The ways we communicate with our doctors have changed dramatically over the past two decades. Gone are the days when the results of an annual physical were delivered over the phone. Now we get that information via patient portals, which come with some trade-offs.
On the one hand, patient portals offer easy access to test results, medical records, and other information. On the other hand, it’s an impersonal delivery system for the most personal details possible. Also, just like with email, it can be more difficult to get a thoughtful response to any questions or concerns. Now a new study published in JAMA Network Open finds that the way we write messages to doctors can determine whether or not we get a reply at all.
When you send your doctor a message through a patient portal, it usually doesn’t get delivered straight to their inbox. Instead, there’s a human layer between you and your physician made up of nurses, medical assistants, and/or other healthcare workers. In most patient portal systems, these professionals are the ones who decide whether or not to escalate your missive to the doctor, triaging them like they might triage patients in the emergency room. And where there are humans, there’s human error, inherent biases, and other issues.
Read more: “Why Doctors Telling Patients Their Symptoms Are ‘Normal’ Can Backfire”
“In many message triage systems, nurses and medical assistants must quickly assess messages and make judgments about their urgency, complexity, and need for escalation,” study author Mitchell Tang of Columbia University said in a statement. “Given those constraints, writing style may understandably influence how messages are perceived and prioritized.”
To find out, Tang and a team of public-health researchers analyzed more than 3.6 million patient portal messages sent by over 500,000 patients from 2021 to 2023. They found that people from marginalized groups like non-whites, those who preferred a non-English language, those with lower education levels, and Medicaid or Medicare-Medicaid dual-eligible patients (the poor and elderly poor) were less likely to get a response.
They then used natural language processing to investigate how the content and writing style of the messages affected response rate. It was by no means trivial. Writing style accounted for 48 percent of the difference between white patients and Black patients, 35 percent for Hispanic patients, 60.5 percent for patients with only a high school education, and 43 percent for Medicaid patients.
Remarkably, the opening few words of the messages—unrelated to the content—proved to be some of the most significant. “One of the most important writing-style features was the message’s salutation,” Tang explained. “Simply beginning a message with ‘Dear Dr. Tang’ seemed to make it much more likely that Dr. Tang would end up responding.”
So what can we do to level the playing field?
Artificial intelligence tools could help, researchers say. For example, a patient-facing AI bot could help beleaguered patients craft their written missives to standardize the style. On the healthcare side, AI tools could be taught to filter out “stylistic noise” and summarize the content of the question or concern.
Still, the researchers caution that these AI tools—which are trained on human inputs—can often perpetuate these same biases. In the end, it may be a matter of alerting healthcare professionals of the problem. “Raising awareness alone could be a meaningful step toward remedy,” the researchers wrote.
Awareness, of course, is important for patients too. So even as healthcare becomes increasingly impersonal and our communication becomes more casual, don’t forget to formally address your doctor in a message.
Apparently, it goes a long way. 
Enjoying Nautilus? Subscribe to our free newsletter.
Lead image: Ghozy Muhtarom / Shutterstock
