Big Data Will Make a Big Difference in Saving Lives

Leveraging data in the healthcare sector to save lives and reduce costs. What will it take? I havent tried to cover every detail but point out a couple of the key elements that will drive new awareness.

Big data doesnt necessarily mean good data. As with all data, big data needs to be extracted and validated, transformed and normalized. And once its in a usable form, it needs analytics, so it can be scrutinized and understood.

Actionable Big Data is possible in health care because of the vast amount of claims, clinical and additional data generated and shared through practice management, billing systems, electronic medical records and data warehouses. Lets take a closer look at the data sources. Before delving in to the data sources let me plug one more time the need for Ontology. Is there any Domain that has a broader mix of terminologies that need to be comprehended, or a greater need for system integration?

Over the years, Doctors developed their own specialized languages and lexicons to help them store and communicate general medical knowledge and patient-related information efficiently. The promise of a global standard for Electronic Health Care Records is still years away. As we know, medical information systems need to be able to communicate complex and detailed medical data securely and efficiently. This is obviously a difficult task and requires a profound analysis of the structure and the concepts of medical terminologies. But while this task sounds daunting it can be achieved by constructing medical domain ontologies for representing medical terminology systems. The most significant benefit that ontologies may bring to healthcare systems is their ability to support the indispensable integration of knowledge and data.

Adding Ontology to the Mix

An ontology is a controlled structured vocabulary to support annotation of data.

By tagging data with meaningful labels which together form an ontology enhancing semantic search. This makes it easier to connect and leverage the right data across the healthcare spectrum without the need to integrate disparate data silos.

As you read through what is possible by utilizing big data, analytics, ontology and the Internet of Things, it will become obvious we can save millions of lives globally. In addition, think of the possibilities to eliminate healthcare waste. Numerous studies have identified that over $300M annually is wasted in providing unnecessary care. This statistic only includes the United States.

Claims Data

The most commonly available form of health care data, claims data, provides great insight for population health discovery and research studies. Claims data helps answer questions about:

• Demographics of population who receive care;
• Care settings they use;
• Categories and types of care delivered.

Also, since claims datas main purpose is for reimbursement, the data is an excellent source for chronicling the cost of care. Claims data helps organizations find retrospective patterns in care. Because most providers submit claims for reimbursement, claims data is useful for seeing the spectrum of care received by a particular patient.

For all its benefits, this data type is hindered by a number of factors. Doctors must submit claims to insurers if they want to be paid for their services. So, only the data necessary to be paid is available on the claim. And since claims are often processed and paid 30 to 60 days after the service is provided, the data is dated by the time it is available. Additionally, the data can be very fragmented caused by patients switching insurance companies. Maybe this will become less of an issue with Obama Care, time will tell.

(EMR) Electronic Medical Records

Electronic Medical Records can provide a complete picture of a patients conditions, with the data being current and not nearly as fragmented.

Health care providers can now access electronic data to round out the patient picture and help improve care quality and patient satisfaction, while controlling costs. Such clinical data is found in Electronic Medical Records (EMRs). The data has been collected for years, but until the EMR came into wider use, most clinical data was bound by paper clips, stored in manila folders, stashed onto shelves, and, over time, locked in storage. Electronic Medical Records provide a rich store of data available for analysis and are found throughout the care process, including the emergency department, hospital inpatient records, physical therapy, radiology areas and outpatient care. And unlike claims data, reflecting how care is actually delivered.

While EMR data can contain much detail, it isnt a perfect data solution. Most EMRs include both structured and unstructured data. Structured data, normally data that is marked, labeled or tagged so that it can be identified and made actionable, makes up only about 20 percent of the EMR. The balance of the data, often found in clinical notes and other free-form text, is unstructured. Such data must be validated, normalized, cleaned and extracted.

Even highly usable clinical data has its limitations though. In fact, the ultimate goal is the integration of clinical and claims data. Marrying claims and clinical data provides distinct advantages for providers, bringing in the strengths of each while working to overcome the limitations of each set. The convergence of these data sets is one many are aspiring to as the next big step. But in taking on risk, its something that needs to be done now rather than later in order to provide accuracy. Further, weaving both socio-demographic and care management data into both sets makes the picture even more powerful. This is exactly where a domain specific ontology can play a key role. The time for interoperability has come. EMR and Healthcare Information Exchange (HIE) represent two of the top technologies now being used to bring more connectivity to health care data.

As we start looking at the key benefits derived by using an Ontology to help define the domain specific vocabulary, we can start seeing the ability to better analyze and cross reference the data. The predictive analytics allow organizations to quickly identify the chronic illnesses most prevalent in their patient population, determine which conditions are costliest to treat and which doctors do the best job managing patient care.

The Benefits of Adding a New Real-time Data Source

With the onslaught of wearable smart sensors coming to market almost daily, monitoring ones vitals in real-time becomes a reality. Fitbit and many others introduced activity and wellness trackers a number of years ago. Going beyond fitness tracking, Sensogram introduces a revolutionary idea that will help users and athletes of all levels to monitor their vital signs and receive valuable information for a healthy and effective workout. Their biosensor technology is more than just a wearable device; it is a solution that combines state of the art data collection with sophisticated big data analytics to analyze and correlate collected information. The device reads, transmits, and stores the following parameters:

• Blood pressure
• Respiration rate
• Oxygen saturation
• Heart rate
• Hydration
• Galvanic skin response
• EKG
• Activity level
• Distance
• Speed
• Fall detection
• Step count
• Calories burned
• Cadence
• Acceleration
• Location and altitude (data acquired from a smartphone or tablet)

Having the vitals simultaneously monitored along with activity parameters allows for a fuller understanding (more breadth and depth) of how ones individual body works, and what needs to be done to reach athletic, weight loss, and other activity-related goals.

