Development and validation of a multi-lingual online questionnaire for surveying the COVID-19 prevention and control measures used in global workplaces

Relatively few infectious individuals (~ 10%) are responsible for most (~ 80%) of local COVID-19 transmission [1]. Environmental factors likely contribute to COVID-19 clusters, or ‘superspreading’ events. Crowded indoor settings, poor ventilation, long contact exposures, and/or frequent interactions with the general public create high or medium-risk working environments for COVID-19 [2, 3]. Workplaces such as hospitals, care homes, schools, building sites, food-processing plants, and high-volume retail outlets possess such environmental features that facilitate explosive COVID-19 clusters [4]. Comprehensive occupational COVID-19 infection prevention and control (IPC) measures can prevent these events, ensuring the protection of workers, their families, and surrounding communities. Effective measures include swift and thorough contact tracing and case isolation, effective personal protective equipment (PPE), testing, worker bubbles, and improved ventilation and air quality [5‐8]. Relying solely on a single control measure like masking is not enough to prevent COVID-19 clusters in high-risk workplace environments [6, 7].

Many countries have been active in creating COVID-19 occupational safety and health (OSH) policies and guidance involving these measures [9]. Nevertheless, data indicate that workers remain at disproportionate risk to COVID-19 in settings where vaccination rates lag and cases surge [10‐12]. The United States’ Occupational Safety and Health Administration (OSHA) issued its first mandatory workplace safety rules aimed at protecting healthcare workers from COVID-19 in June 2021 [13]. Yet these rules do not protect workers in other essential industries who face rising case numbers, stalled vaccinations [14], loosened mask requirements [15], and push back against occupational COVID-19 safety regulations from the national business community [16]. In Europe, as businesses reopen and travel restrictions ease [17], countries continue to experience workplace outbreaks [10, 11]. Case numbers amongst oil sands workers in Canada grew so high in May 2021 that legislators had to declare a state of local emergency [18]; and in India, an unprecedented surge in cases March–April 2021 strained IT firms who struggled to protect the health of employees while maintaining business continuity [19]. These events highlight the ongoing need for disease prevention and control guidance and resources in high-risk occupational settings, particularly as workers have expressed anxiety and concern about becoming infected with COVID-19 and endangering co-workers, customers, and family [20].

Occupational hygienists have called for the production of appropriate tools to effectively transfer safe IPC knowledge and practice to employers and workers [21]. At the international-level, the International Labour Organization (ILO) provides a database with updates on rapidly changing occupational policy responses to the COVID-19 crisis [22]. At the individual-level, to our knowledge, no validated tool exists for surveying workers’ awareness of and satisfaction with the implementation of these policies. As research reveals gaps in policymakers’ and workers’ perceptions of acceptable COVID-19 safety at work, it is important to constructively engage with workers in order to fast track acceptable and feasible OSH solutions [23]. Understanding protective measures in place can, by helping to untangle potential causes of outbreaks, aid governments trying to contain or prevent COVID-19 surges around the world. Identifying gaps in workplace safety response can help to direct the distribution of disease prevention and control resources and raise awareness of effective COVID-19 safety measures.

This study aimed to develop, pilot, and validate an instrument for identifying occupational COVID-19 IPC measures available to workers from diverse geographic and occupational settings. By allowing researchers to measure workers’ perceptions of the level of COVID-19 protections in place, the survey instrument will have important implications for worker protection from COVID-19 and future infectious disease outbreaks.

This cross-sectional, pilot study was conducted in three phases: 1) survey development, 2) pilot testing, and 3) psychometric property evaluation. Specific methodological steps taken during each phase are outlined in Fig. 1. Psychometric properties assessed as part of the study are defined in Fig. 2.

This survey was developed to measure the level of occupational COVID-19 protection available to workers from diverse geographic and occupational settings. Following expert consultation and pre-testing, the survey showed strong face and content validity. Psychometric property evaluation revealed satisfactory levels of internal consistency reliability, and criterion and construct validity. The survey provides a framework for evaluating the level of protection a worker perceives they have from COVID-19 in the workplace. It holds the potential to be of particular use to decision-makers in the distribution of infectious disease prevention and control resources, and to guide recommendations for keeping the global workforce safe and healthy at the onset of emerging infectious disease outbreaks.

Exploratory factor analysis revealed nine key IPC domains that were confirmed by a high percentage of common variance explained. Cronbach’s alpha coefficients showed that items in each domain measured the appropriate construct. By providing a more precise model than originally hypothesized by the researchers, EFA provided insight into how occupational IPC guidelines and programs can be structured. For example, limiting worker movements and contacts emerged as a separate construct from physical distancing between workers. Testing and isolation measures emerged as separate from the cost of testing and isolation incurred. Survey items on universal testing, contact tracing, vaccinations, and signage did not fit into the 9-domain model. However – because research shows the ability of universal testing and timely contact tracing to prevent COVID-19 outbreaks [6]; that the rollout of vaccines plays a crucial part in protecting workers [40]; and that COVID-19 training should be reinforced by the use of signage placed in strategic locations [41] – each of these items were considered important for content validity and left in the survey.

