短信提醒与患者导航:哪种方法更有效提高结直肠癌筛查参与度?
Which is more effective at improving uptake of bowel cancer screening: text message reminders or patient navigation? Results from a three-arm RCT
研究通过一项随机对照试验,比较了短信提醒和患者导航两种方法在提高结直肠癌筛查参与度方面的效果。结果显示,短信提醒和患者导航都能显著提高参与度,但患者导航在特定人群中效果更佳。
研究对比了短信提醒和患者导航两种筛查促进方法的效果,为公共卫生政策提供了数据支持。
Introduction
Colorectal cancer (CRC, also known as ‘Bowel cancer’) is a leading cause of cancer-related mortality in Europe [1]. Several large randomised controlled trials (RCTs) have shown that biennial faecal immunochemical test (FIT) screening can improve CRC outcomes by detecting cases early [2]. As a result, many countries across Europe have implemented programmes based on the test [3].
In England, the National Health Service (NHS) began offering biennial FIT screening to adults, aged 60–74 years, in 2019. The screening programme, referred to as ‘The NHS Bowel Cancer Screening Programme’ (BCSP), is delivered through five national hubs, which send kits to patients, registered with a general practice (GP), within their respective regions. The delivery of invitations is an automated and centralised process, in which information held by the patient’s GP, including their date of birth, NHS Number, name and address, are shared with the BCSP to facilitate the process. Date of birth is used to identify individuals who are eligible for screening, while NHS Number, name and address are used to send invitations to the appropriate individuals. The programme was extended to include adults aged 50-59 in 2024.
As with any screening programme, achieving the public health benefits of FIT-based CRC screening is dependent on high and equitable uptake. However, despite being free at the point of use, uptake of FIT screening is suboptimal, with stark social and regional inequalities [4, 5]. Uptake is particularly low in London, where there is an eight-percentage point difference with the national average (63.8% vs. 71.8% [2024]) [4]. This continues to prevail, despite several interventions having been shown to reduce inequalities, and subsequently implemented. For example, an enhanced reminder letter, sent to non-responders, was previously shown to differentially increase uptake among individuals living in more socioeconomically deprived areas, and was embedded within the programme as a result [6]. Two interventions, which have not been implemented widely, but are being considered or used locally, are text message reminders and patient navigation (PN) [7, 8].
Text message reminders, often referred to as short messaging system (SMS) reminders, are of particular interest to researchers seeking to reduce inequalities in screening, due to their relatively low cost, non-confrontational nature, and the ubiquity of mobile telephones among the adult population (a recent Survey by the Office for National Statistics indicated that over 95% of adults living in the UK own a mobile phone) [9]. A systematic review of text message reminders in cancer screening found them to be effective at improving participation rates by up to 3.3% for bowel cancer (compared with usual care), with the effectiveness of the reminder being contingent on a number of factors, including the content of the message, and the number of messages sent [10].
PN is also a telephone-based intervention, one which involves the provision of personal support, from a trained ‘navigator’, who help patients overcome personal or system-related barriers to healthcare [11]. As with text message reminders, PN is an intervention that has been of interest to researchers seeking to reduce inequalities in screening, due to the potential to deliver tailored information and support to individuals. Previous studies have shown that such navigation can increase screening participation [11], with a review and meta-analysis, conducted by Nelson and colleagues, showing up to a 25% increase in participation in bowel cancer screening [12].
While strong evidence indicates that both text message reminders and PN are effective, and the aggregate data indicate that PN results in larger effect sizes, no studies have directly compared the effectiveness and cost-effectiveness of these interventions with one another. As such, further research is needed to determine the most effective and cost-effective approach to promoting bowel cancer screening uptake. The aim of this study was to directly compare these two interventions, to identify the most effective and cost-effective means to promote uptake of FIT screening in London / reduce inequalities in uptake between London and England.
Methods
Study design
To address the aims of our study, we conducted a single-blind, three-arm RCT.
Setting
The study was conducted across the London boroughs of Barking & Dagenham, Brent, Ealing, Harrow, Havering, Lambeth, Lewisham and Redbridge. These boroughs were selected for their geographic distribution (covering North East, South East and North West London, collectively), low screening uptake and poorer CRC outcomes (see Appendix 1).
Participants
Potentially eligible participants were adults, aged 50–74 years, who had not returned a FIT screening kit within 18 weeks of being sent their invitation. Participants were identified at 18 weeks, on the basis that their screening episode is ‘closed’ at this time, and they are considered ‘responders’ or ‘non-responders’, depending on whether they have completed and returned their FIT kit. Targeting this cohort enabled us to focus interventions on those who do not participate without additional support.
Participant identification
Potentially eligible participants were identified by the London Bowel Cancer Screening Hub, which is responsible for delivering invitations to adults registered with a GP in London. Potential participants were identified using routine information stored on the Bowel Cancer Screening System: an electronic system that provides up to date uptake data for individuals enroled in the BCSP. Specifically, potentially eligible participants were identified by identifying individuals who: (1) had been invited for bowel cancer screening, 18 weeks previously, (2) had not returned their test kit since being invited, and (3) were registered with one of 91 GPs consenting to participate in the study (see ‘GP recruitment’, below, for further information).
Exclusions
Potentially eligible participants were excluded if they had a recorded ‘National Data Opt Out’ (NDOO) request in their GP record. A NDOO is “a request, expressed by a registered patient, logged with a GP Practice, to indicate that personal identifiable information relating to them should not be disseminated or published by NHS Digital” [13].
