AI in healthcare can save time and money on everything from self-scheduling to improved billing accuracy to streamlined rote administrative processes that help patients and providers focus on what matters. But it is important to understand the limitations and risks before jumping on this train.
Grandiose AI promises may dribble away into disappointment, but more realistic applications are already being implemented. This article will navigate the current landscape, anticipate future trends and consider how healthcare organizations can best adopt these technologies.
1. The Future of Jobs
Rather than replacing human clinicians on a broad scale, it seems increasingly likely that gen AI systems will support healthcare workers by taking care of routine tasks and freeing them up to focus on patient care and other aspects of their jobs that are better suited for their unique talents, thereby delivering more personalized care. This could include things like empathetic communication with patients, persuasion skills and big-picture integration. For example, a gen AI system that can synthesize and generate new language from scratch could streamline how electronic health records (EHRs) work by prepopulating visit summaries in the EHR with relevant research or by creating checklists, lab summaries from physician rounds or clinical orders. It may also help improve documentation by automatically scanning and highlighting any gaps or inaccuracies in note-taking or documenting, then suggesting changes to the clinician, who can then review, edit and approve by voice or text.
AI-supported technologies are already making significant contributions in healthcare, including medical imaging analysis and other workflow efficiencies. For example, AI tools can assist radiologists by interpreting medical images such as x-rays or MRIs and identifying lesions, abnormalities and other findings that a radiologist might miss. AI-supported cancer screenings can also produce results much faster than a human radiologist can, potentially saving lives and reducing costs.
Other innovations include virtual health assistants that answer frequently asked questions, provide information on how to manage a condition or connect patients with providers outside normal business hours. An aging population can also benefit from an AI-enabled caregiver who can offer round-the-clock support by responding to basic questions, providing resources and even scheduling appointments.
The emergence of gen AI in healthcare offers many valuable opportunities, but leaders must ensure that they pursue these uses responsibly. This involves understanding how to implement and train these systems, as well as having robust risk-mitigation strategies in place to address ethical issues such as transparency, data ownership and privacy, and how gen AI is used to make decisions that affect human lives.
It is also essential to recognize that the effectiveness of gen AI depends on the availability of substantial, diverse datasets. This includes data that goes beyond what is traditionally captured in EHRs and HIEs, such as a patient’s social determinants of health and their daily activities, which can play a major role in their overall health outcomes.
2. The Future of Healthcare
Whether it’s new medical devices or systems that help manage operations, Lin believes healthcare leaders are increasingly aware of what AI can do for them. In fact, he thinks that, for many organizations, the biggest opportunity will be in operational applications. These may include nurse scheduling, revenue cycle management, and even prior authorization. But he also expects that “big data” and AI will play a role in expanding access to care for underserved populations, reducing racial and ethnic disparities in quality of care, or advancing the cause of health equity.
Despite the promise of these new tools, it is important to recognize that the effective use of AI in healthcare will require rigorous testing and risk mitigations before it can be widely implemented. Without such safeguards, AI systems can make errors that are costly at best and dangerous at worst. They can also introduce biases based on race and gender. And they can be tripped up by lack of standardized, well-maintained, and updated data on key factors that influence patients’ outcomes, like social determinants, lifestyle choices, and daily activities.
To overcome these barriers, the industry needs to work together to ensure that gen AI is safe and effective in clinical practice. This means ensuring that the right datasets are available, that AI is well integrated with clinician workflows and EHRs, and that the appropriate human oversight is in place. In addition, healthcare leaders must educate their staff on how to use gen AI safely and effectively.
As the healthcare industry continues to adapt to a changing landscape, it’s likely that clinicians will shift their skills to those tasks and job designs that rely on uniquely human qualities, like empathy, persuasion, and big-picture integration. At the same time, AI will automate clerical and administrative duties to allow them to focus on more meaningful patient interactions.
