Texting to Improve Colonoscopy Preparation and Adherence Needs More Study

This study by Mahmud and colleagues¹ describes results of their randomized clinical trial to determine whether a text message intervention could reduce colonoscopy nonattendance and improve bowel preparation quality. Among 753 patients in Philadelphia, Pennsylvania, undergoing outpatient colonoscopy, the intervention failed to improve appointment adherence as well as bowel preparation quality compared with usual care. Overall, 195 of 387 patients (53.1%) receiving the intervention texts and 210 of 386 patients (54.4%) in the usual care control group kept their appointment with either good or excellent bowel preparation.¹ Furthermore, no significant differences were found in any secondary outcomes, such as individual components of appointment adherence, bowel preparation quality, colonoscopy reschedule rates, and cancellation lead time.¹

Annually, an estimated 15 million screening, surveillance, and diagnostic colonoscopies are performed annually in the United States,² associated with a decline in colorectal cancer mortality, the second leading cause of cancer-related death.³ Colonoscopies are the most frequently used colorectal cancer screening test in the United States.

Although the findings in the study by Mahmud et al¹ were negative, they inform future research by highlighting critical issues. First, by comparing texting with previous studies in which texting (or a similar intervention) was effective, we could better understand the intervention’s key aspects (and/or study population) and work toward future effectiveness. For example, a pilot study by these authors of a nearly identical text intervention yielded improved colonoscopy show rates.⁴ Additionally, numerous other studies have found text messaging to be effective for appointment attendance.⁵

Second, with colonoscopies already in high demand (and demand expected to increase), interventions to reduce nonattendance are critically needed. This expectation is partly owing to recent recommendations that the starting age for colorectal cancer screening be lowered from age 50 years to age 45 years.³

Unfortunately, many individuals for whom colonoscopy is recommended do not complete the procedure, including more than half of those with a stool test positive for occult blood in some settings.³ Reasons for nonattendance are varied, including some that a text message, by its nature, cannot address. In a telephone survey of patients who missed colonoscopy appointments, Bhise et al⁶ identified a variety of patient-specific and health services barriers. Reported patient-level barriers were categorized as health status–related (eg, priority given to other urgent health issues), procedure-related (eg, difficulty in preparation and uncertainly about sedation and medication discontinuation), and cognitive-emotional barriers (eg, fear of pain, fear of finding cancer or complications, and modesty concerns). Importantly, 78% of respondents reported travel-related issues or scheduling challenges, with lack of a ride or travel companion being the most common travel barrier.⁶

Third, inadequate bowel preparation, which can result in missed pathological findings, including colorectal cancer, affects up to 25% of colonoscopies.⁷ As a result, repeat colonoscopy is recommended within 1 year.⁷ Thus, both unused appointments and incompletely prepared colonoscopies can result in decreased overall colonoscopy efficiency and access, and may lead to delayed diagnosis of important conditions.

Addressing appointment nonadherence and inadequate bowel preparation are especially important given the need to screen the approximately one-third of US adults who are not currently up to date with screening, expand capacity for colonoscopies deferred because of the coronavirus disease 2019 pandemic, and prospectively, individuals who will age into the 45 to 49 years age bracket. When colonoscopy appointments go unused or are ineffectively used (ie, poor bowel preparation), the impact is 2-fold: the patient who should have had an adequate colonoscopy faces potential harm, and the health care system pays a price in terms of inefficiency and reduced colonoscopy capacity. In fact, Mahmud et al¹ found that approximately one-fourth of all colonoscopy slots were lost owing to missed appointments (99 patients [13.1%]), canceled within a timeframe that would make replacement with another patient extremely difficulty (57 patients [7.6%]) or had a poor or inadequate bowel preparation (29 patients [3.9%]).¹ Efficient and effective interventions are needed to address these challenges.

Text messaging is an efficient intervention for patients scheduled for colonoscopy, assuming they have text message-enabled devices. However, to be effective, the message must be received by the patient and contain information that alters behavior to achieve the desired outcome. The intervention delivered in the study by Mahmud et al¹ was not found effective, raising questions about message receipt, content, and intensity. The intervention consisted of 10 messages sent over 7 days after an introductory message offered opting out of the reminder. It is not clear how many patients opted out of receiving the intervention. Furthermore, since the messaging system was not bidirectional, there was no confirmation the messages were received or read. Importantly, the pilot study did include bidirectional text messages, but the method was deemed to be not scalable to all colonoscopy patients. Patients who did read the ensuing messages would have received information that addressed some barriers identified in prior studies (eg, reminders about appointment dates, links to bowel preparation information). In any event, the subsequent messages did not (and realistically, could not) address many of the most common barriers reported, such as transportation challenges. In the pilot study, 76% of patients responded to the texts, often with questions about bowel preparation and food restrictions, and received a telephone call from staff to address their individual questions.⁴ In this study by Mahmud et al,¹ the text messages were the same for all colonoscopy patients; ie, they lacked customization based on patient-specific variables, such as prescribed bowel preparation, presence of diabetes, or use of anticoagulants. For example, a text message stated, “If you take any blood thinners (like warfarin), make sure you have discussed this with your prescribing doctor.”¹ One could speculate that the intervention may have been more effective if only patients using warfarin received information about warfarin discontinuation and if the message provided individualized instructions for discontinuation. Lack of specific information may also contribute to text message fatigue, thereby diluting effectiveness.

With much of the messaging focused on bowel preparation, it may seem surprising that the intervention also failed to improve this outcome. Here, Mahmud et al¹ may have been impeded by limitations to the Aronchick scale, which is applied after the colonoscopist performs washing of the colon throughout the procedure. With the use of water irrigation through the colonoscope, it is possible to clean a significant amount of stool from the colonic mucosa. Without use of a validated prewashing bowel preparation score, it cannot be determined whether text messages were effective at reducing the residual amount of stool at procedure onset.

Overall, Mahmud et al reported that 195 patients (81.6%) in the intervention group and 210 patients (82.7%) in the control group were judged to have a good or excellent bowel preparation. Ultimately, only 13 patients (5.4%) in the intervention group and 16 patients (6.3%) in the control group were deemed to have poor or inadequate bowel preparation.

Finally, it is important to note that this intervention was delivered in addition to usual care reminders already in place. In the particular practice setting, 169 patients (87.6%) in the control group and 327 patients (89.1%) in the intervention group received a call from staff and/or an automated recording. If the objective of sending text messages was to reduce the need for dedicated staff to make reminder calls, the text messages would need to be compared with a group that did not receive these calls.

It is worthwhile to attempt to find an effective intervention for the vexing problem of colonoscopy nonattendance and inadequate bowel preparation. Future studies are urgently needed to find effective solutions to optimize colonoscopy for the health of patients and health care systems.

Published: January 25, 2021. doi:10.1001/jamanetworkopen.2020.35720

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Clancy CM et al. JAMA Network Open.

Corresponding Author: Carolyn M. Clancy, MD, Department of Veterans Affairs, 810 Vermont Ave NW, Washington, DC 20420 (carolyn.clancy@va.gov).

Conflict of Interest Disclosures: None reported.

Disclaimer: The contents of this work do not represent the views of the Department of Veterans Affairs or the US Government.