A randomized, open, multicenter, noninferiority trial was conducted across 24 hospitals in Spain. The country has 17 autonomous communities, and patients from 10 of these regions participated in the study, as illustrated in the map in Figure 1. The project was promoted by GETECCU, and the study design was discussed in successive research meetings before final protocol approval. The trial is registered on ClinicalTrials.gov with the identifier NCT06031038.

As this is a noninferiority study, we conducted a per-protocol analysis. The characteristics of the participants were described using the mean, SD, 95% CI, median, IQR, and range for quantitative variables, as well as absolute and relative frequencies for qualitative variables. Possible baseline differences between the 2 study groups were compared using the Student t test for independent samples or the Mann-Whitney U test for quantitative variables, and the chi-square test or Fisher exact test for qualitative variables. Time in clinical remission was summarized using the mean, SD, median, IQR, and range.

The sample size was estimated based on the results of the pilot trial, which indicated that patients in the control group remained in remission for a median of 14.3 weeks, whereas those in the G_TECCU group had a median remission time of 17.9 weeks. These differences in time to remission were evaluated with a noninferiority limit of –1.5 weeks, considering this difference to be clinically irrelevant within the 12-week follow-up period of this interim analysis. Noninferiority was assessed by calculating the mean and 95% CI for the difference in the variable of interest between the groups. This was done by comparing the limits of the CI with the predefined noninferiority margin using a Student t test. Accepting an α risk of .025 in a 1-sided test, a β risk of .2, and an SD of 3.2, it was determined that 71 patients were needed in each group. Furthermore, assuming a loss rate of 15%, a total of 168 patients was required for the study.

Changes in secondary outcomes from baseline were described by applying appropriate estimators based on the type of variables, as previously outlined. Possible differences in secondary outcomes between the 2 groups were compared using the Student t test for independent samples, the Mann-Whitney U test for quantitative variables, and the chi-square or Fisher exact test for qualitative variables. All analyses were conducted using SAS statistical software version 9.4 (SAS Institute), with a 1-sided significance level of .025 accepted for all analyses.

The study protocol was approved by the local independent ethics committee at La Fe University and Polytechnic Hospital, Valencia (v0.3, 14/07/2020), and by the Spanish Agency of Medicines and Medical Devices (AEMPS/Agencia Española de Medicamentos y Productos Sanitarios: 30/07/2020). The study was conducted in accordance with the following: the “Note for Guidance on Good Clinical Practice” (CPMP/ICH/135/95, May 1, 1996); Royal Decree 223/2004 (February 2004); the Helsinki Declaration on ethical principles for medical research involving human subjects, as adopted by the General Assembly of the World Medical Association (7th revised version, Seoul, 2008); guidelines from the International Conference on Harmonization; and the official regulations imposed by the participating centers.

A total of 169 patients with complex IBD were enrolled in the study between September 15, 2020, and April 28, 2023. Of these, 11 patients were excluded: 4 due to screening failure, 2 due to a lack of baseline data, and 5 who withdrew before randomization. The remaining 158 patients proceeded to follow-up, with 74 (47%) in the G_TECCU group and 84 (53%) in the G_Control group. During the study period, 8 patients (3 from G_TECCU and 5 from G_Control) did not complete the 12-week follow-up. The reasons for this are detailed in Figure 2. As such, the 71 patients in the G_TECCU group and 79 patients in the G_Control group who completed the 12-week follow-up period were analyzed per protocol (N=150). Additionally, all randomized patients (n=158) were analyzed according to the intention-to-treat principle. The per-protocol population was used to summarize patient disposition and baseline characteristics (Figure 2; also see Multimedia Appendix 1 for the CONSORT-EHEALTH [Consolidated Standards of Reporting Trials of Electronic and Mobile Health Applications and Online Telehealth] checklist [35]).

The baseline demographic and clinical characteristics are presented in Table 1. The overall mean age of the participants was 37.8 (SD 18.7) years. Of the 150 patients, 83 (55%) were male and 67 (45%) were female. Among the included patients (n=150), 58 (39%) had UC, and 92 (61%) had CD, all of whom were capable of using the technology. The education levels of the patients were as follows: primary education (12/150, 8%), secondary education (70/150, 47%), and university education (68/150, 45%). Regarding employment status, 100 (66.7%) patients were actively employed, 19 (12.7%) were students, 17 (11.3%) were unemployed, 10 (6.7%) were retired, and 4 (2.7%) were houseworkers. At the onset of the study, 109 (72.7%) patients initiated treatment with biological agents (adalimumab in 55/109, 50.5%, patients), 25 (16.7%) with immunosuppressants (azathioprine in 23/25, 92%, patients), and 16 (10.7%) with combination therapy. No significant differences were observed in the baseline characteristics of patients according to the treatment initiated (immunosuppressants, biological agents, or combination therapy).

