Reducing sitting at work: process evaluation of the SMArT Work (Stand More AT Work) cluster randomised controlled trial

Background: Office based workers accumulate high amounts of sitting time. SMArT (Stand More AT) Work aimed to reduce occupational sitting time and a cluster randomised controlled trial demonstrated it was successful in achieving this aim. The purpose of this paper is to present the process evaluation of the SMArT Work intervention. Methods: Questionnaire data were collected from intervention participants at 6 (n=58) and 12 months (n=55). Questionnaires sought feedback on the different components of the intervention (education, height-adjustable desk, Darma cushion, behaviour feedback, progress chats (coaching) with research team, action planning/goal setting diary) and experiences of evaluation measures. Control participants (n=37) were asked via questionnaire about the impact of the study on their behaviour and any lifestyle changes made during the study. Participants from both arms were invited to focus groups to gain a deeper understanding of their experiences upon completion of 12 month follow up. Results: Focus group and questionnaire data showed a positive attitude towards the height-adjustable workstation with a high proportion of participants using it every day (62%). Most participants (92%) felt the education seminar increased their awareness of the health consequences of too much sitting and motivated them to change their behaviour. Receiving feedback on their sitting time and support from the research team also encouraged behaviour change. The Darma cushion and action planning/goal setting diary were seen to be less helpful for behaviour change. Benefits experienced included fewer aches and pains, improved cognitive functioning, increased productivity, more energy, and positive feelings about general health. Conclusions: Key elements of the programme identified as facilitating behaviour change were: the educational seminar, the height-adjustable workstation, behavioural feedback and regular contact with research staff through regular progress chats.


Abstract
Background: Office based workers accumulate high amounts of sitting time. SMArT (Stand More AT) Work aimed to reduce occupational sitting time and a cluster randomised controlled trial demonstrated it was successful in achieving this aim. The purpose of this paper is to present the process evaluation of the SMArT Work intervention. Methods: Questionnaire data were collected from intervention participants at 6 (n=58) and 12 months (n=55). Questionnaires sought feedback on the different components of the intervention (education, height-adjustable desk, Darma cushion, behaviour feedback, progress chats (coaching) with research team, action planning/goal setting diary) and experiences of evaluation measures. Control participants (n=37) were asked via questionnaire about the impact of the study on their behaviour and any lifestyle changes made during the study. Participants from both arms were invited to focus groups to gain a deeper understanding of their experiences upon completion of 12 month follow up.
Results: Focus group and questionnaire data showed a positive attitude towards the height-adjustable workstation with a high proportion of participants using it every day (62%). Most participants (92%) felt the education seminar increased their awareness of the health consequences of too much sitting and motivated them to change their behaviour. Receiving feedback on their sitting time and support from the research team also encouraged behaviour change. The Darma cushion and action planning/goal setting diary were seen to be less helpful for behaviour change. Benefits experienced included fewer aches and pains, improved cognitive functioning, increased productivity, more energy, and positive feelings about general health. Conclusions: Key elements of the programme identified as facilitating behaviour change were: the educational seminar, the height-adjustable workstation, behavioural feedback and regular contact with research staff through regular progress chats. 3 Background High levels of sedentary behaviour (sitting) have been shown to be detrimentally associated with a number of physical and mental health outcomes [1][2][3]. For example, the U.S. 2018 Physical Activity Guidelines Committee concluded that there was "strong evidence for a direct association between greater amounts of sedentary behavior and higher risk of mortality from all-causes and CVD, and for higher risk of type 2 diabetes and CVD" [4]. Moreover, with trends towards greater sitting in the workplace [5], office workers have been shown to engage in high levels of sitting [6]. The Stand More AT (SMArT) Work programme was an intervention tested in employees of an English East Midlands National Health Service (NHS) Trust. The employees worked predominantly at a seated desk [7][8][9]. A full study protocol is published [7], but, in brief, groups of desk-based staff within the same offices were randomised to either an intervention or control condition. The intervention participants received a multi-component intervention designed to reduce workplace sitting. Those in the control office clusters continued with their usual practice.
The RCT's logic model (see [8]) stated that the intervention was grounded in several behaviour change theories and implemented through the intervention functions of the Behaviour Change Wheel. These included organisational, environmental, and individual and group functions, and informed elements of the process evaluation.
Results at 12 months showed favourable changes in the intervention group relative to controls for occupational sitting time, prolonged sitting, standing time, some musculoskeletal issues, various occupational measures (job performance, work engagement, occupational fatigue, sickness presenteeism), and quality of life [8].
Process evaluation is an important element of an intervention because it provides information on implementation (e.g., fidelity, reach), possible reasons for outcomes (mechanisms, such as participant responses and mediators), and contextual factors shaping intervention outcomes [10]. Given the multi-component nature of this trial, it was important to understand how participants viewed each component as well as the intervention overall. Therefore, to better understand how the SMArT Work intervention operated and was perceived by participants, we undertook several process evaluation assessments with the intervention participants (questionnaires at 6m and 12m follow-up and focus group at 12m follow-up) and control participants (questionnaire and focus group at 12m follow-up only).

