Background:
Psychosocial work environments influence employee well-being. There is a need for an evaluation of organisational-level interventions to modify psychosocial working conditions and hence employee well-being.

Objective:
To test the acceptability of the trial and the intervention, the feasibility of recruitment and adherence to and likely effectiveness of the intervention within separate clusters of an organisation.

Design:
Mixed methods: pilot cluster randomised controlled trial and qualitative study (in-depth interviews, focus group and observation).

Participants:
Employees and managers of a NHS trust. Inclusion criteria were the availability of sickness absence data and work internet access. Employees on long-term sick leave and short-term contracts and those with a notified pregnancy were excluded.

Intervention:
E-learning program for managers based on management standards over 10 weeks, guided by a facilitator and accompanied by face-to-face meetings. Three clusters were randomly allocated to receive the guided e-learning intervention; a fourth cluster acted as a control.

Main outcome measures:
Recruitment and participation of employees and managers; acceptability of the intervention and trial; employee subjective well-being using the Warwick–Edinburgh Mental Wellbeing Scale (WEMWBS); and feasibility of collecting sickness absence data.

Results:
In total, 424 employees out of 649 approached were recruited and 41 managers out of 49 were recruited from the three intervention clusters. Of those consenting, 350 [83%, 95% confidence interval (CI) 79% to 86%] employees completed the baseline assessment and 291 (69%, 95% CI 64% to 73%) completed the follow-up questionnaires. Sickness absence data were available from human resources for 393 (93%, 95% CI 90% to 95%) consenting employees. In total, 21 managers adhered to the intervention, completing at least three of the six modules. WEMWBS scores fell slightly in all groups, from 50.4 to 49.0 in the control group and from 51.0 to 49.9 in the intervention group. The overall intervention effect was 0.5 (95% CI –3.2 to 4.2). The fall in WEMWBS score was significantly less among employees whose managers adhered to the intervention than among those employees whose managers did not (–0.7 vs. 1.6, with an adjusted difference of 1.6, 95% CI 0.1 to 3.2). The intervention and trial were acceptable to managers, although our study raises questions about the widely used concept of ‘acceptability’. Managers reported insufficient time to engage with the intervention and lack of senior management ‘buy-in’. It was thought that the intervention needed better integration into organisational processes and practice.

Conclusions:
The mixed-methods approach proved valuable in illuminating reasons for the trial findings, for unpacking processes of implementation and for understanding the influence of study context. We conclude from the results of our pilot study that further mixed-methods research evaluating the intervention and study design is needed. We found that it is feasible to carry out an economic evaluation of the intervention. We plan a further mixed-methods study to re-evaluate the intervention boosted with additional elements to encourage manager engagement and behaviour change in private and public sector organisations with greater organisational commitment.

Study registration:
Current Controlled Trials ISRCTN58661009.

Funding:
This project was funded by the NIHR Public Health Research programme and will be published in full in Public Health Research; Vol. 3, No. 9. See the NIHR Journals Library website for further project information.