Reducing Physician Jargon with AI
As the Covid-19 pandemic continues to devastate our communities, we find ourselves in a moment where patient adherence has never been more critical, yet doctor-patient communication is more challenging than ever. As healthcare providers we know how crucial establishing connections with patients is for effective care, but these days masks, distance, small video screens, connectivity, and misinformation all pose substantial added barriers. While there may be no way around these additional hindrances, it is essential clinicians work to reduce other factors that negatively impact doctor-patient communication.
When it comes to patient-interaction faux-pas, the use of medical jargon always tops the list. Obviously no clinician is intentionally using terms their patients don’t understand, but 2007 study, 81% of physician-patient encounters involved at least one unexplained piece of medical jargon. The reality of the situation is healthcare providers, usually extremely pressed for time, often don’t even realize when an accidental piece of jargon creeps into their conversation with a patient. It goes without saying that you should explain what a hematoma is, but what about fracture? One study of 249 emergency room patients reported that 79 percent did not know that the word hemorrhage was the same as bleeding and 78 percent did not know that a fracture was a broken bone. Up until this point the only remedy to the situation was to implore doctors to be more mindful, but due to recent advances in medical technology Artificial Intelligence (AI) may offer new tools to reduce the use of accidental jargon and aid in provider-patient communication.
How do we effectively catch ourselves, self-evaluate our communication, when we are more burnt out than ever and frankly fighting for our own lives?
Artificial Intelligence (AI) can help us. We now have tools for medical transcription services with Artificial Intelligence equipped to recognize medical jargon. We know that transcription services can free us to focus on building that rapport with our patients. We can check those transcripts after the fact to ensure that when we introduce new medical terms we immediately followed-up with simple language definitions. But transcription services that utilize AI have the potential to augment our doctor-patient communication--in real time. Transcription services that utilize AI can already synthesize transcripts into notes--they also have the potential to further analyze those transcriptions to provide us with metrics to succinctly evaluate, for example, our ratio of technical to simple language overall, or how well we interspersed technical terms with simple language. As a visit ends, we could check those metrics to ensure that our communication was accessible to our patients, and if not take more time to ensure that our patients understand us.
As Natural Language Processing (NLP) learns to recognize speech patterns, it can be trained to understand when patients’ inflections indicate that they may not have understood us. If incorporated with real-time transcription, these tools could flag not only when that happened but also and what terms potentially triggered that confusion. We could have a live record letting us know that we need to break down a term that we might not have expected, like ‘fracture,’ with simple language for our patient.
With these tools we might be able to be notified when techniques to ensure that patients understand us are effective and when we might need to spend more time following up with our patients--in real time. For example, when we ask a patient to regurgitate an explanation of their diagnosis or plan of care, are they carefully repeating the words we say without really understanding what they mean, or do they introduce new (accurate) language demonstrating that they clearly understand?
AI transcription services can allow us to shift our attention away from note taking to fully engage with our patients so that we don’t miss that masked expression of confusion. These tools have the potential to augment our quality of care and our capacity for care at a moment when we are desperate for both.