Regarding their clinical status (Table 2), 58 of 150 (38.7%) patients were in clinical remission at baseline; however, they were included in the study due to FC levels exceeding 200 µg/g or the presence of moderate to severe endoscopic activity. Concerning biological markers, the mean CRP value was 18.5 (SD 73.4) mg/L, while the mean FC level was 1631.1 (SD 2722.4) µg/g. Among the 24 patients with UC who entered the study following an endoscopic evaluation, 8 (33%) had a Mayo Index of 2, indicating moderate activity, while 16 patients (67%) had a Mayo Index of 3, indicating severe activity. Among the 34 patients with CD who underwent a baseline endoscopy, 27 (79%) exhibited ulcers. Furthermore, 61 of 150 (41%) patients demonstrated good adherence to medication (25/71, 35%, in the G_TECCU group and 36/79, 46%, in the G_Control group). Patient satisfaction, as measured by specific questionnaires, yielded a score of 87.3 (SD 11.2).

The mean time in clinical remission after the 12-week follow-up was 4.26 (SD 3.49) weeks for the entire study cohort. Based on the noninferiority statistics, the follow-up with TECCU was not inferior to standard care, with a mean time in remission of 4.20 (SD 3.73) weeks for the G_TECCU patients compared with 4.32 (SD 3.28) weeks for the G_Control patients (noninferiority P=.02; see Table 3). The mean difference in clinical remission between the 2 groups was –0.12 weeks, with a 95% CI of –1.25 to 1.01 (G_TECCU vs G_Control, P=.02; see Figure 3A).

The subgroup analysis for patients with CD revealed a mean difference in clinical remission of +0.70 weeks for G_TECCU compared with G_Control (95% CI –0.75 to 2.15; P=.003). Given the statistical significance, the noninferiority of TECCU was confirmed in this subgroup. By contrast, the mean difference in clinical remission for patients with UC was –1.65 weeks between G_TECCU and G_Control (95% CI –3.49 to 0.19; P=.87), indicating that noninferiority of TECCU could not be confirmed in this cohort (Figure 3A). Similar results were observed in the intention-to-treat population and after adjusting for multiple tests (see Tables 4-6 and Figure 3B-3D).

After 12 weeks, the mean SCCAI significantly improved in both the G_TECCU (mean change –3.78, SD 3.16, P<.001) and G_Control patients with UC (mean change –4.11, SD 4.31; P<.001). Similarly, the mean HBI value also showed a significant improvement in G_TECCU patients (mean change –1.42, SD 2.69; P<.001) and G_Control patients (mean change –2.07, SD 4.06; P=.002). However, the improvements in SCCAI (P=.27) and HBI (P=.26) scores were not significantly different between the 2 intervention arms.

Disease activity was also assessed based on CRP and FC levels (Figure 4). The mean CRP value significantly decreased after 12 weeks compared with baseline in both groups, with a mean reduction of –15.40 (SD 90.15) mg/L (P=.19) in the G_TECCU group and –13.16 (SD 54.61) mg/L (P=.05) in the G_Control group. However, this improvement in CRP did not significantly differ between the 2 groups (G_TECCU vs G_Control: P=.73). Similarly, the mean FC levels were significantly lower after 12 weeks compared with baseline in both study groups. The G_TECCU group experienced a mean reduction of 832.3 (SD 1825.0) mg/L (P=.003), while the G_Control group had a mean reduction of 1073.5 (SD 3105.7) mg/L (P=.03). However, there were no significant differences between the 2 groups (P=.96).

The difference in the mean change in CRP levels between the groups after 12 weeks was –2.25 mg/L for G_TECCU compared with G_Control (95% CI –28.19 to 23.69); however, noninferiority of TECCU could not be demonstrated (noninferiority P=.73). Similarly, noninferiority was not demonstrated for the reduction in FC levels among G_TECCU patients, with a mean change difference of 241.2 μg/g compared with G_Control (95% CI –806.9 to 1289.2, noninferiority P=.96).

The mean IBDQ-9 scores increased from 59.01 (SD 21.8) to 72.44 (SD 20.3) in G_TECCU patients, indicating a mean improvement of 13.44 points (SD 19.1, P<.001). In G_Control patients, scores improved from 50.36 (SD 19.8) to 68.58 (SD 20.2), reflecting a mean improvement of 18.23 points (SD 22.9; P=.001; Figure 5). However, because the mean difference between the 2 groups was –4.79 (95% CI –12.9 to 3.3; P=.90) and the CI crossed the noninferiority margin, the noninferiority of TECCU in improving the IBDQ-9 score could not be demonstrated.

