There is no arguing that healthcare involves a prodigious amount of data. According to an IDC report, 4.4zb of data was produced in 2013, and it estimated that 44zb of healthcare data will be produced yearly by 2020—a tenfold increase in seven years. In a New England Journal of Medicine article, the authors estimated that a single patient typically generates close to 80 megabytes in imaging and electronic medical record (EMR) data each year. With all the data that is being collected, healthcare organizations have the opportunity to leverage it to drive patient outcomes and lower overall healthcare spend.
When thinking of healthcare data, I frequently recall the Tolstoy quote, “Happy families are all alike; every unhappy family is unhappy in its own way.” Healthcare entities are not homogenous and most have their own unique set of data quirks, concerns, and institutional roadmaps and use cases. Thus, the question becomes, How can healthcare data move beyond static noise to become a business asset to improve patient adherence?
Data does not speak; it is interpreted. In a Stanford Medical School Trend Report, it was estimated that 2,314 exabytes (one exabyte = one billion gigabytes) of healthcare data will be produced by 2020. It is of little wonder that extrapolating clinically useful, interoperable, meaningful, and time-sensitive insight from that prodigious amount of data seems like a daunting task.
As healthcare organizations move from volume to value, data has to be actionable and accurate. Additionally, there are numerous regulatory, privacy, and security requirements that need to be considered. And while there cannot be a one-size-fits-all approach, that should not stop healthcare entities from implementing best practices, specifically as they relate to how to reconcile siloed data, maximize existing stored data, and generate actionable, strategic processes for collecting future patient data.
Parsing through large quantities of data to solve complex, analytical tasks is becoming easier with solutions like artificial intelligence (AI) and machine learning (ML)) tied to a secure cloud. Consider, for example, patient no-shows cost an estimated $150 billion annually in the US and roughly £216 million in the UK. With a growing chronically ill population, patient no-shows can delay disease detection, impact the quality of healthcare delivery, and increase overall healthcare costs.
Healthcare organizations have the data to determine what percentages of their overall practice has a patient no-show problem. Therefore, it is no longer acceptable to say patient no-shows are “part and parcel of doing business.” Let us not forget, the business of healthcare is improving patient outcomes, and reducing no-shows will do just that.
With advances in technology, making sense of healthcare data does not have to be a Sisyphean effort. Healthcare organizations can and should use their data assets to drive patient adherence and better outcomes. Healthcare data is “noisy” and at times, overwhelming. However, with AI and machine learning, healthcare organizations can:
- Utilize data assets to get a holistic view of practice scheduling gaps
- Deliver appointment reminders based on patient engagement preferences
- Detect key practice factors (e.g., provider hours) for patient no-shows
- Leverage internal practice data and external data factors like weather, sporting events, etc., to create probability heat maps to predict patient no-shows
- Utilize data driven recommendations to overbook patient appointments to minimize loss/maximize practice revenue
To improve operational and clinical efficiencies it might be advantageous if healthcare organizations stop vacillating between healthcare data being a blessing and a curse. Healthcare has an abundance of unique opportunities to leverage data to be impactful for both providers as well as patients. There are decidedly numerous use cases on how to leverage healthcare data, but each healthcare organization will need to determine what technologies and use cases are right for their patients and providers.
About the author
Ms. Alison Haughton serves as Senior Product Marketing Manager, Healthcare at Progress. Ms. Haughton focuses on health IT, population health management, and digital care transformation. Previously, she was one of the principal members of AT&T’s ForHealth practice.
Ms. Haughton has served in a series of senior product marketing and leadership roles at IMS Health (now IQVIA), Harvard Medical School, Parexel, and a variety of early stage healthcare companies implementing health IT solutions.