Method
Ethical approval was obtained from Loughborough University, and Research and Innovation approval was obtained from the University Hospitals of Leicester NHS Trust (EDGE ID 34571). All individual participants provided informed consent on entering into the study.
A sequential exploratory mixed methods approach was adopted with data collected from both intervention and control participants using questionnaires and focus groups. Table 1 shows the main methods and data collected.
Questionnaires for the intervention participants sought feedback on the following main elements of the intervention (see Table 1): Educational seminar and leaflet: a 30-minute group educational seminar concerning the health consequences of sitting and the benefits of reducing or breaking up sitting and a leaflet to reinforce the key messages. Feedback was sought at 6 months only, due to one-off nature of the seminar; Feedback on their own sitting, standing and stepping generated from the activPAL monitor; Height-adjustable workstation: participants were given the choice of two models (full electric desk or a choice of 2 sizes of an adjustable platform which sat on their existing desk); Use of a sitting time diary: including action planning and goal setting; Use of the Darma cushion: a cushion placed on office chair which connected to a smart phone via Bluetooth. Real time feedback on sitting is provided along with a vibration prompt to break up sitting regularly; Brief coaching sessions ('progress chats'): These took place every few months throughout the intervention. Feedback was sought at 12 months only.
Questionnaires for the control participants (n=37, 80% of control participants still in study at 12 months) sought feedback on the impact of study measurement sessions and receiving health results. All participants were asked whether other lifestyle changes had been made during the study that might impact on the results, such as moving house or joining a gym.

Focus Groups
Participants were invited to attend a focus group upon completion of 12 month follow-up.
Focus groups were led by one researcher, were semi-structured with a focus group guide devised by the wider research team. Seven focus groups, lasting between 40-64 minutes, took place with 29 intervention participants (46% of intervention participants still in study at 12 months; 72% female; body mass index = 26.1±5.6kg/m 2 ; age = 41.1 ± 12.2 years), representing 16 intervention clusters (84%). The focus group discussion topic guide gathered responses concerning a) experiences of each intervention component, b) the facilitators to take part in the study, if and how their behaviour changed, c) insight on the strategies they used to change their behaviour and their experiences of reducing sitting behaviour, d) benefits and/or negative experiences of the intervention and discussions around sustaining new behaviour, and d) how the messages of SMArT Work could be rolled out (see Table 1).
Questions aimed to gather insight into why they took part, what motivated them to stay in the study once allocated to the control arm, whether they felt supported through the project by their manager and how the measurement feedbacks impacted them in any way.

Data Analysis
Forced choice and Likert scaled questionnaire items were analysed with frequency counts or means and standard deviations using IBM SPSS V25. Open-ended responses from the questionnaire were grouped into coherent themes (by SJHB) using Template Analysis [11].
All statements were entered into Mindgenuis (V6) software and grouped by themes and sub-themes. Audio recordings from the focus groups were transcribed verbatim and also subjected to Template analysis. The first stage of this template analysis was to define the themes relevant to the discussion topics outlined above. Two members of the research team (SOC and FM) independently applied these themes to the focus group data set to develop a template depicting the salient themes. Data that did not fit the initial template, but were relevant to the research aims, were coded and the themes were continuously modified as the data were interpreted, until a final template of five intervention focus group main themes were created and one control group main theme was created (see Table 2). There was agreement between the two researchers on the template themes identified and an additional theme around incidental culture of standing at work. Results were triangulated to integrate findings from focus groups and questionnaires.

