How We Look at Big Data and Bundled Payments in Healthcare

“Big Data” is a buzz word in a lot of industries right now, so what do we mean when we talk about big data in healthcare? We know that big data is, well, big – usually very, very big. When we talk about big data in any context we’re usually discussing something that is more than a single person can analyze and parse. In fact, it might need a team or specialized equipment,and is assembled from multiple data sources.

Of course, those of us in healthcare, and especially healthcare finance, know that more and more, big data is being used to guide health policy decisions.  Our challenge is how to leverage big data as a tool to support decision making.  A very pertinent and timely example is measuring bundled payments as payment trends continue toward multiple variations of bundled payment approaches. 

Bundled payments, put simply, are a method of shifting payment risk between healthcare providers and healthcare consumers, usually third party payers.   Unless a hospital arranges for risk-sharing, it is responsible for controlling costs of care, even in settings over which they do not have any direct control.  Bundled payment methodologies range from individual service arrangements, where each specific service provided is paid separately, to episodic and capitation methodologies.  Service arrangements and encounter arrangements can often be measured within an organization, however assessing and defining costs associated with episodic payments clearly falls into the area of big data.  An episodic payment is made for all encounters associated with a defined episode of care across all providers.

Using big data analytics from Medicare data files we can measure the cost as the sum of all providers involved in providing care.   Once we understand the variation and driving factors of cost we can better understand how much individual providers have control over those costs and whether the reviewed product line or service is likely to be profitable, will require subsidization, or is in a space to consider limiting or exiting.

Clearly big data isn’t going anywhere, so it is imperative that we find ways to access and analyze big data to develop and support our strategies and decision making.  If you have questions, let us know at 888.779.5663. We discuss this issue (with examples) in our whitepaper here.