How conversational AI is transforming personalized and preventive care
How conversational AI is transforming personalized and preventive care16 March 2022 | 11min
Conversational AI used in healthcare has the potential to improve care access while empowering patients to engage in their health in a way that works for them
Healthcare leaders looking to enter this space need to ensure they well understand the needs of patients and offer a simple to use solution that addresses them
The prevention and management of chronic conditions, which require commitment and work from the patient, are where digital health tools may offer the most benefit
COVID-19 highlighted important challenges in healthcare, such as the need for better and more convenient access without losing the touch of personalized care. Conversational AI powered digital solutions are helping to fill this gap while also empowering patients to monitor and take more control of their own healthcare needs.
To help explore this topic further, we sat down with Dr. Jason Paruthi, Medical Director at Lark – a company focused on delivering personalized digital health coaching using deep data knowledge to help users better manage chronic conditions and improve their health with their award-winning, conversational AI healthcare platform.
Read on to learn great insights into how conversational AI is transforming care, important tips for leaders looking to enter this space, and how employers can even benefit from this technology.
The rise of chronic disease and the need for AI-driven solutions
HT: What problem in healthcare is Lark working to solve?
Dr. Jason Paruthi: With 60% of American adults having a chronic condition, and 90% of the US’s annual healthcare expenditures ($3.5 trillion annually) allocated to people with chronic and mental conditions,1,2 we see a need for innovative, evidence-based digital health solutions to engage patients, improve their care, and drive down costs.
What’s more, is that there are simply not enough doctors and nurses to get people the type of care they need. According to the Health Resources & Services Administration (HRSA), 89 million Americans today live in what’s called a “health professional shortage area,” meaning over 15,000 additional primary care practitioners are needed to serve this population.3 That’s not sustainable.
Scalable solutions, such as AI solutions, can meet those patients’ needs without over-burdening the healthcare system, regardless of the geographic barriers they may face.
AI solutions can give people personalized attention and work in concert with providers by giving doctors and nurses tools that connect digital health solutions with providers and that provide more comprehensive care to patients.
Conversational AI improves access to care and helps save resources
HT: What is conversational AI, and how can it be leveraged to drive more efficient and personalized care?
Dr. Jason Paruthi: In healthcare, we are seeing a lot of different uses for AI. Many solutions use the term AI when they’re using predictive analytics or data analysis to help providers with their care decisions, for example.
Conversational AI uses machine learning to support patients 24/7, almost as a personal assistant would. Content is created by experts across various fields such as medical, clinical, and behavioral change specialists, and is then conversationally delivered to patients in a way that’s easy for them to follow and more enjoyable to engage with. This helps them get the day-to-day care they need between visits with their providers while allowing doctors and nurses to focus their time on what matters most.
Conversational AI can also be used to educate members on their condition, giving them feedback in a format that is similar to what a patient would experience with a human being or expert consultant. This technology can be used to deliver care to large populations through an accessible medium, smartphones. The ability to connect patients to their nurses or doctors when needed potentially leads to a higher quality or more acute level of care.
Another powerful feature of conversational AI is the possibility to use it in unison with traditional cognitive behavioral therapy techniques to support patient care more holistically. Such techniques can include stress management skills, reframing, smart goal setting, problem-solving and prioritization, acceptance, and commitment therapy, and even relapse prevention planning.
Certain habits and behaviors can be learned by the way patients interact with conversational AI, which helps improve operational efficiency while reducing overall healthcare costs and reliance on human capital.
How to ensure value with conversational AI tools in healthcare?
HT: How are conversational AI tools being used today, and how do users know which tools have the potential to provide them a real benefit and value?
Dr. Jason Paruthi: We are seeing conversational AI tools being deployed in very different ways not only in healthcare but in other industries as well. For example, an online shopping website with a chatbot, or booking an airline flight with a virtual assistant readily available to help.
The challenge in healthcare, with so many conversational AI-based tools or apps being developed, is knowing which ones are safe and provide a real health benefit for users.
There are certain identifiers and indicators that can inform users of the quality and validity of their apps. For example, being accredited by regulatory bodies such as the Centers for Disease Control and Prevention (CDC) or being recognized by and having the opportunities to work with reputable organizations such as in Lark’s case, the American Heart Association Innovators Network, can help drive progress forward.