Results
Results are presented mainly according to intervention components with quantitative and qualitative data from the questionnaires integrated throughout the results together with the themes identified from the focus groups (shown in Table 2.) Figure 1 shows the flow of participants through the randomised controlled trial.

Educational seminar and leaflet
Only seven intervention participants (14%) did not attend the seminar face-to-face and were sent an audio recording instead. From those who had attended the seminar, questionnaire data (Table 3) showed strong agreement between the respondents that the seminar was delivered at an appropriate level (94% agreed or strongly agreed) and was an appropriate length (95% agreed/strongly agreed). Importantly, 93% felt that the seminar increased their awareness of the health consequences of too much sitting and 95% felt it motivated them to make a change to the amount of time they spend sitting.
Open-ended questionnaire responses were requested concerning understanding the key seminar messages. Of the 52 respondents, 40 provided comments, with three main questionnaire themes emerging: standing and breaking up sitting is desirable to prolonged sitting excessive sitting is bad for your health exercise may not offset the detrimental effects of prolonged sitting.
In the focus groups, some participants discussed how knowledge learned from the seminar For the educational leaflet provided, 89% of questionnaire respondents said that they had read it. The content was seen as appropriate (88% rating agree/strongly agree). Most (88%) felt that the leaflet increased their awareness of the health consequences of too much sitting and 82% said it motivated them to make a change to the amount of time they spent sitting.

Height-adjustable Workstation
Though participants were given a choice of desk, they chose roughly evenly between the two models (60% chose a Varidesk platform and 40% the electric workstation). The purchasing and delivery of the workstations took longer than planned, so many of the participants would have only had the workstation installed for ~1 month prior to the 3 month follow up visit. Table 4 presents the quantitative feedback on the workstation. At 6 months, all those responding said that they have used the workstation at least once during the first 6 months, with all but three respondents reported using the workstation at least a few times per week (33%) or everyday (67%). In the first month of installation, 67% used it every day, with similar rates between desk type. The majority of participants did not find their workstation obtrusive to completing work tasks at 6m (69%) and 12m (67%). Questionnaire data showed that respondents were not self-conscious when using the workstations (84% and 88% at 6m and 12m respectively), nor did they think their colleagues minded when the workstation was being used (98% at 6m; 99% at 12m). These data were supported in the focus groups with participants highlighting that their nonparticipant colleagues did not impact on their use of the height-adjustable workstation [Level 3 sub-theme -social influence].
The people who didn't have the desks, they didn't say anything or, like it wasn't awkward to stand up at any point in the office or anything like that, it was fine (Intervention participant, focus group 3) Questionnaire respondents were asked how they scheduled the use of their workstations, such as specifying certain times or for particular tasks. At 6m, 44% reported scheduling often or very often, with a slight drop to 36% at 12m. The mixed responses were reinforced by the focus groups with some people stating that they had set times/periods when they used their desk, such as first thing in the morning or in the afternoon, but for Additional themes concerned 'comfort and utility' at 6m (e.g., 'more comfortable standing'; 'easy to use'), and 'ergonomics' at 12m (e.g., 'more natural eye line to the screen').
Musculo-skeletal outcomes were also reflected in the data from the focus groups [Level 3 sub-theme -reduction in musculoskeletal problems]: As the SMArT Work Intervention was a cluster design, where groups of people within the same office group were randomised to the same group, if one person was standing up (at their desk or elsewhere), this would often remind other colleagues to also stand up.
Therefore, a knock-on effect of colleagues standing regularly was evident, thus creating a culture shift, as illustrated by these comments from the focus groups [Level 1 Main theme -creating an incidental socio-cultural environment of standing at work]: Questionnaire respondents were also asked to write comments concerning what was negative about using the workstation at both 6m and 12m. The main issue that emerged concerned the lack of space on the desk and concerned papers and files falling off, and a lack of space for handling multiple papers. These comments were exclusively in reference to the Varidesk. Other negative issues mentioned were musculo-skeletal (e.g., 'initial low back pain'; 'initial leg pain'; 'feet can ache'), ergonomic (e.g., 'uncomfortable when typing a lot'; 'sometimes couldn't type when standing'; 'wires would get caught'), social issues (e.g., 'feel awkward when standing and talking to others who are sitting'), and additional work issues (e.g., 'remembering to use the workstation'; 'change to established work pattern'). This was also highlighted in the focus groups by some participants.
Initially it was the lack of space. Because it was a two-tiered system, when you did stand up there was not much space to put your paperwork on (Intervention participant, focus group 2)

