The increase of the elderly population leads to increased prevalence of frailty, risk for poor health outcomes, and related health and social care costs. Lack of physical activity (PA) and established sedentary behaviours (SB) constitute an additional burden, as they are related to progression of chronic disease and disabling conditions. An existing initiative to battle SB and insufficient PA levels are exercise referral schemes (ERS) implemented in primary care, where insufficiently active individuals are referred to a third party service (sports centre or leisure facility) that prescribes and monitors an exercise programme tailored to the patients' needs. ERS had shown improvements in PA in the short-term, but may have limited power to change SB and produce long-term effects. Thus, ERS might be enhanced by self-management strategies (SMS) to promote behavioural change. Such strategies based on social cognitive theory have been shown to increase self-confidence, power to act, and involvement in exercise. In a first stage, a systematic review, focus groups and a feasibility study will be conducted. Then, a three-armed pragmatic randomized controlled trial (RCT) will assess the long-term effectiveness (18-month follow-up) of a complex intervention on sedentary behaviour (SB) in an elderly population, based on existing ERS enhanced by self-management strategies (SMS). It will be compared to ERS alone and to general recommendations plus two educational sessions. The RCT will include 1338 subjects and will have a follow up of 18 months. The effect on SB will be measured as sitting time and the number of minutes spent in activities requiring ≤ 1.5 Metabolic Equivalent Tasks, and PA as daily counts per minute and intensity of exercise, and daily step counts. Secondary outcomes will include: physical function, healthcare use and costs, anthropometry, bioimpedance, blood pressure, self-rated health and quality of life, activities of daily living, anxiety, depressive symptoms, social network, physical activity self-regulation, self-efficacy for exercise, disability, fear of falling, loneliness, executive function, and physical fatigue. In a subsample, the level of frailty-associated biomarkers and inflammation, and sarcopenia-associated markers of muscle quality will be analysed. A process evaluation will be performed throughout the trial. SITLESS will assess policy makers in deciding how or whether ERS should be further implemented or restructured in order to increase its adherence, efficacy and cost-effectiveness.

Inclusion Criteria:

Community-dwelling older persons aged 65 or above.

Able to walk without help of another person during 2 minutes, and without major physical limitations, defined as a score on the Short Physical Performance Battery (Guralnik et al., 1995) of 4 or above. The use of an assistive device will be permitted.

Insufficiently active and/or Self-report sitting too much during a usual day. Answer yes to the following question: 'Do you perform regular physical activity (PA) at least 30 minutes five or more days of the week (please only refer to physical activity that makes the participant run out of breath while doing it or it doesn't allow him/her to maintain a conversation while doing the activity (do not count regular walking)?'

And/or answer yes to question: 'For most days, do you feel you sit for too long (6-8 hours or more a day)? Some examples might include when watching TV, working the computer / laptop or when doing sitting-based hobbies such as sewing'.

Exclusion Criteria:

Dementia assessed with the six-Item Screener. Participants with three or more errors should be excluded.

Unstable medical conditions (e.g. elevated blood pressure after medication) or with symptomatic cardiovascular disease.

Any medical condition which may interfere with the study design (e.g. end-stage disease).

Refuse to wear an accelerometer (primary outcome)

Unable to attend during the study period (80% of the total sessions).

Had participated in an ERS in the 6 months prior the study.