Regarding the impact of the disease on work productivity and daily activities, a significant improvement in WPAI scores was observed at 12 weeks compared with baseline in the G_Control arm (P=.001 for all scores). By contrast, TECCU only demonstrated a significant improvement in impairment while working (mean improvement –17.74, SD 39.65; P=.003) and in activity impairment (mean improvement –14.49, SD 30.69; P=.002). However, no significant differences between the 2 arms were found in work hours missed due to health problems (P=.21), impairment while working (P=.38), overall work impairment (P=.19), or activity impairment (P=.11).

After 12 weeks, the proportion of patients adhering to their medication increased significantly in the G_TECCU group (rising from 25/71, 35%, to 48/71, 68%; P=.001) and in the G_Control group (increasing from 36/79, 46%, to 58/79, 73%; P=.001). However, this improvement in medication adherence did not differ significantly between the 2 arms (G_TECCU vs G_Control: P=.46). Additionally, the MMAS-8 score in G_TECCU patients improved significantly after 12 weeks, with a mean baseline value increasing from 7.07 (SD 1.03) to 7.50 (SD 0.86), representing a mean improvement of 0.48 (SD 1.6, P=.002). Similarly, the MMAS-8 score improved significantly in the G_Control group after 12 weeks, rising from a mean baseline value of 7.25 (SD 1.1) to 7.57 (SD 1.2), reflecting a mean improvement of 0.29 (SD 1.3, P=.09; Figure 6). The difference in the mean change of the MMAS-8 score between the 2 arms was 0.19 (95% CI –0.24 to 0.63; P=.39), indicating a trend toward statistical significance regarding the noninferiority of G_TECCU compared with G_Control (P=.07).

Changes in medication were made for 24 of the 150 patients (16%) during the study period, with no significant differences observed in the number of patients who changed their medication between the groups (G_TECCU: 12/71, 17%, patients vs G_Control: 12/79, 15%, patients; P=.77).

In the G_TECCU group, the mean satisfaction questionnaire score improved from 88.04 (SD 10.3) to 90.51 (SD 9.7; P=.04), while in the G_Control group, it increased from 87.13 (SD 12.3) to 89.57 (SD 8.5; P=.12). The mean improvement in satisfaction scores was 2.47 (SD 8.6) for G_TECCU (P=.04) and 2.43 (SD 11.2) for G_Control (P=.12); however, the noninferiority test between the groups was not significant (P=.97; Figure 7).

Emergency department visits were significantly higher in the G_TECCU group compared with the G_Control group (13/71, 18%, G_TECCU vs 5/79, 6%, G_Control; P=.02). However, the intervention did not result in significant differences in unscheduled outpatient visits (19/71, 27%, G_TECCU vs 16/79, 20%, G_Control; P=.35) or hospitalizations (5/71, 7%, G_TECCU vs 2/79, 3%, G_Control; P=.19). A total of 70 AEs were recorded in the G_TECCU group in 39 of 71 (55%) patients, while 60 AEs were noted in the G_Control group in 31 of 79 (39%) patients. However, the differences between groups did not reach statistical significance (P=.054).

Among these AEs, 5 resulted in patient loss during the follow-up period, with 2 in the G_TECCU group and 3 in the G_Control group. A total of 17 AEs were related to medication, comprising 8 of 71 (11%) patients in the G_TECCU group and 9 of 79 (11%) patients in the G_Control group, with no significant differences between the groups (P=.98). Notably, none of the patients died during this period.

To the best of our knowledge, this is the first trial using a noninferiority design to compare telemonitoring with standard care for patients with moderate-to-severe IBD. In previous studies, telemonitoring was shown to be safe and to improve both health outcomes and costs [8,9,11,13-15]. In fact, telemonitoring can meet the treat-to-target strategy by addressing many variables of the STRIDE-II recommendations [18], but it failed to demonstrate superiority over standard care in many studies [8,10,11,13,14,17]. Nonetheless, noninferiority would also be an acceptable goal if, in return, telemonitoring improves the efficiency of care. In light of the positive results obtained with TECCU in a pilot trial carried out on patients with active IBD [20], here we demonstrate on a nationwide scale that telemonitoring with the TECCU app is not inferior to standard care in inducing and maintaining remission over 12 weeks.

The main outcome of this study was the time patients remained in remission, which is relevant because we evaluated the evolution of clinical activity throughout the follow-up period, not just the clinical status at the end of the trial, as reported thus far [13,15,19,20,36]. The mean period in remission with telemonitoring was 4.20 weeks, a relatively short time that was probably related to the selection of patients during the induction phase of treatment. Then, in this work, we evaluated the first 12 weeks after initiating biological agents or immunosuppressants to manage persistent disease activity or a flare-up. Indeed, the mean time in remission was essentially the same as that achieved through standard care, and its distribution in each group was sufficiently similar to demonstrate noninferiority.

