Open access Protocol Exercise referral schemes enhanced by self-management strategies to reduce sedentary behaviour and increase physical activity among community- dwelling older adults from four European countries: protocol for the process evaluation of the SITLESS randomised controlled trial Laura Coll-Planas, 1,2 Sergi Blancafort Alias,1,2 Mark Tully,3,4 Paolo Caserotti,5 Maria Giné-Garriga,6,7 Nicole Blackburn,3,4 Mathias Skjødt,5 Katharina Wirth,8,9 Manuela Deidda, 10 Emma McIntosh,10 Dietrich Rothenbacher,8 Rodrigo Gallardo Rodríguez,11 Javier Jerez-Roig,7,12 Oriol Sansano-Nadal,7 Marta Santiago,7 Jason Wilson,3,4 Míriam Guerra-Balic,7 Carme Martín-Borràs,7,13 Denise Gonzalez,14 Guillaume Lefebvre,14 Michael Denkinger,9 Frank Kee,4 Antoni Salvà Casanovas,1,2 Marta Roqué i Figuls,1,2 On behalf of the SITLESS To cite: Coll-Planas L, group Blancafort Alias S, Tully M, et al. Exercise referral schemes enhanced by self-management strategies to reduce sedentary behaviour and increase physical Abstract Strengths and limitations of this study activity among community- Introduction SITLESS is a randomised controlled trial dwelling older adults from four determining whether exercise referral schemes can be ►► Process evaluation is an essential part of the testing European countries: protocol enhanced by self-management strategies to reduce of the SITLESS complex intervention. for the process evaluation sedentary behaviour and increase physical activity in of the SITLESS randomised ►► The Medical Research Council guidance on ‘Process the long term, in community-dwelling older citizens. The evaluation of complex interventions’, by Moore et controlled trial. BMJ Open intervention is complex and requires a process evaluation al., is used as a framework for conducting and re- 2019;9:e027073. doi:10.1136/ bmjopen-2018-027073 to understand how implementation, causal mechanisms porting this process evaluation. and context shape outcomes. The specific aims are ►► Mixed methods are used to address the specific ►► Prepublication and additional to assess fidelity and reach of the implementation, aims, and each aim is addressed by several meth- material are published online understand the contextual aspects of each intervention odological procedures applied with a complemen- only. To view please visit the site, evaluate the mechanisms of impact, and explore tary purpose. journal online (http://​dx.​doi.​ org/​10.​1136/​bmjopen-​2018-​ perceived effects. ►► Important information is identified about the differ- 027073). Methods and analysis Following the Medical ences and commonalities regarding implementation, Research Council guidance on complex interventions, a context and mechanisms of impact of a complex in- Received 4 October 2018 combination of qualitative and quantitative procedures tervention aimed at increasing physical activity and Revised 23 April 2019 is applied, including observational checklists and reducing sedentary behaviour across intervention Accepted 22 May 2019 attendance registries, standardised scales (ie, Marcus’s sites in four European countries. Self-Efficacy Questionnaire, Physical Activity Self- ►► Potential limitations are the reduced number of cas- Regulation Scale and the Lubben Social Network Scale) es interviewed from the control arm and the lack of © Author(s) (or their at baseline, postintervention and follow-up assessments, in-depth analysis of the experiences of participants employer(s)) 2019. Re-use semistructured questionnaires gathering contextual with lower adherence to the intervention and of permitted under CC BY. characteristics, and participant observations of the those dropping out of the study. Published by BMJ. sessions. Semistructured interviews and focus groups For numbered affiliations see with the participants and trainers are conducted at end of article. postintervention and during the follow-up to explore effects complement results on impact. The qualitative and Correspondence to their experiences. Outcomes from the standardised quantitative findings will help to refine the logic model Dr Laura Coll-Planas; scales are analysed as moderators within the impact to finally support the interpretation of the results on the ​laura.​coll@​uab.​cat evaluation. Descriptive results on context and perceived effectiveness of the intervention. Coll-Planas L, et al. BMJ Open 2019;9:e027073. doi:10.1136/bmjopen-2018-027073 1 Open access Ethics and dissemination The study design was approved by the to the specific physical location in which the intervention respective Ethical Committee of Ramon Llull University, Southern is put into practice. Denmark, Northern Ireland and Ulm University. Participation is voluntary, In our current ageing societies, promoting physical and all participants are asked to sign informed consent before starting activity has become one of the key healthcare policies for the study. A dissemination plan operationalises how to achieve a social the prevention of age-related disability, risk reduction of impact by reaching academic and non-academic stakeholders. A data non-communicable diseases and improvement of quality management plan describes the specific data sets and regulates its deposition and curation. All publications will be open access. of life.6–8 Moreover, sedentary behaviour has been estab- Trial registration number NCT02629666; Pre-results. lished as a risk