Isabelle White, Lee Smith, Daniel Aggio, Sahana Shankar, Saima Begum, Raluca Matei, Kenneth R. Fox, Mark Hamer, Steve Iliffe, Barbara J. Jefferis, Nick Tyler, Benjamin Gardner

Of all age groups, older adults spend most of the time sitting and are least physically active. This sequential, mixed-methods feasibility study used a randomised controlled trial design to assess methods for trialling a habit-based intervention to displace older adults' sedentary behaviour with light activity and explore impact on behavioural outcomes. Eligibility criteria were age 60-74 years, retired, and ≥6 h/day leisure sitting. Data were collected across four sites in England. The intervention comprised a booklet outlining 15 'tips' for disrupting sedentary habits and integrating activity habits into normally inactive settings, and eight weekly self-monitoring sheets. The control was a non-habit-based factsheet promoting activity and sedentary reduction. A computer-generated 1:1 block-randomisation schedule was used, with participants blinded to allocation. Participants self-reported sedentary behaviour (two indices), sedentary habit, physical activity (walking, moderate, vigorous activity) and activity habit, at pre-treatment baseline, 8- and 12-week follow-ups and were interviewed at 12 weeks. Primary feasibility outcomes were attrition, adverse events and intervention adherence. The secondary outcome was behavioural change. Of 104 participants consented, 103 were randomised (intervention N = 52, control N = 51). Of 98 receiving allocated treatment, 91 (93%; intervention N = 45; control N = 46) completed the trial. One related adverse event was reported in the intervention group. Mean per-tip adherence across 7 weeks was ≥50% for 9/15 tips. Qualitative data suggested acceptability of procedures, and, particularly among intervention recipients, the allocated treatment. Both groups appeared to reduce sedentary behaviour and increase their physical activity, but there were no apparent differences between groups in the extent of change. Trial methods were acceptable and feasible, but the intervention conferred no apparent advantage over control, though it was not trialled among the most sedentary and inactive population for whom it was developed. Further development of the intervention may be necessary prior to a large-scale definitive trial. One possible refinement would combine elements of the intervention with an informational approach to enhance effectiveness. ISRCTN47901994 (registration date: 16th January 2014; trial end date 30th April 2015).