Validation of an Employee Remote Workplace Stress Measure Embedded Within a Novel Employee Wellbeing App

Overview

This study aims to validate an adapted workplace stress scale (American Institute of Stress) for employees working remotely during the COVID-19 pandemic. Our funding company (BPAi) is a global provider of technology enabled business process services. They offer a range of innovative services and bespoke software solutions, to improve the performance of client networks by increasing revenue, reducing cost, enhancing customer satisfaction. Clients have provided feedback that they require a way to manage, monitor and support their employees' wellbeing while working remotely. The MyBuddy web-based app is in development to meet this need. To best support employees, having an effective measure of workplace stress while working remotely will be a core feature of the app. Thus, this research study was commissioned to validate the measure. A theoretical framework has been developed based on recent models. The measure was adapted based on this theoretical framework (see detailed description). The hypothesis is that our measure will be validated using participants from BPAi's clients (technology/automotive corporations). The investigators hope this study will help to contribute to the evidence base and provide a useful tool in assessing remote workplace stress. Participants will be employees of BPAi's clients taking part in the piloting of the novel employee wellbeing app. Participants will be invited to take part in the research when they first log into the app. Participants will be provided with an information sheet and a consent form. Once they have provided informed consent, they will be prompted to complete the embedded questionnaires (including the once to be validated). After one month, they will be prompted to complete the questionnaires again. Once the data has been collected, their access to the app will end, and the data will be analysed. Once the data has been analysed, the measure will either be validated, or require adaptations and re-validation with a new sample. The research will be written up into a manuscript and submitted to peer-reviewed journals for publication.