Because the device is capturing all the vitals listed above, in an integrated fashion, the big data analytics can correlate and predict the likelihood of potential life-threatening events in real time even though no single vital by itself would have created an alert.

There are many dynamics impacting the cost of healthcare. Medicare is moving to accountability measures; insurance costs keep increasing; more complex medical procedures are developed every day and, most importantly, we have a longer living aging population. Because sensors improve health monitoring and disease management overall, they create better quality of care which will help offset rising cost factors.

It is good to see the attention being given by leading industry analysts to the potential of remote health monitoring technologies, sensor-based devices that gather a patient’s health data and relay it to a care provider or even a loved one. I suggest it is a combination of the IoT, smart phones, smart sensors, big data analytics, domain-specific ontologies that have moved the needle from mere potential to reality.

By each of us fully grasping how the availability of real-time vital data can dramatically reduce the development of a chronic illness, like heart disease, heart attacks and strokes. By identifying conditions like hypertension early we can in many cases prevent that chronic condition which is the leading cause of these chronic illnesses. Here are five conditions that could benefit from real-time monitoring:

Why Does Knowing You Have Hypertension Matter?

High blood pressure increases your risk for dangerous health conditions:

• First heart attack: About 7 of every 10 people having their first heart attack have high blood pressure.
• First stroke: About 8 of every 10 people having their first stroke have high blood pressure.
• Chronic (long-lasting) heart failure: About 7 of every 10 people with chronic heart failure have high blood pressure.
• Kidney disease is also a major risk factor for high blood pressure.

Those factors make high blood pressure a topic of interest for Medicare, Medicaid, Medicare Advantage plans, and other private health plans. Health insurers and providers increasingly are focusing their efforts to reduce health care costs for people with chronic conditions, who account for a large portion of health costs and spending. The Centers for Medicare and Medicaid Services (CMS) estimates Medicare beneficiaries with two or more chronic conditions accounted for 93 percent of Medicare spending in 2011, or about $276 billion. Several million of those newly insured under the Affordable Care Act (ACA) starting in 2014 likely have undiagnosed, untreated, or poorly managed hypertension.

Summarizing the benefits of monitoring and analyzing your key vitals

• Improve our long-term health
• Dramatically reduce chronic illnesses
• Reduce the sky-rocketing costs of health care.

Importance of Blood Pressure and Heart Rate Monitoring

First, blood pressure and heart rate are not the same. Quite simply blood pressure is the force the heart exerts against the walls of arteries as it pumps the blood out to the body and heart rate is the number of times your heart beats per minute.

Blood pressure is possibly the most important of all the vitals because the higher your blood pressure is, the higher your risk of health problems in the future.

• If your blood pressure is high, it is putting extra strain on your arteries and on your heart. Over time, this strain can cause the arteries to become to become thicker and less flexible, or to become weaker. If your arteries become thicker and less flexible, they will become narrower, making them more likely to become clogged up. If an artery becomes completely clogged up (known as a clot), this can lead to a heart attack or stroke, kidney disease or dementia.
• There’s a common misconception that people with high blood pressure, also called HBP or hypertension, will experience symptoms such as nervousness, sweating, difficulty sleeping or facial flushing. The truth is that HBP is largely a symptomless condition. If you ignore your blood pressure because you think symptoms will alert you to the problem, you are taking a dangerous chance with your life. Everybody needs to know their blood pressure numbers, and everyone needs to prevent high blood pressure from developing.
• The good news is that there are many ways in which you can lower your blood pressure and put less strain on your arteries and heart if you know the problem exits. The startling facts are that over 2 billion people in the world have high blood pressure and only 46% of them know they have it. Maybe with an easier non evasive way to monitor and measure blood pressure the other 54% or 1.2 billion affected people can do something about it. It is known as the silent killer because there are not usually any symptoms until its too late.

Respiration Monitoring Can Help Predict Health Crises

• Abnormal respiratory rates and changes in respiratory rate are a broad indicator of major physiological instability, and in many cases, respiratory rate is one of the earliest indicators of this instability.
• Regular documentation of the respiratory rate may assist in identifying patients at risk of serious adverse events such as cardiac arrest and unplanned ICU admission, with high levels of specificity.
• Respiratory rate performs at least as accurately in identifying patients at risk of these adverse events as pulse rate and the systolic blood pressure.
• In fact, abnormal respiratory rate is one of the best independent predictors of cardiac arrest.
• A respiratory rate of greater than 24 breaths per minute is able to identify approximately 50% of patients at risk of serious adverse events with 95% specificity.
• A raised respiratory rate is associated with life-threatening conditions such as shock and cardiac failure.

Importance of Oxygen Saturation Monitoring

To really understand the importance of oxygen saturation monitoring, consider this:

• The brain uses about three times as much oxygen as muscles in the body do.
• Brain cells are very sensitive to decreases in oxygen levels and dont survive or function well very long without it.
• The brain is made up of very special cells called neurons, and each of them can do the job of any of the other neurons in the brain. That process gives us the ability to relearn most functions using different parts of the brain. So if we have an injury or a sick part of our brain, we can teach our brain to restructure and control body parts and function with other areas of the brain. That is called brain plasticity.

Importance of Body Temperature Monitoring

Monitoring body temperature is important because it keeps our bodies working and functioning properly. Without it, we would all die. If your body temperature falls too low then respiration happens too slowly and you die. If your body temperature goes too high then the enzymes in your blood denature so they can’t catalyze respiration and other reactions that go on around your body.

There are three main factors which can affect your core body temperature, sending it up or down.

• The energy produced by your muscles during exercise can make your core temperature go up;
• Fevers caused by disease can make your core temperature go up;
• The external temperature rising or falling can make your core temperature go up or down.