A new study suggests that the way patients word their messages in online portals can significantly influence whether their doctor responds. Researchers found that simple adjustments, such as addressing the physician by their last name and crafting messages of a specific length, can boost response rates.
The study, conducted by researchers at Columbia University’s Mailman School of Public Health, analyzed over 3.6 million medical advice requests sent through patient portals by more than half a million adults between 2021 and 2023. Using computer software to examine writing styles, the team compared these patterns with the frequency of patient responses.
Messages that began with a salutation, such as "Dear Dr. [Last Name]," were met with a response about 39% of the time, compared to just 26% for messages without any greeting. This indicates that a formal salutation can make a notable difference in communication.
Beyond greetings, message length also played a role. Shorter messages, containing only four to 25 words, had a doctor response rate of approximately 21%. In contrast, messages ranging from 101 to 200 words saw a response rate of about 43%. The inclusion of multiple question marks was also associated with higher response rates.
These stylistic elements were found to be consistent across various medical topics, from medication inquiries to symptom updates. Researchers theorize that these patterns influence responses because patient portal messages often do not go directly to the physician. Instead, they are frequently triaged by nurses and medical assistants who must quickly assess urgency and complexity. Writing style, therefore, may affect how messages are perceived and prioritized.
Further analysis revealed disparities in response rates among different patient groups. Patients from historically marginalized communities were less likely to receive a direct reply. For instance, Black patients had a 3.7-percentage-point lower reply rate compared to White patients. The study found that writing style accounted for nearly half of this gap among Black patients and about 35% for Hispanic patients.
Factors contributing to lower response rates in these groups included shorter message length, less frequent use of the doctor's name in greetings, and a tendency to start messages with a negative tone. These patients also had lower rates of using multiple question marks and were more likely to have lower educational attainment or be Medicaid recipients.
"Patient portal messages are an increasingly central channel for patient-clinician communication, but there are inequities in how responsive care teams are to messages sent by patients from historically marginalized groups," stated Dr. Mitchell Tang, assistant professor of health policy and management at Columbia University. "The findings could help health systems identify potential sources of bias in portal-message triage and develop more equitable approaches to digital communication."
The study also raised concerns about the increasing use of Artificial Intelligence (AI) in clinical messaging. If AI systems are trained on historical data that reflects existing biases, they could perpetuate or even amplify these disparities in care. The researchers suggested that AI could also be part of the solution, potentially assisting patients in drafting clearer messages or helping clinicians focus on message content while filtering out stylistic elements that might hinder response.