Rationale To the investigator's knowledge, there is no existing psychometric scale to measure employee wellbeing and work-related stress while working remotely. As working remotely is likely to be a core feature for workers for the foreseeable future, either fully or in hybrid form, we believe this is a fruitful area for study. Additionally, there have been very few studies published regarding employee wellbeing while working from home, largely because the pandemic is a new phenomenon. Development and validation of a psychometric measure of employee wellbeing while working remotely is an important step in understanding the impact of remote working on workplace stress and wellbeing. Theoretical Framework The theoretical framework used in this study integrates aspects of the two aforementioned models based on our conceptualisation of remote working and its impact on employee wellbeing. The investigator hypothesises that employee wellbeing is influenced by four domains: (1) environmental; (2) support; (3) social; and (4) boundaries. The environmental domain contains factors such as having an inadequate work environment (e.g. unsuitable chairs, diskspace, interruptions from family, pets, deliveries, etc.). The support domain contains factors such as the ability to access support through one's organization (e.g. supervisors, HR, occupational health, etc.) and one's team of colleagues. The social domain contains factors such as the ability to maintain social relationships with colleagues, and the ability to have informal contact like one would in an office (e.g. chatting while making a drink). The boundaries domain contains factors such as ability to work contracted hours, ability to take breaks, ability to 'switch off' and maintain distance between home and work lives. The investigator further hypothesises that the influence of these domains is moderated by personal factors such as mood, resilience, mental health status, physical health status, etc. For example, if an employee already has an established back condition causing pain, an inadequate workspace lacking a supportive chair will worsen this condition, likely causing some distress, affecting employee wellbeing. Combine this with an inability to contact one's supervisor or HR department to correct the issues, distress is likely to be heightened, causing further decline in employee wellbeing. Quality Assurance and Protocol Compliance The research is designed to be compliant with this protocol. Protocol compliance will be assessed and monitored throughout the study by the lead researcher. Any deviations from the protocol will be recorded and evaluated. The research is being guided by the American Institute of Stress, although they are not the sponsors of this study. This also helps with quality assurance. The principal researcher and author is a psychology graduate with a Master's degree in Psychology (studying for a doctorate in clinical psychology from September 2021), working on a freelance basis. Measures and Scoring: Work and Wellbeing Survey (UWES-9). The UWES-9 is a well-established scale to measure workplace wellbeing in employees. It is a shorter version of the UWES-17. The scale contains nine items comprising three subscales: vigour, dedication, and absorption. The scale uses a seven-point Likert-type scale ranging from "never (0)" to "always (every day; 6)". Scores are obtained by adding up the scores for each item, and then divided by three (subscales) or nine (total score). The maximum score is six and higher scores indicate higher work-related wellbeing. The measure is free to use in research, without permission. This measure is being used to validate our adapted measure. The World Health Organization Five Wellbeing Index (WHO-5). The WHO-5 is a widely used, non-invasive scale validated as a screening tool for depression and as an outcome measure in research. The scale contains five items relating to the previous two weeks. Responses are recorded using a six-point Likert-type scale ranging from "at no time (0)" to "all of the time (5)". The total score is obtained by adding up the scores from each item, with a maximum score of 25. A percentage score can also be gained by multiplying the total score by four. Higher scores indicate higher levels of wellbeing. The measure is free to use in research, without permission. This measure is being used to validate our adapted measure. The Workplace Stress Scale (WSS). The WSS is an eight-item scale validated to measure workplace stress. The scale uses a five-point Likert-type scale ranging from "never" to "always". Three of the eight items are reverse scored. The total score is obtained by adding up the scores from each item with a maximum score of 40. The authors produced score categories as follows: a score of 0-15 indicates the participant is "relatively calm", 16-20 indicates "fairly low stress", 21-25 indicates "moderate stress", 26-30 indicates severe stress, and 31-40 indicates "potentially dangerous levels of stress". This measure will not be used in its original form within this study but will be used in its adapted form (see below). Remote Workplace Stress Scale (R-WSS; measure to be validated). The R-WSS is an adapted version of the Workplace Stress Scale (WSS) to include important domains related to remote working. The measure comprises 11 items relating to the previous week. One item was changed from "conditions at work are unpleasant or sometimes even unsafe" to "conditions for remote working are unsuitable (e.g. uncomfortable workspace noisy environment, etc.)". Three items were added: "I feel isolated from my work colleagues", "I worked my contracted hours", and "I found it difficult to schedule and take breaks while working remotely". The measure uses the same five-point Likert-type scale ranging from "never" to "always", with four items reverse scored. A total score is obtained by adding the scores from each item, with a maximum score of 55. The score categories are as follows: 0-20 indicates the participant is "relatively calm", 21-30 indicates "fairly low stress", 31-38 indicates "moderate stress", 39-48 indicates severe stress, and 49-55 indicates "extreme stress". The American Institute of Stress has approved this adapted version of the WSS for this research study. Standard Operating Procedures Sampling and Recruitment The R-WSS contains 11 items, meaning that a minimum of 220 participants will be required. However, the aim is at least 300 participants, to account for possible attrition and missing data. This is consistent with previous suggestions that a minimum of 200 participants should be sought for a confirmatory factor analysis. The funding company (BPAi) will invite up to three corporate clients with large numbers of employees working remotely to take part in the pilot of the app. Upon agreement from the corporate clients, their employees will be invited to register with the app and will be presented with the information sheet and consent form to take part in this validation study (see 'ethical and regulatory considerations' section below and appendix 1 for more information). Selection of the corporate clients should not impact the results of the study because corporate employees are the target population for both the remote workplace stress measure and the MyBuddy app. Data Collection When participants log into the app for the first time, they will be presented with an information sheet and a consent form, to take part in the research (see below section and subsection 'Sample and recruitment: Consent'). Once the participants have consented to take part, they will be prompted to complete the baseline measures online through the app. The questionnaire responses will be extracted from the app anonymously (i.e. the participant's responses will not be linked to any identifiable information). Anonymous demographic data will also be collected (e.g. gender, age, highest level of education completed), along with app use data (i.e. how often the app was used within the month). The questionnaire responses will then be held securely by the researcher in a password protected Excel spreadsheet ready to be prepared for data analysis. Participants will then be prompted to complete the measures one month after their initial login; they will be able to complete the measures as often as they wish during the app pilot phase. Data Analysis The statistical software package IBM's SPSS will be used to analyse the data from the study. Descriptive statistics will be produced to indicate means and standard deviations for the measures and demographic data. Internal consistency will be measured used Cronbach's alpha coefficient with a minimum level of 0.7. A confirmatory factor analysis and correlation analysis will be conducted to confirm the factor structure and help to validate the measure. If the data analysis suggests that any items or factors are inconsistent, the measure will be adapted, and re-tested with another sample. If the measure is validated, analyses will be conducted with a new sample to confirm construct validity. Assessment and Management of Risk The MyBuddy app is designed to complement existing employee wellbeing risk assessment and management procedures within an organisation. It is designed as a tool for employees to assess, monitor, and track their wellbeing, as well as facilitate access to their organisation's wellbeing support. The app is designed as an assessment tool, rather than a wellbeing intervention. This means it signposts employees to existing wellbeing support in their organisation and notifies employers of their employees' wellbeing scores. As such, we do not anticipate high levels of risk. However, the study asks participants to rate their current wellbeing levels and will provide them with a score and some explanation of their score. There is a risk that people that score highly on the stress measure may become concerned or distressed regarding their wellbeing. In this instance, the app will notify their manager, who is ultimately responsible for managing these risks. Plan for Missing Data Any instances where data is missing or where participants do not complete all measures at baseline and follow-up, will not be included in the study. This will be reported clearly in the form of an attrition rate in the manuscript. Data protection and patient confidentiality As described above, participant data will be held securely in the app, and anonymised questionnaire responses, along with some non-identifiable demographic information, will be extracted from the app for data analysis. This data will be held securely by the lead researcher on a password protected Excel spreadsheet. Participants will be assigned a random participant ID code when they first login, which will be available for them to view again within the app. Their questionnaire responses and non-identifiable demographic information will be linked to their unique participant ID code when the data is extracted from the app. If any of the participants wish to withdraw, they must contact the lead researcher and provide their unique participant code to identify their information for removal from the dataset.