Another key indicator is the National Provider Identifier (NPI) number which is a Health Insurance Portability and Accountability Act (HIPAA) Administrative Simplification Standard and is only issued to covered health providers.4 Having a NPI number means you can educate patients and provide care. For example, Lark has been granted an NPI number, which is in recognition of Lark’s ability to help drive improved outcomes in patient education.
What is also critical to the progression of conversational AI is the ability to understand the patient’s needs and deliver that information to them when they need it so they can move forward on their patient journey.
We are seeing more data as people take more and more personal device readings from the comfort of their home or wherever they are. However, we’re not necessarily seeing better outcomes in patients unless they’re getting real-time, very specific feedback on the data that they’re providing, such as their meals, exercise, medications, and so forth.
The more immediate feedback you can provide, the more engaged users will be with the tools they are using and the better outcomes they will have.
Chronic disease prevention in the workplace
HT: What would you say are the benefits for employers to offer employees digital health solutions that focus on preventing chronic conditions?
Dr. Jason Paruthi: Specifically for employers, digital health solutions need to make sense from a cost perspective. Thinking about their long-term cost expenditures and return on investment (ROI), we’ve seen employers that have implemented a hybrid approach – offering their employees digital solutions to help manage their health in the long-term, in addition to in-person care they might receive through their employer-sponsored health insurance plan.
When you focus on the prevention and management of chronic conditions, employers might see some of their preventive care costs rising in the short term, but in the long term they can save on absenteeism, healthcare costs, and workers’ compensation and disability management claims costs.5
As we move towards a value-based care model throughout healthcare, we need to be sure that digital health solutions are not only providing effective solutions for employers, but also cost-effective solutions so that they can make them available to their employees.
Key tips for startup leaders entering the digital healthcare space
HT: What advice would you give to startup leaders entering the healthcare space, especially with a digital platform where competition is high?
Dr. Jason Paruthi:
|1||Understand the whole patient journey and what makes your company unique. The healthcare industry is riddled with challenges. Startups new to the healthcare space need to reflect upon the whole care patient journey from start to finish and know what patients’ needs are. They should also be clear about what sets them apart from competitors. Is it their technology, their price point? Maybe it’s how scalable their technology and solution is. These factors need to be considered, so that that solution is not just another face in the ground.|
|2||Ensure your offering engages patients. It’s not enough to simply give people the tools they need to manage their condition. We need to provide tools that are engaging and personalized so that people receive the right kind of support when they need it most and are motivated by this.|
|3||Clinically validate your offering or tool. These products or solutions need clinical rigor as well, so when patients’ lives are on the line, you need to know, first and foremost, that your product works and you’ve demonstrated that efficacy that you can solve for their pain points while also reducing fragmentation and cutting costs.|
|4||Aim for simplicity of use and connectivity into existing healthcare infrastructure. From a patient perspective, there is a lot of noise out there which can get overwhelming. Having a simple and comprehensive solution that is personalized to members and is easy for them to use and understand is critical. Connectivity with a patient’s provider is also essential.|
|5||Don’t underestimate people’s ability to engage with digital tools. It’s a common myth that older patients might not be able to engage in digital health solutions. The literature shows that providers might be more hesitant to offer these solutions to older individuals, but when older patients do get these types of digital health tools, we see and strong engagement there.6 Don’t let this be a deterrent.|
Collaboration between startups and corporate partners
HT: How do you see the collaborations between startups and industry-leading corporate partners supporting healthcare transformation?
Dr. Jason Paruthi: Collaborations with corporate partners play a huge role in leading industry-wide healthcare initiatives to make a difference in the lives of millions of people. We’re excited to work with partners to tap into new disease areas and provide more scalable, affordable care to those who need it.
We’re excited to announce a new collaboration with Roche Diagnostics to develop an AI-driven program designed to support the prevention and management of atherosclerotic cardiovascular disease and coronary artery disease, or CAD. Tackling heart disease remains one of the most difficult challenges facing healthcare. Together we hope to provide a personalized solution to better engage patients in preventing future cardiovascular events, but also in managing these conditions at scale.
