JMIR Publications

ABSTRACT

Background:

Children with type 1 diabetes receive diabetes management support while in school, often from school nurses. School-partnered interventions may improve diabetes health outcomes, though there is limited evidence to support their effectiveness and sustainability. Various factors at the school or health system level may interfere with intervention usability, a key aspect of design quality. To identify and address potential barriers during intervention development, we employed user-centered design (UCD) methods to adapt an evidence based psychosocial intervention, the collaborative care model (CCM), to a virtual school-partnered collaborative care model (SPACE) for type 1 diabetes. SPACE is an entirely virtual intervention which establishes a collaborative team of school nurses, parents, diabetes health care providers, and other supports to assist children with an individualized treatment plan and measure progress towards achieving diabetes and academic goals.

Objective:

(1) To adapt the CCM to SPACE for type 1 diabetes through an iterative co-creation process with multidisciplinary community partners; (2) To evaluate and prioritize usability and identify strategies to mitigate potential barriers to implementation.

Methods:

We recruited patient, family, school, and health system partners (n=20) to co-create SPACE through three 90-minute design sessions using a shared whiteboard and video-conferencing. UCD methods included independent and group activities for idea generation, visual voting, and structured critique of an evolving SPACE prototype. We evaluated the prototype with the Usability Evaluation for Evidence-Based Psychosocial Interventions (USE-EBPI) methods. School nurses (n=10) from different districts with varying degrees of experience with diabetes completed individual cognitive walkthroughs of the prototype and associated tasks, followed by the Intervention Usability Scale (IUS). After testing, two members of the research team independently identified and prioritized (1-3 rating scale) discrete usability concerns.

Results:

Design meetings generated over 90 unique ideas for the SPACE content and implementation which were incorporated into the evolving prototype. Cognitive walkthroughs identified 17 discrete usability issues which were identified by 10-60% of school nurses. Usability issues were categorized as related to technology, available infrastructure, workflow, complexity, value, and training. Average priority rankings ranged from 1.0-3.0, with two issues receiving the highest priority: ability to access the virtual platform (30% of users) and mechanism for data sharing between nurses and providers (60% of users). Average IUS ratings (77.8, SD=11.1, 100-point scale) indicated adequate usability. There was no difference in IUS ratings by school nurse experience, suggesting that other factors may be more likely to influence usability.

Conclusions:

Cross-sector interventions are complex in nature, and perceived usability is a potential barrier to implementation. Using virtual co-creation methods with vested community partners promoted high-quality intervention design which is aligned with the priorities of end-users. Quantitative and qualitative assessments indicated an appropriate degree of usability, suggesting that SPACE is ready for more formal testing.