Darma Cushion
The questionnaire findings (Table 5) showed the use of the Darma cushion and associated 'app' was moderate at 6m; 55% reported using it since it was given to them and few planned to use it in the future. Assessing over the past 6m, users of the Darma cushion reported varied responses, with 39% reporting infrequent use, while 37% reported daily use. Only a small percentage (15%) of participants viewed their feedback on the app frequently. Only 11 (20%) reported using the cushion in the last 6 month at the 12m time point, with 18% reporting infrequent use, 36% using it 'a few times per week', and 45% reporting daily use. The use of the Darma cushion was initially reasonable with 68% reporting daily use in the first month. The cushion vibration function was used by 62% and 46% of those that reported using the cushion in the past 6 months at the 6m and 12m time points respectively, with most (87% and 70% at 6m and 12m respectively) reporting it to be useful.
Ratings were provided on a number of characteristics of the Darma cushion at 6m and 12m. These are shown in Table 6. Data from the 32 participants using the cushion in the first 6 months suggested that it was easy to use (71% agreement), was not obtrusive (54%), it increased awareness (69%), and encouraged less sitting (66%). The small sample using the Darma cushion at 12m reported it as easy to use, largely unobtrusive, it increased awareness, although only 54% agreed it decreased sitting.
Focus group discussions on the topic suggested participants found other ways to set prompts, including using the Varidesk computer/phone app and Google Chrome Stand Up! At 6m and 12m, reasons given in the questionnaires for not using the cushion centred on lack of comfort, technological issues with the app and phone (e.g., syncing, storage and battery problems), length of charging lead and other reasons. The latter included a perception by some that it was not needed and that they could implement their own behaviour change without it. Similar comments were also made in the focus groups [Level 3 sub-theme -cushion]: It was very uncomfortable [and] it ran out of batteries so I never recharged it (Intervention participant, focus group 3) The lead is really short, you had to plug it in, I think once it died…. (Intervention participant, focus group 3).
Some also reported that they used it initially but didn't need it once using the heightadjustable workstation became more of a habit [Level 2 sub-theme -factors that promote standing and habit formation]: I did at the very start but then after that I actually found, because I was generally pretty good with my standing desk that I didn't really see the requirement for the cushion (Intervention participant, focus group 7) I think to start with, I had to use those timers and things to remind myself to stand up, but now it is just so natural…I just stand up until I feel like sitting down again or I stand up when I feel like I need to stand up" (Intervention participant, focus group 3) Table 7 presents the quantitative responses to the diary. Most questionnaire respondents reported that they either never used or no longer used the diary to keep a record of their sitting (91%) nor used it for goal-setting (93%) within the first 6m. Similar data were found for 12m.

Sitting Time Diary
Reasons for not using the diary, including for goal-setting, were given in open-ended comments and included perceived lack of time and time pressure of their job (e.g., 'work pressures -didn't think about it'), forgetting, not finding it useful (e.g., 'didn't see point'; 'doesn't work for me'), and motivation (e.g., 'effort of completing outweighs benefits').
Similar responses emerged from the focus groups (Level 3 sub-theme -diary): I think realistically you are probably not going to carry a paper diary around with you…it became another thing to either forget, like keep up with. And you always have your phone on you, so it's easier just to write things on your phone (Intervention participant, focus group 3) Some participants, however, stated in their questionnaire open-ended responses that they did not use the diary because they felt they did not need it. Some stated that their heightadjustable workstation was enough to encourage them to sit less, while others had created their own routine and habit (e.g., 'stand when work allows me'; 'I usually stand in the morning').