To remove NDOOs, the London Bowel Cancer Screening Hub shared the NHS numbers of potentially eligible participants with NHS England, using the ‘Messaging for Social Care and Health (MESH)’ system. NHS England then removed the NHS numbers of those with a recorded NDOO request, in their GP record, and returned the cleansed list to the London Bowel Cancer Screening Hub.
The process of identifying potentially eligible participants / excluding those with a NDOO request was repeated on a weekly basis (from 23rd June 2025 until 14th July 2025 [the study period]), with new batches of potentially eligible participants identified each week, over a 4-week period (see Appendix 2 for data flow diagram).
Randomisation
Once potentially eligible participants had been identified, and NDOOs excluded, the London Bowel Cancer Screening Hub randomly allocated the eligible participants to one of three trial arms, in a 1:1:1 ratio, using computerised, pseudo-random allocation (i.e. Microsoft Excel’s ‘RAND’ function). Specifically, participants were randomly allocated to receive either:
-
1.
Usual care (‘Usual care’)
-
2.
Text message reminders only (‘SMS’)
-
3.
An initial text message reminder, followed by PN telephone calls (‘SMS + PN’)
Interventions
Participants randomised to the usual care group received no further communication from the London Bowel Cancer Screening Hub, as per standard practice. Participants randomly allocated to the ‘SMS’ group received text message reminders 19, 21, 23 and 25 weeks after their initial invitation. Participants in the SMS + PN group, meanwhile, received a text message reminder, 19 weeks after their initial invitation*, followed by PN calls at 21, 23 and 25 weeks.
*Participants in the SMS + PN group received a text message reminder at 19 weeks, prior to the PN calls. This allowed the navigators to confirm which participants’ telephone numbers were active, and thereby call those with an active telephone number.
Participants successfully contacted by telephone in the SMS + PN group also received an SMS directly following their call. The SMS contained a link directing them to a short online survey, which collected data on acceptability and satisfaction with the intervention. The wording of the text message and survey items is provided in Appendix 3.
The wording of the items used for the survey was consistent with messages routinely used for service evaluation by iPlato (the company delivering the interventions – see below for details).
The CONSORT diagram presented in Fig. 1 presents an overview of the study design.
CONSORT diagram.
SMS
The text message reminders used for both the SMS and SMS + PN groups were the same. They reminded recipients that they had not yet returned their kit and could request a new kit via Freephone. A link to opt out of receiving further text messages and a link to relevant information in different patient languages were also included (See Appendix 4).
PN
The PN telephone calls similarly reminded participants that they had not yet completed their screening test and could request a new kit, should they wish to take part, via the London Bowel Cancer Screening Hub. In addition, the PN telephone calls elicited barriers to participation, which the navigator provided advice on, so as to support the individual to complete the test (see Appendix 5 for a copy of the PN script used for this study).
Where a participant did not answer the first PN call, a second PN call attempt was made at a different time on a different day of the week. If the participant did not answer the second PN call, a third PN call attempt was made on a day and time that was different to the previous two.
Intervention development
The development of the wording for the text message reminders, and PN script has been described previously, and can be found in the published manuscript for the study protocol [14].
Intervention delivery
The SMS and PN interventions were delivered by iPlato: an mHealth company with experience delivering SMS reminders and PN for cancer screening programmes (e.g. breast). iPlato were selected as the vendor for this study, on the basis that they provide mHealth services (e.g. flu vaccine reminders) to over 90% of London’s GPs, meaning that no new software or systems had to be installed [15].
The navigators were all experienced in delivering PN, having previously delivered PN to breast cancer screening non-attenders, as well as having completed training modules provided by Health Education England, including: (1) Behaviour Change Literacy for Individuals and Workforce; (2) Cancer in the Community; (3) Communicating with Empathy (Making the most of listening); (4) Data Security Awareness; (5) Population screening; (6) Shared decision making; (7) The changing story of cancer and (8) cultural competence.
To deliver the interventions, the London Bowel Cancer Screening Hub transferred lists of participants randomised to the SMS and SMS + PN groups to iPlato, on a weekly basis. These lists were sent in the form of a database, which included participant NHS numbers (required for iPlato to find the mobile numbers of the participants, and deliver the interventions to them), as well as items for iPlato to populate as they delivered the interventions. These items included: (1) the date and time any text message reminders were sent, (2) the time and date any PN call attempts were made, (3) whether the text message reminder failed to send, (4) whether the participant answered the phone during the PN call attempt, (5) whether the participant opted out from receiving further text messages (6) reasons for non-participation discussed during the PN call, and (7) participant responses to the PN survey.
Data extraction
At the end of the study, once all interventions had been delivered, iPlato transferred the populated databases to the London Hub. The transfer took place 26 weeks after the cohort received their screening invitation. The reason for this is that test kits completed within 26 weeks of invitation contribute toward uptake, while test kits returned after this contribute toward coverage.
Upon receiving the databases from iPlato, the Hub extracted data on: (1) whether the participant completed their test kit within 26 weeks, (2) whether the test result was normal or abnormal and (3) known confounders (including age, sex, area-level deprivation [derived from the participants postcode, using the index of multiple deprivation: IMD], whether it was their first screening episode and the geographic region of their home address). These data were extracted from the Bowel Cancer Screening System and were added to the databases.
