Artificial Intelligence for Inflammatory Bowel Diseases (IBD); Accurately Predicting Adverse Outcomes Using Machine Learning.

BACKGROUND: Inflammatory Bowel Diseases with its complexity and heterogeneity could benefit from the increased application of Artificial Intelligence in clinical management. AIM: To accurately predict adverse outcomes in patients with IBD using advanced computational models in a nationally representative dataset for potential use in clinical practice. METHODS: We built a training model cohort and validated our result in a separate cohort. We used LASSO and Ridge regressions, Support Vector Machines, Random Forests and Neural Networks to balance between complexity and interpretability and analyzed their relative performances and reported the strongest predictors to the respective models. The participants in our study were patients with IBD selected from The OptumLabs® Data Warehouse (OLDW), a longitudinal, real-world data asset with de-identified administrative claims and electronic health record (EHR) data. RESULTS: We included 72,178 and 69,165 patients in the training and validation set, respectively. In total, 4.1% of patients in the validation set were hospitalized, 2.9% needed IBD-related surgeries, 17% used long-term steroids and 13% of patients were initiated with biological therapy. Of the AI models we tested, the Random Forest and LASSO resulted in high accuracies (AUCs 0.70-0.92). Our artificial neural network performed similarly well in most of the models (AUCs 0.61-0.90). CONCLUSIONS: This study demonstrates feasibility of accurately predicting adverse outcomes using complex and novel AI models on large longitudinal data sets of patients with IBD. These models could be applied for risk stratification and implementation of preemptive measures to avoid adverse outcomes in a clinical setting.

The effects of inflammatory bowel disease on caregivers: significant burden and loss of productivity.

BACKGROUND: Caregiver burden is the emotional, physical, practical, and/or financial burden associated with taking care of a patient with a chronic condition. Limited literature on caregiver burden in Inflammatory Bowel Diseases (IBD) has accounted for some predictors, but its effect on work productivity (absenteeism and presenteeism) is unknown. METHODS: In a prospective study, patients and their respective caregivers were surveyed from November 2015 until July 2017. Data on demographics, work productivity, quality of life, disease activity, caregiver burden and productivity were collected. The burden on caregivers was assessed and associations between caregiver productivity and caregiver burden were analyzed. Additionally, predictors for caregiver burden were identified. RESULTS: One hundred two IBD patients and their respective caregiver were included. In total, 39% of IBD caregivers experienced burden. Caregivers with burden experienced significantly more absenteeism and presenteeism (65 and 85% respectively). Furthermore, 51% of caregivers felt that they should be doing more for their care recipient and felt they could do a better job at caregiving. Predictors of burden included race/ethnicity, history of fistulas, diagnosis of ulcerative colitis, higher caregiver education, and hours spent caregiving. CONCLUSION: Caregivers with burden had significantly more productivity decrease compared to those without burden. Additionally, the majority of caregivers feel they should be providing more and better care for their recipients. The development of strategies to address caregiver's distress and perceived burden when caring for IBD patients is warranted.

An Exploration Into the Use of a Chatbot for Patients With Inflammatory Bowel Diseases: Retrospective Cohort Study.

BACKGROUND: The emergence of chatbots in health care is fast approaching. Data on the feasibility of chatbots for chronic disease management are scarce. OBJECTIVE: This study aimed to explore the feasibility of utilizing natural language processing (NLP) for the categorization of electronic dialog data of patients with inflammatory bowel diseases (IBD) for use in the development of a chatbot. METHODS: Electronic dialog data collected between 2013 and 2018 from a care management platform (UCLA eIBD) at a tertiary referral center for IBD at the University of California, Los Angeles, were used. Part of the data was manually reviewed, and an algorithm for categorization was created. The algorithm categorized all relevant dialogs into a set number of categories using NLP. In addition, 3 independent physicians evaluated the appropriateness of the categorization. RESULTS: A total of 16,453 lines of dialog were collected and analyzed. We categorized 8324 messages from 424 patients into seven categories. As there was an overlap in these categories, their frequencies were measured independently as symptoms (2033/6193, 32.83%), medications (2397/6193, 38.70%), appointments (1518/6193, 24.51%), laboratory investigations (2106/6193, 34.01%), finance or insurance (447/6193, 7.22%), communications (2161/6193, 34.89%), procedures (617/6193, 9.96%), and miscellaneous (624/6193, 10.08%). Furthermore, in 95.0% (285/300) of cases, there were minor or no differences in categorization between the algorithm and the three independent physicians. CONCLUSIONS: With increased adaptation of electronic health technologies, chatbots could have great potential in interacting with patients, collecting data, and increasing efficiency. Our categorization showcases the feasibility of using NLP in large amounts of electronic dialog for the development of a chatbot algorithm. Chatbots could allow for the monitoring of patients beyond consultations and potentially empower and educate patients and improve clinical outcomes.