Feedback on Sitting Time
A large majority of the questionnaire respondents were in agreement, at both time points, that receiving feedback on their sitting time helped them think about their sitting, highlighted that they could be sitting too much, motivated them to change, helped plan and set goals, and was useful for reviewing progress ( Table 8). Some of these findings were also discussed by focus group participants. [Level 3 sub-theme -feedback].

'Progress Chats' (i.e., coaching) with Research Team Staff
All but 1 of the participants who were left in the study at 12 months had all 4 coaching sessions (n=62). There were 72, 65, 65, and 63 participants participating in the 1 st , 2 nd , 3 rd and final coaching sessions, respectively. At 12m only, participants were asked in the questionnaire to reflect on the coaching and support provided by research staff through the 'progress chats' that were provided (see Table 9). Participants reported that the chats helped them formulate plans (90%), helped them stay on track (90%), motivated them (94%), help them find solutions (87%), and provided support often enough (93%).

Other Lifestyle Changes
The questionnaire results showed other lifestyle changes were made by 39% of intervention participants in the first 6 months. Of those reporting the nature of such changes, 10 were positive (e.g., signing up for gym membership), and seven were negative (e.g., illness).

Facilitators and Barriers to Behaviour Change
Though the desk appeared to positively impact on behaviour change by providing participants the opportunity to stand whilst working, the lack of space on the Varidesk platform did appear to put participants off changing the desk position. This was mentioned by 50% of those reporting negative issues at 6m and concerned papers and files falling off, and a lack of space for handling multiple papers.
However, during the focus groups, many reported that this led to strategies to enhance the tidiness of their desk, thus creating a positive outcome [Level 3 sub-theme -changes in work style]. Other barriers to standing at both 6 and 12m from the questionnaire data were musculoskeletal (e.g., 'initial low back pain'; 'initial leg pain'; 'swollen ankles and feet'), ergonomic (e.g., 'uncomfortable when typing a lot'; 'sometimes couldn't type when standing'; 'wires would get caught'), and additional work issues (e.g., 'remembering to use the workstation'; 'change to established work pattern'; 'feel awkward when standing'). Regarding barriers to behaviour change, focus group participants felt that if they were having a stressful day or were engrossed in a certain tasks then they felt that they either needed to sit down or would simply forget to break up their sitting [Level 2 sub-theme -aspects of the job]. Others reported an emphasis on moving more, reflecting sub-themes of using selfmonitoring (e.g., 'I count my steps daily'), incidental (e.g., 'increased my stair use'), and exercise (e.g., 'I try to walk at lunchtime'). Being part of the project seemed to create greater awareness in some control participants, mainly around sitting. This was reflected in a reduction for daily sitting time in control participants at 3 month follow up, although not at further follow-up [8].

I think it is hard to stand on stressful days. That's what I found. You know, if there is a lot of stress in the office then it is difficult because you tend to sit. You
Finally, participants in the control group were asked whether any changes were made to their lifestyle after receiving health test results from the assessments. From 37 responses, 22 (60%) said that the tests did not have any impact on their lifestyle.
It became apparent from the focus groups that the feedback they received from the health measures at baseline, 3, 6 and 12 month follow up was a key motivator to staying in the study [Level 2 sub-theme -feedback from health measures].
It does make you more aware of, you know, the BMI and everything really…Its just to be aware of the whole , like, you've sort of had an MOT, haven't you…Every four months you have one, which I think is good (Control participant, control focus group 2)

Discussion
The process evaluation showed that participants had positive attitudes towards the height-adjustable workstation, with many using it on a daily basis. Most participants felt the education seminar increased their awareness of the health consequences of too much sitting and motivated them to change their behaviour. Receiving feedback on their sitting time and support from the research team also encouraged behaviour change. The Darma cushion and action planning/goal setting diary were seen to be less helpful for behaviour change. Several benefits were perceived by participants, including fewer aches and pains, improved cognitive functioning, increased productivity, and more energy. Additionally, behaviour change seemed to be enhanced by behavioural feedback and regular contact with research staff through regular progress chats.
From this process evaluation, we can draw on the following categories recommended in Medical Research Council (MRC) guidance [10]: intervention context: the contextual factors that might affect the implementation and outcomes of the intervention implementation: the implementation of the trial itself mechanisms of impact: any mechanisms helping to explain the impact of the trial.