Anonymisation
Once the data had been extracted from the Bowel Cancer Screening System, the databases were pseudonymised, with identifiers removed and replaced with a participant ID. The pseudonymised dataset was then sent to a member of the health promotion team, within the London Bowel Cancer Screening Programme, for analysis.
Blinding
The health promotion officer analysing the data was blinded to the participants group allocation until the end of the trial, when all data had been collected and anonymised. Due to the overt nature of the interventions, it was not possible to blind participants to the interventions they had been assigned; however, participants were not aware they were receiving the interventions as part of a trial, minimising the Hawthorne effect (a phenomenon where individuals modify their behaviour when they are aware they are being observed [16]) and self-selection bias (see ‘Consent’, below, for details).
GP recruitment
As the London Bowel Cancer Screening Hub did not have access to patient mobile telephone numbers, these had to be sourced from GP clinical systems. iPlato has access to these, where a GP is using their mHealth services; however, they can only use them for specific purposes where permission has been provided. In order for iPlato to use these numbers for the purposes of the study, consent had to be obtained.
To obtain GP consent, an information sheet, consent form, and invitation letter were sent to GPs, via email, from the Joint Screening and Immunisation Lead for Cancer Screening in London (Dr Josephine Ruwende, NHS England). GPs were required to sign and return the consent form to the chief investigator (Dr Robert Kerrison, UCL and University of Surrey), in order to participate in the study.
A total of 323 GP Practices were eligible to participate in the study (see Appendix 1); however, to ensure workforce capacity was not unduly strained, we capped recruitment to approximately 30% of practices (n = 92). The practices included represented a broad range of geographies across London, with approximately even distribution across North East (n = 34), South East (N = 30) and North West (n = 27) London. As stated above, the aim was to recruit practices with lower uptake and more deprived populations. The recruitment strategy resulted in a cohort where only 5% of participants lived in the least deprived quintile of areas (see Table 1).
Primary outcome
The primary outcome of the study was uptake within each trial arm (uptake is defined as the return of a completed FIT kit within 26 weeks of invitation [kits returned after 26 weeks contribute to coverage]).
Secondary outcomes
The secondary outcomes were kit result (defined as ‘normal’ or ‘abnormal’), text-message op-out rates, and reasons for non-participation discussed during the PN calls (documented after each call, by the PN).
PN process evaluation
Process evaluation assessed the optimal day and time to call patients, acceptability and satisfaction receiving a PN call (assessed through the online survey) and navigator experiences delivering PN for the study (assessed through an end of study meeting).
Sample size
The sample size requirement for the study was 901 participants per trial arm (2703 total). The sample size calculation for the study has been described previously and can be found in the published manuscript of the study protocol [14].
Data analysis
Descriptive statistics (i.e. frequencies and percentages) were used to report the demographic characteristics, and primary and secondary outcomes, overall and by study group.
Differences in demographic characteristics, between study groups, were assessed using the chi-square test. Differences in primary and secondary outcomes, meanwhile, were assessed using univariate and multivariate logistic regression, before and after adjusting for covariates. As there were three study groups, pairwise comparisons were performed to test for differences between any two groups (an adjusted p value of 0.025 was used as the threshold for statistical significance, as per the Bonferroni correction method).
The analyses were performed on an intention-to treat basis and per-protocol basis, so that both the ‘real world’ impact and the ‘potential’ impact of the interventions could be assessed. To conduct the per-protocol analyses, we included individuals who received a text message reminder for the SMS group (confirmed digitally), and individuals who answered the phone in the SMS + PN group (confirmed manually). To minimise risk to confidentiality, the NHS numbers of individuals in the control group were not sent to iPlato, and so it was not possible to determine whether they had a recorded mobile number, which would have allowed us to compare the intervention groups to a more comparable control in the per-protocol analysis.
We also performed subgroup analyses, to assess the effectiveness of the interventions, by demographic subgroups, on an intention-to-treat and per-protocol basis, using univariate and multivariate logistic regression. An exploratory analysis examined which reasons for non-participation PN was effective at addressing (descriptively).
Finally, we calculated the cost, per additional person screened, for the text message reminders and PN. This involved dividing the total cost of the interventions, by the number of additional people screened in each intervention group.
All analyses were performed using SSPS statistics (version 27.0).
Missing data
There were no missing data for the primary outcome or co-variates. As such, complete case analysis was conducted.
Consent
Consent to participate in the study was not obtained from participants. This was to ensure the ecological validity of the study (i.e. avoid the Hawthorne effect and self-selection bias). The requisite approvals to include adults, without their consent, were obtained through the Confidentiality Advisory Group (CAG), who review ethics applications seeking to process patient data without their consent.
As part of the approval, participants were notified they had been included in the study, without their consent (and the reasons why), at the end of the study. This was achieved through a debrief letter (Appendix 6), mailed directly to the participants. The letter was sent by the London Bowel Cancer Screening Hub and included details for the chief investigator, should they wish to discuss the matter. The debrief letter also included an invitation to participate in a short follow-up interview, for those who received a text message reminder or PN telephone call reminder. None came forward for interview.
Trial registration
This trial was registered prospectively with the International Standardised Randomised Controlled Trials Number Registry (ISRCTN): https://www.isrctn.com/ISRCTN17245519. The study protocol was published in BMJ Open [14].
Transparency
This study has been reported in accordance with CONSORT guidelines. Please see Appendix 7 for a copy of the CONSORT checklist.