factor for mortality, which is independent of low physical activity, and reducing it is increasingly becoming a relevant public health concern, especially for older adults.9–12 Therefore, the SITLESS project was Introduction designed to determine whether the effects of exercise Randomised controlled trials have been traditionally referral schemes (ERS) may be enhanced by self-manage- focused on outcome evaluations by linking cause and ment strategies (SMS), to reduce sedentary behaviour and effect and have been criticised as oversimplistic for increase physical activity with the final goal of long-term ignoring relevant aspects, such as the crucial role of behaviour change, an improvement in health, physical context in shaping outcomes.1 Therefore, the guidance function, quality of life as well as psychosocial outcomes on the design and evaluation of trials of complex inter- in community-dwelling older European citizens. In order ventions has emphasised the relevance of the processes to to achieve these goals, SITLESS was designed as a three- provide greater confidence in the conclusions about the armed pragmatic randomised controlled trial to assess effectiveness and thus improve how they inform future the effectiveness and cost-effectiveness of existing phys- intervention development.2 Complex interventions are ical activity interventions (ERS) enhanced with SMS, composed of multiple active components that might, for compared with ERS alone and also with general recom- instance, interact, be directed to change behaviour, allow mendations about physical activity (control group).13 14 a degree of tailoring or require a level of skills by those The ERS intervention aims to improve strength, aerobic delivering and/or receiving the intervention. In addition, capacity and balance during 16 weeks, with two 60 min the effects of the intervention might be context-depen- sessions per week. The SMS consists of one one-to-one dent, and the causal pathway between the intervention session, six group-based sessions and four telephone and the outcome might be long, variable or include more prompts to encourage behaviour change. Participants are than one causal pathway.2 followed up for 12 and 18 months postintervention. The The Medical Research Council (MRC) has recently control group receives two sessions consisting of general proposed guidance to structure the process evaluation information and recommendations on healthy ageing. of clinical trials of complex interventions.3 4 This frame- A total of 1338 participants are expected to be included work is organised around three categories—(1) context, (446 in each arm). (2) implementation and (3) mechanisms of impact— The SITLESS intervention fulfils the definition of and considers their mutual relationships. This guideline a complex intervention, mainly since it comprises recommends assessing fidelity and quality of implemen- different active components that allow for tailoring to the tation, clarifying causal mechanisms (ie, to get a better personal and local characteristics, it is aimed at changing understanding of the complex pathways between the behaviours, and has a wide potential range of short-term intervention and the outcomes, and to identify unex- and long-term physical, social and psychological effects.15 pected mechanisms) and identifying contextual factors Accordingly, it was built on a logic model presented in associated that might explain why different results could figure 1, and the SITLESS trial requires a process evalua- be found across sites. Likewise, the Context and Imple- tion as recommended. mentation of Complex Interventions (CICI) framework In addition, for the SITLESS study, conducted in the has been proposed to assess context in detail.5 According European research framework Horizon 2020, it is a to the CICI framework, context is defined as ‘anything key issue to use the approach of Responsible Research external to the intervention that may act as a barrier or and Innovation (RRI). RRI is understood as an approach facilitator to its implementation, or its effect’.5 According that anticipates and assesses potential implications and to Moore et al.,4 ‘complex interventions work by intro- societal expectations with regard to research and inno- ducing mechanisms that are sufficiently suited to their vation, with the aim to foster the design of inclusive and context to produce change, while causes of problems sustainable research and innovation.16 RRI has been targeted by interventions may differ from one context to defined as an attempt to rethink research as a public good another’. The CICI framework comprises three dimen- and thus conducted with and for society.16 In practice, sions—context, implementation and setting—which the SITLESS project has considered specifically public interact with one another and with the intervention involvement, gender, ethics, open access and governance dimension. Hence, context comprises seven domains— along the project. geographical, epidemiological, sociocultural, socioeco- Therefore, the SITLESS project has designed a nomic, ethical, legal and political—whereas setting refers process evaluation with the aim of understanding how 2 Coll-Planas L, et al. BMJ Open 2019;9:e027073. doi:10.1136/bmjopen-2018-027073 Open access Figure 1 The logic model of the SITLESS intervention. BMI, Body Mass Index; HRQoL, Health-related Quality