Mapping, co-creating, and trialling workplace physical activity interventions: a route map for developing feasible workplace physical activity interventions for employee mental health

Tumblety, Craig (2026) Mapping, co-creating, and trialling workplace physical activity interventions: a route map for developing feasible workplace physical activity interventions for employee mental health. PhD thesis, University of Glasgow.

Abstract

Poor mental health, including depression and anxiety, is a world-leading cause of illness and disability worldwide. Higher levels of physical activity are strongly associated with better mental health, while PA interventions have been demonstrated to be as effective in treating and preventing severe depression as medication. Workplaces have been identified as important settings for potential interventions; however contextual factors play a role in the outcomes of each intervention. This PhD aimed to review the literature on workplace physical activity interventions for mental health, evaluate the mental health and physical activity behaviours in a small-to-medium Scottish enterprise (ACS Clothing Ltd) through a mapping process, then use the findings from both the review and mapping to codevelop an intervention for the whole workplace, and finally test the feasibility of this tailored intervention.
Study one (Chapter 3) reports a three-year workplace mapping process designed to characterise mental health, wellbeing, productivity, and physical activity patterns within a small‑to‑medium enterprise and to inform the development of a workplace physical activity intervention. Annual Wellbeing Evaluations (AWEs) were delivered across 2022-2024, combining physical health assessments, seven‑day accelerometery, and validated surveys of wellbeing, depression/anxiety, stress, sleep, productivity, and work engagement. Linear models examined differences across years and job types (manual vs. desk-based), while compositional data analysis assessed how physical activity behaviours contributed to these patterns. Across 181 evaluations, most outcomes remained stable over time, except for a marked worsening of depression and anxiety in the third year, coinciding with organisational restructuring. Manual workers consistently exhibited poorer mental health, sleep, and engagement than desk‑based workers. Physical activity compositions during work and across the full day meaningfully contributed to differences in depressive symptoms and job‑type disparities. The AWEs, originally conceived as observational, evolved into an engagement tool as personalised feedback prompted behaviour change and organisational adjustments.
Study two (Chapter 4), was a systematic review of evidence on the effectiveness of workplace physical activity and exercise interventions for improving employee mental health and wellbeing. Following PRISMA guidelines, searches to June 2024 identified 54 eligible studies (58 interventions) spanning randomised controlled trials, quasi-experimental designs, and pre-post studies. Validated mental health outcomes were extracted and synthesised using effect-direction plots due to methodological heterogeneity. Risk of bias was assessed using ROB-2, ROBINS-I, and NIH tools, with certainty of evidence evaluated through GRADE. Certainty of evidence was low for stress, mood/affect, and general wellbeing, and very low for anxiety, depression, and quality of life, primarily due to heterogeneity, methodological weaknesses, and inconsistent reporting. Mind–body activities such as yoga and Tai Chi, group-based delivery, and higher intervention dose were more frequently associated with favourable effects in some outcomes.
In study 3 (Chapter 5), potential interventions were developed through a cocreation process. Co-creation involves engaging end-users and stakeholders throughout the research process increasing the likelihood of intervention success and of sustaining an effect beyond the research project. This approach was delivered through workshops with managers and employees. These workshops involved feeding back the results from the review and mapping exercise and facilitating participants to create their own workable intervention solution. Participants’ proposals were recorded using a participatory drawing exercise, where participants drew their ideas before discussing the practicalities, and discussion points were recorded via researcher note taking. Findings from these workshops were delivered to a project steering group consisting of a representative sub-set of employees and managers who, along with the research
team, assisted with the final development of the intervention.
Study 4 (Chapter 6) tested two co‑developed physical activity interventions (job rotation and walking meetings) designed to support employee mental wellbeing. The mixed‑methods feasibility trial applied Framework Thematic Analysis of semi-structured interviews to explore feasibility, acceptability, and perceived impact, while linear mixed‑effects models with compositional analysis of physical activity behaviours were used to assess potential effectiveness. Both interventions were acceptable within the respective target work groups, with 80% adherence for job rotation and 100% for walking meetings. For job rotation participant reports indicated perceived benefits to workforce flexibility and self-efficacy, while acceptability was shaped by capacity for rotation, resistance to change, and organisational agility. Acceptability of the walking meetings was influenced by factors such as group size, physical and social environment, and leadership support. Job rotation had no detectable effect on wellbeing in this sample (p = 0.703), while walking meetings showed potential benefits (p = 0.019). The findings suggest both approaches are feasible but require further adaptation to workplace context to maximise uptake and impact.
Beyond generating new empirical insights, this PhD contributes a methodological model for integrating mapping, co‑creation, and feasibility testing within live workplace systems, offering a replicable approach for future workplace health research.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: This research was supported by the PhD studentship of Medical Research Scotland (PHD-50209-2020).
• Subjects: R Medicine > RC Internal medicine > RC1200 Sports Medicine
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Cardiovascular & Metabolic Health
• Funder's Name: Medical Research Scotland (MEDRESSC)
• Supervisor's Name: Ryde, Dr. Gemma, Demou, Dr. Evangelia and Kelly, Dr. Paul
• Date of Award: 2026
• Depositing User: Theses Team
• Unique ID: glathesis:2026-86149
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 23 Jul 2026 10:27
• Last Modified: 23 Jul 2026 10:32
• URI: https://theses.gla.ac.uk/id/eprint/86149

Related URLs

• Article DOI
• Article DOI
• Article DOI