A process evaluation of a return-to-work intervention to improve Cooperation between Sick-listed employees and their Supervisors (COSS)

Abstract

BACKGROUND: This is a process evaluation of an intervention to improve Cooperation regarding return-to-work (RTW) between sick-listed employees and their supervisors (COSS, which consists of a conversation roadmap, monitoring of cooperation using questionnaires and, if necessary, extra support by an occupational physician (OP).

OBJECTIVE: Objectives were to study (1) the adoption of COSS by a banking organization and (2) its implementation among individual employees, supervisors and OPs.

METHODS: We used quantitative data (online questionnaire, project administration, conversation minutes, emails) and qualitative data (semi-structured interviews). We analyzed quantitative data descriptively (by calculating sum scores, percentages, mean scores and standard deviations). The coding system to analyze the qualitative data was data-driven.

RESULTS: The organization’s representatives reported positively (e.g. fit with existing policy) and negatively (e.g. high intensity) about COSS. At least one OP (out of five) used the monitoring information. Project administration data show a modest reach of COSS among employees and supervisors. The roadmap was used by a minority (35% of the employees and 25% of the supervisors). Relatively many (40% of the employees and 100% of the supervisors who used COSS to evaluate conversations) were satisfied with COSS as a tool to evaluate conversations with the employee/supervisor afterwards. Interview results indicate that the roadmap was considered useful in specific situations (e.g. psychological complaints). All employees and supervisors participated in the monitoring. The majority of the responding employees and supervisors received OP support and was satisfied about this support.

CONCLUSIONS: Despite the good adoption of COSS by the organisation, it was only partially implemented by professionals, employees and supervisors. We hypothesize that our implementation approach did not fit completely with the culture at the bank. Also, the results illustrate the need for other intervention methods for improving cooperation between employees and supervisors and a more specific target population.

Keywords

Sick leave sickness absence implementation work resumption

1 Introduction

Early return-to-work (RTW) after sick leave is important to increase sustainable participation. Employees are considered to benefit in terms of health and quality of life, while employers gain financially [1]. However, there are many bottlenecks in the cooperation between sick-listed employees and their supervisors during the RTW process [2 –5]. Based on former studies [5 –7], five bottlenecks were identified. First, in many cases there is no planned and/or time contingent approach of meetings between employees and employers [5], even though research has shown that early intervention is effective for all employees and that time contingent interventions, gradual RTW or taking a decision as to when and/or how RTW will take place are effective for employees with physical complaints [6]. Second, there is a lack of open communication between employees and supervisors/HR professionals about RTW [5, 7]. Third, employees and supervisors/HR professionals tend not to take joined decisions about employees’ RTW (e.g. supervisors/HR professionals may decide on their own) [5, 7]. Fourth, employees and supervisors/HR professionals do not feel symbiotically dependent for their primary goals (employees: medical recovery/supervisors/HR professionals: RTW) [5]. Strict norms and beliefs play a role in these goals (employees: RTW after medical recovery, HR professionals: RTW during recovery) [5, 7]. Fifth, there is distrust between employees and supervisors/HRprofessionals [5].

Resolving these bottlenecks may facilitate employees’ RTW. Surprisingly, few RTW interventions involve supervisors. One of the few interventions we found focused on burned out employees and included a convergence dialogue between the employee and the supervisor. This intervention improved RTW after 1.5 and 2.5 years of follow up (the latter for younger participants only) [8, 9]. To our knowledge, there is no such intervention that is generic, i.e. for all employees regardless of their diagnoses, which is accessible via the workplace. Generic interventions at an organizational level are important for the Netherlands because Dutch supervisors are not allowed to ask employees about their diagnoses.

Based on the studies mentioned above, we developed such a generic intervention to improve the cooperation between sick-listed employees and their supervisors (COSS). This article reports on a process evaluation of COSS in a Dutch bankingorganization.

1.1 Description of the intervention (COSS)

We developed COSS for the Dutch situation in which the Improved Gatekeeper Act prescribes employees and employers to cooperate in RTW (for example, to write an action plan for RTW). Employers should pay at least 70% of the income during two years of sick leave [10, 11]. Nevertheless, cooperation still appears to be sub-optimal in practice, for example because of distrust between employees and supervisors/HR professionals [5].