Collaborations with corporate partners are critical to making a difference in the lives and millions of patients. Startups often have innovative solutions, innovative tools, but they often need access to resources and capital. These large partnerships can accelerate these technologies coming to market and enable leveraging the technology to reach additional chronic conditions, reach additional patients that wouldn’t be possible independently by either party.
A push towards prevention with digital health solutions
HT: Was there a reason that you started with diabetes as a chronic condition with your solution?
Dr. Jason Paruthi: The first, is the magnitude of the need. About 70 – 80% of Americans are overweight or obese, putting them at high risk for prediabetes or diabetes development. This is an extremely challenging condition for a patient to manage and it is tough on their health because it’s multifactorial. It affects not just your blood sugar, but also all different components of your body such as your heart, kidneys, and eyes.
The truth is that the well is drying up for safe, effective, and cost-effective medications in the cardiometabolic space, especially diabetes. We fervently researched for a pill or a treatment that could cure or greatly impact people’s diabetes. Although there are excellent medications out there, diabetes is one of the key conditions that does require a lot from the patient. The patient does need to exercise to improve their insulin sensitivity, eat right and lose weight. No pill or medication can do that.
Because the needs of a patient with diabetes are so comprehensive, we felt that Lark’s conversational AI technology was an excellent healthcare tool to help patients manage their disease. It could be there for a patient at all times of the day to help inform them what to do about their blood sugar, help them remember to take their medications, to nudge them to keep exercising, for example.
HT: For chronic conditions that require patient engagement, is that where you see digital tools being most valuable in the future?
Dr. Jason Paruthi: I would say so because the paradigms are shifting as well in healthcare.
For example, with hypertension, we would always tell people that diet and exercise are critically important to manage blood pressure. But, if you go to a provider presentation at a medical conference they’ll typically say those lifestyle modifications don’t work for hypertension. What’s most important is picking the right medications and helping patients adhere to those medications.
Lifestyle modifications are hard. They require effort and the patient needs to be engaged in their health so they are motivated to do the work. Before digital health solutions, this was more difficult to achieve but now – with patients getting non-stop data from their Apple Health and Google Fit, activity bands, and so forth – it’s easier.
Looking forward, with digital health solutions we will see more efficacy of lifestyle modifications because we’ve found a better way to help members engage and adopt those behavior changes.
The paradigm is shifting from “find the right medicine and prescribe it” to helping empower patients to take control of their healthcare and take preventive measures. Helping people understand what they can do today to prevent needing those medications or having that heart attack or stroke or high blood pressure down the road.
Jason Paruthi, MD With extensive experience in clinical medicine, translational research, and health technology business development, Dr. Paruthi collaborates with Lark’s Health Committee and product teams to incorporate the latest evidence and clinical guidelines into Lark’s AI coaching platform. Additionally, he educates partners and prospects on how to best leverage Lark for optimized health outcomes, care coordination, and cost savings in their patient populations. He received his BA and MD from Boston University and was a trainee and translational researcher at Harvard Medical School where he studied clinical interventions for the prevention and management of type 2 diabetes, cardiovascular disease, and obesity-related chronic conditions. Outside of work, Dr. Paruthi enjoys spending time with his family and trying out new healthy, spicy recipes.
- Centers for Disease Control and Prevention (CDC). (2022). Article available from https://www.cdc.gov/chronicdisease/resources/infographic/chronic-diseases.htm [Accessed February 2022]
- Centers for Disease Control and Prevention (CDC). (2022). Article available from https://www.cdc.gov/chronicdisease/about/costs/index.htm [Accessed February 2022]
- Health Resources & Services Administration (HRSA). (2022). Data dashboard available from https://data.hrsa.gov/topics/health-workforce/shortage-areas [Accessed February 2022]
- Centers for Medicare & Medicaid Services (CMS). (2021). Article available from https://www.cms.gov/Regulations-and-Guidance/Administrative-Simplification/NationalProvIdentStand [Accessed February 2022]
- Centers for Disease Control and Prevention (CDC). (2015). Article available from https://www.cdc.gov/workplacehealthpromotion/model/control-costs/index.html [Accessed February 2022]
- Graham et al. (2021). Front. Digit. Health 3:642818. Article available from https://www.frontiersin.org/articles/10.3389/fdgth.2021.642818/full [Accessed February 2022]