Patient Experiences and Outcomes of a Telehealth Clinical Care Pathway for Postoperative Inflammatory Bowel Disease Patients.

Background:Despite advancements in treatment for inflammatory bowel disease (IBD), surgery remains inevitable for patients and IBD management is costly.Introduction:Frequent postoperative monitoring is needed for early detection of both short-term complications and long-term disease recurrence. We developed a care pathway for postoperative home monitoring of IBD patients using telehealth applications.Materials and Methods:We performed a retrospective cohort study with a matched control group to assess the efficacy of the Tight Control Surgery Scenario (TCSS), a 4-week postoperative care pathway. IBD patients aged 18 or older who underwent an IBD-related intestinal operation between October 2013 and December 2015 were eligible. Enrolled participants submitted postsurgical questionnaires and wound photos through e-mail. We measured patient satisfaction with the care pathway and assessed its impact on 30-day postoperative hospital readmission rates, emergency department (ED) visits, and gastroenterologist (GI)-related office visits.Results:Sixty-four (n) cases were enrolled in TCSS and matched to 64 historic controls. Patients who completed the additional evaluation survey expressed overall satisfaction. Readmissions, 30-day ED rates, and GI visits were numerically higher in cases compared with controls, but this difference was not statistically significant.Discussion:TCSS demonstrates the feasibility of implementing a telehealth care coordination platform for postsurgery IBD management. Patients with more complications may have sent in more photos due to greater concern for maintaining their health.Conclusions:Implementation of TCSS for easy home monitoring is feasible. While we did not see reductions in ED visits, GI follow-up visits, or readmissions, patient satisfaction was high, thus demonstrating its feasibility for telehealth applications.

Development and Validation of an Inflammatory Bowel Diseases Monitoring Index for Use With Mobile Health Technologies.

BACKGROUND & AIMS: Mobile health technologies are advancing rapidly as smartphone use increases. Patients with inflammatory bowel disease (IBD) might be managed remotely through smartphone applications, but no tools are yet available. We tested the ability of an IBD monitoring tool, which can be used with mobile technologies, to assess disease activity in patients with Crohn's disease (CD) or ulcerative colitis (UC). METHODS: We performed a prospective observational study to develop and validate a mobile health index for CD and UC, which monitors IBD disease activity using patient-reported outcomes. We collected data from disease-specific questionnaires completed by 110 patients with CD and 109 with UC who visited the University of California, Los Angeles, Center for IBD from May 2013 through January 2014. Patient-reported outcomes were compared with clinical disease activity index scores to identify factors associated with disease activity. Index scores were validated in 301 patients with CD and 265 with UC who visited 3 tertiary IBD referral centers (in California or Europe) from April 2014 through March 2015. RESULTS: We assessed activity of CD based on liquid stool frequency, abdominal pain, patient well-being, and patient-assessed disease control, and activity of UC based on stool frequency, abdominal pain, rectal bleeding, and patient-assessed disease control. The indices identified clinical disease activity with area under the receiver operating characteristic curve values of 0.90 in patients with CD and 0.91 in patients with UC. They identified endoscopic activity with area under the receiver operating characteristic values of 0.63 in patients with CD and 0.82 in patients with UC. Both scoring systems responded to changes in disease activity (P < .003). The intraclass correlation coefficient for test-retest reliability was 0.94 for CD and for UC. CONCLUSIONS: We developed and validated a scoring system to monitor disease activity in patients with CD and UC that can be used with mobile technologies. The indices identified clinical disease activity with area under the receiver operating characteristic curve values of 0.9 or higher in patients with CD or UC, and endoscopic activity in patients with UC but not CD.

Patient Experience and Satisfaction with an e-Health Care Management Application for Inflammatory Bowel Diseases.

BACKGROUND: Rising healthcare expenditures have been partially attributed to suboptimal management of inflammatory bowel diseases (IBD). Electronic health interventions may help improve care management for IBD patients, but there is a need to better understand patient perspectives on these emerging technologies. AIMS: The primary aim was to evaluate patient satisfaction and experience with the UCLA eIBD mobile application, an integrative care management platform with disease activity monitoring tools and educational modules. The secondary objective was to capture patient feedback on how to improve the mobile application. METHODS: We surveyed IBD patients treated at the UCLA Center for Inflammatory Bowel Diseases. The patient experience survey assessed the patients' overall satisfaction with the application, perception of health outcomes after participation in the program, and feedback on educational modules as well as areas for application improvement. RESULTS: 50 patients were included. The responses indicated that the patients were greatly satisfied with the ease of patient-provider communication within the application and appointment scheduling features (68%). A majority of respondents (54%) also reported that program participation resulted in improved perception of disease control and quality of life. Lastly, a majority of participants (79%) would recommend this application to others. CONCLUSIONS: Mobile tools such as UCLA eIBD have promising implications for integration into patients' daily lives. This patient satisfaction study suggests the feasibility of using this mobile application by patients and providers. We further showed that UCLA eIBD and its holistic approach led to improved patient experience and satisfaction, which can provide useful recommendations for future electronic health solutions.

