Challenges in the design of a Home Telemanagement trial for patients with ulcerative colitis

Nonadherence, inadequate monitoring, and side-effects result in suboptimal outcomes in ulcerative colitis (UC). We hypothesize that telemanagement for UC will improve symptoms, quality of life, adherence, and decrease costs.

This article describes the challenges encountered in the design of the home telemanagement in patients with UC trial.

In a randomized trial to assess the effectiveness of telemanagement for UC compared to best available care, 100 patients will be enrolled. Subjects in the intervention arm will complete self-testing with telemanagement weekly; best available care subjects will receive scheduled follow up, educational fact sheets, and written action plans. Telemanagement consists of a home-unit, decision support server, and web-based clinician portal. The home-unit includes a scale and laptop. Subjects will respond to questions about symptoms, side-effects, adherence, and knowledge weekly; subjects will receive action plans after self-testing. Outcome variables to be assessed every 4 months include: disease activity, using the Seo index; quality of life, using the Inflammatory Bowel Disease Questionnaire; adherence, using pharmacy refill data and the Morisky Medication Adherence Scale; utilization of healthcare resources, using urgent care visits and hospitalizations.

We encountered several challenges during design and implementation of our trial. First, we selected a randomized controlled trial design. We could have selected a quasiexperimental design to decrease the sample size needed and costs. Second, identification of a control group was challenging. Telemanagement patients received self-care plans and an educational curriculum. Since controls would not receive these interventions, we thought our results would be biased in favor of telemanagement. In addition, we wanted to evaluate the mode of delivery of these components of care. Therefore, we included written action plans and educational materials for patients in the control group ('best available care'). Third, we could not blind subjects to group assignment. In an attempt to decrease bias, staff was masked to group assignment to decrease measurement bias. Fourth, we selected outcome measures that were not invasive to decrease risks to subjects and to enhance recruitment.

Our results may not be generalizable as our program is a tertiary center. Further, subjects are not blinded to the intervention potentially resulting in bias; we attempt to minimize this bias by having staff masked to treatment group at the time of assessment of outcome measures.

To the best of our knowledge, our trial will be the first randomized controlled trial to evaluate telemedicine in subjects with gastrointestinal disease. We describe several issues encountered in design and implementation of our trial that will aid investigators when planning telemedicine trials in inflammatory bowel disease.