The Benefits of Using Text Messaging for Research
One of the biggest challenges in clinical research is how to engage participants. From recruitment to delivery of results and everything in between, researchers must communicate with subjects and coordinate a multitude of tasks. The time and tools can be burdensome for both study teams and participants. More and more, researchers are utilizing technology readily available and familiar to participants. One such technology, text messaging, also known as short message service (SMS), is an increasingly popular and affordable communication tool for research. More than 95% of the U.S. population owns a cell phone. Response time to text messages is unparalleled; greater than 90% of text messages are read within three minutes. (1) Moreover, text messaging traverses nearly all socioeconomic classes and reaches across geographic locations. In this post, we explore options for how and when to use text messages in clinical research studies.
Texting can be a powerful research tool. It can be used for straightforward tasks, such as sending reminders to take a medication and for more complex tasks, such as collecting data on pain scores and delivering real-time interventions. Text messages allow researchers to reach more participants easily. While only 20-30% of emails are opened, 98% of texts are opened. On average, it takes 90 minutes for a response to an email, but only 90 seconds for a response to a text message. (2) In addition, the response rate for text messaging is eight times higher than for email. That could represent a 6,000% increase in communication efficiency.
Text messaging can be used in a variety of ways: to contact a single participant, families or communities. One-way automated text messages are a great way to remind participants of an upcoming visit or a research task to complete. Two-way texting is an easy way for participants to confirm their appointments or provide study data. A text can be more useful than a call, especially if a participant does not answer and information must be left on voicemail. According to a recent study, 19% of millennials do not check their voicemail, (1) whereas 97% of Americans use text messaging at least weekly. (3)
Text messaging can also be used in conjunction with surveys. For short and simple daily data collection, such as medication use, hours of sleep, or number of sodas consumed, surveys can be texted. Short surveys can be sent one question at a time, and participants can respond via text. Branching logic can be built within texting software if needed to help keep surveys as brief as possible. For example, if a participant reports he or she has no pain, then a follow-up question asking about the location of that pain will not be sent. Cell phones do have small screens that make it challenging to respond to complex surveys or tasks. If you want to collect longer or more detailed data, consider texting a link to a web survey in a program such as REDCap. While that may be a limitation for participants without smartphones, for those with them it can be more effective than e-mailing or mailing a survey.
If you decide to collect data via text, be thoughtful about the process. Think about the number of back-and-forth texts, who will respond to messages, and when participants will expect a response. Set expectations for responses after work hours, on weekends, and holidays.
There are options for how to send texts. If you have a small study or only plan to send a few messages, you could use a Duke-issued smartphone to send messages manually. If you have more messages to send or want to automate the task, consider a commercial texting service, such as Twilio or Mosio. You may even be able to use REDCap to schedule text messages and manage data you receive from participants. Finally, if you have no budget for texting, but want to incorporate some semi-automated texting, you can use the SMS gateway to send an email to a mobile device. There are some drawbacks to this, but it is a start.
A few notes about regulatory approvals and security, especially for the Duke University community. The use of texting needs to be spelled out in your consent form, and consent should be obtained before sending identifiable information to a commercial entity. Texting is not a secure form of communication, so be careful about the information you include in your text messages, and encourage your participants to do so as well. Using text messaging is often an overlooked means of interacting with participants in the world of mobile apps but we encourage you to consider it! The engagement metrics speak for themselves and the costs are significantly less than building a native app.
If you want to learn more about how you can use text messages in your study, contact the Duke Mobile App Gateway for consultation.