COSS aims to improve Cooperation between Sick-listed employees and their Supervisors. It consists of a ‘conversation roadmap’ to structure and intensify their cooperation. Another component is the monitoring of the quality of cooperation (using an assessment instrument that consists of questionnaires focused at bottlenecks A-E mentioned earlier, filled out by employees and supervisors) at the start of COSS, in week eight of sick leave, and every twelfth week thereafter until complete RTW. When employees and supervisors started COSS in e.g. the sixth week of sick leave (the maximum inclusion/commencing time was 10 weeks), they did not receive the second monitoring. Based on our analysis of these questionnaires, we advised OPs (who were trained) about improving the cooperation.

1.2 The current study: A process evaluation

An economic- and effect evaluation of COSS showed that the intervention was cost-effective and produced a (non-significant) trend towards effectiveness regarding work resumption [12-13]. It is important to understand under which conditions COSS produced a relatively small effect, in order to formulate recommendations for future practice [14 –16]. We performed a process evaluation among the intervention group of the field study, using quantitative and qualitative data. Following several innovation theories and models, we focused on the adoption (attitude towards innovation) and implementation (in daily routines) [17 –20]. Our research questions were: 1) How was COSS adopted by the organisation? 2) How was COSS implemented among individual employees, supervisors and OPs?

2 Method

2.1 Design

COSS was studied in a large Dutch banking organization. We based our process evaluation on data triangulation (i.e., quantitative online questionnaire, project administration, conversation minutes of telephone calls, composed by the first author of this study, emails; as well as qualitative semi-structured interviews). This study included multiple stakeholders (i.e. organization’s representatives; OPs, employees, supervisors). According to Dutch law, our study did not require ethical committee approval (correspondence dd. 7 November 2011, registration number: METC 11-4-115/Dutch trial register: 3151). All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was obtained from all study participants.

2.2 Measures and population

Table 1 describes the evaluation topics, stakeholders, outcome variables and methods for data collection per research question.

We collected quantitative data about reach, use, satisfaction and fidelity of COSS. We collectedqualitative data to gain insight into participants’ attitudes and/or experiences regarding the adoption, use, satisfaction and fidelity.

2.3 Procedure

Interview participants were recruited from the stakeholders participating in COSS using convenience sampling [21]. Employees and supervisors were invited by email and were telephoned by NH afterwards. We included employees who had resumed work or supervisors who supported employees who had resumed work. All interviews (30 minutes each) took place 10 months after the start of COSS. We interviewed two OPs, but we were also interested in whether the other three participating OPs provided extra support and thus we emailed them (and if needed, send out reminder mail(s)).

We emailed an access code to an online questionnaire about the use of- and satisfaction with COSS to employees and supervisors after each employees’ RTW. If needed, we send out reminder mail(s).

NH built the project administration during the implementation of COSS.

We also made minutes of the content oftelephone calls with employees who were sick-listed for two to three weeks (during the recruitment for COSS).

2.4 Data analyses

First, the semi-structured interviews were transcribed and analyzed using data-driven coding strategies [22]. Consistent and meaningful fragments (open coding) were searched. Then, concepts were abstracted, defined and delineated and decisions about their relevance were made (axial coding). Constant comparisons were made between and within participant cases to refine concepts before themes were defined [22]. Researcher triangulation (i.e. frequent discussion) was applied.

Questionnaire data were dichotomized (only where items were not dichotomous already) and analyzed descriptively (by calculating sum scores, percentages, mean scores and standard deviations) with IBM SPSS Statistics for Windows, Version 22.0. Further, project administration data were analyzed by counting employees who were invited and appeared to use COSS. Data regarding the reach and fidelity of COSS in the conversation minutes and emails were counted and summarized.

3 Results

Table 2 describes the characteristics of the interview participants.