Patient and public involvement
Patient and public involvement was embedded throughout the research. In the development phase, the wording of the text message reminder and PN script were co-created through online discussion [14]. In the analytic stage, meanwhile, members of the public contributed to the interpretation of data, resulting in the development of a plain English summary (see Appendix 8).
Ethical approval
Approvals to conduct the research were obtained from University College London’s Joint Research Office (Ref: 150666), the Screening Research, Innovation and Development Advisory Committee (Ref: 2223 014 BCSP Kerrison), the Health Research Authority (Ref: 22/WM/0212) and the Confidentiality Advisory Group (Ref: 22/CAG/0140). All methods were performed in accordance with the relevant guidelines and regulations.
Results
Sample characteristics
In total, 3129 participants were eligible for inclusion in the study, following the removal of NDOOs. Of those eligible for inclusion, the majority were aged 50–59 years (n = 1805, 57.7%), male (n = 1761, 56.3%) and lived in the first or second most deprived quintile of areas (n = 1793, 57.4%). Half had been invited for bowel cancer screening previously (n = 1568, 50.1%).
There were no statistically significant differences in the distribution of demographic characteristics between study groups (see Table 1).
Intention-to-treat analysis
In the intention-to-treat analysis, uptake of bowel cancer screening was 2.0%, 6.5% and 7.5% in the usual care, SMS and SMS + PN groups, respectively (see Table 2). Uptake was statistically significantly higher in the SMS and SMS + PN groups, compared with the usual care group (OR: 3.43, 95% CIs: 2.09, 5.64, p < 0.001 and OR: 3.99, 95%CIs: 2.44, 6.52, p < 0.001, respectively); however, there was no statistically significant difference in uptake between the SMS and SMS + PN groups (OR: 1.16, 95% CIs: 0.83, 1.16, p > 0.025).
There were no statistically significant differences in uptake by demographic characteristics, including age, sex, IMD quintile or area (all p values > 0.05). Results for the multivariate analysis were the same, with one exception: people living in IMD quintile 4 were statistically more likely to participate in bowel cancer screening than people living in IMD quintile 1 (see Appendix 9).
Per-protocol analysis
Of the 1051 individuals allocated to the SMS group, 307 (29.2%) either did not have a mobile telephone number recorded on the GP clinical system, or the number was not up-to-date, and messages failed to send. As such, only 744 (70.8%) received the text message reminders as intended and were thereby eligible for inclusion in the per-protocol analysis.
Of the 1023 individuals allocated to the SMS + PN group, 524 (51.2%) either did not have a mobile telephone number recorded on the GP clinical system (n = 286, 28.0%) or did not answer the phone when called (n = 238, 23.2%). As such, only 499 (48.8%) received the PN telephone call as intended and were thereby eligible for inclusion in the per-protocol analysis. The sample characteristics of participants included in the per-protocol analysis are presented in Table 3.
Inferential statistics revealed that groups were unbalanced in the per protocol cohort, suggesting that some populations were less likely to have a valid mobile number. To test this hypothesis, we ran descriptive statistics and adjusted regression to explore possible variations (see Appendix 10). The analysis revealed adults aged 70–75 years were significantly less likely to have a valid mobile number stored on their GP’s system, as were men and individuals registered with a practice in specific regions (e.g. Ealing).
In the per-protocol analysis, uptake of bowel cancer screening was 2.0%, 8.6% and 12.6% in the usual care, SMS and SMS + PN groups, respectively (see Table 2). Uptake was statistically significantly higher in the SMS and SMS + PN groups, compared with the usual care group (OR: 4.62, 95% CIs: 2.79, 7.63, p < 0.001 and OR: 7.09, 95% CIs: 4.27, 11.76, p < 0.001, respectively). Uptake was also significantly higher in the SMS + PN group, compared with the SMS group (OR: 1.54, 95% CIs: 1.06, 2.22, p < 0.025).
While there were differences in the sample characteristics, between groups in the per-protocol cohort (Table 3), these characteristics were not associated with uptake (all p values > 0.05; see Table 2). As with the intention-to-treat analysis, results for the multivariate analysis were the same, with the exception that people living in IMD quintile 4 were statistically more likely to participate in bowel cancer screening than people living in IMD quintile 1 (see Appendix 9).
Subgroup analysis
Subgroup analyses explored the effectiveness of the interventions to improve uptake, by demographic subgroup, on an intention-to-treat and per-protocol basis. The intention-to-treat analysis indicated that, on an intention-to-treat basis, uptake was statistically significantly higher in the SMS and SMS + PN groups, for most subgroups, when compared with usual care (see Table 4; see Fig. 2). When comparing subgroup differences in uptake, between the SMS and SMS + PN groups, specifically, only sex was statistically significant, with uptake being statistically significantly higher among men in the SMS + PN group, compared with the SMS group (7.6% vs. 4.8%; p > 0.025).
a Variation in uptake group allocation (Intention-to-treat). b Variation in uptake by age and group allocation (Intention-to-treat). c Variation in uptake by sex and group allocation (Intention-to-treat). d Variation in uptake by IMD and group allocation (Intention-to-treat). e Variation in uptake by region and group allocation (Intention-to-treat). f Variation in uptake by invitation and group allocation (Intention-to-treat).
