By: Daisy Liu
Introduction: what is “Big data”?
From the number of clicks on a website to the height of an average resident of downtown Toronto, almost everything in the world is describable in terms of data. However, confusion remains around what differentiates this form from “Big data,” another term used in data analytics.
Contrary to its name, big data is not necessarily distinguishable from traditional datasets by size. Instead, it is characterised by the way it is used, requiring highly scalable analytics processes, flexibility with regards to the format of data, real-time results, modern storage platforms, high data quality, and applications to machine learning. In the context of the healthcare system this can include massive volumes of information gathered from sources ranging from patient records to search engine data following the adoption of digital technologies. By taking advantage of the huge volumes of information generated by everyday interactions, researchers and analysts have the potential to create tools that revolutionise personalised treatments and give clinical practitioners the in-depth knowledge they need to give patients the best care possible.
Applications of big data analytics to healthcare
In their 2014 article, Bates et al. present six use cases that they believe represent the clearest opportunities to reduce costs incurred by the clinical system through the use of big data, thereby freeing funds for other purposes. This includes: (1) high cost patients, (2) readmissions, (3) triage, (4) decompensation, (5) adverse events, and (6) diseases affecting multiple organ systems. In each case, information gathered from the clinical experiences of many patients can be used to predict the risk of a single subject for developing a certain condition. Aside from avoiding the need for expensive procedures, this can also protect the patient from possible injury or debilitation by providing earlier and possibly more accurate diagnoses.
Within the realm of public health, big data can be used to improve public health surveillance and assessment and facilitate compilation of large amounts of diverse data once the appropriate infrastructure has been set in place. The PASSI surveillance project in Italy, for example, provides wide-ranging lifestyle information about almost 90% of the population. This creates a powerful tool for those planning public health actions. Another notable application of Big data is in the movement towards value-based healthcare where users are charged on a fee-for-value basis rather than the current fee-for-service system present in most countries. Under this model, medical care would exist as integrated patient-centred medical homes rather than siloes of specific services coordinated and led by a patient’s primary physician. Given their increased involvement, big data can provide practitioners with the extra insight needed to make informed decisions while connecting disparate organisations through a shared database.
Despite holding great potential for revolutionising healthcare, the path towards efficient incorporation of big data analytics into the health system remains riddled with ethical, legal, and technical challenges. In collecting large volumes of information about individuals, the risk of compromising privacy and the public demand for transparency, trust, and fairness must be balanced carefully in order to protect the rights of patients. However, the current lack of appropriate infrastructure for this purpose makes it difficult to afford sufficient protection, especially with regards to sensitive information (e.g. genomics data). Similar problems translate to the technical aspect of big data usage, with people expressing concerns about the “black box” that often arises when using highly trained AI in clinical processes. Indeed, while it is not always necessary to possess intimate knowledge of the intricate workings of specific applications, moral imperatives to provide transparency to users make the problem difficult to navigate. Another technical problem that is currently being addressed is the extremely diverse set of sources from which information can be gathered as well as the sheer volume, making it difficult to process and merge into conventional databases. Without a proper analytical strategy, finding meaningful information and trends can resemble looking for needles in a haystack. Siloed data and budget constraints represent further challenges.
Despite the present challenges, many researchers and organisations around the world are striving to harness the power of big data in healthcare. Aside from diagnostics and preventative care, other areas where it is predicted to have a huge impact include precision medicine and wearable devices, both of which allow for further personalisation of care. However, due to legal and ethical problems like the ones discussed above, usage is currently limited to low-impact applications where wrong diagnoses or advice have fairly low impact. In Canada, current initiatives include that of Health Canada through BORN (Better Outcomes Registry & Network) and a partnership with SickKids. By taking advantage of the increasingly digitalised lifestyle of most citizens, health systems around the world can make use of the huge volume of data to revolutionise the ways in which they provide care, resulting in better health outcomes as a whole.
This article is written in collaboration with the Health and Human Rights (HHR) subcommittee of the University of Toronto International Health Program. If you found its contents interesting, please consider attending the 2021 HHR Conference and/or submitting an abstract to the 2021 HHR Research Poster Fair.
More information on seminars, speakers, and scheduling can be found on our website: https://www.hhrights.org/
Event: UTIHP HHR Research Poster Fair 2021
Time: March 11th, 2021 - March, 13th 2021
Topic: The Future of Healthcare Accessibility Through Telehealth
Presentation format: Online poster fair
Abstract submission: https://forms.gle/8EC3pNZbQmgm3WZi6
Bates, DW., Saria, S., Machado-Ohno, L., Shah, A., Escobar, G. 2014. Big Data In Health Care: Using Analytics To Identify And Manage High-Risk And High-Cost Patients. Health Affairs. 33(7): 1123-1131. https://doi.org/10.1377/hlthaff.2014.0041.
Bresnick, J. 2018. Can Healthcare Avoid “Black Box” Artificial Intelligence Tools? [Internet]. Danvers (MA): HealthITAnalytics; [cited 2021 Feb 6]. Available from: https://healthitanalytics.com/news/can-healthcare-avoid-black-box-artificial-intelligence-tools.
Durcevic, S. 2020. 18 Examples Of Big Data Analytics In Healthcare That Can Save People [Internet]. Berlin (DE): The datapine Blog; [cited 2021 Feb 6]. Available from: https://www.datapine.com/blog/big-data-examples-in-healthcare/.
Landau, J. 2019. 4 Areas Where Big Data is Transforming Healthcare Right Now [Internet]. HIT Consultant Editorial; [cited 2021 Feb 6]. Available from: https://hitconsultant.net/2019/09/09/4-areas-big-data-transforming-healthcare/.
NEJM Catalyst. 2018. Healthcare Big Data and the Promise of Value-Based Care [Internet]. Waltham (MA): New England Journal of Medicine Catalyst; [cited 2021 Feb 6]. Available from: https://catalyst.nejm.org/doi/full/10.1056/CAT.18.0290.
Pastorino, R., Vito, CD., Migilara, G., Glocker, K., Binenbaum, I, Ricciardi, W., Boccia, S. 2019. Benefits and challenges of Big Data in healthcare: an overview of the European initiatives. European Journal of Public Health. 29(3): 23-27. https://doi.org/:10.1093/eurpub/ckz168.
Precisely Editor. 2019. Big Data vs Traditional Data: What Defines Big Data? [Internet]. Pearl River (NY): Precisely; [cited 2021 Feb 6]. Available from: https://www.precisely.com/blog/big-data/big-data-definition-data-what-defines-big-data