In total 29 employees (10 male and 19 female) and 32 supervisors (14 male and 18 female) filled out the questionnaire and 10 participants gave a semi-structured interview. These persons provided us with information regarding the adoption (paragraph 3.1) or implementation (paragraph 3.2) of COSS.

3.1 Research question 1: Adoption

The organization’s representatives reported positive and negative attitudes towards COSS.

Regarding the positive attitudes, a representative noted that COSS’s aim fitted well to the organiza-tion’s policy. “.. we increasingly paid attention to how the OP can coach the supervisor..” (representative 1)

Further, this representative noted: “.. you incorporated our suggestions (into COSS).. this made that I wanted to go for it in the management team..” (representative 1) The representative reported satisfaction about the cooperation with the research team.

The representatives also reported negatively towards COSS. The roadmap and the monitoring of the cooperation could imply a lot of work for supervisors: “Then, it (intervention) becomes a burden..” (representative 2)

Besides, providing COSS was said to be complex as “..many actors are involved..” (representative 2) There would be a risk of failure.

Further, they doubted that COSS would be supported by the organisational culture, in which employees were used to little top-down regulation of their work. “Employees are usually provided with many ideas.. only few are obligatory..” (representative 1) The representatives doubted that every employee and supervisor would be interested in COSS, as this was not the case with the organisations’ sick leave policy (which, e.g., recommends meetings between employees and supervisors). “In case.. this (sick leave) takes longer, than I think that people will be curious.” (representative 1) Regarding COSS, a representative noted: “We can only try to convince them (employees and supervisors) of how helpful it may be.” (representative 1) The organisation did not impose policy on employees andsupervisors.

Finally, a representative reported to lack an overview of stakeholder contacts in RTW, which made it difficult to estimate whether employees and supervisors would need COSS.

3.2 Research question 2: Implementation

3.2.1 Reach

Invitations were sent to employees sick-listed for five or ten working days. A considerable part of these invitations were sent to employees who already returned to work, were about to resume work, or were not sick-listed in the first place. So these employees were not eligible for participation in COSS. For another considerable part of the invited employees, we do not know their reasons for non- participation, for example because we could not contact them. A minority of the employees refused to participate for example because they felt too ill or did not trust the confidential treatment of questionnaire data. In total, 39 employees filled out the first monitoring questionnaire for COSS and were included. Some employees and supervisors received log in information for the questionnaire but were excluded later on, e.g. because of being terminally ill.

3.2.2 Use

The online questionnaire results regarding the use of COSS are displayed in Table 3.

The roadmap consisted of three parts (preparation, conversations, evaluation), of which the preparation for conversations was used most often (employees: 34.5%, supervisors: 25%). All employees and supervisors participated in the monitoring at the start of COSS. For 26 cases (i.e. couples of employees and supervisors), 1 or 2 more monitoring reports were sent to the OPs later during sick leave. In total 96.6% of the employees and 81.3% of the supervisors who filled out the online questionnaire, received OPsupport.

3.2.3 Satisfaction

The questionnaire results (Table 3) show that 40% of the employees and all supervisors who used the roadmap were satisfied with the evaluation guidelines in the roadmap. They were less often satisfied with the roadmap as a tool for preparing- and agenda setting for meetings. Overall, 85.7% of the employees and 88.0% of the supervisors who received OP support, were satisfied with the OP. Less than half of the employees and supervisors indicated that the OP enhanced their mutual trust. 35.7% of the employees and 60.0% of the supervisors thought that the OP acted independently.

It appeared from the interviews with employees and supervisors that the usefulness of the conversation roadmap depended on the situation.

First, employees and supervisors differed regard-ing whether they considered the concept of a road-map to be appealing. Some employees were enthusiastic: “It is a handy tool.” (employee 3) This employee liked the stepwise approach: “..both the employee and the supervisor fill out their own roadmap and compare afterwards.” (employee 3)

Supervisors, however, considered the roadmap to be too administrative: “I do not want a lot of extra administration activities.” (supervisor 1). This supervisor noted: “The risk is that it (roadmap) becomes a checklist..” (supervisor 1) The supervisor thought that the roadmap may provoke superficial conversations about RTW.