In the per-protocol analysis, similar observations were made, with uptake being statistically significantly higher in the SMS and SMS + PN groups, for most subgroups, when compared with usual care (see Table 4; see Fig. 3). When comparing uptake between the SMS and SMS + PN groups, sex was, again, the only statistically significant variable, with uptake among men being statistically significantly higher among men in the SMS + PN group, compared with the SMS group (14.5%vs. 6.7%; p < 0.025).
a Variation in uptake by group allocation (Per-protocol). b Variation in uptake by age and group allocation (Per-protocol). c Variation in uptake by sex and group allocation (Per-protocol). d Variation in uptake by IMD and group allocation (Per-protocol). e Variation in uptake by region and group allocation (Per-protocol). f Variation in uptake by invitation and group allocation (Per-protocol).
Results for the multivariate analysis were the same (see Appendix 11).
Clinical outcomes
All test kits returned in the control (n = 21), SMS (n = 69) and SMS + PN group (n = 77) had a definitive normal result (i.e. <120 μg Hb/g). As such, there was no evidence that those completing the test kit, as a result of the intervention, were more or less likely to have pathology detected (e.g. adenomatous polyps, colorectal cancer, etc.)
PN Process evaluation
Of the 1,243 adults who received a text message reminder across the SMS and SMS + PN groups, 11 (0.9%) opted out of receiving further text message reminders (nine in the SMS group and two in the SMS + PN group). Of the nine individuals who opted out in the SMS group, two opted out after receiving the first text message, three opted out after the second, one opted out after the third and three opted out after the fourth.
When it came to contacting participants, they were most likely to answer the phone on a Friday (n = 10, 55.6%), followed by a Monday (n = 56, 48.7%) or a Tuesday (n = 96, 46.4%). Participants were most likely to answer the phone between 1 and 2 pm (n = 60, 54.1%; Appendix 12).
Among those who answered the phone, the most common reasons given for non-participation were: ‘Thought they were up to date’ (n = 97, 19.4%), ‘Did not receive the test kit’ (n = 97, 19.4%), ‘Been busy/not got around to it’ (n = 67, 13.4%), ‘Not interested in taking part’ (n = 46, 9.2%), ‘Away when kit arrived/moved address’ and ‘Lost the kit’ (n = 32, 6.4%). There were no apparent differences between men and women (Appendix 13).
Uptake was highest among those whose reason for non-participation was: ‘Thought they were up to date’ (uptake: n = 29, 29.9%), followed by ‘Been busy/not got around to it’ (uptake: n = 12, 17.9%), ‘Lost the kit’ (uptake: n = 5, 15.6%) and away/moved address (uptake: n = 5, 11.6%). One notable difference was that uptake was higher among men whose reason for non-participation was ‘Thought they were up to date’, compared with women reporting the same reason (39.6% vs. 20.4%; Appendix 14).
Of 499 individuals who received a PN telephone call, 22 (4.4%) completed the online survey. Among responders, the mean acceptability and satisfaction scores were 4.27 (SD = 1.16) and 4.55 (SD = 0.91), respectively. These scores indicate that acceptability and satisfaction with the PN telephone call were high (5 being the highest score). These findings were corroborated by the navigators, who stated: ‘the majority of patients were grateful for the reminder’.
Feedback from the navigators on their own experiences was also positive. The navigators were asked to rank, from 1 to 5, whether they enjoyed making the calls (5 being the highest, 1 being the lowest). All navigators responded 5. Navigators also reported they were ‘concerned about calling patients out of the blue, but wasn’t a problem” and that “The training that [you] provided for the team to do before making the calls was really useful. It gave an understanding of the programme and how important screening is, and they were able to pass that information on to patients when they spoke to them’.
Cost per additional person screened
To determine the cost of text-message reminders, for each additional person screened, we multiplied the cost of one text-message, by the total number of text messages sent, and divided this by the additional number of people screened (calculated by subtracting the proportion screened in the usual care group from the proportion screened in the SMS group). This gave a figure of £1.82 (see Appendix 15).
To determine the cost of PN, for each additional person screened, we divided the total cost of delivering PN by the additional number of people screened (calculated by subtracting the proportion screened in the SMS group from the proportion screened in the SMS + PN group). This gave a figure of £62.38 (see Appendix 15).
Discussion
Summary of main findings
The results of this study demonstrate that text message reminders and telephone PN are effective at increasing bowel cancer screening uptake among non-responders. The results also demonstrate that, while telephone PN is not more effective than text message reminders on an intention-to-treat basis, it is more effective on a per-protocol basis (indicating that, if reach could be improved, PN could improve overall uptake more than TMRs).
Interestingly, the results indicate that telephone PN is more effective than text message reminders for men, specifically, with significant improvements in uptake observed on both an intention-to-treat basis and per-protocol basis. Conversely, no improvement in uptake was observed among women receiving telephone PN, compared with women receiving text message reminders, on either an intention-to-treat or per-protocol basis.
The results of this study also suggest that PN telephone calls and text message reminders are highly acceptable, with positive feedback from patients and navigators for PN, and low opt-out rates for those receiving text message reminders (these results should be interpreted with caution, however, due to the low response rate).
Finally, the results suggest that the cost per additional person screened, using text message reminders, is fairly low, while the cost per additional person screened using PN is comparatively high.
Comparison with previous literature
To our knowledge, this is the first study to compare the effectiveness of text message reminders against telephone PN, with a recent review (2023) finding no such similar studies [17]. The results of our study are consistent with previous studies independently comparing the effectiveness of text message reminders and telephone PN against usual care. For example, a systematic review of studies testing the effectiveness of text message reminders, against usual care, found them to be effective at improving uptake by up to 3.3% (in the context of bowel cancer screening) [10]. Similarly, a meta-analysis found that PN can increase bowel cancer screening participation by up to 25% [12].
