It is common knowledge now that the amount of data worldwide is growing exponentially across all industries, and healthcare is certainly one of them. I came across a few figures recently that convey the pace of data growth pretty well: Ninety percent of the world’s data is less than two years old, total data collected will grow by 40% next year, and that per IBM’s estimates 2.5 quintillion bytes of new data is generated each day. For those who are wondering, a quintillion is 10¹⁸ bytes. Major contributors to data volumes include imaging data, social media data, geocoding data, sensor data, web traffic data and RFID data.

There are various definitions of big data out there and no tipping point in terms of size beyond which data is considered “Big”. I particularly like Gartner’s definition that defines big data as “high-volume, high-velocity, high-variety information assets that require new forms of processing to enable enhanced decision making, insight discovery and process optimization. Industries like financial services, insurance, retail, telecom and transportation will likely be among the first to leverage big data. The healthcare industry is traditionally slow in adopting new technologies, but the potential business value is tremendous.

Until recently, the interests of payers, providers and patients were not aligned, even though they were heavily intertwined. That being said, the healthcare industry is going through its own evolution at a pace not seen before. Accountable care organizations (ACOs) are here to stay and forcing stakeholders to work closer together with common objectives. Most executives in the business agree that leveraging data assets is a key part of the future success for ACOs. It is interesting to look at the confluence of the big data wave and the ACO wave to understand how the healthcare industry could commercialize the value of big data in the coming few years. There are about a dozen potential big data applications in healthcare that I have come across, from which I picked three based on my personal judgment. I feel these three applications have the most business promise in the near term; these are continuing to evolve quickly as forward thinking providers and payers get beyond the initial proof of concepts and put these ideas into action.

1. Member Retention and Patient Engagement.  Payers are under intense pressure to retain their membership and this problem is expected to intensify significantly when the health insurance exchanges are rolled out. Providers are already figuring out new ways to survive under the ACO model. There is potential to leverage currently untapped data assets such as web clickstream data, medical device data, CRM data and social media data in a complementary manner to the traditional data sets such as claims, eligibility and clinical. For example, by mining new information from these sparse but valuable data sets, Payers can identify members that are at highest risk of leaving their current health plan and put in place a proactive mitigation strategy. Providers can use the same data to create a patient outreach program that is more relevant and timely.

2. Remote monitoring of patients. There is much promise in monitoring patients with chronic conditions such as diabetes and heart disease by analyzing data collected at home using a variety of connected devices. Such devices can easily upload relevant medical readings to a remote server for constant monitoring. Big data applications can scan this data on a near real time basis and immediately identify needs for patient intervention.

3. Comparative Effectiveness and Cohort Analysis. Outcomes analysis uses cohorts of patients to determine treatments most likely to benefit patients with a combination of specified characteristics. Healthcare practices often vary widely across the country and thorough analysis of large data sets can bring more uniformity to the treatment process, likely improving outcomes. There is also potential for cost reduction if big data can be used to eliminate procedures and lab work that results in unnecessary overutilization.

My verdict on big data is that there is tremendous potential in this space. However, the healthcare industry is just beginning to scratch the surface. Right now, most healthcare analytics applications are not true “big data” applications and the term is often wrongly used. It will be interesting to see how this space evolves and I will be posting additional thoughts as I work with healthcare practitioners going forward.