Doctors May Ignore Your Messages If They’re Not Well-Written

In recent years, the way we communicate with our doctors has evolved significantly. Instead of receiving annual physical results over the phone, we now receive them through patient portals, which provide easy access to test results, medical records, and other information. However, this impersonal method of communication can lead to challenges in receiving thoughtful responses to questions and concerns.

A new study published in JAMA Network Open reveals that the writing style of messages sent to doctors through patient portals can significantly impact whether or not a response is received. When a doctor receives a message via a patient portal, it typically doesn't go directly to their inbox. Instead, it passes through a human layer comprising nurses, medical assistants, and/or other healthcare workers who triage the message like they might triage emergency cases.

The study's author, Mitchell Tang from Columbia University, explained that in many message triage systems, these professionals must quickly assess messages and make judgments about their urgency, complexity, and need for escalation. Due to these constraints, writing style may understandably influence how messages are perceived and prioritized.

Using natural language processing, the researchers analyzed over 3.6 million patient portal messages from 2021 to 2023, involving more than 500,000 patients. They discovered that marginalized groups, including non-whites, non-English language speakers, those with lower education levels, and Medicaid or Medicare-Medicaid dual-eligible patients, were less likely to receive a response.

Writing style accounted for significant disparities in response rates, with an impact of 48% for white patients compared to Black patients, 35% for Hispanic patients, 60.5% for those with only a high school education, and 43% for Medicaid patients.

The opening few words of messages proved to be particularly crucial in determining the response rate. Tang noted that beginning a message with "Dear Dr. Tang" significantly increased the likelihood of a response. While artificial intelligence tools could potentially address this issue by helping patients craft their messages with standardized styles, the researchers cautioned that these tools may inadvertently perpetuate biases.

They suggested raising awareness and educating healthcare professionals about the issue as a potential remedy.