Colorectal cancer (CRC) is a major cause of morbidity and mortality throughout the world. It is the second most common malignant disease [1,2], with 90% survival when it is identified early and immediate surgical intervention is performed. In Israel, the screening policy for average-risk individuals aged 50-75 years is an annual fecal occult blood test (FOBT) [3]. A patient with positive FOBT result requires immediate follow-up with colonoscopy, and surgery should be performed when CRC is detected. A delay in follow-up markedly undermines the benefits of CRC screening, including incidence, mortality, life-years saved, and net costs of screening [4-7]. Recommendations regarding the time between a positive result and colonoscopy vary across countries, ranging between 30 and 180 days [8-13]. In Israel, the Ministry of Health guidelines define the standard period between a positive FOBT result and a follow-up colonoscopy as 90 days [3,14]. From the literature, we learned that 40%-60% of individuals who undergo FOBT screening do not continue with follow-up after a positive result [7,15,16]. In the Israeli population, follow-up rates after a positive FOBT result are 71%, and the proportion of patients who complete follow-up varies across health care organizations, with the proportion in Meuhedet being about 50%. The median time to follow-up, nationally, is 112 days, which is significantly longer than the recommended 90 days [17].

Barriers to follow-up after a positive FOBT result have been identified in the literature [18-23] and are divided into 4 general groups based on whether they are related to patients, physicians, providers, or information technology [24-26]. Intervention programs have been developed to target each of these groups [27]. Patient-targeted interventions include educational strategies, peer counselors or navigators, reminders, and coupons. The interventions were effective for short-term, but not long-term, follow-up. Several studies have described interventions aimed at increasing follow-up rates and have showed mixed results [13,20,28-41]. When the patient-physician relation is good and based on clear communication and trust, the patient is more likely to adhere to and fulfill the physician’s instructions [42,43].

Studies addressing patient interventions have used letters, emails, telephone calls, and nurse navigators. Mailed invitations were found to be as effective as telephone reminders and increased follow-up rates by approximately 30% in an Italian study [40]. In a Scottish study, a reminder phone call that included making an appointment for colonoscopy increased adherence by 4.7% [13]. In a British study aimed at minority populations, the study nurse called patients repeatedly, invited them to the clinic, and scheduled a colonoscopy appointment for each one; this intervention increased follow-up by 8.4% [39]. All these interventions were found to be effective, but some are extremely labor-intensive and time-consuming.

To date, few studies have examined the effectiveness of short message service (SMS) text messaging as an intervention tool within community medical settings. SMS text message reminders have been shown to be effective in improving preparation for colonoscopy in Korea [44]; they increased Streptococcus pneumonia immunization in a primary care setting in Lebanon [45] by 7.2% (less than phone calls, but more than emails). A literature review examining the effect of SMS text messages and emails to improve diabetes management showed that simple phone calls, letters, or SMS text message reminders can have a positive impact on clinical and behavioral outcomes [46]. A systematic review [47] indicated that SMS text messaging interventions improved patients’ medication adherence rate. In a review of the factors associated with nonadherence to oral antiplatelet therapy in acute coronary syndrome and interventions that modify these factors, only reminder-based interventions, including SMS text messages, had consistently beneficial impacts on adherence outcomes at 3 and 12 months [48]. In a randomized controlled trial designed to assess the effects of SMS text message reminders on adherence to a healthy diet, medication, and smoking cessation among adult patients with cardiovascular disease, researchers found that SMS text messaging was effective in improving adherence to a healthy diet and medication but not smoking cessation [49]. A different study reported increased success for smoking cessation among patients attending control visits as a result of scheduled clinic appointments following SMS text message reminders. The smoking cessation rate was 24% in patients who did not respond to SMS text message reminders at all and 28.6% (n=28) in patients answering any SMS text message at least once (P=.001) [50]. In contrast, in an Australian study, SMS text message reminders were used to improve hepatitis B vaccination among high-risk sexual health center attendees, and it was found that this intervention was not effective [51].

To the best of our knowledge, no study has used SMS text messaging technology to increase adherence to recommendations for CRC screening follow-up after a positive FOBT result. Therefore, we aimed to examine the effectiveness of sending SMS text messages to patients as an automated tool to increase adherence to colonoscopy follow-up and to examine the influence of physicians’ attitude toward FOBT as a screening test for CRC and patient adherence to follow-up.

