For providers and even payers that have not implemented strong practices for data governance and management, now may be the time to make the investment to get your data house in order. Otherwise you could be on the outside looking in as new artificial intelligence (AI) / business intelligence (BI) capabilities enable strategic shifts in healthcare.
Modern Healthcare just published an interesting set of articles on artificial intelligence – Realizing AI. Realizing AI: AI's role in healthcare starts small, gets bigger is the first of three and the series discusses different aspects of the application of artificial intelligence in healthcare including a number of use cases in place today. Examples of the areas covered include: predictive modeling, augmenting providers in various ways, proactive/cost effective patient engagement, precision medicine, speech to text capabilities. Most of these are examples of where providers and payers have gotten beyond the hype of AI and applied it (albeit in varying levels of depth) to play value-added roles in models that addressed some combination of cost, improved outcomes and patient experience.
When we get beyond the hype and the one-off pilots I believe AI combined with business intelligence (BI) will be key strategic differentiators for both payers and providers. AI (paired with BI) at varying levels will be very important if not critical enabler to addressing a number of the challenges in US healthcare which make it “the tapeworm of the American economy” according to Warren Buffet. From helping providers work at the top of their license to precision medicine to optimizing brick and mortar investments to “nudging” consumers to better health and keeping seniors safe in their homes – AI and BI will have a lot to add.
There is only one minor detail to being positioned to leverage this key strategic opportunity – most organizations do not have trustworthy, clean, consistent data to take advantage of the new capabilities. Here are a couple of quotes from the articles and the survey:
- “While healthcare is awash in data, those data are not often consistent, clean or in sets large enough to “teach” AI algorithms enough to be trustworthy”
- “But in the end, it depends primarily on one thing: ”
- “Whether AI succeeds depends in large part on how available the necessary data are”
- “The data are in a place where the know-how does not exist (EHRs/vendors), and the know-how is in a place where the data doesn’t exist”
- A recent Deloitte survey cited analytics as a critical asset for long term health strategies but data quality was the most common barrier cited.
Now there are certainly other factors which will impact the roll-out of AI/BI within healthcare – change management as a key one – but I believe the “community standard” used to determine whether a quality care was provided (e.g. in malpractice cases) will incorporate AI/BI tools. For example, if a radiologist missed an early stage growth on an image that turned into cancer the case could hinge on whether an AI tools was used to augment the read.
One of the key reasons I joined Amitech after years in consulting and healthcare innovation with Ascension is their experience/expertise in applying business intelligence and data management to enable it. Amitech has a great deal of experience on both the provider and payer sides in helping organizations clean up their data assets and structure them for use. Let us know if you would like to revamp your data strategy and execution.