The finding that telephone PN is more effective at increasing uptake among men appears to be consistent with a single-arm study testing PN in Lancashire, which also reported higher uptake among previously non-responding men than women [18]. Conversely, the finding that text message reminders are not more effective at facilitating uptake among first-time bowel cancer screening invitees, compared with non-first-time invitees, contrasts with a previous study by Hirst et al. [8]. One possible explanation for this discrepancy is that, since Hirst et al.’s study, the programme has been extended to include adults 50 years of age (as opposed to 60 years of age), resulting in a cohort that is demographically different.
The findings that text message reminders are cost-effective and acceptable are also consistent with previous literature. For example, a recent study conducted in the USA found that 89% of people interviewed ‘considered it important and useful to receive SMS text message reminders for FIT completion’, and that replacing postal reminders with text message reminders would result in a ‘reduction of $2.40 per participant gained’ [19].
Similarly, the findings that telephone PN is acceptable, but costly, are also consistent with previous literature. For example, a US study found that >50% of people contacted by a patient navigator, regarding breast, bowel or cervical screening, agreed or strongly agreed it was ‘helpful’ and ‘gave information that was useful’ [20].
It is important to note, however, that our findings exploring acceptability may be biased. The response rate was very low and reported by those who answered the telephone when called (incurring self-selection bias). The views of those who do not answer the phone were not captured, and it is likely that this cohort would perceive the intervention less favourably. Indeed, recent evidence suggests that approximately 25% of UK adults aged 55+ years feel unsafe answering the phone, due to fraudulent calls [21].
Finally, the finding that ‘Thought they were up to date’, ‘Did not receive the test kit’, ‘Been busy/not got around to it’, ‘Not interested in taking part’, ‘Away when kit arrived/moved address’ and ‘Lost the kit’ were common reasons for non-participation has been documented previously [22], as has the finding that many patients do not have a valid mobile number recorded with their GP [8]. For example, a study by Hirst et al. found that 51% of individuals sent a text message reminder did not have a valid mobile number recorded [8]. Our study finds this figure has dropped to around 27% (both studies were conducted in London). One possible explanation is that GPs are using text messaging services more frequently today, and have consequently increased efforts to update these records. One thing which has not changed, however, is that older adults and adults registered are less likely to have a valid telephone number stored on their GP’s system, and there is considerable variation between regions within London, as previously documented by Hirst and colleagues [8].
Implications for research, policy and practice
This study has a number of implications for policy and practice. It indicates that both text message reminders and telephone PN could be used to improve uptake of bowel cancer screening in London/reduce regional inequalities in uptake between London and the rest of the country. We estimate that facilitating uptake among an additional 4.5–5.5% of non-responders (36.2% of the invited population) would translate to an absolute increase in uptake of approximately 1.6–2.0 percentage points with TMRs and PN, respectively (see Appendix 16). In absolute terms, this would equate to an additional 19,000 –23,200 people screened per annum, which would translate to an additional 32–39 screen-detected cancers annually (see Appendix 16).
In addition, this study indicates that PN could be used to improve uptake among men, specifically, and could be used to reduce sex/gender-based inequalities in bowel cancer screening uptake [5]. Targeting this cohort with PN would be the most effective and cost-effective use of PN in a resource-constrained system.
Finally, this study indicates that text message reminders and PN can be implemented without generating or widening existing inequalities. Both appear to be equitable, with no differential effects by age or deprivation.
This study also has a number of implications for future research. First, studies are needed to assess whether text message reminders and telephone PN could be used to improve uptake in other regions, with uptake below the national average. These regions will be demographically distinct from London, with differing ethnic and socioeconomic compositions. Second, studies are needed to assess whether text message reminders or PN is more effective for other screening programmes. Bowel cancer is distinct in that participants complete the test from home, while other programmes, such as lung and breast screening, require adults to attend in-person appointments (one such study for lung cancer screening is underway [23]. Finally, further research is needed to assess whether text message reminders and telephone PN reduce, perpetuate or exacerbate inequalities in uptake experienced by other marginalised groups, such as people with severe mental illness [5].
Strengths and limitations
This study has several strengths. First, it used a RCT design, reducing selection bias. Second, participants were blinded to the fact that they were participants in a study, minimising the Hawthorne effect. Finally, we assessed the effectiveness of the interventions on an intention-to-treat and per-protocol basis, allowing us to assess real-world and potential effectiveness.
This study also has several limitations. First, we were limited to data stored on the bowel cancer screening system. As such, were unable to control for/conduct subgroup analysis for other known confounders, such as ethnicity. Second, the study was conducted exclusively in London, limiting its generalisability. Third, the navigation was conducted exclusively by women, and we did not keep a record of which navigators navigated which participants. This meant that we were unable to explore whether the effectiveness of the intervention was modified by navigator characteristics (previous research suggests that men and women are more likely to agree with advice from female practitioners than male practitioners—the same may also be true for advice from navigators [24]. Finally, the navigators did not know the participants’ primary language, and the interventions were delivered in English, which may have disadvantaged those with limited English proficiency.
Conclusion
This study demonstrates that telephone PN is more efficacious than text message reminders at facilitating uptake among men, but not women. Given that men are less likely to participate in bowel cancer screening, we believe telephone PN could be used to help reduce inequalities in bowel cancer screening participation between men and women.