of Life; MVPA, Moderate and Vigorous Physical Activity; SMART, Specific Measurable Achievable Relevant Time-Oriented. implementation, causal mechanisms and the contextual and comparability between and within intervention sites; factors shape the short-term and long-term outcomes. and (3) assessing the generalisability of the intervention’s The primary outcomes of the trial include (1) sedentary effectiveness results. Third, the SITLESS project aims to behaviour as sitting time and time spent in activities which understand the mechanisms of impact. Fourth and finally have been traditionally reported as requiring ≤1.5 Meta- it aims to explore the perceived effects of the intervention. bolic Equivalent of Task (MET), and (2) physical activity The specific research questions corresponding to each as daily counts-per-minute and intensity of exercise aim are shown in figure 2. performed, both objectively measured by ActiGraph accelerometer. Furthermore, the secondary outcomes include other health-related aspects, quality of life, func- Methods and analysis tion and psychosocial outcomes at postintervention and Theoretical frameworks informing this process evaluation at 12 and 18 months follow-up. The specific aims of the The process evaluation of the SITLESS project follows process evaluation are first to assess the implementation the guidance on ‘Process evaluation of complex inter- of the intervention in terms of (1) fidelity to the inter- ventions’ from the MRC.4 According to this guidance, vention design, variability of what was delivered and how process evaluation is understood as an essential part of between intervention sites; and (2) reach (adherence) of the design and testing of the SITLESS intervention. In participants to the intervention. Second is to understand addition, context will be specifically evaluated according the role of context through the following: (1) exploring to the CICI framework.5 whether contextual factors regarding the characteristics of the setting, facilitator, the neighbourhood, healthcare The SITLESS process evaluation framework system and other location-specific aspects affect imple- Figure 2 illustrates the application of the MRC guid- mentation and outcomes; (2) assessing the variability ance to the SITLESS project. This figure highlights the Coll-Planas L, et al. BMJ Open 2019;9:e027073. doi:10.1136/bmjopen-2018-027073 3 Open access Figure 2 The process evaluation framework for the SITLESS trial. **IDEA refers to the IDEA (Identify Develop Evaluate Analyse) problem solving framework.24 ERS, exercise referral schemes; PA, physical activity; SB, sedentary behaviour; SMS, self-management strategies. research question in the categories of implementation, Qualitative methods require specific considerations context, mechanisms of impact and perceived effects, regarding the research team, reflexivity and the study and relates them with each methodological procedure design. applied. In addition, it shows the intervention character- A multicountry and multidisciplinary research team istics and the outcomes. including a variety of health professionals (medical doctors, epidemiologists, exercise physiologists, physio- Mixed methods used to address the specific aims therapists and so on) will lead the analysis of the qual- A combination of quantitative and qualitative procedures itative data gathered in Barcelona, Odense, Ulm and will be applied. This process evaluation was planned prior Belfast. All members of the team involved in the quali- to implementation of the intervention, except for the tative research will be encouraged to keep a research semistructured questionnaire on contextual data that was diary to record their reflexive notes, impressions of the added more recently. Nevertheless, observational check- data and thoughts about the analysis throughout the lists, attendance registries and guidance for the qualita- process. Researchers should remain flexible and adap- tive procedures were tested with the first participants and tive throughout the process and not be attached to improved accordingly to ease their use and capture the certainty to generate rich findings that could explain the most relevant data. complexity of social issues. Researchers with less experi- Quantitative procedures quantify the fidelity to the ence in qualitative research will be trained to code, index intervention manual and participants’ adherence to and chart data and to think reflexively about how their the intervention through observational checklists and identities and experience affect the analysis process in attendance registries, respectively. They also involve order to help to interpret with accuracy the meaning and standardised scales to assess the mediating factors (ie, significance of the data. Marcus’s Self-Efficacy Questionnaire (MSEQ),17 the We will apply purposeful sampling of participants with 12-item Physical Activity Self-Regulation Scale (PASR- a maximum variation sampling method as a strategy to 12)18 and the Lubben Social Network Scale (LSNS)).19 select a small number of cases that maximise the diver- Hence, contextual characteristics will be collected in a sity relevant to the research question. By doing so, table format with prespecified topics (eg, geographical we aim to explore the differences and commonalities aspects). regarding gender, functional level, ethnicity and socio- Qualitative