The effectiveness of mass cancer screening programs can be compromised by lack of follow-up of abnormal findings. Incomplete FOBT follow-up with colonoscopy is a significant problem that has been studied extensively. This study shows that sending SMS text message reminders to patients following a positive FOBT result is an effective way to increase adherence rates to follow-up colonoscopy. In this study, there was a relative increase of 49.2% in adherence in the intervention group using a simple, inexpensive means of communication. This surprisingly large increase in adherence could be a result of a combination of simplicity, repetition, and timeliness. Reminders were simply worded, with a clear message, sent immediately after results were obtained, and then repeated twice over the next month. Another potential advantage is that unlike telephone calls, which require active responses, the messages are “pushed.” Probably, patients who responded to the message were those who were more likely to complete the follow-up but required a “nudge.” Another possibility is that the message created cognitive dissonance in some patients who chose to ignore the results until they received the reminder. Future research should focus on how different contact methods are effective for different types of patients.

In this study, age, gender, SES, and geographic location were not significantly associated with adherence to follow-up of positive FOBT results either in the bivariate or multivariable analysis. Increasing age was previously associated with lack of follow-up in some studies [39], but not in others [21]. Similar to our findings, several studies did not find strong associations between gender and complete diagnostic follow-up [21,29,54]. However, others suggested that women are less likely than men to undergo follow-up testing [16,25,55]. Although the association between SES and follow-up was not significant, there appears to be a trend toward increasing follow-up rates at higher SES. This needs to be investigated further.

The physician plays an important role in a patient’s decision regarding follow-up tests. Trust in the physician and good communication between the patient and physician will positively influence the adherence rates of patients after a positive FOBT result. In this study, 95.1% (254/267) of physicians declared that FOBT is a reliable screening test for CRC and that they would not advise their patients to undergo FOBT again to ensure that the positive result is reliable. The physician’s attitude toward FOBT as a screening test significantly influences the patient’s adherence to colonoscopy.

A new finding of this study is that physicians’ attitude toward FOBT has a major influence on the rates of colonoscopy after a positive FOBT result. This finding may explain why intervention studies that provide physicians with knowledge and feedback have been effective. For example, Myers et al [56] showed that one-on-one physician training, audit, and feedback (physicians received lists of their patients with incomplete diagnostic evaluations) resulted in improved completion of diagnostic testing. Singh et al [57] assessed a clinic-based quality improvement activity that included provider education, a positive FOBT registry, and feedback; they found that it significantly decreased the time to colonoscopy referral and completion and increased colonoscopy completion by 18.7%. In spite of the fact that FOBT has been a standard practice for early detection of CRC, >10% of physicians in our study expressed mistrust in this method; this is possibly an underestimate, as some physicians may hesitate to express this view openly, which is contrary to organizational guidelines. It is, therefore, essential to educate physicians regarding the reliability and effectiveness of FOBT as a screening tool for CRC.

Owing to recent developments that have made digital communication within the health care sector readily available and inexpensive, SMS text message reminders could potentially be implemented in other areas. The finding of a large increase in follow-up after a positive FOBT result illustrates the need to further investigate different aspects of SMS text message usage, such as patients’ age, gender, the type of action they aim to increase, and the wording of the message. Changing technology and patient preferences with regard to contact communication should be considered when determining future interventions to improve usage and effectiveness.

A limitation of this study is that only patients who agreed to receive SMS text messages were included, which created a biased population in terms of age, cultural beliefs, and SES. An additional limitation is that SES was measured using zip code rather than a direct measure, and the SES of 56 participants was missing, although this is the standard method for measuring SES in Israel.

This study was conducted in 1 of the 4 health care provider organizations in Israel. Selection of providers was voluntary, and the member distribution in terms of age, gender, and SES was similar to that of the Israeli population. In addition, the Israeli population is very ethnically diverse and includes immigrants from many countries. Our findings are potentially generalizable to other populations.

In conclusion, this study is the first to directly evaluate SMS text message reminders for improving colonoscopy follow-up among Israeli CRC screening program participants following a positive FOBT result. We have shown that a simple, inexpensive intervention for patients improves colonoscopy follow-up after a positive FOBT result. It is important to maximize the potential of these findings by increasing the acceptance of SMS text messages within the population and to examine their use in other screening programs. In addition, it is important to examine, in future studies, the reasons because of which patients refuse to receive SMS text messages. The physician’s attitude toward FOBT as a screening test significantly influences patient adherence to colonoscopy. Therefore, further work needs to be done among physicians to increase adherence.