To real time or not to real time – that is the analytics question

May 20, 2016 / By AJ Dandrea

When we consider the complexity of systems and the challenges in making data-driven systems usable, it is easy to forget the principle of “real time.” As an industry, we hear that clinicians are fatigued with all the technology tools they must use when providing care. So is it truly possible to have one technology system that encompasses analytics, care delivery, interaction and reimbursement – and all in real time?

Why not?

Electronic medical record (EMR) systems are close to real time; reimbursement systems are too. However, analytics systems need some help to achieve real-time status. There are more than a few data sources required if analytics are to be holistic, yet actionable so interventions can happen in real time. Today’s EMR systems represent clinical data as close to real time as possible (without bio-implant science fiction, that is). But what about other data sources, including payer, benchmarks, ad-hoc, readmission and various other feeds? How do these data flows become real time? Only if we exclude systems that are not capable of real time. True innovation depends on market rejection of standard, 100-day after-the-fact spreadsheet analytics products. We have all seen products where data is processed too slowly or with dashboards that cater to those fortunate enough to have an 80 inch monitor in their office.

As technologists, we have learned from our caregivers that they would like one system, preferably a part of their EMR to manage care, to predict risk and the need for interventions, to foster interaction and to handle reimbursement. The optimists believe we do not have to wait for full adoption of a care-driven cost model before this can happen. The pessimists would have us believe it is impossible within the fee-for-service model. I side with the optimists – we already have real-time EMRs and reimbursement technologies, and we can force innovation to provide real-time prediction, interaction, and actionable analytics.

This notion of an all-encompassing system can be achieved, but we need help from caregivers to promote the necessary innovation. If clinicians refuse to adopt products that are not real time, the industry will seek out new solutions and partnerships to provide a real-time platform. The ideal analytics platform will:

  • Allow for real-time predictive, actionable interventions and follow up
  • Be accessible by patient, caregiver, payer and data scientists
  • Integrate with coding and reimbursement
  • Be mobile, simple and secure

To realize this goal will require time – real time – but it is possible.

AJ Dandrea is operations manager of the cloud hosting organization at 3M Health Information Systems