procedures address the same research ques- economic background, next to the type of intervention tions but according to participants’ perceptions, trainers’ arm (SMS+ERS group, ERS group and control group). experiences and researchers’ observations. Qualitative Each qualitative procedure targets a specific purposeful procedures include participant observation, semistruc- sample of participants from each of the four intervention tured interviews and focus groups and are detailed in sites (Barcelona, Belfast, Odense and Ulm) according table 1. to the characteristics specified in table 1. Moreover, 4 Coll-Planas L, et al. BMJ Open 2019;9:e027073. doi:10.1136/bmjopen-2018-027073 Open access Table 1 Specific qualitative procedures, preselected criteria for the purposeful sampling Preselected criteria for the purposeful Qualitative sampling and numbers of each profile in each procedure Target intervention site Timeframe Participant Participants in the SMS Criteria: During the SMS intervention. observation group 1 from low socioeconomic neighbourhood (or medium, if no low). 1 medium neighbourhood (or high, if no low). Semistructured Participants in the SMS Criteria (priority low and medium socioeconomic End of intervention. individual group status): interviews 1 frail man and 1 frail woman. 1 robust man and 1 robust woman. 1 for each ethnic minority (if applicable). Criteria: 12-month follow-up. 1 man and 1 woman (priority is low-to-medium 18-month follow-up. socioeconomic status). Participants of the ERS Criteria: 12-month follow-up. group 1 man and 1 woman (priority is low-to-medium 18-month follow-up. socioeconomic status). Participants of the Criteria (priority is low-to-medium socioeconomic End of intervention. control group status): 12-month follow-up. 1 man. 18-month follow-up. 1 woman. Focus groups Participants of the SMS Criteria: End of intervention. intervention groups 1 low socioeconomic neighbourhood (or medium, if no low). 1 medium neighbourhood (or high, if no low). Participants of the ERS 1 (priority is low-to-medium socioeconomic status). End of intervention. intervention groups ERS and SMS trainers 1 with all trainers involved in the ERS and SMS End of intervention. training. ERS, exercise referral schemes; SMS, self-management strategies. several procedures will be applied at different time- outdoors, and whether and how the trainers tailored the points (postintervention and follow-up) to match with ERS sessions to the frail participants). the relevant assessment time-points of the trial. The aims regarding context, mechanisms of impact and perceived Attendance registries effects will be explored at the end of intervention and All SMS, ERS and control group sessions have a corre- both follow-up periods, whereas implementation is only sponding attendance registry recorded by the trainer assessed at the end of the intervention. after each session. It contains each participant’s atten- Online Supplementary appendices 1–6 show exam- dance, reasons for not attending and adverse events due ples of the templates used as observational checklists and to the intervention. The attendance is used to calculate attendance registries, and guidance for the qualitative the adherence as an equivalent concept of dose and procedures (semistructured interviews and focus groups). considers the compliance with the intervention (eg, did Each aim is addressed by several methodological proce- only part of the exercises due to pain). dures applied with a complementary purpose as illus- trated in figure 3 and described in detail below. Qualitative procedures Methods to evaluate aim 1 Semistructured interviews and focus groups with partici- Observational checklists pants will include questions about their perception of the After each session of all SMS and ERS groups, the trainer implementation of the specific intervention elements. records the degree of fulfilment of each task and each Focus groups with trainers will have specific questions on aim. Hence, the observational checklist of the ERS their perception of the implementation challenges. In sessions also covers the implementation characteristics addition, the participant observation of the SMS and ERS that might vary across intervention sites or across sessions sessions by researchers will focus on the implementation (eg, whether the sessions are conducted indoors or of the intervention. Coll-Planas L, et al. BMJ Open 2019;9:e027073. doi:10.1136/bmjopen-2018-027073 5 Open access Figure 3 Specific aims, procedures and outcomes of the process evaluation and their triangulation with impact evaluation outcomes. ERS, exercise referral schemes; SMS, self-management strategies. Methods to evaluate aim 2 Methods to evaluate aim 3 Prespecified contextual characteristics Standardised scales The evaluation of this aim is addressed with a table with The MSEQ,17 PASR-1218 and LSNS19 are assessed within prespecified domains of context. Specifically, all inter- the baseline, postintervention and follow-up evaluations vention sites will provide information on the specific to explore the role of self-efficacy, self-monitoring, goal intervention setting and context. Setting characteristics setting, eliciting social support, reinforcements, time will comprise neighbourhood socioeconomic level (low/ management, relapse prevention and social network as medium/high), urban/semirural/rural