In addition, this study found that both text message reminders and telephone PN were effective and acceptable at improving uptake in London. As such, their local implementation could help reduce regional disparities in participation.
Data availability
The data are available from the corresponding author upon request.
References
Ferlay J, Colombet M, Soerjomataram I, Dyba T, Randi G, Bettio M, et al. Cancer incidence and mortality patterns in Europe: Estimates for 40 countries and 25 major cancers in 2018. Eur J Cancer. 2018;103:356–87.
Zheng S, Schrijvers JJA, Greuter MJW, Kats-Ugurlu G, Lu W, de Bock GH. Effectiveness of colorectal cancer (CRC) screening on all-cause and CRC-specific mortality reduction: a systematic review and meta-analysis. Cancers. 2023;15:1948.
Schreuders EH, Ruco A, Rabeneck L, Schoen RE, Sung JJY, Young GP, et al. Colorectal cancer screening: a global overview of existing programmes. Gut. 2015;64:1637–49.
Public Health England. Public Health Profiles (2020) [Available from: https://fingertips.phe.org.uk/search/bowel%20screening#page/0/gid/1/ati/6/iid/91720/age/280/sex/4/cid/4/tbm/1/page-options/car-do-0_ovw-do-0.
Kerrison, Jones RR, Peng A, Price J, Verne G, Barley J, et al. Inequalities in cancer screening participation between adults with and without severe mental illness: results from a cross-sectional analysis of primary care data on English Screening Programmes. Br J Cancer. 2023;129:81–93.
Raine R, Moss SM, von Wagner C, Atkin W, Hans IK, Howe R, et al. A national cluster-randomised controlled trial to examine the effect of enhanced reminders on the socioeconomic gradient in uptake in bowel cancer screening. Br J Cancer. 2016;115:1479–86.
Muliira JK, D’Souza MS. Effectiveness of patient navigator interventions on uptake of colorectal cancer screening in primary care settings. Jpn J Nurs Sci. 2016;13:205–19.
Hirst Y, Skrobanski H, Kerrison RS, Kobayashi LC, Counsell N, Djedovic N. Text message Reminders in Colorectal Cancer Screening (TRICCS): a randomised controlled trial. Br J Cancer. 2017;116:1408–14.
Office for National Statistics. Mobile Phone Usage Survey [Available from: https://webcache.googleusercontent.com/search?q=cache:dLsp7W7Gt2cJ:https://www.ons.gov.uk/file%3Furi%3D/aboutus/transparencyandgovernance/freedomofinformationfoi/mobilephonedata2016to2017/document3surveyresults.pdf+&cd=16&hl=en&ct=clnk&gl=uk.
Uy C, Lopez J, Trinh-Shevrin C, Kwon SC, Sherman SE, Liang PS. Text messaging interventions on cancer screening rates: a systematic review. J Med Internet Res. 2017;19:e296.
Wang ML, Gallivan L, Lemon SC, Borg A, Ramirez J, Figueroa B, et al. Navigating to health: evaluation of a community health center patient navigation program. Prev Med Rep. 2015;2:664–8.
Nelson HD, Cantor A, Wagner J, Jungbauer R, Fu R, Kondo K, et al. Effectiveness of patient navigation to increase cancer screening in populations adversely affected by health disparities: a meta-analysis. J Gen Intern Med. 2020;35:3026–35.
NHS Digital. Opting out of sharing your confidential patient information 2019 [Available from: https://digital.nhs.uk/about-nhs-digital/our-work/keeping-patient-data-safe/how-we-look-after-your-health-and-care-information/your-information-choices/opting-out-of-sharing-your-confidential-patient-information.
Duffy T, Gil N, Siddique B, Duffy S, Prentice A, Marshall S, et al. Comparing the effectiveness and cost-effectiveness of text message reminders and telephone patient navigation to improve the uptake of faecal immunochemical test screening among non-responders in London: a randomised controlled trial protocol. BMJ Open. 2024;14:e079482.
Primary Care News. 6 CCG to use iPlato representing 1.6 million patients 2017 [Available from: https://htn.co.uk/2017/05/22/6-ccgs-use-iplato-representing-1-6-million-patients/.
Elston DM. The Hawthorne effect. J Am Acad Dermatol. 2021;95:383–4.
Mosquera I, Todd A, Balaj M, Zhang L, Benitez Majano S, Mensah K, et al. Components and effectiveness of patient navigation programmes to increase participation to breast, cervical and colorectal cancer screening: a systematic review. Cancer Med. 2023;12:14584–611.
Stoffel ST, McGregor L, Hirst Y, Hanif S, Morris L, von Wagner C. Evaluation of the Call for a Kit intervention to increase bowel cancer screening uptake in Lancashire, England. J Med Screen. 2022;29:166–71.
Vives N, Travier N, Farre A, Binefa G, Vidal CJose, Pérez Lacasta M, et al. Effectiveness and acceptability of targeted text message reminders in colorectal cancer screening: randomized controlled trial (M-TICS Study). JMIR Public Health Surveill. 2024;10:e57959.
Rawl SM, Baltic R, Monahan PO, Stump TE, Hyer M, Ennis AC, et al. Receipt, uptake, and satisfaction with tailored DVD and patient navigation interventions to promote cancer screening among rural women. Transl Behav Med. 2023;13:879–90.
Button M, Shepherd D, Hawkins C, Tapley J. Fear and phoning: Telephones, fraud, and older adults in the UK. Int Rev Victimol. 2025;31:117–34.
