“Big data” and “big data analytics” are some of the hottest buzzwords around at the moment, and big data has been likened to teenage sex: everyone talks about it, nobody really knows how to do it, everyone thinks everyone else is doing it, so everyone claims they are doing it. Healthcare industry is not exempt from being shaken by big data. McKinsey estimated that new clinical pathways enabled by big data analytics could create cost savings between $300 – $450 billion in the US only. Estimates of the fiscal impacts of exploiting big data analytics in healthcare seem to typically range in hundreds of billions of dollars. In addition to the financial benefits, big data also promises to deliver better care outcomes.
Big data analytics are a very generic tool that can be applied to many different ends in healthcare context. Examples include reducing frauds, predictive analytics to improve outcomes, real-time monitoring of patients, information sharing between different systems, genomic analytics, evidence-based medicine, etc. If you have Google and five minutes at your disposal, you can easily find many more ends to which big data has been applied in health care. However, all that is just a preview of what we know we can do with big data – what about all the applications we will discover in the future? The future applications may make current estimates of the impact of big data seem conservative.
Four Requirements for Realising the Potential of Big Data
- An incentive to exploit big data. Traditionally, healthcare providers have been compensated on a fee-for-service model. This does not incentivise moving to a more pre-emptive care, if it means diminishing financial returns. A “fee-for-value” compensation model would drive healthcare providers to optimise the outcomes.
- Commercially available, suitable tools and services. Healthcare providers range from huge university hospitals to private practices. Not everyone can hire PhDs as data scientists. Commercially available and affordable tools and services enable scaling up the use of big data analytics in healthcare.
- Technology to collect, store and process vast amounts of different types of health data. Health data comes in many different shapes: electronic medical records, imaging, claims information, lifestyle and activity data, nutrition and diet… These data are contained in different systems in different (structured and unstructured) format, thus aggregating the data is no small challenge. To make matters even more difficult, the body of the data is huge and it is growing at an ever-increasing pace.
- Existence of and access to relevant bodies of data. In healthcare, these data exist, but accessing them is another matter. Healthcare IT is notorious for the legacy IT systems with limited or non-existent interfaces. Furthermore, privacy is a major concern when sharing health data to external operators.
Where does Netmedi fit in all this?
In the above listing, our main focus is on point 2, to offer our customers means to collect patient-reported outcomes and turn them into actions that benefit everyone involved. This, of course, necessitates that we also work to make points 3 and 4 happen. In practice, this means developing Kaiku Health to be able to collect data directly from patients and integrating Kaiku Health to other medical IT systems.
As we have seen, the term “big data analytics” cover myriad different applications in healthcare. We have chosen collecting and using patient-reported data as our angle, because it holds great potential for improving outcomes for patients. Personally, I feel it is very exciting to be working in this emerging field. The change in the awareness of and the attitude towards using big data in healthcare compared to four years ago, when we started, can be felt at every meeting with our current and future customers. And no wonder – there is scientific evidence that collecting and using patient-reported outcomes in routine care has a positive impact on the quality of the life of the patients. Our mission is to provide tools that make this change happen.
Written by Joel Lehikoinen, CTO of Netmedi, former researcher at CERN, M.Sc. Applied Physics
Tags: big data, Data science, digital health, eHealth, Patient-Reported Outcomes, Preventive medicine, Software