The Impact of AI and Telemedicine on Behavioral Health Services
The behavioral health landscape faces several major challenges, primarily due to a severe shortage of providers and increasing demand for services. As recent years have shown, behavioral health needs are increasing across all demographics.
This mismatch between supply and demand leads to long waiting times, problems with access to care and, in some cases, missing out on necessary treatment.
Andy Flanagan is CEO of Iris Telehealth, a provider of telepsychiatry technology and services. He holds a Master of Science in Health Informatics from Northwestern University’s Feinberg School of Medicine. His previous experience includes serving as CEO three times, as well as founding a SaaS company and holding senior-level positions at Siemens Healthcare, SAP, and Xerox.
We interviewed Flanagan about the challenges facing mental health care, how mental health providers can leverage AI risk models to ensure patients are matched with the most appropriate clinician at the right time, how AI can significantly improve the efficiency of the already overstretched mental health workforce, and how AI can increase the profitability of delivering mental health services, including telemedicine services.
Q. What are the challenges in behavioral health today? And where do telehealth and AI fit in?
A. One of the most pressing issues is the inefficient allocation of resources. Currently, our healthcare system often operates on a first-come, first-served basis, which does not always match clinical urgency.
We are not effectively prioritizing patients based on their risk levels or severity of need. This means that someone with a critical mental illness may be in line behind others with less urgent needs, potentially leading to poorer outcomes and more visits to the emergency department.
This is where telehealth and AI come in as potential game-changers. Telehealth has already proven its value, particularly in behavioral health. Approximately 55% of behavioral health encounters are now taking place virtually and this has not decreased after the pandemic, as has happened in other sectors of healthcare.
This trend is happening because telehealth removes many barriers to care: patients don’t have to take time off work, travel to appointments, or deal with the stigma that can come with visiting a psychiatric clinic in person. It’s a patient satisfaction metric and a facilitator of better clinical outcomes.
AI, on the other hand, is still in its infancy but shows enormous promise. One of the most exciting applications in healthcare is patient triage and resource allocation. AI algorithms can analyze patient data to determine risk levels and prioritize care accordingly, meaning we can move away from the current first-in, first-out model and toward one where patients who need care most urgently are seen first.
This approach can significantly improve outcomes and reduce the pressure on emergency services.
Additionally, AI can help predict outpatient access gaps and supply-demand imbalances within a health system or clinic population based on provider type, time of day, and acuity level. This predictive capability can help health systems optimize staffing and scheduling to increase productivity and patient satisfaction.
Finally, AI could help address the shortage of healthcare providers by augmenting the capabilities of existing clinicians. For example, AI could handle routine administrative tasks, freeing up clinicians to interact with patients. It could also help clinicians make more informed decisions about patient care.
AI and telehealth offer enormous potential, but they are not magic bullets. We need to think carefully about how we implement these technologies. We need to be careful about generative AI applications that could compromise patient privacy or data security.
Instead, we should focus on machine learning applications that use discrete, anonymized data to improve healthcare delivery without compromising patient data.
Telehealth has already proven its value in increasing access to care, but when combined with effective, responsible use of AI, it holds the promise of more efficient, effective, and personalized mental health care. We must use these technologies to augment, rather than replace, human care, always with a focus on improving patient outcomes and experiences.
Q. How can behavioral health providers use AI risk models to ensure patients are matched with the most appropriate clinician at the right time? And how does telehealth fit into this?
A. AI risk modeling in behavioral health analyzes a wide range of patient data to assess clinical urgency and care needs, taking into account factors such as prior diagnoses, medication history, frequency of care utilization, social determinants of health, and even real-time data from wearable devices or patient-reported outcomes.
By processing this complex information, AI can generate a comprehensive risk score for each patient, providing a nuanced understanding of the patient’s current mental health status and potential future risks.
This risk stratification enables providers to move beyond the traditional first-come, first-served model of care delivery. Instead of having patients wait in a queue based on when they requested an appointment, AI can help prioritize based on clinical need.
For example, a patient with a history of suicide attempts and recent crisis events may be flagged for immediate intervention, even if they requested an appointment after someone with milder symptoms. This approach ensures that limited clinical resources are allocated where they can have the most significant impact, potentially preventing mental health crises and reducing emergency department visits.
AI can also match patients with the most appropriate clinician based on their specific needs and the clinician’s expertise. For example, a patient struggling with both depression and substance abuse could be matched with a clinician who specializes in treating dual diagnoses. This strategy can lead to more effective treatment outcomes and higher patient satisfaction.
Additionally, telehealth enables more flexible scheduling, which complements the AI risk model’s ability to prioritize urgent cases. If a high-risk patient needs to be seen quickly, telehealth makes it easier to fit them into a provider’s schedule, perhaps even on the same day. This rapid response capability could be critical to preventing mental health crises and ensuring continuity of care.
As these AI risk models become more sophisticated and widely adopted, we may see a shift toward more proactive, preventative behavioral health care. Instead of waiting for patients to reach out when they’re in crisis, providers can use AI to identify patients who may benefit from early intervention and reach out proactively.
Q. How can AI significantly improve the efficiency of already overburdened behavioral health workers? And where does this help telehealth providers?
A. One of the most promising applications for AI-enhanced workforce efficiency is in administrative and documentation tasks. Behavioral health professionals spend a lot of time on paperwork, charting, and other administrative tasks.
AI-powered tools can streamline these processes, potentially using natural language processing to generate clinical notes from recorded sessions or automating insurance coding. This allows clinicians to focus more of their energy on direct patient care, potentially increasing the number of patients they can see without sacrificing quality.
AI can also serve as a powerful decision support tool for clinicians. By analyzing clinical data and keeping up with the latest research, AI systems can provide evidence-based treatment recommendations tailored to each patient’s unique circumstances. But AI systems should not replace clinical judgment.
For example, an AI system can flag potential drug interactions or suggest alternative treatments based on a patient’s history and symptoms. However, it is always up to the clinician to determine the appropriate level of care.
Specifically for telehealth providers, AI-powered chatbots and virtual assistants can handle initial patient intake by gathering basic information and performing preliminary assessments before a patient meets with a clinician. These clinical support tools ensure that the provider already has a comprehensive view of the patient’s situation as soon as the telehealth session begins.
Q. Can you explain how AI can improve the profitability of delivering behavioral healthcare services, including telemedicine services?
A. AI improves operational efficiency, optimizes resource allocation, and increases access to care—all of which impact a healthcare system’s profitability. AI algorithms can analyze patient data, historical patterns, and real-time factors to optimize appointment scheduling and clinician workloads. This optimization can reduce no-shows and improve clinician efficiency.
AI can even help identify patients who are at risk of discontinuing their treatment or who might benefit from more intensive care, enabling proactive interventions.
We also know that leveraging this technology effectively increases profitability by automating many time-consuming administrative tasks using algorithms to support documentation, billing, and coding processes. This reduces the administrative burden on clinicians, minimizes errors, and improves revenue cycle management.
AI can streamline the entire virtual care workflow – from patient intake to follow-up care coordination – allowing providers to focus more on direct patient care and potentially see more patients in a given time frame.
AI-driven predictive analytics identify trends in patient demand, treatment outcomes, and operational metrics to help guide strategic planning, resource allocation, and service expansion. Telehealth providers can leverage this capability to identify underserved markets or optimal times to offer certain services, leading to increased market share and revenue growth.
Follow Bill’s HIT reporting on LinkedIn: Bill Siwicki
Send him an email: bsiwicki@himss.org
Healthcare IT News is a publication of HIMSS Media.