AI in Healthcare III: Continuous Care Domain
Introduction
Imagine an approximate $360 billion in cost savings occurring over a 5-year period through the implementation of sustainable Artificial Intelligence (AI) applications (Pifer, 2023). This huge savings becomes a rare possibility if the challenges around communication, resource management & organization in the healthcare industry is addressed. The healthcare sector grapples with pervasive challenges of fraud and inefficient referral management; and the integration of advanced technologies such as AI or generative AI offers an opportunity to help plug these holes. In this release of our series of AI in Healthcare, we dive into the influence of AI within the Continuity of Care domain.
Continuity of care surpasses organizational and disciplinary boundaries as it is referred to as the relationship between the individual healthcare provider and patient beyond the single occurrence of a health condition. In primary literature it is distinguished by other healthcare domain attributes by how the patient is cared for and the care provided over time, of which both need to exist when establishing precise guidelines for measurement (Haggerty et al, 2003). There are 3 types of continuity of care which depend on the form and setting of the care provided and received: Informational Continuity, Management Continuity, and Relational Continuity, (Haggerty et al, 2003). Informational continuity refers to using past events and personal details in order to provide appropriate care for each individual. Management refers to a reliable and unified approach to handling a health condition that adjusts to a patient’s evolving needs. Relational is about maintaining an ongoing therapeutic connection between a patient and health providers. For the longevity and existence of continuity there needs to be care that is experienced in the form of coordination and connection.
AI influence within the spectrum of this domain can result in improved clinical operations, standards, and safety, leading to cost savings. These advantages also apply to physician groups who can leverage algorithms in the continuous care domain, with a practical use case in referral management (Pifer, 2023). AI systems can have the capacity to provide recommendations that help improve the process of locating the appropriate service providers.
Conversational chatbots, for example, that can comprehend complicated phrases, dialects, and mispronunciations create an outcome that optimizes patient engagement and operational efficiency by assisting both providers and patients at the point of service delivery i.e., hospitals, clinics or any other facilities.
Applications within Healthcare Maintenance
Healthcare fraud costs the US approximately 300 billion annually, with health insurance fraud accounting for about 100 billion. This substantial sum contributes to higher insurance premiums for customers. Artificial intelligence (AI) models offer effective fraud detection and prevention by automating manual processes.
Health insurers traditionally review claims manually to identify fraud, which is both time-consuming and inefficient. Up to 70% of claims are considered unusual or potentially false based on company policies, leading to extensive reinvestigation. Due to the inefficiency of manual fraud detection, firms are prioritizing investment in AI-based fraud detection, particularly those utilizing natural language processing (NLP).
NLP models significantly reduce the time needed to identify suspicious claims and become more effective as they analyze datasets of previous fraudulent claims. Consequently, they have the potential to lower the rate of suspicious cases from the current 70%.
Deloitte reports that chatbots can automate up to 75% of customer service requests, resulting in a 40% reduction in customer service costs and freeing up to 97% of employees’ time. These benefits make chatbots a valuable addition to the insurance sector.
Chatbots also play a significant role in handling insurance claims, including first notices of loss (FNOL) and tracking claim progress. Insurers, especially after the COVID-19 pandemic, have increasingly adopted digital channels like mobile apps, with some companies handling over 50% of claims through these platforms. Chatbots can efficiently engage with customers on mobile and messaging apps, facilitating digital and automated claims processing.
Source: company website
Applications within Referral Management
Generative AI can play a valuable role in primary care referral management by automating and optimizing various aspects of the process. Primary care referral management involves coordinating and facilitating patient referrals from primary care providers to specialists or other healthcare professionals. For example, Central Ohio Primary Care (COPC) has embraced digital transformation to streamline its referral process, partnering with LeadingReach for data analytics. Prior to this, COPC relied on fax referrals, leading to delays and duplication of effort. LeadingReach allows digital referral submission, enabling quicker updates on referral status. Referral coordinators spend less time on follow-ups, thanks to digital messaging and data analytics. This transformation has not only improved the referral process but also generated insights, improved communication between primary care physicians (PCPs) and specialists, and encouraged collaboration to provide high-quality, timely care at lower costs.
Final Thoughts
The potential of AI to enhance referral management service levels as well as delivering efficiency within many health maintenance organizations is not in doubt. However, issues around governance, data privacy, ethics remain a big bug as with any applications of the technology within the healthcare systems. We do think that the benefits of leveraging the power of the systems do outstrip any of the potential pitfallls in the efficient delivery of healthcare services to the population.
Written by Ayowole Delegan, Promit Ghosal, Prosenit Kundu, Joyce Musweu
References:
https://www.frontiersin.org/articles/10.3389/frai.2022.962165/full
https://medcitynews.com/2017/11/ai-chatbots/
Haggerty JL, Reid RJ, Freeman GK, Starfield BH, Adair CE, McKendry R. Continuity of care: a multidisciplinary review. BMJ. 2003;327(7425):1219–1221. doi:10.1136/bmj.327.7425.1219. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC274066/