Health outcomes and experiences of direct-to-consumer high-intensity screening using both whole-body magnetic resonance imaging and cardiological examination.

BACKGROUND: Alongside a clinical and research setting, whole body magnetic resonance imaging (WB-MRI) is increasingly offered as a direct-to-consumer screening service. Data is needed on the clinical relevance of findings and associated psychological impact of such screening. Therefore, we conducted a prospective follow-up study to provide insight in the effectiveness and psychological impact of direct-to-consumer screening using both WB-MRI and cardiological examination. METHODS AND FINDINGS: The study population consisted of 3603 voluntary, primarily middle-aged participants who underwent commercial WB-MRI and cardiological screening at one of 6 study clinics in Germany or the Netherlands between July 2014 and March 2016. MRI investigation consisted of directed scans of the brain, neck, abdomen and pelvis. Cardiovascular examination included pulmonary function, resting electrocardiogram, transthoracic echocardiogram and a bicycle exercise stress test. Findings were assessed by experienced radiologists and cardiologists. In addition, participants were inquired about several (psychological) domains, including the expectations and consequences of the screening procedure. Out of 3603 individuals, 402 (11.2%) demonstrated abnormal MRI (n = 381) and/or cardiological findings (n = 79) for which they were advised to undergo further consultation <3 months in regular healthcare. In 59.1% of cases of abnormal MRI findings which were consulted, fully completed consultations were available in 87.1%. After consultation, 77.6% of initial MRI outcomes were adopted. In 40.9% of cases of abnormal MRI findings, recommendations for consultation were not adhered to during the study period. 71.1% of adopted MRI-findings required treatment or monitoring, including 19 malignancies. For abnormal cardiological findings, 70.9% of cases were consulted in regular healthcare. Of these, 91.1% had a completed follow-up procedure of which 72.5% of initial findings were adopted and 83.8% of these findings required treatment or monitoring. The most frequently reported psychological consequences of the screening procedure were getting reassurance (72.0%) and insight into one's own health status (83.0%). 5.0% reported to feel insecure about their health and 6.2% worried more about their health as a consequence of screening. Main limitations of the study were considered the telephonic follow-up of referred clients and the heterogeneity of screening equipment and assessment of radiologists and cardiologists. CONCLUSIONS: Direct-to-consumer screening using whole-body MRI and cardiological testing is feasible and effective for the detection of clinically relevant and treatable abnormalities. Psychological harm was not frequently reported in study participants.

Presenteeism in Inflammatory Bowel Diseases: A Hidden Problem with Significant Economic Impact.

BACKGROUND: Indirect costs associated with impaired productivity at work (presenteeism) due to inflammatory bowel disease (IBD) are a major contributor to health expenditures. Studies estimating indirect costs in the United States did not take presenteeism into account. We aimed to quantify work limitations and presenteeism and its associated costs in an IBD population to generate recommendations to reduce presenteeism and decrease indirect costs. METHODS: We performed a prospective study at a tertiary IBD center. During clinic visits, work productivity, work-related problems and adjustments, quality of life, and disease activity were assessed in patients with IBD. Work productivity and impairment were assessed in a control population as well. Indirect costs associated with lost work hours (absenteeism) and presenteeism were estimated, as well as the effect of disease activity on those costs. RESULTS: Of the 440 included patients with IBD, 35.6% were unemployed. Significantly more presenteeism was detected in patients with IBD (62.9%) compared with controls (27.3%) (P = 0.004), with no significant differences in absenteeism. Patients in remission experienced significantly more presenteeism than controls (54.7% versus 27.3%, respectively, P < 0.01), and indirect costs were significantly higher for remissive patients versus controls ($17,766 per yr versus $9179 per yr, respectively, P < 0.03). Only 34.3% had made adjustments to battle work-related problems such as fatigue, irritability, and decreased motivation. CONCLUSIONS: Patients with IBD in clinical remission still cope with significantly more presenteeism and work limitations than controls; this translates in higher indirect costs and decreased quality of life. The majority have not made any adjustments to battle these problems.