Research Warns Against Too Many Automated Patient Outreach Messages

March 30, 2021 - Healthcare organizations that over-leverage their automated patient outreach technologies may run the risk of high attrition, according to new data from the Institute for Health Research at Kaiser Permanente Colorado.

The data, published in JAMA Network Open, found that patients who got a lot of automated messages from their medical providers were more likely to opt out of receiving future messages. These results could provide insights into how healthcare organizations deploy patient outreach messaging in the future, the IHR researchers said.

Automated patient outreach technologies have become integral to provider patient engagement strategies. In an effort to create efficiency and reduce staff burden, organizations have specifically tapped automated text messaging tools and interactive voice response (IVR), which rely on virtual technology instead of human support.

These tools allow organizations to send out generalized messages—“Don’t Forget to Receive Your Annual Flu Shot”—to their entire patient population. Organizations can also send targeted messages to a specific population to drive patient education and access to care.

But like automated messages in other service sectors, the ones providers send out might not be getting read or even lead patients to unsubscribe. Many marketing professionals are familiar with the notion that a consumer might opt out of receiving messages altogether if message volume gets too high. In the case of healthcare, providers want to avoid attrition to ensure patients get the messages they need.

READ MORE: Can Text Message Patient Outreach Improve Appointment Adherence?

The organizations delivering on that goal are usually ones striking a great balance between too many and not enough messages, the team at IHR found. In an analysis of over 428,000 adult patients at an integrated healthcare system, the researchers found a link between high message volumes and high attrition rates.

There indeed did prove to be such a thing as too many patient outreach messages, the researchers reported, and organizations need to be careful they don’t fall into that territory.

Most patients received some form of automated patient outreach during the one-year study period, with about 84 percent getting at least one automated text message from the provider, and about 68 percent getting at least one IVR message.

Opt-out rates weren’t alarmingly high; 2.5 percent of patients opted out of text messages, while 1.5 percent opted out of IVR messages. The researchers said this was consistent with marketing campaigns in other service sectors.

But when breaking down those who opted out of automated patient outreach, the researchers observed a key trend. Those who received higher message volumes were more likely to have opted out.

Those who received more than 10 text messages from the provider were more likely to opt out of the outreach campaign, with those receiving more than 20 annually being three-times more likely to opt out than those who received only two texts each year.

A similar trend appeared for those receiving IVR messages; individuals who got more than 10 automated phone calls each year were more likely to opt out of the program than those receiving two or less. Those getting more than 20 phone calls were the most likely to opt out of the program.

Importantly, there was also a link between those receiving both text messages and IVR messages and high attrition. Getting both forms of automated patient outreach increased the odds that a patient would unsubscribe.

Although old age was positively associated with message opt-out rates, few other sociodemographic factors were. The researchers did not that patients were likely to opt out of message campaigns after receiving messages that were mostly informational and did not offer personalized insights into their health. In other words, patients want messages that will be important to them and that do not read like a marketing blast.

The researchers also recommended healthcare organizations make note of how patients would like to be notified. This will ensure messages come to patients in the medium they prefer and reduce the risk of message fatigue.

The team also suggested organizations risk stratify patient messages. For example, appointment reminders could run through an algorithm flagging the patients most likely to miss their appointment. This targeted, automated approach is more likely to get the right kinds of messages to the right kinds of patients, the team said.

Moving forward, the researchers encouraged further investigation into this trend at other healthcare organizations. Additionally, future study could look at patient attitudes about automated patient outreach messaging and what types of messages would be acceptable to patients.

Healthcare’s hurdle toward automation will likely not subside. Automated patient outreach is both time- and cost-efficient, both of which are invaluable factors for organizations trying to boost patient engagement in a fiscally responsible way. But organizations need to take a patient-centered approach to this messaging, the researchers concluded, if they want to avoid unintended consequences of patient attrition and unsubscribes.