The experiences with COSS also varied with the types of health complaints. In case of psychological complaints, the roadmap helped to structure the RTW process: “I used to forget what I said within a day. That is why I wrote everything down (in the roadmap)..” (employee 3) The roadmap facilitated the employee to agree with the supervisor on RTW: “We only had to follow the steps..”(employee 3)

In contrast, one of the supervisors evaluated the roadmap as less useful for employees with physical- and/or less severe health complaints: “She had a brace. Still, she could use her other arm.. It (roadmap) becomes a burden if it is over the top for the health complaint.” (supervisor 1) This supervisor thought that in this case, the roadmap would only imply extra work.

The clarity of the medical prognosis was considered relevant as to whether the roadmap was useful. The roadmap was particularly regarded as useful in case of an unclear prognosis. “It (roadmap) gave me the feeling that I still focused on RTW.” (employee 2) Once the prognosis was clear, the employee stopped using the roadmap: “..the wound needed to heal and I had to start exercising again.”(employee 2)

The degree of mutual trust between employee and supervisor also played a role. An employee told how the roadmap may help in case of a lack of trust: “..the roadmap may give instructions to employees on how to deal with the supervisor.” (employee 4). A supervisor noted: “If one experiences a problem with an employee, then one wants to do this (roadmap).. to build documentation.” (supervisor 1) Both supervisors noted that documentation might limit the risk of a financial sanction by the social insurer after two years of absence (i.e. an extra year of wage payment in case of insufficient effort to support RTW). “.. one can say: we agreed to do this, but she (employee) does not want to (resume work).” (supervisor 1)

However, in case of mutual trust, the roadmap was considered to be less useful. “I have a good relation with my employees, so that they will tell about their issues openly.” (supervisor 2) This supervisor thought that the employees’ open communication would facilitate RTW. We also interviewed the employee of this supervisor, who noted: “I am lucky with a supervisor who empathizes a lot with what happens.. My supervisor actually knows everything of me.” (employee 3) The employee confirmed the supervisor’s experience.

Further, whether contact between the sick-listed employee and the supervisor comes about sponta-neously was considered relevant. An employee noted about situations where contact was not initiated by either the supervisor or the employee: “..it (roadmap) could be quite good (to initiate contact).”(employee 1)

In contrast, an employee who had spontaneous contact with the supervisor noted: “.. I did not see a reason to use it.” (employee 4)

Further, the supervisor’s experience with support-ing absent employees played a role. The roadmap was considered useful in case the supervisor lacked experience: “We did not know what to discuss. The roadmap provided us with ideas.” (employee 2) A supervisor mentioned: “It is not always easy to discuss the essence of the problem..” (supervisor 2)

This supervisor had a lot of experience in supporting absent employees: “If one empathizes with the employee, one does not need the roadmap..” (supervisor 2) This quote illustrates the supervisor’s self-confidence in supporting absent employees.

Finally, the flexibility in the use of the conver-sation roadmap played a role. The roadmap recommends that employees and supervisors evaluate their conversation separately, before the next conversation. It went well in cases where employees and supervisors fitted the roadmap to their own situations: “During meetings we discussed the questions (in the roadmap). During the next meeting, we evaluated the previous one.” (supervisor 2)

In contrast, an employee who literally followed all steps experienced an issue: “.. we thought that we should hurry up. There is a timeline in the roadmap..” (employee 2) Thus, if used flexibly, the roadmap was considered more useful than in cases where the steps were followed strictly.

3.2.4 Fidelity of extra OP support

We interviewed two OPs. One used the reports about the quality of cooperation. “I discussed the information with the employee.” (OP 2) This OP was satisfied: “The reports were very clear..” (OP 2) Yet, the reports did not describe which issues (e.g. a lack of trust) were experienced by whom. “Then (when the reports would describe which issues were experienced by whom) I would know whether I have to advise the employee, the supervisor or both.” (OP 2)

The other OP reported: “I did not use them (the reports).” (OP 1) This OP experienced that the information often did not match with the OPs own impression of the situation: “You often stated that this (OP action) was not needed. Meanwhile, supervisors asked for a lot of support.” (OP 1) Yet, this OP thought that the reports focused on the employees’ employability instead of inability: ‘..that was very appealing..” (OP 1)

Finally, one OP replied to our email about the fidelity to COSS and mentioned not to provide this support.