By design, this study focused on survey development and validation. Nevertheless, survey results and hypothesis testing applied in the context of psychometric property evaluation revealed important considerations for occupational health. In terms of country-level differences, Canadian workers identified a higher number of implemented COVID-19 IPC precautions than Irish respondents despite representing a lower percentage of health care workers (42% of Canadian respondents worked in health care or social assistance vs. 60% of Irish respondents). This may be explained, in part, by an older sample of Canadian respondents. Research shows that older and more experienced workers report greater comfort with IPC skills [42], corroborating our finding that older workers were twice as likely to feel protected as younger workers. However, with both countries reporting similar vaccination rates and the implementation of return-to-work COVID-19 safety policies [9, 14], it will be important to examine survey data for specific variations in workplace IPC response. Though the total number of protective measures did not vary significantly between UK and Irish respondents, UK respondents reported feeling significantly less protected at work. One explanation relates to measures’ adherence. For example, over one third of UK respondents reported that colleagues ‘rarely’ or ‘never’ wore masks correctly, compared to just 10% of Irish respondents. This demonstrates how individuals may perceive and prioritize safety differently, particularly when COVID-19 safety precautions have the potential to create physical or social discomfort [9, 43], add to the workload [42], or result in a loss of income (32% of survey respondents reported not receiving paid time off to be tested for COVID-19). Consequently, voluntary adherence to instituted IPC measures will require clear and systematic communication on the necessity for and strategy in place behind them [44]. Though relatively few workers responded from developing countries, we chose to report findings from Nigeria and Poland to demonstrate the importance of targeting workers in less-studied, low- and middle-income countries in future survey distribution. Nigerian participants reported some of the lowest numbers of protective measures despite a surge in cases at the time of survey distribution, indicating that factors such as financial resources and public health infrastructure may be contributing to differences in occupational COVID-19 response.

Feeling protected at work varied according to gender, professional role, and occupational sector. In line with reports that women are more likely to feel concerned over their own health and the health of others’ [45], survey results indicated that men felt significantly more protected at work than women despite higher rates of vaccination amongst female workers (77% of women were vaccinated, compared to 66% of men). The COVID-19 pandemic has exacerbated gender inequalities. Despite being more likely to work on the frontline and in essential sectors, women are severely underrepresented in COVID-19 decision-making bodies [46]. We found that twice as many male respondents were in managerial roles as female respondents, and that managers felt more protected than employees after accounting for sector effect. Mediation analysis will be of value in the next phase of survey roll-out to assess the role of various factors on the pathway between gender and feeling protected from COVID-19 at work. Educators and retail workers felt significantly less protected than healthcare workers, with health care workers reporting the highest average protective score and retail workers the lowest. While IPC response has been widely studied in the healthcare setting [6], few studies have examined COVID-19 safety amongst retail, transport, and other essential workers. Our finding that 55% (197/359) of non-health care respondents felt unprotected at work, coupled with reports of COVID-19 outbreaks in international bars, building sites, food-processing plants, hotels, shops, and transportation settings [4], underline the value in studying the COVID-19 safety resources available to geographically and occupationally diverse groups of workers. By successfully capturing gender and occupational inequalities, the survey is an effective tool for tailoring disease prevention and control response to those most in need of protection.

Similar to findings from Ireland and Canada [47, 48], 10% of survey respondents reported an unwillingness to be vaccinated for COVID-19. This group did not feel significantly less protected at work than vaccinated participants. Misinformation is likely to be contributing to a sense of false security. Research shows that vaccine-hesitant individuals consume significantly less information about COVID-19 from newspapers, television, radio, and government agencies, reporting higher levels of trust in social media [49]. Recommendations for addressing vaccine hesitancy include collaborations between health agencies and multiple societal stakeholders to avoid the feeling that public health measures are for the benefit of government authorities [50]; and public health messaging that is targeted, clear, direct, repeated, positively orientated, and emphasises the personal benefits of vaccination against COVID-19 [49]. Similar recommendations can be applied in the workplace to ensure adequate awareness of and adherence to occupational safety precautions that are important in addition to vaccination campaigns. Of 456 vaccinated survey respondents, 38% still did not feel adequately protected from COVID-19 at work.

Pre and pilot testing resulted in additional lessons learned regarding survey distribution. The multi-modal recruitment strategy used resulted in greater uptake than reported in other, similar studies (McRobert et al. received 387 complete responses from a multidisciplinary international sample of clinicians using social media posts and email distribution over a three-month period [27]). Nevertheless, we found that posting in relevant occupational LinkedIn groups garnered little to no response. Posting in occupational Facebook groups was slightly more successful, however researchers were required to post from personal profiles creating a level of individual exposure not identified by other studies that used Facebook groups to recruit for health research [51, 52]. The most successful recruitment method involved partnering with relevant organizations who could distribute the survey link via their email network (e.g., EU-OSHA, UCD Centre for Safety and Health at Work, BC General Employees’ Union). However, ensuring partner participation required extensive effort. Of 132 emails sent by the researchers to international trade unions and occupational safety and health organizations, seven eventually agreed to distribute the survey (5%). Reasons for declined participation included precarious political and/or pandemic-related circumstances, overworked staff, and survey fatigue. As a next step, and to ensure as robust and unbiased a study sample as possible, the research team plans to roll out the survey in select target countries using Qualtrics^XM Research Services; to post the survey link to personal social media profiles and encourage snowballing with the aid of a promotional video (Additional file 7); and to continue working with external partners to solicit participation via email recruitment.

A multi-lingual online survey to measure the level of occupational COVID-19 protection available to workers in diverse geographic and occupational settings was developed, piloted, and validated. The survey showed strong face and content validity in English, as well as internal consistency, criterion validity, and construct validity. Translated versions of the survey into Polish, Spanish, French, Hindi, and Chinese demonstrated potential for adaptability across working cultures and countries. The survey instrument is suitable for measuring occupational COVID-19 IPC response in varied global and workplace settings, providing valuable insight for COVID-19 occupational safety and health guidelines and future infectious disease preparedness. As a follow-up to this research, survey data will be used to map and compare occupational COVID-19 IPC measures used in international workplace settings.