In 2011 according to reports, data from the U.S. healthcare system alone exceeded 150 exabytes. With the current growth rate, this will soon reach the zettabyte-scale. A value-focused organization based in California has more than 9 million members and is believed to have between 27-44 petabytes of potentially rich data from varied sources.
It’s clear that the healthcare industry every year generates a significant amount of data from medication administration records, electronic health records, laboratory and information systems, and other various sources. Now with added adoption of eHealth and mHealth wearable technologies, the need for capturing and leveraging data to support a wide range of healthcare functions including medical decision support, disease tracking & monitoring, and most importantly population health management, is only going to increase.
Why does population health management need big data?
The notion of Population Health Management (PHM) is now not a new one. It evolved from an idea and is now a clinical discipline that works on developing and continually refining measures to improve the health status of populations. To achieve true PHM, we cannot limit ourselves just to the traditional way of using claims and clinical data by itself and predicting the valuable, necessary insights to create a better healthcare world.
There are numerous ways big data can be leveraged to achieve better clinical and financial outcomes:
Even with mountains of data acting as raw material for informed and intelligent decision-support, a few challenges remain. One of the biggest challenges is the storage of medical data incoming from disparate silos across different states, hospitals, and administrative departments. Considering the great potential that big data holds, bridging the gaps and developing the technology to capitalize on data to deliver improved and informed outcomes could be revolutionary.
Applying big data to capture value
Leveraging data connections and integrating disparate data sets is especially important in the transition healthcare organization is already underway: value-based reimbursement. It’s crucial to incorporate a big data strategy to harness the value big data holds and deliver value in care. To start with, some of the measures that could be put into action are –
1) Creating a Holistic patient-centered dataset – Most of the healthcare data is stored in disparate silos is almost inaccessible and unanalyzed. For using these value-rich data elements, they need to be first brought together and integrated into a standard, unified patient-level dataset in a very meaningful way. The goal of creating these patient-level datasets by aggregating data from different sources is to be effectively able to visualize and document any interaction a physician has had with the patient.
2) Data-Driven Interventions – With an efficient big data strategy in place, providers can work on attending to patients facing any immediate health or financial risk, and constantly monitor the outcomes of their interventions. Risk stratification, quality reporting, and outcome monitoring with incorporated analytics can help identify patients and segment them by patterns and similar interactions.
3) Stratifying the population – After segmenting population by different attributes, the next process is to assign a risk score to each patient and segment the population into high-need, medium-risk, and low-risk patients. After risk stratification, providers can focus on high-need patients and make continuous efforts to lower the risk score for the population over the time.
4) Waste and Care Variability Reduction – One of the major factors behind the ever-increasing healthcare cost is patient variation and the waste associated with it. Variation and Waste reduction remained out of focus till now, but with value-based reimbursement being the agenda of healthcare, patient variation and the patient waste associated with current healthcare delivery models have gained traction and so have the technologies associated with it.
5) Predictive Analytics – Predictive analytics in any field is to predict the unknown and healthcare is no different. With well-structured stratification of patients based on risk, providers can employ logical case-based analysis and systematic arrangement of the population through drilled-down patient-focused datasets to gain close and near observations.
6) Change with time – The amount of data will only increase as time passes. Healthcare would have to be updated in their methods of data storing, analytics, research, and experimentation of how can we exploit the current technology to constantly raise the bar.
The Road Ahead
Since there is so much data out there, creating an analytics system well-suited for big data, providers should understand the complexity they face and find the right data sets to aggregate and fuse together to deep dive into the problems. Big data has the potential to change the way care is delivered, to say the least. In addition to improving clinical and financial outcomes, a well-employed big data strategy can save millions of dollars and lives. Promoting and incorporating big data into healthcare is the need of the hour and it will not only bring new opportunities on the front but will lead the industry into a new, data-driven era.
For more updates , Subscribe
If you want to see our efforts in the area , schedule a quick demo