Skip to main content

Table 1 Crucial aspects to consider in clinical studies with medical devices that rely on behavioral changes and options to reduce potential limitations including their challenges

From: Diabetes management intervention studies: lessons learned from two studies

Aspects to consider Options to reduce limitations Challenges
Engaging participants in the intervention group Optimal engagement of involved HCPs:
 • Investigator meetings
 • Repeated training sessions of study sites
 • Peer-to-peer review
• Cost and time expensive
• Consider optimal timing of training activities
Support in device handling
on individual levels:
 • Training for HCPs
 • Technical trainers should be familiar with study aims
 • Visualization of device-based intervention messages for subjects
• Cost and time expensive
• Consider optimal timing of training activities
Detailed assessment of adherence:
 • Questionnaires (e.g., about therapy adaptations, perceptions of tools)
 • Result analysis of both: “per protocol population” and “intention to treat population”
• Adherence to some instructions are difficult to trace
Control group management Cluster-randomization • Limited statistical power
• Calculated cluster numbers have to be feasible, ethically justifiable, and affordable
Minimum attention to control group • Disease management is complex
• Standard of care cannot be fully standardized
• Standard of care is often less defined and monitored than the interventional treatment
Pilot trials • Cost and time expensive
Pretest periods • Cost and time expensive
Third study arm • Reduced statistical power ➔ requires more subjects ➔ cost and time expensive
Use of historical controls; retrospective data collection • Identification of a suitable control data
• Consider progress in treatment standards
Using two separate protocols for the two groups • Informed consent forms should be well designed to avoid inclusion of interventional aspects
Two-stage randomization • Ethical concern
Adaptive study designs • Changes have to be planned and defined in advance
• Implementation of adaptations has yet to be investigated