area, type of mediators of the decrease in sedentary behaviour and centres involved with the recruitment, type of centres in increase on physical activity as predicted by the SITLESS which the ERS intervention, SMS intervention and the logic model (figure 1). The MSEQ has been shown to be control group intervention are conducted and existing a reliable, valid and accurate measure of self-efficacy for ERS in these settings. This information will be collected physical activity participation in populations with chronic by researchers at each intervention site. Data on context diseases.20 21 The PASR-12 has been shown to be a reli- will include those geographical, epidemiological, socio- able and valid scale for the measurement of self-regula- cultural, socioeconomic, ethical, legal and political tion for physical activity among older adults.18 The LSNS information that might be relevant as a barrier or facil- is commonly used as a measure of social networks. It has itator to the intervention implementation, or its effect, high levels of internal consistency, stable factor structures for instance existing national programmes on physical and high correlations with criterion variables, and clinical activity. These contextual data will be first collected by the cut-points show good convergent validity among Euro- leaders of each intervention site and complemented with pean community-dwelling older adults.22 the knowledge and views of the members of each local advisory board. Qualitative procedures This will be used to get a better understanding of complex Qualitative procedures causal pathways and unexpected mechanisms between Semistructured interviews and focus groups with partic- intervention and outcomes. Specifically, semistructured ipants and trainers include specific questions about interviews with participants and focus groups with partic- the context. We will specifically explore the role of the ipants and trainers at the end of the intervention and at physical environment and social context, with the latter follow-up will include questions about perceived effects comprising the personal circumstances, the personal according to participants’ experiences and trainers’ network and the group dynamic of the SMS+ERS and ERS observations. In addition, the participant observation of intervention arms. the SMS and ERS sessions by researchers will focus on 6 Coll-Planas L, et al. BMJ Open 2019;9:e027073. doi:10.1136/bmjopen-2018-027073 Open access how participants actively respond and interact with the groups, and the trainers’ notes from the participant intervention. observation. The framework method will be applied to analyse these data.23 The framework method belongs Methods to evaluate aim 4 to the thematic analysis or qualitative content analysis. Qualitative procedures These approaches first identify commonalities and differ- This aim is addressed through semistructured interviews ences in qualitative data, and then focus on relationships with participants and focus groups with participants and between different parts of the data to draw descrip- trainers at the end of the intervention and at follow-up. tive and/or explanatory conclusions clustered around Perceived impacts will be considered as complementary themes. Accordingly, we will apply a combined approach to the results of the effectiveness evaluation assessed with to the analysis, enabling themes to be developed both standardised scales.13 inductively from the accounts (experiences and views) Within the RRI framework, the SITLESS process eval- of research participants and deductively from existing uation explores if there are differing responses to the literature. Regular team meetings will facilitate our crit- intervention by disadvantage categories related with ical exploration of participant responses, discussion of gender, disability, ethnicity and socioeconomic position. deviant cases and agreement on recurring themes. It also considers how these categories affect implemen- The analysis will be conducted following these steps: tation and mechanisms of impact and, consequently, the transcription, familiarisation with the interview, coding of effectiveness of the intervention.16 the transcripts and diaries of the participant observation, developing a working analytical framework, applying the Analysis plan analytical framework, charting data into the framework Data from the process evaluation will be analysed matrix, and interpreting the data. according to two main procedures: quantitative and qual- The initial coding of the first four transcripts, one itative analyses. per intervention site, will be conducted by two indepen- Quantitative analysis dent researchers, adding coding labels, notes and ideas. Quantitative process evaluation data, recorded in the As a next step, a working analytical framework will be attendance registries and the observational checklists, developed by discussing codes assigned, similarities and will be used to conduct a descriptive analysis, overall and differences to achieve an agreement on a set of codes to by intervention site. The SITLESS observational checklists establish an initial analytical framework. This framework will be used to calculate the percentage of fidelity of the will be refined by coding of further manuscripts until no implementation regarding the tasks and aims of the SMS new codes are generated by two independent