One of the most promising areas for AI is in interpreting clinical data and helping doctors find the most accurate and relevant information. For example, when it comes to diagnosing disease, AI could provide a more complete picture of a patient’s symptoms and history by using machine learning to analyze the full spectrum of data in an EHR. Similarly, AI could assist in the screening process by analyzing radiology images to predict tumor location and grade. This helps to speed up diagnostics and reduce the need for manual intervention.
3. The Future of Medicine
Healthcare providers face constant pressure to do more with less. Many of these challenges stem from the fact that facilities and staff are finite. It can be challenging to provide round-the-clock patient support, respond to all phone calls, and keep up with the latest technology advances that improve care delivery. AI applications can help streamline these tasks, freeing up humans to focus on more compassionate face-to-face professional care.
Patients also want to be able to access information and resources when they need them. AI-powered chatbots can be used to answer frequently asked questions, and flag data that needs physician review. These tools can provide a more efficient response than a phone call, and help patients avoid waiting on hold or missing appointments.
AI can be helpful in reducing costs, too. It can identify potential errors in prescriptions, and help providers reduce waste through a process of “machine learning” that compares actual and predicted outcomes of a test or procedure. It can also detect anomalies that a human would miss, such as when a patient’s blood test is above or below the normal range.
Streamlining drug development is another major benefit of AI. It can identify and match drugs with new targets more quickly than traditional methods, cutting years from the process. This saves money and lives.
In addition, gen AI can help detect and prevent health changes that might otherwise go unnoticed. A company called SELTA SQUARE, for example, is using an AI system to automate the drug safety process known as pharmacovigilance (PV). PV is a legally mandated discipline that requires a tremendous amount of time and effort from pharma producers to monitor adverse reactions in patients, track them, and understand how they happen so they can be prevented.
However, some Americans worry that a more automated, data-driven approach to medicine will hurt personal connections between patients and their doctors. 57% of those surveyed say they believe that using AI in their health and medical care would make the relationship worse, while just 13% say it would make it better. It’s likely that the majority of these concerns will be addressed as gen AI continues to develop, and converge with other emerging technologies.
4. The Future of Data
In the age of big data, AI is being hailed by technologists and healthcare professionals for its ability to unlock just a piece of the $1 trillion in improvement potential that exists in the healthcare industry. It can do so by automating tedious and error-prone operational work, bringing years of clinical data to clinicians’ fingertips in seconds, and by modernizing health systems infrastructure.
Ultimately, the impact of AI in healthcare will depend on how well it is integrated into existing clinical practice. This will require overcoming challenges like data quality, privacy, and bias—as well as building trust among patients and physicians. It will also be important to ensure that healthcare workers’ needs are fully addressed and that the benefits of AI are rolled out in ways that are scalable.
The good news is that healthcare leaders are already starting to address these issues. One example is the use of a “cognitive translator” that can translate patient conversations into plain English and provide feedback to doctors, helping improve communication between clinicians and patients. Another is the use of a virtual assistant that can help patients manage their health and wellness by tracking medication adherence, scheduling appointments, and answering questions.
Other tools that have been used to support clinical care include a machine learning algorithm that can identify patterns in medical imaging, such as mammograms or lung cancer screenings, that may indicate the presence of a disease. The software can then highlight those results so a physician can take further action. In addition, an AI tool being used by cardiologists at the Mayo Clinic is able to identify people who are at risk of a heart attack or stroke in five or 10 years, even when they have no symptoms.
Lastly, an AI system is being developed that can help patients better understand their medical diagnosis and treatment options by converting text-based education materials into multiple languages and adjusting for different reading levels. This can help increase patient satisfaction and adherence to their treatment regimens.
While gen AI has the potential to make a significant impact in the field of medicine, it will be critical for healthcare organizations to carefully consider the social impacts and ethical implications of these new tools. BBB’s National Programs can be a valuable resource for companies to leverage in order to protect consumers and shape superior AI-powered healthcare experiences—both online and in-person.
