1. Introduction
Research exploring the impact of employee wellbeing in the workplace has advanced substantially over the last two decades. There is growing evidence linking workplace performance and productivity with mental and physical health issues among employees [
1]. Research has shown that implementing interventions in the workplace designed to improve employee health and wellbeing have frequently resulted in health status improvements and enhanced work performance [
2,
3]. While such interventions vary in duration, composition, and intensity, they are all designed to promote an increase in healthy lifestyle behaviours, including stress management, improved nutrition, and reducing sedentary behaviour (defined as any waking behaviour that occurs in a sitting or lying position and results in an energy expenditure of 1.5 METs or lower) [
4,
5].
As society has progressed and technology has advanced, there has been a decline in demand for manual labour-based jobs, with these being replaced by more office or sedentary-based occupations. In a sample of UK full-time office workers, 65% of time at work was sedentary and sitting at work accounted for 63% of total daily sitting time [
6]. With such a large proportion of an adult’s day being spent at work, the importance of reducing sedentary behaviour in the workplace is highlighted. Indeed, sedentary behaviour has a negative impact on health, being associated with an increased risk of diabetes, obesity, cardiovascular problems, as well as mental health issues and some types of cancer [
7]. Research suggests metabolic health is compromised in those who spend the majority of their days engaged in sedentary behaviour, even if they are engaging in moderate/vigorous physical activity [
8]. In other words, sedentary behaviour is an independent risk factor for obesity and chronic disease. Therefore, interventions designed to target sedentary behaviour in the workplace are specifically needed.
Research suggests that breaking up prolonged sedentary time can have positive health outcomes for an individual. Short-term laboratory-based experiments have reported that when sitting is interrupted every 30 min by brief activity breaks (i.e., two minutes of treadmill walking or light resistance activity), postprandial glucose and insulin levels are significantly reduced [
9,
10]. Moreover, research has demonstrated a significant relationship between health outcomes (i.e., triglycerides, glucose, waist circumference) and the total number of breaks from sitting, independent of total sedentary time [
11]. Research has also demonstrated that implementing interventions designed to reduce sedentary behaviour in the workplace can not only reduce sitting time, but can also indirectly lead to improvements in mental wellbeing (fatigue, tension, depression, and anxiety) [
12]. Therefore, when examining interventions, it is important to consider a range of factors that may also be impacted as indirect effects are common.
Recent reviews of interventions to reduce sedentary behaviour in the workplace include strategies such as educational/behavioural, environmental, and multi-component, including the use of computer-based or mobile health technologies [
13,
14,
15]. All of these strategies have shown some success at improving sedentary behaviour and physical and mental wellbeing. Indeed, research has documented that computer-based, mobile, and wearable technology-based interventions targeting workplace sitting, compared to non-work-based sitting, were more effective at medium-term (3–6 months) follow-up [
14]. A number of limitations have been documented in these published reviews including limited description of the intervention, in addition to a lack of qualitative measures to explore participant experiences. Cost of intervention delivery is also a limiting factor for large scale implementation and impact.
Digital-based interventions have the potential to reach large populations at a low cost and offer potential to tailor interventions to the needs of individuals or specific groups of office workers. A recent systematic review of occupational-based digital health interventions displayed moderate evidence for interventions in improving employee health (i.e., psychological wellbeing, sleep, physical activity, and sedentary behaviour) [
16]. Nevertheless, a large decline in technology use and engagement is often observed with digital interventions [
15]. Yet, there is evidence that a nudge or prompt driven approach can yield high engagement in intervention components as well as reduce sitting time and increase light activity at work [
17,
18].
The aim of this study was to pilot test a nudge-based digital intervention (Welbot) to reduce sedentary behaviour and improve physical and mental wellbeing in the workplace. Unlike many other digital interventions, Welbot has been developed with academic and professional input and in collaboration with industry specialists. It is also a more holistic approach as it aims to target many of the different areas that contribute to one’s overall wellbeing at work (e.g., reducing sedentary behaviour, increasing physical activity, having regular screen breaks, improving mental wellbeing, promoting positive work environments, ensuring that you are staying hydrated, and eating healthily, etc.).
4. Discussion
This study evaluated the effectiveness of a new and innovative digital intervention—“Welbot”—that aims to improve physical and mental wellbeing at work. Following the intervention, participants self-reported significantly less time sitting and more time standing and objectively recorded more steps at the week-1 follow-up (compared to baseline) and less prolonged sedentary behaviour (>30 min). A number of outcomes relating to objective measurement of sedentary behaviour (sedentary time, standing time, bouts of sedentary behaviour) showed a trend towards improvement at the week-1 follow-up, however these improvements were not maintained by the week-2 follow-up and hence were not coherent with subjective measures. These trends for improvements in sedentary outcomes are comparable with previous research [
33] reporting that computer-delivered prompt interventions are an effective way to not only reduce sedentary behaviour, but also promote physical activity in desk-bound employees. The finding that improvements in sedentary outcomes were not maintained by week 2 follow-up is also consistent with a recent systematic review of mobile health interventions to promote physical activity and reduce sedentary behaviour in the workplace, which emphasised the need to explore the reasons for decline in engagement with interventions [
15]. Future research would benefit from including a larger sample of participants, particularly for the objective ActivPAL component. Future research with the Welbot intervention should explore how participants use the programme over time and which components are most effective in promoting sedentary behaviour change to enhance the effect of the intervention.