4 Discussion

This study aimed to describe the adoption and implementation of COSS, an organizational RTW intervention to improve cooperation between sick-listed employees and their supervisors. Despite its good adoption by the organization, COSS was only partially implemented by OPs, employees and supervisors. Several factors can explain these results.

The first explanation may have to do with the misalignment of the implementation approach and the organizational culture. The culture at the bank is characterized by a high level of professional autonomy and little top-down pressure to comply with the sick leave policy. Consequently, there is freedom for employees and supervisors to decide about actions to take during sick leave and they may already experience a considerable amount of self-direction. These employees and supervisors probably do not expect that they can benefit much from an intervention to promote self-direction regarding RTW or they may not want to use COSS when their self-direction focuses on activities that do not necessarily facilitate a return to work. Particularly given the lack of steering and control from the top at the bank, a high degree of non-compliance with COSS can come about.

Although the researchers engaged in bottom-up implementation activities (e.g. visits to several local offices of the bank), COSS was implemented mostly top-down; i.e. introduced by a manager as ‘the’ new way to support absent employees’ RTW. This implementation approach was chosen in agreement with this manager in the bank and seemed to be the most appropriate one at the start of this project. However, it can be taken from the above that this top-down communication approach does not fit with this organization [23].

A second explanation that is to some extent linked to the first, is that our findings suggest that supervisors were not convinced of the added value of COSS. These supervisors reported they had extensive experience with supporting absent employees (which suggests that they developed their own style and procedures); hence, they would not need COSS. Alternately, they did not consider sick leave management to be an important management task. Research also shows that managers postpone actions related to RTW interventions [24]. Sick leave guidance takes a lot of the already scarce time of supervisors (supervisors in various Dutch companies, personal communication, 2015), which may be intensified by COSS. In general, our findings illustrate how difficult it is to change work practices of professionals.

Further, we found that the extra OP support related to COSS (in addition to the regular OP support) was hardly provided. An OP noted that the monitoring reports deviated from the OP’s own perception of the cooperation. Also, COSS standardizes the evaluation of the cooperation between employees and employers and hence partially takes over the OP role. An OP representative who was part of our project team, did not seem to consider this as problematic during the development phase. In the end, OPs may have ignored COSS as they may have perceived COSS as a threat to their professional autonomy.

Moreover, employees and supervisors did not always use COSS in those situations where they could have benefited from it. For example, an employee and both interviewed supervisors mentioned that they would use the roadmap only in case of distrust. In case of trust, the roadmap was not considered to have an added value for their cooperation. The roadmap, however, was developed to prevent (not resolve) distrust, because research has shown that supervisors’ distrust can arise during sick leave [5].

Also, COSS may have been too generic. Our findings show that satisfaction with COSS differed with the employees’ and supervisors’ situations. Generally, COSS (which is a very structured intervention), was considered particularly valuable in uncertain situations.

Finally, external factors most likely played a role. During the study, Dutch media reported about OPs who did not adequately protect employees’ confidential information [27]. This, and the economic climate in which employees experience job insecurity, may have made employees feel unsafe and unwilling to participate in (research about) COSS.

4.1 Methodological reflection

For several reasons, this was a very comprehensive process evaluation. It included data- and stakeholder triangulation, which is important to acquire a complete image of a study topic [21]. Another strength is that we evaluated both the adoption and implementation of COSS. We interviewed male and female employees with physical and psychological complaints. Furthermore, many employees and supervisors of the intervention group in the effect- and economic evaluation filled out the process evaluation questionnaire, which suggests that our study population was a good reflection of the intervention group. The process evaluation results are necessary to understand the results of the effect- and economic evaluation of COSS [12, 13].

