Futuristic healthcare interface with the text: input, output

Introduction to eHealth records

One of the biggest promises for big data analytics lies perhaps in the health care sector. Hospitals, clinics, pharmacies, and allied health providers are all capturing large amounts of data on their patients. However, to effectively analyse these records they need to be aligned to a common format and shared across providers.

Electronic health records

Electronic health records (EHRs) aim to combine health information on patients in a single repository. Many countries have been working on establishing EHR databases, where health data from multiple sources is stored in a way that facilitates processing. EHRs contain information such as medical histories, diagnoses, hospital discharge summaries, medications, allergies, immunisations, radiology images, and test reports1,2.

The V’s that characterise EHRs

EHRs are big data in several ways:

  • they are large in volume as they contain data from millions of people with dozens of entries each
  • they exhibit variety as medical information can be encoded in different ways due to lack of standards
  • the veracity of entries is paramount as wrong information could lead to incorrect treatment of patients.3

The challenges

While the collection and analysis of health data in EHRs has the potential to advance the state of medicine, researchers face a range of practical issues in using EHRs to derive insights.

  • Firstly, the quality of available data may not be sufficient for the investigation purpose. For instance, the study sample may have lots of missing or inaccurate data.
  • Secondly, few clinicians and health care professionals have sufficient data science skills to build advanced computer models.
  • Thirdly, only few tools for locating relevant data exist, making it difficult for researchers to know what datasets are available.
  • Finally, there is a lack of international frameworks that regulate uniformly how EHRs are shared, and what the legal and ethical responsibilities for using EHRs are.4

Addressing the challenges

In order to address the latter challenge, initiatives like Australia’s National Data Linkage Demonstration Project (NDLDP) to improve cardiac care, and the Fast Healthcare Interoperability Resources (FHIR) standard have recently been proposed.

NDLDP

The NDLDP consolidated health transaction records from Australia’s two most populous states. This has resulted in a register of over ten million people (almost half of the country’s population) with over seven billion records.5

FHIR

FHIR is a standard for exchanging EHRs. It defines a common data format to store data and a set of functions to access data programmatically.6,7 FHIR now provides the foundation for records stored in Apple’s Health app for iPhone.8

In the next step we will discuss the future for automated analysis in the medical field.

Your task

Given the choice to decide if your details were stored within an EHR, what would you do? Opt-in or opt-out? Why?

Share your thoughts in the comments.

References

  1. Australian Digital Health Agency. What’s in a My Health Record? [Internet]. Australian Government [cited 2018 Dec 22]. Available from: https://www.myhealthrecord.gov.au/for-you-your-family/whats-in-my-health-record 

  2. HealthIT.gov. What information does an electronic health record (EHR) contain? [Internet]. The Office of the National Coordinator for Health Information Technology (ONC); 2013 [updated 2013 Mar 17; cited 2018 Dec 22]. Available from: https://www.healthit.gov/faq/what-information-does-electronic-health-record-ehr-contain 

  3. Hemingway H, Feder GS, Fitzpatrick NK, et al. Using nationwide ‘big data’ from linked electronic health records to help improve outcomes in cardiovascular diseases: 33 studies using methods from epidemiology, informatics, economics and social science in the ClinicAl disease research using LInked Bespoke studies and Electronic health Records (CALIBER) programme. Southampton (UK): NIHR Journals Library; 2017 Jan. (Programme Grants for Applied Research, No. 5.4.) Available from: https://www.ncbi.nlm.nih.gov/books/NBK414778/ doi: 10.3310/pgfar05040 

  4. Hemingway H, Asselbergs FW, Danesh J, Dobson R, Maniadakis N, Maggioni A, van Thiel GJ, Cronin M, Brobert G, Vardas P, Anker SD. Big data from electronic health records for early and late translational cardiovascular research: challenges and potential. European heart journal. 2017 Aug 29;39(16):1481-95. 

  5. Falster M, Schaffer A, Jorm LR, Wilson A, Brieger D, Nasis A, Emerson L, Pearson S. Using Australia’s National Data Linkage Demonstration Project (NDLDP) to improve cardiac care: Towards a national, whole-of-population linked data resource for evidence-informed health policy. International Journal of Population Data Science. 2018 Aug 23;3(4). 

  6. HL7.org. HTTP - FHIR v3.0.1 [Internet]. HL7.org; 2017 [updated 2017 Apr 19; cited 2018 Dec 22]. Available from: https://www.hl7.org/fhir/http.html 

  7. HL7.org. Overview - FHIR v3.0.1 [Internet]. HL7.org; 2017 [updated 2017 Apr 19; cited 2018 Dec 22]. Available from: https://www.hl7.org/fhir/overview.html 

  8. Apple. Apple announces effortless solution bringing health records to iPhone [Internet]. Apple; 2018 [updated 2018 Jan 24; cited 2018 Dec 22]. Available from: https://www.apple.com/newsroom/2018/01/apple-announces-effortless-solution-bringing-health-records-to-iPhone/ 

Share this article:

This article is from the free online course:

Big Data Analytics: Opportunities, Challenges and the Future

Griffith University