of their contracting specific diseases and recommending preventive measures that would work for them.

Population health management (PHM) makes use of new technologies that enable practitioners to gather relevant data related to a patient in different care settings and helps them use data analytics to derive meaningful insights about the patient’s condition. This approach has been proven to help practitioners provide the right treatment to improve patient conditions in the most affordable way without compromising quality.

Below are 3 key components of a PHM approach to affordable and relevant patient care:

To know a patient population, the first step is to gather all relevant data related to them at all points to get a complete clinical picture. This data should include both clinical data and patient-generated health data (PGHD). Clinical data fall into 6 major categories:

PGHD, as the name suggests, is any health data generated by the patient. It may include biometric data, wearables data, treatment history, symptoms and health history. With the rise of digital apps, gathering of PGHD is becoming easier to accomplish.

 published by IQVIA, 40 percent of the digital health apps downloaded today are focused on managing a health condition, and about 55 percent of these health apps use sensor data driven by wearables like Fitbit and Jawbone. All of this means that there will be more data to leverage, manage and analyze to inform a patient’s medical/health condition.

Data sharing is important for the success of any PHM program since it allows providers to eliminate process silos and see data holistically, which can result in actionable insights.

Ensuring interoperability across systems enables data sharing with other stakeholders to deliver care across the continuum. For example, when caring for a patient in a home setting, the attending nurse should have access to the same actionable insights or comprehensive view of the patient data as the doctor sitting far away in a hospital.

A great example of what interoperability can achieve at a regional level is a major initiative by the Louisiana Public Health Institute (LPHI). The Institute conducts clinical research using patient data pulled from different health systems through 

, a partnership between LPHI and other regional healthcare entities that provides access to data on more than 3 million patients across Louisiana and Texas. Through a unique infrastructure for engagement between clinicians, researchers, and patients, REACHnet allows partnering entities to conduct clinic-based research activities while keeping patients engaged in their treatment over a longer period of time. The results of these research projects, along with holistic patient health data, enable providers to deliver improved care to their patients.

The other important factor to consider when sharing data is the importance of data governance to ensure the quality of data gathered and to avoid duplicate, or even worse, incomplete information.

For the success of PHM, applying predictive analytics enables providers to make the most of the data and better target patient populations likely to develop a serious medical condition. Predicting these conditions allows them to make timely interventions, and ultimately reduce healthcare costs.

 — an initiative to make communities they serve 20 percent healthier by 2020. Humana identified the populations living with food insecurity, social isolation, and physical inactivity. Working closely with physicians, 2 of the 4 Bold Gold communities witnessed a substantial decrease in "unhealthy days” in 2016, surpassing 2017 projections. Through a holistic approach to PHM, Humana has been able to improve living conditions of its members.

By aggregating patient data across multiple sources, analyzing that data and sharing it to make informed treatment decisions, providers in the U.S. are elevating the standards of care and improving financial outcomes.

Dr. Sawad Thotathil is a clinician and Senior Director of Healthcare at 

. He directs product strategy and management for the company’s Digital Healthcare group, helping health systems achieve their digital transformation goals.

Taking a holistic view of a patient population can help deliver more personalized care and a preventive approach to reducing healthcare costs.