Big Data in Healthcare
In class, we were introduced to how big data is used in the professional context, particularly within LinkedIn connecting people to opportunities. But there’s also a huge application of big data for healthcare.
Dr. Eric Topol, executive vice president at Scripps Research Institute, emphasized that we can now collect environmental data that is greatly informative about the human body that survives within the context of such an environment. We can also reap data from any part of the human body through new sensor technology from the usual scans down to molecular information like our DNA [1]. Numerous medical technologies have come about in the last decade that collect and analyze big data in order to advance medicine. For example, Dignity Health “uses a big data and advanced analytics platform to predict potential sepsis cases at the earliest stages… Dignity Health monitors 120,000 lives per month in 34 hospitals and manages 7,500 patients” [1]. Or more well known companies like 23AndMe collect DNA information to track ancestry as well as correlations between diseases and DNA markers. In more academic contexts, the American Gut Project is one of the world’s largest projects to map the microbiome inside human organs [2].
Healthcare is jumping into big data as we delve deeper into this data-inundated era. However, it is still in the rudimentary phases in healthcare given its slow adoption in that realm. In particular, big data in medicine is slow because of the lack of “technical expertise required to use it and a lack of robust, integrated security surrounding it” [3]. Most data scientists in the medical field are only acquainted with relational data (i.e. data neatly organized into and represented by easily analyzable tables). However, big data is, for the most part, unstructured and raw, meaning the first big step before even attempting to analyze the data is figuring out how to organize the massive amount of information. As Dr. Anil Jain mentioned, “It’s not about the data, it’s about what you do with the data in terms of making sense of it” [4]. Given the vast amount of data, which continues to grow rapidly (as can be seen by an estimation of 2,314 exabytes of healthcare data by 2020 [5]), it becomes increasingly important to ask the right questions in order to interpret data that is unstructured. The perspective in which the data is seen makes a difference in the outcome of analyses. Such skills are not common in data analyzers who obtained their skills when data was limited to small, simple structures. Especially in healthcare, with the added layers of complex research protocols, it undoubtedly takes a long time to sift through the data in ways that hopefully positively impacts healthcare.
The second mentioned reason for the delay in impact of big data in healthcare is security. All health data, at least in the US, must be HIPAA compliant. Given that big data, by its very definition, is stored across multiple devices and servers, it is all the more crucial that all of these sensitive health data are stored securely.
Despite these two obstacles, big data is definitely playing an increasingly important role in healthcare, from research to business. For example, “companies like IBM Watson Health and Flatiron Health are taking data from electronic health records, cleaning it up, and using that data to identify potential interventions and cost-saving opportunities… they offer a longitudinal context of a person’s medical history and life” [4]. Having so much information is a challenge, but the information is certainly valuable given that they contain patterns from which we can learn more about the human body and thus, improve the quality of longer lives of humankind. Big data in healthcare will help to not only provide more personalized information about each individual human body, but also discover effective treatments and medicine and influencing the healthcare system from business to governmental domains. It will be exciting to see how many lives will be saved with the growth of research and innovations on healthcare using big data.
[1] – https://www.sas.com/en_us/insights/articles/big-data/big-data-in-healthcare.html
[3] – https://www.healthcatalyst.com/big-data-in-healthcare-made-simple
[4] – http://fortune.com/2018/03/20/big-data-finally-changing-health-care/
Users who have LIKED this post:
One comment on “Big Data in Healthcare”
Comments are closed.
I really like the way you approach some of the barriers we still have to trespass in order to develop the analysis of big data in Healthcare systems, such as the lack of “technical expertise” and the need of high security given the sensitiveness of the data. I would also like to complement your post by adding some other challenges I think are relevant in this segment:
Firstly, regarding the capture of data, I would like to argue that many healthcare data providers don’t always do it in a clean, complete, accurate, and formatted way so this could have an impact on the usage of this data. Additionally, the storage of it it’s a critical cost, security, and performance issue for the IT department, and many organizations cant afford such costs. Another big challenge is the management and update of such data, in the sense that, reused or reexamined data from the past may not be accurate, need frequent update – given the non-statistic nature of healthcare data – and can be under possible violation. The last challenge we have to overcome, which I think is perhaps the most important, is the sharing of that information. Given the different ways health records are designed, implemented presented by the big organizations, it could often lead to a scenario of clinicians without information they need to make key decisions, follow up with patients, and develop strategies to improve overall outcomes.