Study setting and population

The study is conducted among public primary and upper-secondary schools (n=55 schools) in four municipalities in Sweden. Two municipalities are located in an urban area and two municipalities in a rural area. The municipalities give a good representation of different geographical areas and socioeconomic as well as urban and rural areas.

Eligibility criteria

All personnel employed by the schools are eligible to participate, including teachers, administrators, and support personnel (e.g., reading specialist, teacher’s aide, and paraprofessional). Individuals not employed by the school (e.g., external cleaning and maintenance personnel) will be excluded, as they do not fall under the management of the school.

InterventionsGuideline to be implemented

The object that will be implemented is the Guideline For The Prevention Of Mental Ill-Health At The Workplace [5]. The guideline includes recommendations how employers in cooperation with their personnel can prevent mental ill health within their organization. The guideline includes the following recommendations: (1) workplaces have well-established routines/policies related to organizational and social risk management, (2) employers have knowledge of the relationship between organizational and social risks and mental ill health, and (3) workplaces regularly assess their organizational and social work environment and intervene on identified risk factors. Personnel involvement is strongly emphasised in the guideline. For example, it is recommended to conduct group discussions with personnel to prioritize work environment risks that need changing to prevent mental ill health and to involve personnel in the development of action plans describing changes that need to be made. The guideline was systematically developed in a collaboration between researchers, employer representatives, and occupational health service staff and includes recommendations that are based on the best available evidence in the field (e.g., [33,34,35,36]. The guideline complies fully with the Swedish Work Environment Authority’s organizational and social work environment (AFS 2015:4) provisions. Since 2018, the guideline is disseminated through the Swedish Agency for Work Environment and Expertise (https://sawee.se/). A full description of the recommendations has been published previously [9].

Implementation strategies

The strategies to be evaluated are based on those developed in our previous study [9]. An additional strategy was added, namely an internal facilitator. Refinements were made to the content of the educational meeting and workshops, with more focus on knowledge provision regarding the guideline recommendations and Plan-Do-Study-Act methodology. Moreover, educational material to support the formation of Plan-Do-Study-Act cycles was added in the form of templates to be used by the implementation team. To provide a deeper understanding of the mechanism of change, the original COM-B pathways were reassessed and refined based on the Theoretical Domains Framework (TDF) [37]. The domains of the TDF have previously been successfully mapped onto the COM-B, with excellent agreement [37]. Implementation strategies specified in accordance with Proctor recommendations for specification [38] are described in Table 1. The implementation logic model is depicted in Fig. 2.

Table 1 Specification of the implementation strategiesFig. 2

Implementation logic model

Implementation teams

At each school the headmaster will form an implementation team consisting of 4–5 people depending on the size of the school. The team will include the principal, teacher (union)—representative, occupational health and safety officer, and support staff representative. The members of the team should represent the actual mix of staff who will be involved in the implementation of the guideline. To support the principal with the formation of the team, instructions will be sent by email, including a template to specify team members’ names, roles, and motivation for inclusion. The team will be encouraged to meet regularly. The schools will choose the length and frequency of the meetings.

Educational meeting

At each municipality, implementation teams and school-district representatives (e.g., director of education, HR-specialist) will participate in a 1-day educational meeting conducted by an implementation researcher (LK) and a researcher in occupational health (CB). The educational meeting will be a mixture of lectures, discussions, activities, and group exercises. During the first lecture, the researchers will provide information on mental ill health, the guideline, and how working in accordance with the guideline can prevent mental ill health. Each team will conduct exercises aimed at reflecting on their current adherence to the guideline recommendations, identifying the benefits of adhering to the guideline, setting an implementation goal, and planning for implementation. Instructions related to the exercises will be given through several short lectures given by the implementation researcher and complementary materials, including templates for goal formulation. At the end of the meeting, a lecture will be given on what is needed to succeed with the developed plan. Potential barriers and facilitators are also introduced.