Goodwin BC, Myers L, Ireland MJ, March S, Ralph N, Dunn J, et al. Barriers to home bowel cancer screening. Psycho Oncology. 2021;30:1756–64.
McInnerney D, Simmonds I, Hancock N, Rogerson S, Lindop J, Gabe R, et al. Yorkshire Lung Screening Trial (YLST) pathway navigation study: a protocol for a nested randomised controlled trial to evaluate the effect of a pathway navigation intervention on lung cancer screening uptake. BMJ Open. 2024;14:e084577.
Schieber AC, Delpierre C, Lepage B, Afrite A, Pascal J, Cases C, et al. Do gender differences affect the doctor–patient interaction during consultations in general practice? Results from the INTERMEDE study. Fam Pract. 2014;31:706–13.
Acknowledgements
The authors would like to acknowledge the helpful contributions made by their patient and public collaborators.
Funding
The study was funded by NHS England and the NIHR Policy Research Programme (reference: PR-PRU-NIHR206132. The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care. Robert S Kerrison is supported by a Cancer Research UK Population Research Fellowship [RCCCEA-Nov24/100001]. Funders were not involved in the study.
Author information
Authors and Affiliations
Department of Behavioural Science and Health, University College London, London, UK
Benzeer Siddique, Ghalia Kayal, Aradhna Kaushal, Christian von Wagner & Robert S. Kerrison
St Mark’s Bowel Cancer Screening Centre, Harrow, UK
Benzeer Siddique, Andrew Prentice & Sarah Marshall
Centre for Evaluation and Methods, Wolfson Institute of Population Health, Queen Mary University of London, London, UK
Thomas Duffy
School of Health Sciences, University of Surrey, Surrey, UK
Natalie Gil & Robert S. Kerrison
St Mark’s Hospital, Harrow, UK
Natasha Djedovic & Abhinav Sundriyal
University of Birmingham, Birmingham, UK
Michael Lewis
NHS England (London), London, UK
Josephine Ruwende
Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia
Ghalia Kayal
Applied Health Research Hub, University of Lancashire, Lancashire, Preston, UK
Yasemin Hirst
Centre for Cancer Screening, Prevention and Early Diagnosis, Wolfson Institute of Population Health, Queen Mary University of London, London, UK
Stephen Duffy
Authors
- Benzeer Siddique
- Thomas Duffy
- Natalie Gil
- Andrew Prentice
- Sarah Marshall
- Natasha Djedovic
- Abhinav Sundriyal
- Michael Lewis
- Josephine Ruwende
- Ghalia Kayal
- Aradhna Kaushal
- Yasemin Hirst
- Stephen Duffy
- Christian von Wagner
- Robert S. Kerrison
Contributions
Benzeer Siddique. Conceptualisation, methodology, formal analysis, writing—review and editing. Thomas Duffy: Conceptualisation, methodology, writing—review and editing. Natalie Gil: Methodology, writing—review and editing. Stephen Duffy: Conceptualisation, methodology, writing—review and editing, funding acquisition. Andrew Prentice: Conceptualisation, methodology, writing—review and editing, funding acquisition. Sarah Marshall: Conceptualisation, methodology, investigation, writing—review and editing, funding acquisition. Natasha Djedovic: Conceptualisation, methodology, investigation, writing —review and editing, funding acquisition. Abhinav Sundriyal. Methodology, investigation, formal analysis, writing—review and editing. Michael Lewis: Conceptualisation, methodology, writing—review and editing. Josephine Ruwende: Conceptualisation, methodology, writing—review and editing, funding acquisition. Ghalia Kayal: Supervision, formal analysis, writing—review and editing. Aradhna Kaushal: Supervision, writing—review and editing. Yasemin Hirst: Methodology, writing—review and editing. Christian von Wagner. Conceptualisation, methodology, supervision, writing—review and editing. Robert Stephen Kerrison: Conceptualisation, methodology, investigation, supervision, formal analysis, writing—review and editing, funding acquisition, guarantor.
Corresponding author
Correspondence to Robert S. Kerrison.
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethics approval and consent to participate
The study was conducted in accordance with the principles of the Declaration of Helsinki. The study was reviewed and approved by University College London’s Joint Research Office (Ref: 150666), the Screening Research, Innovation and Development Advisory Committee (‘RIDAC’, Ref: 2223 014 BCSP Kerrison) and the Health Research Authority (Ref: 22/WM/0212). Participants were not consented to participate in the research. Approval to include individuals, without their consent, was obtained from the Confidentiality Advisory Group (Ref: 22/CAG/0140). Participants were notified about the research, and given the option to withdraw their data, on completion of the study. None opted to withdraw their data and national data opt outs were excluded from the study altogether. All methods were performed in accordance with the relevant guidelines and regulations.
Consent for publication
All authors have reviewed the manuscript and consent to its publication.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Siddique, B., Duffy, T., Gil, N. et al. Which is more effective at improving uptake of bowel cancer screening: text message reminders or patient navigation? Results from a three-arm RCT. Br J Cancer (2026). https://doi.org/10.1038/s41416-026-03645-1
Received: 14 November 2025
Revised: 04 September 2026
Accepted: 29 September 2026
Published: 05 November 2026
Version of record: 05 November 2026
DOI: https://doi.org/10.1038/s41416-026-03645-1
来源:Nature 健康护理主题 · nature.com