In patient portal surge, Black patients wait longer for responses

Background

Patient portal messaging has become a key feature of the digital healthcare experience. During the pandemic, patients and providers alike turned to secure direct messaging to keep open a line of communication even when they could not physically meet in person.

Today, the patient portal supports a wraparound digital consumer experience, allowing individuals to connect with their care teams to ask questions, book appointments and request prescription refills.

But as patient portal utilization trends upward, questions of health equity and digital haves and have-nots emerge. Beyond disparities in digital health access, researchers have questioned disparities in digital health engagement. Do all patients get equitable patient portal treatment?

In this latest study, which assessed asynchronous patient portal messages sent by nearly 342,000 patients to their primary care providers, revealed steep differences across racial lines, payer type and language preference.

PCPs respond quicker to messages from White patients

First, the researchers analyzed patient portal messages that received an answer within one day, finding that White patients were far more likely to get a fast response than their Black and Hispanic counterparts.

For example, of the 48,983 messages sent by Black patients, only 32,165 (65.7%) got a response within a day. For Hispanic patients, that number was 63.9%. Meanwhile, 68.5% of the messages sent by White patients got a one-day response.

But disparities fell beyond racial lines.

When comparing insurance type, researchers found that folks enrolled in commercial insurance were more likely to get a one-day response than dual-eligible patients. Moreover, 68.4% of messages sent by patients preferring to speak English received a one-day response, compared to 58% of those sent by folks who preferred to speak another language.

After adjustment, one-day response rates were 1.1 percentage points lower for both Black and Hispanic patients compared to White patients.

The steeper disparities emerged when looking at insurance type and language preference. The one-day response rates were 4.9 percentage points lower for dual-eligible patients compared to commercially insured patients. Additionally, they were 4.1 percentage points lower for English-speaking patients than for those with limited English proficiency.

Structural barriers limit patient portal equity

According to the researchers, structural barriers are likely driving these digital health disparities.

"Clinics more likely to care for patients from minoritized racial and ethnic groups, patients who do not have commercial insurance, and patients who do not prefer English were less likely to have the resourcing or workflows to ensure fast responses to asynchronous patient messages or to meet patients’ messaging-related language access needs in a timely fashion," the researchers explained.

Many clinics treating traditionally underserved operate under razor-thin margins. With few resources, they might not have the staff or workflows in place to answer patient portal messages in a timely manner. Moreover, they might lack the resources to prioritize hiring bilingual staff, invest in medical interpreters or adopt technologies that help translate patient portal messages sent not in English.

The data also suggest differences in how various patient populations engage in digital technologies. For example, low-income, dual-eligible patients might not have the flexibility to send patient portal messages during business hours.

Many providers have set-up dedicated time to answer patient portal messages that align with clinic time in an effort to achieve some work-life balance, meaning those sent at night and on weekends might take longer to get a response.

"Although much attention to patient messaging has emphasized workforce well-being, our results emphasize its importance for the delivery of equitable care as well," the researchers emphasized.

Many industry experts are pointing to generative AI for a fix here, as overburdened providers work to address patient concerns equitable. Indeed, the researchers acknowledged GenAI's potential for drafting responses quickly and in another language.

However, they concluded by cautioning against AI overreliance, as the technology could also fall prey to "unanticipated biases" and perpetuate unequal patient experiences.

Sara Heath has reported news related to patient engagement and health equity since 2015.