researchers. and ERS sessions in each intervention site. The adher- Next, the final analytical framework will be applied to the ence to the ERS, SMS and control group sessions will be remaining transcripts. The use of a software to conduct calculated as a percentage of sessions attended with each the analysis will be discussed. Once all the data are coded, respective attendance registry. In addition, this registry we will summarise the data in a matrix for each theme.23 will allow estimating the frequency of reasons for not The matrix will comprise one row per participant and one attending. Percentage of fidelity of the implementation column per code with a separate sheet for each category, and adherence will be compared across intervention sites using also illustrative quotations into the corresponding to identify whether differences may exist between them. cells. To interpret the data, themes are generated from Comparison will be conducted applying analysis of vari- the data set by reviewing the matrix and making connec- ance models. tions within and between participants and categories. Specific data will be entered in the main study database. Lastly, analytical memos and team discussions will be This is the case for the variable fidelity and for adherence used to develop themes aimed at offering possible expla- that will allow us to assess whether a better implemented nations for what is happening within the data agreed by intervention or a higher dose of the intervention is consensus. related with more effect, respectively. The potential medi- ator role of adherence/dose, the MSEQ, the PASR-12 and Triangulation with impact evaluation outcomes the LSNS on the effect of the intervention on physical As shown in figure 3, outcomes of the process evalua- activity and sedentary behaviour will be explored by cova- tion are triangulated with impact evaluation outcomes riance analyses, adding these scales as covariates to the (ie, quantitative outcomes on effectiveness) to support mixed linear models testing the effect of the interven- their interpretation, thus reinforcing each other. Specif- tion. The statistical efficacy analysis of SITLESS has been ically, quantitative outcomes on fidelity, adherence and described previously.13 The information collected in the from standardised scales are analysed as moderators prespecified table on context characteristics will be used with the impact evaluation outcomes. Descriptive results to describe each intervention site. on context and qualitative findings on perceived effects are compared with impact results to complement them. Qualitative analysis Qualitative results and the overall quantitative results will The SITLESS qualitative data analysis will be based on help to refine the logic model to finally support the inter- the transcripts of the semistructured interviews and focus pretation of impact evaluation results. Coll-Planas L, et al. BMJ Open 2019;9:e027073. doi:10.1136/bmjopen-2018-027073 7 Open access Patient and public involvement All publications from the SITLESS project, including SITLESS, as an RRI project, has considered in the study those related with process evaluation, will be open access. design and involved from its onset different types of stake- holders as primary, secondary and tertiary end users.13 Stakeholders comprise female and male older adults, Discussion representatives of older adults’ associations, primary The SITLESS project has developed a process evalua- healthcare and sports professionals, policymakers and, tion to understand how implementation, causal mecha- where relevant, other local stakeholders such as health nisms and the contextual factors shape the outcomes of insurance. Specifically, a total of four local advisory boards a complex intervention that enhances ERS with SMS to were created at the beginning of the project, one in each reduce sedentary behaviour and increase physical activity intervention site (Barcelona, Odense, Belfast and Ulm). among community-dwelling older adults of four Euro- They met periodically with the local researchers’ team pean countries. Accordingly, this trial has the purpose of and participated in a number of decisions of the project facilitating future intervention development by informing to consider patients’ priorities and motivations, experi- policymakers, practitioners and researchers about imple- ence and preferences. Initially, a literature review was mentation, mechanisms of impact and contextual factors conducted, which included how older adults perceive in detail. In case the intervention is proven as being effec- physical activity and sedentary behaviour in order to tive and cost-effective, the information will be focused design an intervention with the potential to achieve on the delivered intervention to allow replication in sustained changes of healthy behaviour. At each site, we new contexts considering the relevant core components, initially explored older adults’ experiences, preferences context characteristics and implementation challenges. and priorities regarding behaviour change through focus In case of being ineffective, process evaluation results will groups. Their views were used to adapt the intervention contribute to define whether the failure can be attributed design. Local advisory boards contributed to the chal- to the intervention itself or to a poor implementation. lenges