Development of the SELFBACK smartphone app to support self-management of non-specific low back pain using Intervention Mapping

Research Aim/Objective:
To describe the systematic step-by-step process of designing and developing the SELFBACK smartphone app
using Intervention Mapping (IM).
Research Methods:
IM was used in a two-year process of designing and developing the intervention content and technical features
of delivery of the SELFBACK app. In five steps, we used IM to ensure embedding of low back pain (LBP) selfmanagement
evidence, behavior change theory, target group participation and stakeholder inclusion in
developing the app. Data to support the development were collect from interviews with people with LBP and
health care professionals, literature reviews, clinical experience and group discussions among the more than
forty involved researchers, clinicians and specialists within musculoskeletal health, exercise physiology,
behavioural science, app development, machine learning and health innovation management. During the
development, we conducted user tests and revision, two feasibility studies and a pilot study.
Results:
The SELFBACK intervention aimed to reduce pain-related disability among people with non-specific LBP by
supporting them to use evidence-based self-management strategies via an app. Uptake and utilization of selfmanagement
strategies for LBP are affected by personal determining factors, e.g. lack of knowledge, low selfefficacy,
or low outcome expectations. Methods embedded in behavior change theory were outlined for all
determining factors to ultimately increase use of evidence-based self-management strategies. A three-pronged
program of strength and flexibility exercises of progressing difficulty levels; educational components in the form
of advice, quizzes and tools; as well as monitoring and feedback of step counts, was developed and adapted to
app format. The SELFBACK app created weekly self-management plans with goals for all three components. Each
week, a user’s plan was updated based on achievements, a tailoring questionnaire, preferences and a
sophisticated case-based reasoning system that compared the current case (user) with past similar successful
cases.
Discussion:
We describe a detailed example of using IM to systematically develop a theory-driven, complex, and digital
intervention designed to support self-management of LBP. The underlying decision support system using
artificial intelligence adds to the complexity of the selfBACK intervention.
While IM is a time-intensive collaborative process, the transparent process adds to the sparse literature on
development of complex interventions for self-management and should ensure effective implementation and
inform future researchers in this evolving field.