Digital fall prevention for older adults : Feasibility of a self-managed exercise application and development of a smartphone self-test for balance and leg strength

Sammanfattning: As the numbers of older adults grow, fall prevention is vital to reduce health care needs due to falls and to increase quality of life. Balance and strength exercises have been found to be effective in fall prevention, however, long-term adherence is often poor. The growth of digital technology in society has generated opportunities for fall prevention with eHealth. The aim of this thesis was to evaluate the feasibility and use of a new digital fall prevention exercise programme, and to develop and investigate a smartphone self-test application for balance and leg strength. Three different studies included community-dwelling older adults ≥ 70 years, who were able to rise from a chair and stand without support. A feasibility study evaluated a new digital exercise programme (DP) compared to a paper booklet exercise programme (PB) for self-managed fall prevention in a four-month controlled participant preference trial (n = 67) (Paper I & II). Self-reported data on adherence, falls efficacy, and functional ability were collected and analysed, along with performance-based measures of gait speed, balance, and chair stand test. In Paper II the feasibility was explored of using the self-reported scales and performance-based outcome measures. A self-test application was also developed (Paper III) in co-creation with 10 participants, who met during five sessions to design the application’s instructions and user interface. The participants’ preference for and their contribution to the application design was analysed with qualitative content analysis with a deductive-inductive approach. A concurrent validity study (n = 31) (Paper IV) assessed the correlations between variables from the self-test prototype and outcome measures from clinical instruments. Results from the feasibility study show that 43% chose the DP and 57% PB, and the attrition rate was 17% and 37% respectively. Both groups had similar adherence, but for the subgroup that exercised most, participants in the DP group reported significantly more exercise time (Paper I). Participants in both groups reported a boost in balance after the intervention, and in the DP group also improved leg strength. Significantly more participants continued to use the DP at 12 months. The self-managed exercise intervention (Paper II) resulted in improvements in functional leg strength, which positively correlated with exercise time, but no other performance-based outcomes showed any significant improvements. Performance-based measurements of balance as well as the self- reported balance confidence and fear of falling revealed ceiling effects. Pre-assessments of self-reported outcomes and performance-based measures showed significant but low correlations, no such correlations were seen in change scores. The deductive-inductive analysis of the co-creation process resulted in 17 subcategories within the seven facets of the Optimized Honeycomb model for iii user experience (Paper III). The main results were that participants desired clear and appropriate information to understand why things were done in a certain way, and their contributions enhanced the user experience of the self-test. The concurrent validity testing of the self-test prototype (IV) showed low to moderate correlations for the strength test but limited correlations for the balance test. In conclusion the DP group showed comparable adherence to the programme as the PB group, as well as to previous studies, indicating it was feasible to use the new DP. DP participants also reported better exercise maintenance after 12 months. Positive self-reported effects were expressed in addition to leg strength improvement. Outcome measures for balance and falls efficacy revealed ceiling effects, consequently, these instruments might not be suitable for assessments in all community-dwelling older adults. In particular, for balance related outcomes there is a need for new more sensitive measurements. The co-creation of the smartphone self-test was feasible and valuable for user experience, but further validity and reliability testing are needed before it can serve as an independent assessment tool.