faced regarding recruitment, retention of partic- Hence, process evaluation will provide useful data for ipants in the study and the dissemination actions. Once systematic reviews to compare interventions considering the intervention design was finished, a feasibility study their implementation and contextual characteristics. was conducted in the four sites and evaluated with qual- itative interviews to the participants. Their perceptions Trial status were gathered and used to improve the intervention for SITLESS started in May 2015. The feasibility study was the definitive trial. conducted from December 2015 until June 2016. The Qualitative procedures undertaken in the trial include recruitment for the trial started in July 2016. At the time questions to assess whether participants perceive any of submission of this protocol, all 1390 participants have burden due to the intervention and to their participation been included in the study. To date, follow-up assess- in the study. ments are being conducted and are planned to finish by The dissemination plan comprises sharing the results September 2019. of the trial at each healthcare, social or leisure centre involved in the recruitment and/or intervention to reach Author affiliations 1 end users, health professionals and other relevant stake- Fundació Salut i Envelliment (Foundation on Health and Ageing)- UAB, Universitat holders. Further dissemination activities will be held to Autònoma de Barcelona, Barcelona, Spain 2 Institute of Biomedical Research (IIB Sant Pau), Barcelona, Spain involve participants of each site as much as possible. 3 Institute of Mental Health Sciences, School of Health Sciences, Ulster University, Newtownabbey, UK 4 UKCRC Centre of Excellence for Public Health (NI), Centre for Public Health, Queen's University Belfast School of Medicine Dentistry and Biomedical Sciences, Ethics and dissemination Belfast, UK 5 Dissemination Department of Sports Science and Clinical Biomechanics, Center for Active and Healthy Ageing (CAHA), Syddansk Universitet, Odense, Denmark The SITLESS dissemination plan operationalises the main 6 School of Health and Life Sciences, Glasgow Caledonian University, Glasgow, UK dissemination goal, which is to achieve a social impact 7 Department of Physical Activity and Sport Sciences, Faculty of Psychology, by reaching academic and non-academic stakeholders, Education and Sport Sciences (FPCEE) Blanquerna, Ramon Llull University, that is, primary care and healthcare professionals, older Barcelona, Spain 8 adults, policymakers, professionals related to physical Institute of Epidemiology and Medical Biometry, Universitat Ulm, Ulm, Germany 9 Agaplesion Bethesda Clinic, Geriatric Research Unit Ulm University and Geriatric activity promotion and so on. Specifically, process eval- Center Ulm Alb-Donau, Universitat Ulm, Ulm, Germany uation results will enhance social impact by providing 10 Health Economics and Health Technology Assessment (HEHTA), University of implementation details that might be of interest for poli- Glasgow Institute of Health and Wellbeing, Glasgow, UK 11 cymakers. In addition, qualitative results (eg, quotations Departamento de Ciencias del Deporte y Acondicionamiento Físico, CEADE, of participants) will support reaching non-academic stake- Universidad Católica de la Santísima Concepción, Concepción, Chile 12 Research Group on Methodology, Methods, Models and Outcomes of Health and holders. A data management plan has been developed for Social Sciences (M3O), Faculty of Health Science and Welfare, Centre for Health and the SITLESS study describing each specific data set in the Social Care Research (CESS), Universitat de Vic - Universitat Central de Catalunya, study and regulating its deposition and curation. Catalunya, Spain 8 Coll-Planas L, et al. BMJ Open 2019;9:e027073. doi:10.1136/bmjopen-2018-027073 Open access 13 Department of Physical Therapy, Faculty of Health Sciences (FCS) Blanquerna, 5. Pfadenhauer LM, Gerhardus A, Mozygemba K, et al. Making Ramon Llull University, Barcelona, Spain sense of complexity in context and implementation: the Context 14 Sport Initiative et Loisir Bleu Association, Strasbourg, France and Implementation of Complex Interventions (CICI) framework. Implement Sci 2017;12:1–17. 6. Lee IM, Shiroma EJ, Lobelo F, et al. Effect of physical inactivity on Acknowledgements We would like to thank the advisory board members from major non-communicable diseases worldwide: an analysis of burden Barcelona, Odense, Ulm and Belfast and the participants of the feasibility study and of disease and life expectancy. Lancet 2012;380:219–29. the trial for their valuable contributions. 