Clinicians — including doctors, nurses, and other healthcare professionals — face high levels of stress and burnout due to the nature of their work, with about half of all healthcare workers reporting burnout in a recent Harvard study. High workload, long hours, and patient demands can all contribute to this burnout, potentially leading to decreased job satisfaction and quality of care.

Electronic health records (EHRs) are a major contributing factor to the already high workload facing healthcare workers. In 2022, the U.S. Surgeon General reported that each hour of care requires approximately two hours of EHR work, while a 2022 survey published in JAMA Internal Medicine found that the average physician spends 1.7 hours on EHR tasks outside of normal working hours. (Another study, from Medscape’s 2019 Physician Compensation Report, found that clinicians spend 10 to 19 hours a week just on paper and administrative tasks, depending on their specialty.)

Recent advances in natural language processing (NLP), generative AI, and ambient voice technology offer potential solutions to this problem and burnout contributor. By leveraging these technologies, healthcare organizations can reduce the burden on clinicians and improve their well-being, ultimately resulting in better patient outcomes. This is already a large and growing field, with Grand View Research reporting a market size of $17.7 billion in 2022, which is expected to grow to $53.6 billion by 2030.

Natural language processing is a subfield of artificial intelligence (AI) that focuses on the interaction between computers and human language. This technology allows computers to understand and interpret natural language in the same way that humans do. One of many possible applications of NLP in healthcare, and one with the direct potential to reduce burnout, is analysis and synthesizing of data points within EHRs. EHRs contain a wealth of information about patients’ medical histories, including symptoms, diagnoses, treatments, and outcomes. However, this information is often difficult for clinicians to extract and analyze, as it is buried in unstructured text amongst different file types. By identifying and extracting medically-relevant information from EHRs, NLP algorithms can simultaneously reduce the amount of work necessary to surface critical information and uncover additional insights to help clinicians make informed decisions about patient care — and with roughly 80% of all healthcare data being unstructured and therefore underutilized, there’s enormous potential for gains. For example, solutions are focused on pulling summarized data into succinct formats for quick updates on where patient acuity levels are as part of daily rounding exercises.

Ambient voice technology is another emerging technology that has the potential to revolutionize healthcare and which has already seen growth, with healthcare representing the largest market share of AI voice technology and already accounting for $1.7 billion in revenue. Ambient voice technology uses voice-activated devices like smart speakers and virtual assistant devices to interact with patients and healthcare providers, allowing them to automate actions and routine tasks like scheduling appointments and refilling prescriptions, freeing up clinicians’ time to focus on more complex tasks and top-of-license skillset. These technologies can also capture the conversations of patient encounters, freeing up time that is currently spent on manual data entry in patient charts. Ambient voice technology can even be used to provide patients with personalized health coaching, reminders, and educational content, potentially improving health outcomes and reducing the burden on clinicians.

In addition to reducing workload, NLP and ambient voice technology can also improve the quality of care that clinicians provide. By analyzing EHRs and other patient data, NLP algorithms can identify potential health risk factors and recommend preventative measures. This can help clinicians to identify health problems earlier and provide more targeted interventions, ultimately leading to better patient outcomes.

Together, these technologies have enormous potential to reduce burnout among clinicians by reducing administrative workload, improving healthcare workers’ efficiency, and creating additional capacity by freeing up time that can be used for other tasks. Automating routine tasks such as data entry and appointment scheduling also reduces the administrative burden on clinicians, allowing them to spend more time on patient care — reducing stress, improving job satisfaction, and ultimately leading to better quality of care.

However, the adoption of NLP and ambient voice technology in healthcare is not without challenges, beginning with the need for accurate and reliable data. Because NLP algorithms rely on large amounts of data to learn and improve their accuracy, healthcare organizations must ensure that their EHRs are accurate and up-to-date so that NLP algorithms can provide reliable insights. Ambient voice technology also requires patients to be comfortable with sharing their health information with virtual assistants and other voice-activated devices, placing the impetus on healthcare organizations to ensure that patients are fully informed about the privacy and security risks associated with these technologies before implementing them.

Further, while NLP and ambient voice each hold great promise, the performance of these technologies are best implemented in a “land and expand” approach based on encounter type and service line complexity. The accurate capture of information is critical for patient care and the lexicon of language capture and interpretation must achieve the highest standard of accuracy. Only in the last few years have we begun to see intelligent voice systems approach and exceed the accuracy of a human worker. This will only continue to improve, alongside technical integration of data capture into core EHR systems.

The need for effective, personal communication between clinicians and patients is critical. While NLP and ambient voice technology can automate routine tasks and provide patients with personalized health coaching, they cannot—and should not—attempt to replace the human connection that is so essential to effective healthcare. Healthcare organizations must ensure that clinicians are still able to communicate effectively with their patients, even as they adopt these technologies. This may involve providing additional training and support to clinicians to help them adapt to new technologies and maintain effective communication with their patients. Nonetheless, NLP and ambient voice technology offer promising solutions to the problem of clinician burnout and represent a strong opportunity for strategic investment.

For health systems ready to begin investing in these technologies, the first and most critical step is to engage with clinicians and nurses to find out how they’re spending their time and identify areas where NLP and ambient voice could add value. This will make it possible to align the interests of clinical, business, and technology stakeholders, and translate this general need into an immediate, actionable opportunity.

From a technology standpoint, it is also essential to understand that NLP and ambient voice are not solutions themselves, but capabilities that could be served by a variety of solutions. Navigating build vs. buy decisions and identifying the right way to acquire these capabilities—whether that be through an intelligent documentation solution, workflow automation, or robotic process automation—is vital to creating readiness, prioritizing opportunities, and successfully capturing ROI.

Although NLP, ambient voice, and AI are emerging technologies, they are already becoming vital tools for reducing burnout and improving patient experience and outcomes. For health systems that are prepared to act, the payoff will be considerable.

Photo: gpointstudio, Getty Images