Context
There was a mix of positive and negative changes made during the lives of intervention and control group participants during the course of the trial. There was no apparent systematic bias in this regard. However, taking part in the study did appear to influence controls, at least in the short term. It is clear that any assumptions that control group participants remain stable in their behaviours during the trial are unfounded. In the present study, 40% of controls who responded felt that feedback from their assessments led to either confirmation of their situation, a greater awareness of issues, or actual changes to behaviour. Given that changes in primary and some secondary outcomes in the trial were largely in the desired direction [8], any differences seen in the trial outcomes between intervention and control participants may be an underestimation.

Implementation
Five key implementation elements of the intervention assessed were the seminar and leaflet, the workstation, the Darma cushion, diary, and coaching 'chats'. The seminar and leaflet achieved good reach; most read the leaflet and attended the seminar.
The Darma cushion was chosen based on feedback from participants in our development work [9]. However, in the intervention study, responses to the cushion were mixed in terms of its usefulness. Some participants sought out their own methods for receiving prompts to break up their sitting. This highlights that 'one size does not fit all' and future interventions may wish to consider flexibility in the tools offered to participants. It is likely that diaries for action planning and goal setting were considered too difficult and an extra task not worth doing. If greater use of the Darma cushion is to be encouraged, issues concerning comfort and enhanced technology are priorities to address. It is unlikely that one self-monitoring or prompting tool will satisfy everyone, therefore there is a need to offer a greater choice of devices and tools for self-monitoring and prompting.
Comments from participants reflected low uptake of the diary and highlight that behaviour change techniques (BCTs) and other strategies provided by researchers may not always be seen in the same light by participants. Goal-setting as a BCT will not be effective if adherence is low. This will more likely be the case for BCTs and tasks that require greater cognitive effort and time.

