Big Data: Is it Having an Impact on Healthcare?

Published December 8, 2016

There is a common saying in business: you can have it good, fast, or cheap. Pick two.

The saying has always applied to the world of healthcare. But this world is changing before our eyes. Sometime in the not-so-distant future, it may be possible for ordinary Americans to access high-quality, efficient healthcare at a fraction of the current cost.

How is this possible? The answer, in short, is big data.

What is Big Data?

The expression “big data” refers to sets of information that are so large and so complex that they cannot be interpreted, or even measured, by traditional means. For example, in healthcare, information traditionally was gathered by analyzing insurance claims data, or by mining relatively small databases from healthcare systems. It was generally acknowledged that these methods could not deliver an accurate picture of an individual patient’s health, nor could it tell us the best way to deliver services to that individual. The old system gave us the best it could, given its limitations. Data were extracted from spreadsheets that could, at best, relate only the most basic pieces of information, such as age, gender and diagnosis. Big data, by contrast, analyzes many more data points, allowing more appropriate treatment options for patients.

Big data in healthcare refers to the reams of data that are now being generated from wearable devices and mobile apps, as well as from portable monitoring devices and digital health advisors. Devices such as blood pressure monitors and continuous blood sugar monitors are designed to be worn at home and easily used by the patient. Together, these devices and the information they provide has been called the “medical Internet of things (mIoT)”. These devices promise to fill in the blank spaces in the healthcare picture.

The Medical Internet of Things

The Internet of things refers to the network of physical devices embedded with sensors that connect to a network, and from there to the Internet. A famous new example is the refrigerator that collects and transmits data to the end-user alerting them that their milk is about to go bad.

In healthcare, the Internet of things refers to everything from continuous blood sugar monitors for diabetics, to smart wheelchairs. The advantage of these devices is that they deliver quality (good) data in real time (fast), at a fraction of the traditional cost (cheap). In fact, these devices offer a degree of connectedness via data that was simply unattainable in the traditional healthcare delivery system. Patients are now connected to their own health data that is shared with family, providers, and home care companies, completing an integrated data-sharing loop.

The mIoT in Action

Here is an example of how the mIoT might function in the real world: Imagine an elderly woman recovering at home after a hip operation.  She is linked to a combination of several connected services streaming data toward different parties, such as family members, a digital health coach, and her doctors. Her vital signs are checked by a home care nurse and she receives medication reminders from virtual nursing aide. Even her activity level is measured. All the stakeholders involved in her care, particularly the patient herself, are kept up to date in real time as to her progress.

How will we know if the mIoT is improving outcomes and reducing costs? The answer is simple: analyze the data. The beauty of big data and its godchild “predictive analytics”, is that outcomes and cost data can be evaluated in real time, and tweaks can be applied to the system to achieve constant improvement. For patients with diabetes, for example, predictive analytics can crunch through millions of real-time blood sugar measurements and evaluate whether the interventions made in response to these data improves outcomes for the patient.

Big Data and Security Breaches

Data security has been a concern of medical informatics experts for over 30 years. One fallout of the AIDS epidemic in the 1980s was a clamp-down on transmission of protected health data. This movement culminated with congressional passage of the Health Insurance Portability and Accountability Act (HIPAA) of 1996. The advent of the mIoT has revitalized the concerns that HIPAA was meant to address.

The Internet has already generated at least two solutions to security concerns. The first is encryption. Encrypted networks in the banking and e-commerce communities have largely, though not completely secured transfers of money. The second solution is decentralization of data, i.e., individual patient control over what data is available and who can access it.

The Future

Will big data and the mIoT have a positive impact on healthcare? Time will tell, but the trend appears to be unstoppable. A published study of a diabetes prevention study demonstrates that the concept is solid, and that big data is working today. It is incumbent upon us to harness the power of big data and the medical Internet of things in order to achieve the goal of good, fast and cheap healthcare.