However, there were some limitations, particularly related to the questionnaires. First, the process evaluation questionnaire did not distinguish between regular- and extra OP support, as part of COSS. This made it difficult to have an overview of the use of- and satisfaction with the extra OP support. Consequently, our quantitative findings regarding the OP possibly refer to their regular support (not COSS). Second, with respect to the process evaluation questionnaire and the monitoring questionnaire (as part of COSS), employees and supervisors noted that the suitable response options were missing. They could not always fill out the questionnaires truthfully. This can also help to understand why the monitoring reports deviated from the OP’s perception of the cooperation. Third, the monitoring questionnaire asks for sensitive information. It can be difficult to fill out such questions about a person with whom you cooperate very much and are dependent on, particularly since information is given to a thirdparty (OP).

There also were study limitations regarding the interviews. We noticed a certain bias due to social desirability. For example, interviewees who did not use COSS themselves usually told the interviewer about hypothetical situations where COSS would have been useful to them. The interviewees may have done this to not disappoint the interviewer who also was one of the developers of COSS. Further, recall bias may have occurred as some employees resumed work about six months before the interview.

Furthermore, of the employees and supervisors who filled out the process evaluation questionnaire, five indicated at the end of the questionnaire that they did not receive the roadmap. In one case it seems that the email was not addressed correctly. It remains unclear why the others did not receive the roadmap. The recipient’s spam filter probably filtered out our email.

Finally, this study was performed in only one for-profit organization (a large banking organization that has offices all over the Netherlands), which implies a limited generalizability to other for-profit sector organizations, and non-profit organizations. The description of the setting and participants allows readers to evaluate the applicability of the results to their organizations though [28].

4.2 Implications for further research

Methods such as responsive evaluation or action research can be useful to acquire a more complete picture of an intervention in practice, especially in a more participative procedure [28, 29]. Responsive evaluation is based on negotiation and focuses on the claims, concerns and issues of stakeholders. In multiple ‘circles’ stakeholders’ own constructions of reality are taken into account in the evaluation. In the end, consensus must be reached [28]. Action research is a cyclic process of observation, reflection, planning and acting regarding change. Its strength is a cooperation between researchers and practiceprofessionals [29].

These methods strengthen a process evaluation with stakeholder triangulation but rule out most of the limitations mentioned in the methodological reflections paragraph.

4.3 Implications for practice

This process evaluation (focussing in the adoption and implementation of COSS) was performed to understand under which conditions the intervention produced a relatively small effect, in order to formulate recommendations for future practice [14 –16]. Our results showed that the implementation of interventions to improve self-direction in RTW should fit with the organizational culture. Thus, projects should start with a diagnosis of this culture. Depending on the degree of professionalism of an organizational culture (i.e. the degree to which employees perform their work using considerable task autonomy), more intensive participation of important stakeholders (i.e. employees, supervisors, OPs) may be necessary already from the beginning of intervention development, i.e. defining what important aspects are in the cooperation between sick-listed employees and their supervisors. Also, the need for a tool like COSS by the stakeholders must get more attention in the starting period. In highly professional cultures, RTW interventions can be realized based on mutual professional control, i.e. reflection among professionals [30]. Leaders should focus on inspiring professionals to make the proposed changes in work procedures [31]. Attention should also be paid to considering sick leave as a management task, as supported by Dutch legislation concerning sick-leave and RTW [10, 11]. This is a prerequisite for interventions such as COSS (for example by organizing meetings between managers to exchange bestpractices).

Further, our findings suggest that interventions should be generic (COSS was considered useful for a range of situations), but can best be applied in uncertain situations (i.e. psychological health complaints, unclear medical prognosis, lack of trust or spontaneous contact between employees and supervisors, lack of supervisors’ experience with supporting absent employees). Overall, cooperation interventions should allow for flexibility. COSS can be further developed into a toolbox with parts of COSS to be used as intervention instruments that can be applied by employees and supervisors. They themselves can decide about what tools to apply, depending on the needs of the situation.

Moreover, the monitoring questionnaire should be replaced by an interview by the OP regarding how sick-listed employees and their supervisors experience their mutual cooperation. Although it requires a time investment, an interview allows more space to employees and supervisors to provide in-depth and situation-specific information [21]. Such information can also facilitate the OPs’ role in COSS. Also, an interview gives more room for the OPs’ professional autonomy.

