Introduction
Patients with inflammatory bowel disease (IBD) require long-term, intensive care. To optimise future healthcare delivery, identifying patients suitable for remote monitoring is crucial. Patient-reported outcome measures integrated within eHealth systems, such as the 4-item Mobile Health Index (mHI)1, offer potential for predicting disease activity. This study explores the predictive value of the mHI within the e-health system DEARhealth2 at a Dutch university medical centre.
Aims & Methods
This multi-centre prospective cohort study tracked IBD patients within the e-health system. Patients were categorised into high-, intermediate-, and low-risk groups based on current and past biochemical disease activity. High-risk patients had current biochemical disease activity (FCP > 150 µg/g and/or CRP > 5 mg/L), intermediate-risk patients were in remission but experienced a biochemical flare within the last three years, while low-risk patients had been in biochemical remission for at least three years. A personalised care pathway, considering both risk group and medication usage, dictated the frequency of outpatient visits and lab tests. Over a 24-month period, patients are monitored according to standard of care (biochemical disease activity; endoscopic disease activity) and with mHI scores at both pre-defined time points and patient-initiated moments. These preliminary results focus on assessing the predictive value of the mHI, above a validated cut-off value1, at baseline for predicting a flare within six months. Sensitivity, specificity, and predictive values are calculated alongside simple logistic regression and Kaplan-Meier analyses.
Results
Out of 319 patients (56.7% Crohn’s Disease (CD), median age 43 years, median disease duration 15 years), 89 exhibited biochemical disease activity at baseline and were excluded from the current analysis. Among the 230 patients in remission at baseline, 51 (21.3%) scored above the mHI cut-off point at baseline and 79 (34.3%) experienced a biochemical flare-up within six months of follow-up. The relative risk of experiencing a disease flare in Ulcerative Colitis (UC) patients with a baseline mHI score above the cut-off point was 2.44 (p = 0.04) with sensitivity of 42.1%, specificity of 82.4%, positive predictive value of 40.0%, and negative predictive value of 83.6%. In CD patients, this relative risk was 1.62 (p = 0.39) with sensitivity of 38.5%, specificity of 73.4%, positive predictive value of 14.7%, and a negative predictive value of 90.9%. Logistic regression analysis revealed a significant association between a mHI score above the cut-off point and disease flares in UC patients (OR = 3.394, 95% CI: 1.110-10.330, p = 0.03), whereas in CD patients, such an association was not significant (OR = 1.724, 95% CI: 0.487-5.603, p = 0.37). Similarly, Kaplan-Meier analysis demonstrated a longer flare-free survival in UC patients with a baseline mHI score below the cut-off point (p = 0.03), while no significant difference was observed in CD patients (p = 0.14).
Conclusion
This study indicates that the mHI could be a valuable tool for predicting disease activity, and particularly quiescent disease, in UC patients. However, this predictive capability was not observed in patients with CD. These results underscore the potential utility of mHI in enhancing the management of IBD patients, with implications for personalised care and resource allocation within healthcare systems. Further research should explore repeated mHI measurements in combination with biomarkers like calprotectin for a more comprehensive approach.
References
1 Van Deen, W. K., van der Meulen-de, A. E., Parekh, N. K., Kane, E., Zand, A., DiNicola, C. A., ... & Hommes, D. W. (2016). Development and validation of an inflammatory bowel diseases monitoring index for use with mobile health technologies. Clinical gastroenterology and hepatology, 14(12), 1742-1750.
2 DEARhealth: Software as a medical device using Dynamic & Risk-based care pathways and Smart Decision Support Recommendations. DEARhealth. (n.d.). https://dearhealth.com/hospital/
Disclosure
PWJM reports consulting fees from Abbvie and Takeda, and payment for presentations from Galapagos and Takeda. PWV reports consulting fees from Abbvie, Galapagos, BMS, Janssen, Lilly, and Takeda, and payment for presentations from Galapagos and Takeda. AEvdMdJ received research grants from Nestle, Norgine, Cablon, Galapagos, and ZonMw, including speaker’s fees from, Tramedico, Janssen Pharmaceuticals, Takeda, Galapagos, Vedanta, Ferring, and AbbVie. These are not related to this work. All other authors have no conflicts of interest to declare.
This study has received funding from Galapagos/Alfasigma