Patients may enjoy the convenience of automated texting while cursing its curtness

Many if not most provider organizations use automated texting to let patients instantly confirm or cancel upcoming appointments. And many patients mistake the incoming messages as opportunities to interact with the sender. When this happens, patients often try to communicate about more than just appointment scheduling. 

On the one hand, such misconstruals are an unfortunate but generally harmless effect. On the other, they symbolize lost opportunities to “inform the design of more transparent, responsive [communication] systems that improve patient engagement and care.”

What’s more, in large health systems, every patient’s failed attempt at meaningful texting may represent “substantial numbers of [would-be] patient interactions and reflect recurring patterns rather than isolated cases.”

So state the authors of a study conducted at Kaiser Permanente Colorado and published June 8 in JAMA Network Open. 

Bidirectional communication in demand 

Shane Mueller, PhD, MSW, and fellow researchers arrived at their conclusions after analyzing the content of 743 patient text responses to automated messages sent by Kaiser. These included not only appointment reminders but also preventive care outreach and opt-in/opt-out commands. 

The authors note that Kaiser Permanente Colorado sends around 4.4 million automated texts to more than 500,000 patients a year—twice the volume of emails it sends. 

The 743 analyzed responses came from a field of 38,400 texts sent by 28,500 patients. 

The team organized the studied texts into eight overarching domains—opt-out or opt-in requests, appointment management, requests for help or information, communication preferences, corrections of inaccurate data, comments on usefulness and usability, expressions of frustration and uninterpretable responses. 

“Replies reflected expectations for bidirectional communication, highlighted errors in patient information or pointed to usability barriers,” the researchers report. “These communications underscored limitations of exclusively automated processing and revealed opportunities to make communication systems more responsive.”

In their discussion section, Mueller and co-authors underscore the commonality of patients seeing automated text messages as opportunities for bidirectional communication. 

This finding reveals needs that “extend beyond current system capabilities,” they write. “Addressing these responses could reduce communication errors, improve appointment management and strengthen trust in healthcare communication.”

Are not LLMs at the ready for virtual 1-on-1 exchanges?  

The researchers cite prior research focusing on the effectiveness of texted appointment reminders and other basics of communication. 

By contrast, the present study shows how patients respond to automated message programs unfit to receive replies consisting of unstructured diction.

“Rather than providing predefined or easily classifiable responses, patients frequently used these messages to communicate diverse needs, preferences and concerns, revealing a broader range of communication behaviors than had been described in prior work,” Mueller et al. write. 

“These findings highlight automated messaging as a sociotechnical system shaped by the interaction between technical design and patient interpretations,” they add. In doing so, the findings “underscore the limited guidance available for interpreting or responding to these diverse communication patterns.”

Taking up the question of whether large language AI might help patients and providers communicate via text more effectively, the authors comment:

‘Prior research suggests that incorporating AI in healthcare systems can assist in patient communication, reduce the volume of messages clinicians need to address, improve patient satisfaction and reduce demand on clinician time by automating routine interactions. However, these tools require oversight to ensure safety, accuracy and data security.’

Read the full study.