7. WHO. Global recommendations on physical activity for health. WHO Library Cataloguing-in-Publication Data Geneva, 2010:60. Contributors LC-P, MT, MG-G, PC, EM and DR contributed to the conception 8. Ding D, Lawson KD, Kolbe-Alexander TL, et al. The economic burden of the study. LC-P, SBA, KW, MSk and NB contributed to the development of the of physical inactivity: a global analysis of major non-communicable specific procedures. LC-P will coordinate the overall development of the process diseases. Lancet 2016;388:1311–24. evaluation. LC-P, KW, MSk and NB will coordinate the process evaluation in their 9. Owen N, Sparling PB, Healy GN, et al. Sedentary behavior: emerging respective intervention sites, and LC-P, SBA, KW, MSk and NB will contribute to evidence for a new health risk. Mayo Clin Proc 2010;85:1138–41. the acquisition of process evaluation data. LC-P, SBA, MT, PC, MG-G, CM-B, FK, 10. Koster A, Caserotti P, Patel KV, et al. Association of sedentary time MSk, KW, NB, MDei, EM, DR, RGR, JJ-R, OS-N, MSa, JW, MG-G, DG, GL, MDen, with mortality independent of moderate to vigorous physical activity. PLoS One 2012;7:e37696. ASC and MRiF contributed to the design of the final process evaluation and will 11. Harvey JA, Chastin SF, Skelton DA. Prevalence of sedentary behavior contribute to the analysis and interpretation of data and writing of the manuscript. in older adults: a systematic review. Int J Environ Res Public Health LC-P and MRiF contributed to the data management plan and analysis. All authors 2013;10:6645–61. provided feedback on the drafts of this paper and have read and approved the final 12. Ekelund U, Steene-Johannessen J, Brown WJ, et al. Does manuscript. physical activity attenuate, or even eliminate, the detrimental association of sitting time with mortality? A harmonised meta- Funding This work was supported by the European Union Horizon 2020 analysis of data from more than 1 million men and women. Lancet programme with grant number H2020-Grant 634270. 2016;388:1302–10. Competing interests None declared. 13. Giné-Garriga M, Coll-Planas L, Guerra M, et al. The SITLESS project: exercise referral schemes enhanced by self-management strategies Patient consent for publication Obtained. to battle sedentary behaviour in older adults: study protocol for a Ethics approval The study design was approved by the Ethics and Research randomised controlled trial. Trials 2017;18:221. 14. Deidda M, Coll-Planas L, Giné-Garriga M, et al. Cost-effectiveness of Committee of each intervention site: the Ethics and Research Committee of Ramon exercise referral schemes enhanced by self-management strategies Llull University (reference number: 1314001P) (Fundació Blanquerna, Spain), the to battle sedentary behaviour in older adults: protocol for an Regional Committees on Health Research Ethics for Southern Denmark (reference economic evaluation alongside the SITLESS three-armed pragmatic number: S-20150186) (University of Southern Denmark, Denmark), Office for randomised controlled trial. BMJ Open 2018;8:e022266. Research Ethics Committees Northern Ireland (ORECNI reference number: 16/ 15. Lewin S, Hendry M, Chandler J, et al. Assessing the complexity NI/0185) (Queen’s University Belfast) and the Ethical Review Board of Ulm of interventions within systematic reviews: development, content University (reference number: 354/15) (Ulm, Germany). Participation is voluntary, and use of a new tool (iCAT_SR). BMC Med Res Methodol and all participants will be asked to give written informed consent before the start 2017;17:1–13. of the study. 16. Bardone E, Lind M. Towards a phronetic space for responsible research (and innovation). Life Sci. Soc. Policy 2016;12:5. Provenance and peer review Not commissioned; externally peer reviewed. 17. Marcus BH, Selby VC, Niaura RS, et al. Self-efficacy and the stages of exercise behavior change. Res Q Exerc Sport 1992;63:60–6. Open access This is an open access article distributed in accordance with the 18. Umstattd MR, Motl R, Wilcox S, et al. Measuring physical activity Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits self-regulation strategies in older adults. J Phys Act Health others to copy, redistribute, remix, transform and build upon this work for any 2009;6(Suppl 1):S105–S112. purpose, provided the original work is properly cited, a link to the licence is given, 19. Lubben JE. Assessing social networks among elderly populations. and indication of whether changes were made. See: https://​creativecommons.​org/​ Fam Community Health 1988;11:42–52. licenses/​by/​4.​0/. 20. Garratt AM, Løchting I, Smedslund G, et al. Measurement properties of instruments assessing self-efficacy in patients with rheumatic diseases. Rheumatology 2014;53:1161–71. 21. Mendoza-Vasconez AS, Marquez B, Benitez TJ, et al. Psychometrics References of the self-efficacy for physical activity scale among a Latina women 1. Berwick DM. The science of improvement. JAMA 2008;299:1182–4. sample. BMC Public Health 2018;18:1097. 2. Campbell NC, Murray E, Darbyshire J, et al. Designing and evaluating 22. Lubben J, Blozik E, Gillmann G, et al. Performance of an abbreviated complex interventions to improve health care. BMJ 2007;334:455–9. version of the Lubben Social Network Scale among three European 3. Moore G, Audrey S, Barker M, et al. Process evaluation of complex community-dwelling older adult populations. Gerontologist interventions. UK Med. Res. Counc. Guid. Prep 2014;19-45:64–75 2006;46:503–13. http://www.​popu​lati​onhe​alth​sciences.​org/​MRC-​PHSRN-​Process-​ 23. Gale NK, Heath G, Cameron E, et al. Using the framework method evaluation-​guidance-​final-​2-.​pdf. for the analysis of qualitative data in multi-disciplinary health 4. Moore GF, Audrey S, Barker M, et al. Process evaluation of research. BMC Med Res Methodol 2013;13:117. complex interventions: Medical Research Council guidance. BMJ 24. Marcus B, Forsyth L. Physical Activity Intervention Series. Motivating 2015;350:h1258–7. people to be physically active. 2 Edn: Human Kinetics, 2009. Coll-Planas L, et al. BMJ Open 2019;9:e027073. doi:10.1136/bmjopen-2018-027073 9