Finally, it is important to inform employees and supervisors about situations (such as in cases of mutual trust) where COSS can be very useful, even if they do not expect that the intervention will be of added value to them. Moreover, employees and supervisors should be informed about the confidentiality of interventions such as COSS as this may encourage them to participate. In the current process, confidentiality was maintained by, among others, having a third party (in this case a researcher) analyze the results of the monitoring questionnaires.

5 Conclusion

This study aimed to describe the adoption and implementation of COSS, an organizational RTW intervention to improve cooperation between sick-listed employees and their supervisors. COSS consists of a conversation roadmap, monitoring of cooperation using questionnaires and, if necessary, extra support by an OP. We conclude that despite its good adoption by the organization, COSS was only partially implemented by OPs, employees and supervisors. Our findings point at the importance of fitting the implementation of RTW interventions to organizational culture.

Conflict of interest

The authors declare that they have no conflicts of interest.

Author information

NH is a working as an assistant professor at the Department of Work and Organizational Psychology, Faculty of Psychology and Educational Science, Open University, Heerlen, The Netherlands. Her areas of study are sick leave and work resumption after sick leave. She completed a Master of Work and Health as well as a PhD at Maastricht University. The research described in this article was part of this PhD study.

NB is a health scientist and is working as an assistant professor at the department of Social Medicine at Maastricht University. Her research topics are in the field of organization, management and work, with special interest in the work of health care professionals.

IH (PhD) is a health scientist and she works as an assistant professor at the department of Social Medicine, Maastricht University, The Netherlands (Faculty of Health, Medicine and Life Sciences). Her current research interests encompass various topics in the area of work and health as well as research methodology. She supervises PhD students who perform research in these areas and is (co-)author of various scientific papers in these fields. In addition, she is involved as a teacher in the bachelor programs Health Sciences, European Public Health, Medicine and the master programme Work, Health and Career at Maastricht University.

FN† was a professor of Inclusive Labour Organisation at the Department of Work and Social Psychology of the Faculty of Psychology and Neurosciences, Maastricht University. Before being a professor of Inclusive Labour Organisation, he was professor of Psychology of Labour and Health. His research made a big contribution to the broad field of work and health. His research focused on, among others, sick leave, re-integration and redesign ofwork (processes) for the purpose of creating more inclusive employment.

Footnotes

Acknowledgments

The authors wish to thank the Dutch Foundation ‘Stichting Instituut Gak-SIG’ for providing funding for this study. We also thank the organisation and the study participants.

References

Waddell

, Burton

. Is work good for your health and well-being? London: TSO; 2006.

Friesen

, Yassi

, Cooper

. Return-to-work: The importance of human interactions and organisational structures. WORK 2001;17:11–22.

Wrapson

, Mewse

. Supervisors’ responses to sickness certification for an episode of low back pain: Employees’ personal experiences. Disabil Rehabil 2011;33:1728–36.

Lysaght

, Larmour-Trode

. An exploration of social support as a factor in the return-to-work process. WORK 2008;30:255–66.

Hoefsmit

, de Rijk

, Houkes

. Work resumption at the price of distrust: A qualitative study on the return to work legislation in the Netherlands. BMC Publ Health 2013;13:153.

Hoefsmit

, Houkes

, Nijhuis

FJN

. Intervention characteristics that facilitate return to work after sickness absence: A systematic literature review. JOOR 2012;22:462–77.

Hoefsmit

, Houkes

, Nijhuis

FJN

. Environmental and personal factors that support early return-to-work: A qualitative study using the ICF as a framework. WORK 2014;48:203–15.

Karlson

, Jönsson

, Pålsson

, Åbjörnsson

, Malmberg

, Larsson

, Österberg

. Return to work after a workplace-oriented intervention for patients on sick-leave for burnout –a prospective controlled study. BMC Publ Health 2010;10:301.

Karlson

, Jönsson

, Österberg

. Long-term stability of return to work after a workplace-oriented intervention for patients on sick leave for burnout. BMC Publ Health 2014;14:281.