Ongoing training in the form of workshops

To support the implementation teams with their implementation process, five 2.5-h workshops will be held at each municipality by the researchers (LK and CB) over a 12-month period. Each workshop contains a lecture aimed at providing detailed information on guideline recommendations (one recommendation per workshop). The workshops will mix lectures with discussions, activities, and exercises. Each workshop is divided into three modules. During the first module, the teams will present the progress they have made with their PDSA-cycle and adjustments that need to be made to continue (see description below). The second module is aimed at providing detailed information on the guideline recommendations. This module includes lectures combined with discussions and exercises. During the exercises, teams compare how they currently work with the prevention of mental ill health with what is recommended by the guideline. At the end of the module, teams will have identified what needs to be changed. The third module is aimed at providing teams with information and skills regarding how to implement the guideline within their workplace. Lectures will be given on Specific, Measurable, Achievable, Realistic, and Timely (SMART)—goals and how to conduct PDSA cycles. During this module, teams will conduct exercises aimed at formulating goals and making plans for implementation in accordance with the PDSA cycles. During each workshop, implementation teams receive complementary materials, including handouts of the lectures, material, and templates for the exercises.

Plan-Do-Study-Act cycles

The implementation teams will start with their first PDSA cycle during the first workshop. In line with the PDSA structure, the teams will compare their current situation with the guideline recommendations, formulate a goal based on the recommendations, identify barriers to achieving the goal, identify what changes are needed to achieve the goal, and develop an implementation plan that describes how the changes are to be implemented and who is responsible for the changes. Between workshops, the teams are encouraged to implement the changes and measure the effects, compare the results with the formulated goal, and, if necessary, adjust the changes. To support the teams with their PDSA cycles, lectures will be given during the workshops explaining how to conduct PDSA cycles; moreover, teams will during each workshop receive PDSA planning templates. The template is based on the planning tool developed by the National Implementation Research Network and translated to Swedish for the purpose of the study [26]. At the end of each workshop, the teams will hand in a copy of their template to the research team. Between workshops, teams will be reminded to complete their template encouraging them to reflect over their implementation progress.

Internal facilitator

At each municipality a representative of the school district will be selected in collaboration with the research team to act as internal facilitator. The internal facilitator should be familiar with the school-level organizational structures and procedures. In this study, the functions of the facilitator can include supporting implementation teams with identifying changes to be made, helping to prioritize, supporting with identification, and understanding barriers, helping with problem-solving if needed, and providing positive reinforcement. Overall, the facilitator will provide support needed for to the implementation teams to work according to the recommendations of the guideline, for example, by providing resources and technical support. The internal facilitator will participate during the educational meeting and each workshop.

Outcomes

Table 2 describes the measurement variables, method of data collection, data source, and time-point for each measure.

Table 2 Measurement variables, method of data collection, data source, and time pointsParticipant timeline

The time schedule of enrolment, implementation strategies, and assessments is described in Fig. 3.

Fig. 3

The time schedule of enrolment, implementation strategies, and assessments

Sample size and power

For the mediation analysis, the power calculation by Lee et al. [44] is used due to the lack of information on the mediators to conduct own calculation. In the study by Lee and colleagues, the same TDF constructs as well as a similar modeling strategy were used to assess mediators as in the present study. Their calculation indicated that a sample of 121 provides 80% power to detect moderate treatment mediator and mediator outcome effects. With 55 participating schools and 4 to 5 implementation team members per school, we are expecting a sample size of around 220 participants, clustered in their respective schools, for mediation analysis. However, for the implementation outcome, we will have data for all school personnel, with an estimated sample size of around 1500 participants (equivalent to 28 participants per school).

Recruitment

Recruitment of municipalities started in September 2020 by contacting municipalities that had previously shown interest in participating in research projects, advertising the project via stakeholder-related channels, and contacting municipalities by email. Between September 2020 and April 2021, the research team met with four municipalities that showed interest in participation. All municipalities agreed to participate. Next, informational meetings were held with principals and union representatives to inform them about the background, rationale, and logistics of the project. Finally, school personnel were recruited. An informational film recorded by the research team describing the project and what participation entails was disseminated to all personnel by the principals. An informational letter describing the study objectives, research approach, voluntary participation, data collection process, and that participation can be stopped at any time on request by the participants was sent by email. School personnel were given the opportunity to contact the research team if anything was unclear. Individuals who decided to participate in the study accordingly completed an informed consent form.

Assignment of interventions

Due to practical necessity (spread of recruitment), schools were randomized in two rounds. The first randomization was stratified by municipality and clustered, so that all schools in a school district with same central leadership or alternatively all schools with the same headmaster were randomized as a cluster. Upper secondary schools were randomized as a separate stratum in the first randomization round to guarantee their equal distribution among the two groups. The second round of randomisation included only one municipality. Two school pairs (four schools in total) had the same principles and were randomized as clusters. These schools were randomized as a separate stratum from the other schools to maintain balance in the number of pupils and staff. No further stratifications were made. The randomization was done by randomly ordering the clusters within the strata and assigning group allocation based on order. A seed was set for replication. Randomization was conducted by an independent statistician, blind to the identity of the schools and not involved within the project. The principal investigator was not involved in the group assignment. Due to logistical reasons (i.e., planning workshops within the school’s schedules) randomization occurred prior to baseline measurements. The principal investigator informed the municipalities and schools of their group allocation. Blinding of principals or school personnel is not possible within the chosen study design.