A subgroup analysis confirmed the noninferiority of TECCU for patients with CD, but it was inconclusive for patients with UC. Furthermore, the improvement in clinical activity was consistent with the reduction in the markers of inflammation (CRP and FC) over time. These data imply that patients with active IBD benefit from both telemonitoring and standard care, and while telemonitoring may not improve disease activity compared with in-person visits, it may enhance other health outcomes, reduce costs, decrease time off work, and lower the number of hospital visits, among other benefits.

In this sense, the TECCU app previously demonstrated greater cost-effectiveness than standard care in a similar subset of patients [24,25], as subsequently reproduced in other settings [37,38]. Therefore, telemonitoring stands out as an interesting alternative to standard care in the follow-up of patients with IBD, whether they are experiencing active disease or not. Either of these options could be used complementarily during a patient’s follow-up, based on the preferences of both patients and health care providers in each case. In this regard, telemonitoring apps can leverage the advancements achieved during the COVID-19 pandemic [39].

As witnessed here, enhanced QoL and medication adherence are generally inversely correlated with disease activity [7]. The IBDQ-9 and the MMAS-8 medication adherence scores improved in both the TECCU and control groups after 12 weeks, indicating a trend toward statistical significance in the noninferiority of G_TECCU compared with G_Control in medication adherence. The improvement in HRQoL could be attributed to the reduction in disease activity in both groups following the escalation of medication, rather than solely to the follow-up method itself. Furthermore, the enhancement in HRQoL was similarly linked to a significant improvement in work productivity and a reduction in activity impairment (as measured by the WPAI) at 12 weeks compared with baseline in both groups.

Another notable finding was that a higher proportion of patients in the G_TECCU group had emergency department visits. However, there were no significant differences between the groups concerning unscheduled outpatient visits, hospitalizations, or AEs. Taken together, these results indicate that the telemonitoring follow-up schedule should be tailored to each patient, and the learning curve for implementing this approach may vary between centers. This aspect will be further analyzed after the completion of the 12-month follow-up period of this trial.

The efficacy of telemonitoring lies in its capacity to deliver interventions tailored to the varying clinical scenarios presented by patients who are open to using remote care, rather than solely focusing on how these interventions are provided. In fact, various patient profiles have been identified as predictors of failure in a telemedicine context, partly due to the diverse demographic and socioeconomic characteristics of the patients [40]. In this context, patient satisfaction remained stable in both the G_TECCU and G_Control groups in our trial, with values around 90% after 12 weeks. It is important to note that the G_TECCU group had received standard care before being enrolled in the study, and the switch to telemonitoring did not diminish their level of satisfaction, even though their disease activity was uncontrolled at that time.

The main strength of this study is that it was designed as a randomized controlled trial, allowing us to assess the impact of a web-based telemonitoring system on disease outcomes in the follow-up of IBD and to compare this with standard care. Additionally, we selected patients during the induction phase of biological agents or those initiating immunosuppressants due to active IBD or a flare-up, a specific population of IBD that has been underrepresented in previous studies [8,13,15,19,41].

Moreover, the use of validated indices of activity and biological markers (CRP and FC) that correlate well with endoscopic activity [42-45] added value to the assessment of remote monitoring. Additionally, the follow-up schedule was consistent across both arms and was adapted to daily practices in the management of IBD, which enhances the reproducibility of the results obtained. Finally, this nationwide study includes patients recruited from 24 hospitals across Spain, further enhancing the external validity of our findings.

Among the limitations of this study, the follow-up period was only 12 weeks. While this time frame may be adequate to evaluate the induction phase of therapies, it may be too short to fully assess the maintenance period, especially because patients and physicians need some time to learn how to use the telemonitoring platform. Therefore, trials with longer follow-up periods will be necessary to confirm the efficacy of this web-based system in improving long-term disease outcomes. Indeed, the follow-up of the patients enrolled in this study continues, and there are intentions to extend this period to 52 weeks for as many of these patients as possible.

Besides, the sample size may be too small to compare disease outcomes in specific patient subgroups (eg, patients with UC or CD and those receiving different medications). In addition, given the nature of the interventions assessed, neither the patients nor the researchers were blinded to the intervention; however, the data obtained were analyzed by an independent statistician who was blind to group identification. Finally, we excluded patients with active perianal disease, ileo-rectal/ileo-pouch anal anastomosis, or stoma; therefore, the impact of TECCU on surgical patients, in whom physical examination is very important, remains unclear.

This Spanish multicenter trial conducted on behalf of GETECCU studied patients with active IBD who began using biological agents or immunosuppressant drugs, demonstrating that the TECCU app is not inferior to standard care for achieving and maintaining remission in the short term. Noninferiority could represent a meaningful advancement if, as seen in other studies, telemonitoring improves additional health outcomes and reduces costs. The TECCU app is also associated with improved medication adherence and HRQoL; however, a reduction in costs still needs to be demonstrated over a longer follow-up period.