Following the intervention, participants reported significant improvements in depression, anxiety, and stress; post-intervention all participants were in the “normal” to “mild” range. Such results are congruent with previous research, which have exhibited digital interventions as an effective avenue for improving employees’ mental wellbeing at work [
16], particularly for employees’ who are experiencing less complex psychological difficulties. This is crucial given the significant levels of psychological difficulties in the current working population, with occupational-related stress, anxiety, and/or depression affecting 526,000 employees living in Britain in 2016/17 [
34]. Perhaps Welbot functions to improve employees’ depression, anxiety, and stress via its emphasis on mindfulness. An abundance of previous literature has highlighted the benefits of mindfulness in the workplace, with improvements in depression, fatigue, stress, anxiety, burnout, job performance, and work–life equilibrium [
35,
36,
37,
38]. Contrastingly, such improvements may also be resultant from its focus on decreasing sedentary behaviour. Indeed, research has exhibited that prolonged occupational sitting is associated with higher levels of psychological distress [
39]. Hence, a nudge-based digital intervention that aims to reduce this adverse behaviour may secondarily also improve mental wellbeing. Alternatively, these improvements could merely be due to the simple act of taking a break from work tasks. Such breaks have been shown to reduce levels of fatigue and improve employees’ sense of vitality [
40]. Regardless of what specific elements are driving this change, it appears that Welbot, as a digital intervention, can enhance some areas of psychological functioning (i.e., depression, anxiety, and stress). Contrastingly, results demonstrated that levels of mental wellbeing failed to significantly change following the intervention. Hence, although it is positive that the digital intervention did not impair functioning, this limits our ability to make any firm conclusions about Welbot’s overall impact on mental wellbeing outcomes. Therefore, further research utilising larger sample sizes and a control group is needed.
Results exhibited that levels of procrastination and work engagement failed to significantly alter post-intervention. This is a novel finding in the emerging field of nudge-based digital interventions and although it was somewhat unexpected, it indicates that the introduction of Welbot to the workplace did not impair occupational functioning. In particular, the lack of change with regards to employees’ work engagement is of importance. Despite Welbot interrupting employees every 40-min and asking them to participate in a nudge, this “time-out” from work tasks did not correspond to a detriment in work engagement. Previous research has shown that digital interventions are in fact capable of enhancing job performance and employee health [
16,
41]. Therefore, although it is positive that Welbot did not impair employees’ functioning in these areas, prospective research should now focus on investigating how the digital intervention can be further refined to optimise its effect on occupational functioning.
Several positive attributes of the digital intervention were identified by the participants, including a perception of positive behaviour change as a result of using Welbot. These perceptions were partially supported by the quantitative findings where self-reported sitting time during the working day significantly decreased over the two weeks. Participants also reported that their awareness of unhealthy behaviours at work increased as a result of using the digital intervention, which has been echoed in a recent systematic review exploring factors affecting patient and public engagement with digital interventions [
42]. In relation to the negative attributes of the digital intervention, feeling self-conscious whilst carrying out the activities at their desk was reported in previous studies that specifically used health interventions in the workplace. Indeed, research has found that not having a private space within the workplace to access a digital mental health intervention and feeling exposed using the intervention whilst sat at their desk were barriers to engagement [
43]. Whilst participants reported on the positive and negative attributes of Welbot, they did offer useful suggestions for improvements to the digital health intervention. Several participants advocated for Welbot to allow individualised preferences in relation to the frequency and timing of nudges and the option to sync these to online calendars. Personalised tailoring of information within digital health interventions has been reported by users in previous research as an important facilitator of continued engagement [
42,
43,
44]. Indeed, more recent versions of Welbot than what was used in the current study offer this function. Nevertheless, the qualitative component of this pilot study utilised a small sample size (
n = 6), which may not necessarily be representative of the sample as a whole. Therefore, further qualitative research into people’s experiences of using Welbot are needed in order to make it as practical, accessible, and user-friendly as possible.
The main limitation of this pilot study was that it utilised a small sample size (n = 41), which was even further compromised in the ActivPAL (n = 6) and qualitative (n = 6) components. This largely limits the interpretation of findings and hence, caution is needed. Therefore, future larger studies would benefit from conducting a priori power analysis to ensure an adequate number of participants to detect an effect and to explore the potential efficacy of Welbot in reducing sedentary behaviour and improving wellbeing in the workplace. Other limitations to this study relate to the participant demographic characteristics and study design. The majority of participants were white, female, full time workers. Therefore, prospective studies utilising larger sample sizes (particularly for the objective ActivPAL component) and more diverse participant demographical characteristics are needed. The study used a single-arm repeated measure design and there was no control group. This limits our ability to attribute any improvements in wellbeing to the digital intervention as it is plausible that simply delivering regular notifications to users without any associated task may perform just as well. Therefore, further research adopting a two arm, repeated measures design with a control group and participants randomly allocated to each arm of the study is essential. A further limitation is the short intervention period (i.e., 2 weeks), hence prospective research would benefit from using longer intervention periods. It would also be helpful to have a longer follow up period (e.g., 6 months) to ascertain whether any benefits of using the digital intervention are maintained over time.
A further limitation may be the use of the ActivPAL research device, which may have encouraged participants to change their behaviour. The possibility of aligning Welbot with personal activity trackers could be explored as a means of capturing and integrating such feedback. Finally, it remains unclear which elements of this multi-component intervention are most important for maximising health and wellbeing gains. An in-depth analysis of user engagement patterns alongside the use of standardised outcome measures and improved reporting of “active” components of Welbot would enhance the future evaluation of this digital intervention.