Mechanisms of impact
The SMArt Work intervention was developed based on the Behaviour Change Wheel [9 12].
A key element of this approach is the 'COM-B' framework where behaviour (B) is considered to be a function of the capability (C) of the individual, the opportunity (O) they have, and their motivation (M). These can be seen as mechanisms of behaviour change and are considered in this discussion.
The educational seminar and leaflet were well received. They appeared to increase awareness of the health consequences of too much sitting and provided motivation to make changes to the amount of time spent sitting. This addresses the motivation element of the COM-B framework and is more associated with 'reflective' forms of motivation, requiring participants to process information prior to decision making. In addition, the seminar and leaflet are likely to enhance perceptions of capability. One belief endorsed was 'exercise may not offset the detrimental effects of prolonged sitting'. The belief that exercise does not offset the deleterious health effects of too much sitting is a controversial point in the contemporary literature and is probably a reflection of the development of the research field. Early epidemiological studies and meta-analyses suggested that higher levels of sedentary behaviour were associated with negative health outcomes when controlling for levels of moderate-to-vigorous (MVPA) or leisure-time physical activity [e.g., 2 13]. However, research has suggested that high levels of MVPA attenuate the effects of sitting on mortality [14 15]. At the time of the development of the SMArT Work project, beliefs were more aligned with the comments emanating from the open-ended comments of participants. If we repeated the education, we would advise that the message reflect a more balanced view.
The height-adjustable workstation was also well received and was reported to have had numerous benefits. The two Varidesk models was viewed positively but some reported issues of a lack of space on the platform for papers. However, there was evidence that people adapted to this and it became a positive feature (i.e., they became tidier). The provision of such desks enhances participant's capability and opportunity to reduce their sitting time.
Some of the qualitative findings support our quantitative results [6] concerning positive changes for musculoskeletal problems. Other process evaluations have also found participants reporting improvement in musculoskeletal issues [16]. Our qualitative findings also support our quantitative results around job performance, work engagement and recovery from occupational fatigue [8]. Importantly, most of the participants discussed how regularly standing benefited their work performance including concentration, confidence, and creativity, and they also mentioned a positive impact on energy levels. These findings have also been reported in other qualitative studies evaluating small scale height-adjustable workstations [16][17][18]. Specifically, Leavy and Jancey [18] found their participants reported that they felt height-adjustable workstations helped to create energy within work spaces and increased work performance.
The process evaluation also highlighted how standing at desks not only improved interaction between colleagues related to work tasks, but it also had a wider positive influence on engaging other employees not involved in the study in terms of reducing their sitting. Our intervention therefore provides new insights into how the development of social norms of regular standing has a widening influence on the workforce. Future trials could evaluate the reach of the effect of these types of interventions and assess changes in behaviour among non-participants.
Facilitators to changing sitting behaviours at work were explored during focus groups. It appeared that the most important components of the intervention to change sitting behaviour were the educational seminar and the provision of the height-adjustable workstations. The seminar was considered a strong influence in using the workstation and shows the importance of providing some education alongside the provision of heightadjustable workstations.
Very few barriers were reported by the intervention participants in adhering to the intervention. The ones that were reported included a lack of space on the height adjustable desk platform that sits on top of an existing desk when raised to the standing position. However, participants often found ways to work around this during the intervention. The seminar session at the start of the intervention encouraged participants to identify other strategies in addition to using the desk to break up their sitting time.
However, standing in meetings was considered difficult because of the wider predominant work culture of sitting and feeling self-conscious in the presence of senior staff. This is consistent with work by Mansfield et al. [19] and suggests that wider social behaviour change strategies are needed to make standing in meetings acceptable and the norm.
A large majority of the intervention participants reported very positively on their interaction with research staff, and especially for the 'progress chats' (coaching) offered.
These were reported as being helpful for motivation and planning, and appear to support the development of the processes in the COM-B framework, and in particular motivation and capability. Of note is that being part of the trial seemed to have positive consequences for just under half of the control group participants. These controls felt that they had made changes to their sitting behaviour, physical activity and nutrition.
In conclusion, the SMArT Work programme was successful in reducing sitting time for desk-based employees [8], and this process evaluation has provided valuable information on elements of the intervention and study that appear to have facilitated such behaviour change. These include the educational leaflet and seminar, the height-adjustable workstation, and behavioural feedback and interactions with research staff.

Strengths and Limitations of Process Evaluation
The main strengths of this process evaluation were the multiple methods used and two time points assessed through the questionnaires. A comprehensive set of indicators was assessed to judge context, implementation and impact of the intervention and RCT.
Limitations included the willingness of participants to respond fully to open-ended questions in the questionnaires. However, even though less than half the intervention participants took part in the focus groups, 84% of the clusters were represented. Not everyone completed the process evaluation questionnaires. Individuals taking part in the process evaluation could be biased. Taking part in the study did appear to influence the behaviour of the control group participants.

Consent for publication
Not applicable

Availability of data and material
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.       Table 7. Responses concerning the Diary and goal setting 6 months f/% Recording sitting/standing time in diary In the past 6 month(s) have you used the diary to keep a record of your sitting and/or standing time?
Yes 5/9% Used to but not anymore 9/16% No 42/75% How often do/did you use the diary the record the time you spend sitting and/or standing? (only answered by those answering yes or used to in previous question) Every day 5/31% Few times a week 6/38% Once a week 1/6% Infrequently 4/25% How useful is keeping a written record of your daily sitting and/or standing in helping you change your behaviour? (only answered by those answering yes or used to in previous question) 5=Extremely useful 1/7% 4 6/40% 3 4/27% 2 1/7% 1 = Not at all useful 3/20% Goal setting element in diary In the past 6 month(s) have you used the goal setting element in the diary? Yes 4/7% Used to but not anymore 19/33% No 35/60% How often do/did you use the goal setting element in the diary?
(only answered by those answering yes or used to in previous question) Every week 8/38% Every couple of weeks 8/38% Once a month or less 5/24% How useful is/was the goal setting in encouraging you to reduce your sitting time? (only answered by those answering yes or used to in previous question) 5=Extremely useful 2/10% 4 8/38% 3 6/29% 2 3/14% 1 = Not at all useful 2/10%