10.

Ministry of Social affairs and Employment. Wet verbetering poortwachter (Improved gatekeeper act). [cited 2013 March 6]. Available from: http://www.arboportaal.nl/onderwerpen/arbowet–en–regelgeving/verzuim/wet-verbetering-poortwachter.html 2013.

11.

Ministry of Social affairs and Employment. Ziekteverzuim van het werk (Sickness absence). n.d. [cited 2013 March 6]. Available from: http://www.rijksoverheid.nl/onderwerpen/ziekteverzuim-van-het-werk

12.

Hoefsmit

, Houkes

, Boumans

, Noben

, Winkens

, Nijhuis

FJN

. The effectiveness of an intervention to enhance cooperation between sick-listed employees and their supervisors (COSS). JOOR 2016;26:229–236.

13.

Noben

, Hoefsmit

, Evers

, de Rijk

, Houkes

, Nijhuis

. Economic evaluation of a new organizational RTW intervention to improve cooperation between sick-listed employees and their supervisors (COSS): A field study. JOEM 2015;57:1170–1177.

14.

Durlak

, DuPre

. Implementation matters: A review of research on the influence of implementation on program outcomes and the factors affecting implementation. AJCP 2008;41:327–50.

15.

Linnan

, Steckler

. Process evaluation for public health interventions and research: An overview. In: Steckler

, Linnan

, editors. Process evaluation for public health interventions and research. San Francisco: Jossey-Bass; 2002. pp. 1–23.

16.

Oakley

, Strange

, Bonell

, Allen

, Stephenson

. Process evaluation in randomized controlled trials of complex interventions. BMJ 2006;332:413–6.

17.

Grol

, Wensing

. Implementatie: Effectieve verbetering van de patiëntenzorg (Implementation: Effective improvement of patient care). Maarssen: Elsevier gezondheidszorg; 2006.

18.

Green

, Kreuter

. Health program planning: An educational and ecological approach. 4th ed. New York: McGraw-Hill; 2005.

19.

Kotter

. Leading change 1996–Harvard Business School Press Boston.

20.

Rogers

. Diffusion of innovations. 5th ed. New York: Free Press; 1995.

21.

Polit

, Beck

. Nursing research: Generating and assessing evidence for nursing practice. 8th ed. Philadelphia: J.B. Lippincott; 2008.

22.

Boeije

. Analysis in qualitative research. London: Sage Publications Ltd; 2010.

23.

Mintzberg

. Structure in fives: Designing effective organizations. Prentice Hall: New Jersey; 1992.

24.

Wrapson

, Mewse

. Supervisors’ responses to sickness certification for an episode of low back pain: Employees’ personal experiences. Disabil Rehabil 2011;33:1728–36.

25.

Berg

, Horstman

, Plass

, van Heusden

. Guidelines, professionals and the production of objectivity: Standardisation and the professionalism of insurance medicine. Sociology of Health and Illness 2000;22:765–91.

26.

Dodier

. Clinical practice and procedures in occupational medicine: A study of the framing of individuals. In: Berg

, Mol

, editors. Differences in medicine. Durham: Duke University Press; 1998. pp. 53–79.

27.

Agra

. De verzuimpolitie: Het gebeurt recht onder je neus. (The sick leave police: It happens right in front of you). [cited 2014 July 2]. Available from: http://www.artsennet.nl/blogs/Artsennetbloggers/Dolf-Algra/Blogbericht-Dolf-Algra/45/De-Verzuimpolitie-het-gebeurt-recht-onder-je-neus.htm.

28.

Guba

, Lincoln

. Fourth generation evaluation. Newbury Park: Sage Publications Ltd; 1989.

29.

Koshy

. Action Research for improving educational practice. A step-by-step guide. 2nd ed. London: Sage Publications Ltd; 2011.

30.

de Bruijn

. Managing professionals. Abingdon, Oxon: Routledge; 2011.

31.

Mintzberg

. Covert leadership: Notes on managing professionals. Knowledge workers respond to inspiration, not supervision.. Harvard Business Review 1998;76:140–7.