Is your doctor leaving you on read?
A new study suggests that the secret to getting your provider to respond may come down to something as simple as how you write your patient portal message.
Researchers found that using one phrase in particular could significantly boost your chances of hearing back, especially from the person you were trying to reach.
It’s a timely tip as more Americans turn to their phones and computers to manage their health. At least 12% of US adults now use secure online platforms or apps to message their doctors about appointments, test results and ongoing treatments.
And those digital exchanges are on the rise. Between 2020 and 2025, online portal messages sent to US providers more than doubled, jumping from an average of 2.2 messages per patient per year to 5.4.
But sending a message doesn’t guarantee you’ll get one back. Roughly one-third of patient portal messages don’t receive a direct response, with rates even lower for some racial and ethnic groups.
Eager to better understand why, researchers from Columbia University’s Mailman School of Public Health took a closer look.
The team analyzed more than 3.6 million medical advice requests sent through patient portals by 511,020 adults between 2021 and 2023.
They used computer software to examine the wording and writing styles in nearly 990,000 English-language messages, then compared those patterns with how often patients got a response.
The researchers found that messages that greeting the doctor by their last name got a reply about 39% of the time, compared with just 26% of those with no greeting at all.
“One of the most important writing-style features was the message’s salutation,” Dr. Mitchell Tang, assistant professor of health policy and management, said in a press release. “Simply beginning a message with ‘Dear Dr. Tang’ seemed to make it much more likely that Dr. Tang would end up responding.”
But the greeting wasn’t the only writing choice that appeared to play a role. Messages containing multiple question marks also had noticeably higher response rates.
Length mattered, too. Messages containing just four to 25 words had a doctor response rate of about 21%, while those between 101 and 200 words had a response rate of roughly 43%.
The same patterns held true across a range of medical topics, from simple questions about medications to more serious updates about symptoms.
Researchers are still trying to understand why these factors may affect whether a patient gets a reply, but they have a theory.
“Unbeknownst to many patients, these messages do not usually go directly to their doctor’s inbox,” Tang said.
“In many message triage systems, nurses and medical assistants must quickly assess messages and make judgments about their urgency, complexity and need for escalation,” she explained.
“Given those constraints, writing style may understandably influence how messages are perceived and prioritized.”
There may be other factors at play as well. The team observed that patients from historically marginalized groups were less likely to get a reply from their care team.
Compared with White patients, for example, Black patients had a 3.7-percentage-point lower rate of replies from the doctor they were trying to reach.
When researchers dug deeper, they found that writing style accounted for nearly half of that gap among Black patients.
The same factor explained about 35% of the gap for Hispanic patients, roughly 61% for patients with only a high school education and approximately 43% for Medicaid recipients.
These patients tended to write shorter notes and were less likely to use the doctor’s name in a greeting or include question marks. Medicaid patients and those with lower education levels were also more likely to start with a negative tone, which was linked to lower reply rates.
“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,” Tang said.
“The findings could help health systems identify potential sources of bias in portal-message triage and develop more equitable approaches to digital communication.”
But the study author’s warned that there’s another piece of the puzzle that could complicate matters even further. AI tools are increasingly being used in clinical messaging, helping with everything from triage and summarizing messages to drafting responses.
The researchers warned that if those systems are trained on historical data, they could learn and repeat the same biases that have already shaped human decisions, potentially hurting patient care.
“AI may also be part of the solution, with thoughtful implementations reducing rather than perpetuating bias,” the team wrote.
For example, they suggested that AI could help patients draft their messages, while also helping doctors focus on the substance of those messages and filter out stylistic quirks that might make them less likely to respond.
