During National Health IT Week, champions from across the industry are uniting to share their voices on how health IT is catalyzing change in U.S. healthcare. The following post from a National Health IT Week Partner is one of the many perspectives of how information and technology is transforming health in America.
The day of big data has arrived, and in healthcare, it’s landed on our desks with a resounding thud. Our challenge lies with discerning how to analyze information and use it to effectively improve patient outcomes, costs and efficiencies.
Many of us are already influenced by machine learning and artificial intelligence (AI). For example, if I buy hiking boots online, suggested items pop up that I can use as well, like bug spray or sunscreen. The data analytics behind those recommendations includes a wealth of information about me – my demographics, such as age, gender, education and income level, as well as where I live and other factors that influence my buying decisions. My prediction is that we will be able to apply the same principles to healthcare data.
Imagine a doctor who can review operational and clinical data in real time for a patient who’s had knee replacement surgery. After the patient goes home, she is given a wearable to monitor her step count. If her steps trend downward, it’s probably time for someone to intervene because she’s in pain or not ambulating correctly. That same physician could also see where she’s received care, the cost of the care and who performed the surgery. They can also compare her progress against others with similar demographic and health backgrounds using machine learning and streaming analytics that not only gather relevant data across the entire care continuum – from hospital to rehab facility to home – but draws inferences from that information in real time to truly influence cost and care outcomes. In addition, if the patient had three MRIs that cost $2,000 each and someone with similar demographics and health conditions had one MRI that cost $500 – caregivers can explore why that happened and work toward more uniformity.
This idea inspires me, but I think we can take a more practical look at how AI can support the business operations of healthcare as an achievable first step, along with connecting that operational data with remote care, device data and patient electronic health records. Here are next steps for creating efficiencies with the power of AI and interoperability, specifically using chat bots.
As a recent Advisory Board report states, “AI works best when paired with humans.” I agree, and the goal is to use it to create efficiencies across the care continuum that not only help staff in their roles, but that free clinicians, caregivers and office staff to focus on more valued activities. AI can help augment and automate human tasks and functions where appropriate, and we’re hoping sooner rather than later it can offer advice, with a goal of allowing caregivers to focus entirely on patient care.
AI can quickly answer employee queries on days outstanding, by supply, such as bandages from a certain supplier. AI can also track unused supplies to minimize excess inventory. In addition, AI can help alleviate the amount of time – and frustration – nursing and clinical staff spend searching for supplies by not only providing location, but automating future order and delivery.
For those healthcare employees without regular access to a computer, such as lab technicians, AI can quickly and accurately empower cross-functional self-service. All employees need to do is ask to get answers about anything from paid time off balances to company holidays.
AI can augment the payment process, detecting payment, vendor and invoice patterns, and suggesting automating payments for a specific invoice that is approved 99 percent of the time.
Employing AI to help manage and maintain hospital equipment can provide staff with needed insights, quickly, and thus ensuring something as simple as too-bright hallway lights can be adjusted during patient sleeping hours. This not only contributes to patients getting the restful sleep they need, but may in turn boost those patient satisfaction scores that can have an impact on a hospital’s reimbursement levels.
Having the ability to capture supply demand automatically from your clinical systems, automate your patient refund process, including status updates of checks, and flow patient information automatically between the most commonly used HL7 message type, point of use, scheduling, case scheduling, and billing systems will help your organization achieve higher cost and care efficiencies. Combining all information through AI or machine learning will help you quickly draw inferences between all that data and give you the power to positively influence the entire continuum of care.
We are just beginning to unlock the potential of AI in healthcare, and making health data interoperable. And while we aren’t quite there with using machine learning to help us adjust treatment and care protocols in real time, we are well positioned to take the next step with computing power and streaming analytics to uncover how AI can impact healthcare.
The views and opinions expressed in this blog or by commenters are those of the author and do not necessarily reflect the official policy or position of HIMSS or its affiliates.
Healthcare Transformation | Access to Care | Economic Opportunity | Healthy Communities
Share your story, provide insights and help develop healthcare policy during NHIT Week – focused on catalyzing actionable change within the U.S. health system through the effective use of health IT.
View this year’s calendar of events
Explore this guide on supporting #NHITWeek through social media
Submit an #IHeartHIT Story
Participate in the Virtual March
Explore 10 ways to celebrate in your hometown
See the 2018 NHIT Week partners list, and sign up today