Data collection methodsFidelity to the guideline

Fidelity to the guideline will be assessed by a checklist and web survey at baseline and 12- and 24-month follow-ups. The checklist contains three sections and 12 statements with one section for each guideline recommendation. Per recommendation, the school management indicates their agreement with the related statement, e.g., “At our school, we have updated work environment policies”. Respondents indicate whether they agree with the statement and if so, provide a detailed description of how and when the activity was performed, and attach the related documents (e.g., work environment policies). The checklist is electronic, developed for the purpose of the study, and was pilot-tested among a small sample of principals not participating in the study.

The web survey provides a measure of fidelity to the guideline from the perception of the school personnel. It is hypothesized that if schools adhere to the guideline recommendations, then the school personnel will be exposed to the related activities. The web survey was developed for our previous study [9]. For the present study, cognitive testing of the questions and response categories was conducted among teachers (n=5), to ensure that the questions successfully capture the scientific intent and that they make sense to the respondents. The survey contains 12 statements related to the recommendations of the guideline. Respondents indicate on a 5-point Likert Scale ((1) “strongly disagree”- (5) “strongly agree”) the extent of their agreement with the statements. For example, “During the last work environment survey results of the survey were presented to personnel (e.g., by email or during a joint meeting)”. A link to the web survey will be sent by e-mail by the project team. The survey can be obtained upon request from the corresponding author.

Hypothesized mediators

Mediators will be assessed with the Determinants of Implementation Behaviour Questionnaire based on the Theoretical Domains Framework (DIBQ [39]). For this study, items were translated and back-translated from English to Swedish by an independent researcher. The items were pilot-tested among participants of our previous school study. The questionnaire will be completed on four occasions by all participants exposed to the implementation strategies. The following domains will be assessed, each with three items, on a 7-point scale ranging from (1) strongly disagree to (7) strongly agree: knowledge, skills, beliefs about consequences, beliefs about capability, social influences, intention, goals, behavioural regulation, and environmental context and resources. Table 2 provides an overview of the time points when each specific domain will be assessed.

Descriptive variables

Descriptive variables, including age, gender, number of years working at the school, and number of years working within the profession, will be assessed by web survey completed by principals and school personnel at baseline and 12- and 24-month follow-ups. Self-reported stress and self-rated health will be assessed with a single item with 5-point response anchors ranging from (1) “not at all” (5) “very much” [40, 41]. Self-perceived health is assessed with a single question from the SF-12 Health Survey [42] with 5-point response anchors ranging from (1) excellent to (5) bad. Self-reported psychosocial safety climate (PSC) will be assessed with the 4-item Psychosocial Safety Climate Scale with 5-point response anchors ranging from (1) strongly disagree to (5) strongly agree [43].

Readiness to implement

Readiness to implement will be assessed with the Leader/Staff Readiness to Implement Tool during the educational meeting and during workshop 5. The tool was developed for use in the school context and has a leader and staff version (Cook et al., submitted). Both versions contain 14 items with 5-point response anchors ranging from (1) strongly disagree to (5) strongly agree.

Process outcomes

The implementation process will be evaluated by collecting information on implementation outcomes as defined by Proctor and colleagues [45]. Implementation outcomes include penetration and fidelity. Moreover, we will explore participants’ perspectives regarding how or why the implementation strategies worked or failed (and how they might be optimized in subsequent efforts) via observations and semi-structured interviews.

Penetration

Penetration will be assessed through attendance lists, completed by the research team during the educational meetings and during each workshop. Penetration will be operationalized as the absolute number and proportion of individuals participating in the educational meeting and workshops in relation to those expected to participate. Information on penetration is collected as a low penetration could influence the functioning of the implementation mechanisms.

