Leveraging interpretive intelligence in clinical workflows
Automation has been making human workers superfluous for centuries, but until recently, workers whose jobs required high-level cognitive skills have been able to rest easy, confident no machine could possibly replace them when it came to making nuanced decisions based on the evaluation of complicated, sometimes contradictory data.
But that was before Artificial Intelligence (AI) came along, stepping out of the pages of science fiction and into our daily lives. It now seems possible — even probable — machines will replace many mid-level knowledge workers and the question arises whether someday robots will replace doctors and nurses.
It’s a provocative question, certainly, but not the most interesting one facing our industry. A more critical question is will the healthcare ecosystem — the vendors and the solution providers — be able to survive without AI? I ask this because I believe doctors and healthcare administrators will increasingly demand answers to questions, and solutions to the challenges, that are difficult, if not impossible, to solve without the aid of AI-driven solutions.
These questions will range from practical issues of practice management to vital questions of patients’ health. For example:
- How much will it cost to treat this patient?
- How much and how fast will I get paid?
- Is medication or surgery the best treatment option for this particular patient?
- Where and/or when should I schedule this surgery?
- When will this patient be able to return to his/her normal routine?
Some of these, of course, are the perennial questions that have always faced healthcare practitioners, but the truth is recent changes in technology have made innovative solutions possible in a way never before imaginable. For example, all kinds of data are now readily available in consumable (discrete) forms — from PHI to financials to protocols — and storing and managing this data is getting cheaper every day.
Additionally, healthcare providers are beginning to understand the shift from service to value-based care and are seeing how it can work for them, both clinically and financially. Finally, the healthcare practitioners themselves are changing: computer- and technology-savvy clinicians who got their medical education and training in the 1990s and early 2000s (the so-called Generation X and Y) are now entering into leadership positions where they can affect change.
In other words, there is both a greater supply of data than ever before and a greater demand for it. However, this demand isn’t simply for large data-dumps of undigested information. What’s necessary is for healthcare providers and administrators to have the critical data they need, and only the data they need, when and where and in the form they need it. This is where AI can help make critical decisions about amalgamating and filtering data.
There’s enormous potential for AI (or “smart solutions”) to optimize clinical protocols by drawing on a huge pool of evidence-based results. As we move toward a value-based environment, AI will be increasingly necessary to proactively and dynamically manage patient outcomes. This, in turn, will optimize the treatment experience, leading to greater patient engagement — and this greater continuity of care will promote both healthier patients and healthier practices. Clinicians will also gain insights into how to manage risks, which leads to lower costs and better margins.
Will robots replace healthcare providers? It seems unlikely, but care teams will start to leverage interpretive intelligence in daily clinical workflow. Machine learning, along with AI, will become an integral part of the healthcare mix because the vast resources of critical data will only be truly available when clinicians have tools to track real-time data embedded in their daily workflows resulting in better patient care at a lower cost.
As seen in Health IT Outcomes.
Latest posts by Khal Rai (see all)
- Will Robots Replace Healthcare Providers? - June 13, 2017
- The Year’s Innovations – Wrap-up & What’s to Come! - January 11, 2017
- Data Done Differently - October 11, 2016