Fidelity

Fidelity to the educational meeting and workshops will be assessed by the research team during the meeting and workshops by using a checklist to note whether they were implemented in accordance with the study protocol. A research log is kept describing deviations and possible reasons for deviations. Fidelity to the implementation teams will be assessed during the educational meeting by examining whether each school has formed an implementation team that is comprised of the recommended representatives. Throughout the study, information will be collected on whether the formation of the teams has changed along with possible reasons for those changes. Fidelity to the PDSA cycles will be assessed by collecting the implementation teams’ PDSA templates after each workshop and comparing the templates against the key principles of the strategy [46]. Fidelity to the internal facilitator will be assessed by checking whether each municipality has appointed an internal facilitator. Information on fidelity is collected as low fidelity to the different strategies could influence the functioning of the implementation mechanisms.

Implementation process

The functioning of the implementation strategies will be assessed by observation and by semi-structured interviews. Observations of the implementation teams will be made during workshops 2 and 5 by using an observation protocol (based on [47]) focusing on the following themes: planning and organisation, interest and engagement, productivity, process, and group-dynamic and climate. The collected information will be used to assess the functioning of the implementation teams. Semi-structured interviews will be conducted at a 12-month follow-up with all principals and a purposive selection of implementation team members based on their role in the implementation process. An interview guide will be developed for the purpose of the study covering the following themes: activities related to the implementation process, how the implementation strategies have supported the implementation process, challenges experienced during the implementation process, how implementation strategies might be optimized in future efforts, additional strategies that might be needed, and other areas for improvement. Semi-structured interviews will also be conducted with school district representatives to assess the extent to which they have acted as internal facilitators supporting the implementation process among their schools according to their functions. The interviews will be conducted by the research team, audio-recorded, and transcribed verbatim.

Context

Six months after the educational meeting the research team will conduct a telephone follow-up (20–30 min) with all principals to assess contextual factors potentially influencing the functioning of the implementation mechanisms. For this study, an interview guide was developed based on an existing guide [24], which was adapted to fit the school context. The guide includes two questions and follow-up prompts. First, the principals will be asked to summarize the activities that they have undertaken to implement the guideline since the educational meeting. Second, they will be asked to identify any circumstances that may have occurred at their school and/or school district and may have influenced the implementation process, such as personnel turnover and organizational changes. The follow-ups will be audio-recorded and transcribed verbatim.

Data management

Data will be collected by using the secure web platform Research Electronic Data Capture (RedCap [48]) and stored electronically in a password-protected folder on a secure server at our institute. All data will be deidentified, by removing all person-identifiable information from the database and replacing it with a code. The code key is saved, which will enable individuals to request an extract of the collected data and demand that information on him/her will be destroyed without any given reason. Only the research team will have access to the identification code, which is stored separately from the data.

Data analysis

The study will use mixed methods to fulfil the complementarity function for evaluation purpose [49]. Thus, causal linkages between the strategies, mediators, and implementation outcomes will be tested with the help of quantitative methods, while data from semi-structured interviews will be analyzed to provide further insight into the process of the mechanism functioning and experiences of those involved. The results of the qualitative study of the implementation process will be nested within the quantitative study of implementation mechanisms.

Statistical analysis

Categorical variables will be presented with count and percentage, continuous variables with median and interquartile range, unless the mean and standard deviation is deemed more informative. The primary outcomes of interest are the mediated effects between the outcome fidelity and the implementation strategy. The mediators are the domains of the Determinants of Implementation Behaviour Questionnaire (DIBQ) [39]. All the domains consist of three items. The first step will be to use a confirmatory factor analysis to confirm that the hypothesized domains are reasonable and can be used. We will then use a mediation analysis to estimate the average mediation effect, average direct effect, and average total effect as well as the proportion mediated. The mediation analysis will be done within the Structural Equations Modeling framework to best take advantage of the data structure. A detailed statistical analysis plan will be developed in collaboration with a statistician prior to starting the data analysis.

Ethics

The study has been approved by the Swedish Ethical Review Authority (No. 2021-01828). The study complies fully with current ethical requirements regarding the handling and storage of personal data and regarding the informed consent process in accordance with General Data Protection Regulation. An information letter is sent to potential participants describing the study aim, research approach, data collection process, and voluntary participation, including the possibility to withdraw at any time. The letter also states that data is only collected for the purpose of the study, presented at the group level and that no personal data will be shared with the school or school district. The letter allows participants to make informed decisions about whether to participate. Informed consent will accordingly be collected from all participants.

Plans for dissemination

Results will be disseminated within the scientific community through peer-reviewed open-access papers and scientific conferences. Results will also be disseminated outside of the research community, for example, through social media, popular scientific reports, and national seminars for key stakeholders, including municipalities and schools.