"Dis-able bodied" or "dis-able minded": stakeholders' return-to-work experiences compared between physical and mental health conditions.

PURPOSE
This study aimed to explore if and why the return-to-work (RTW) experiences of various workplace stakeholders in the Netherlands and Denmark differ between physical and mental health conditions, and to understand the consequences of potentially different experiences for the RTW process in both health conditions.


METHODS
We studied 21 cases of long-term sickness absence, and held a total of 61 semi-structured interviews with the various actors involved in these cases.


RESULTS
Physical cases were seen as "easy" and mental cases as "difficult" to manage, based on the visibility and predictability of health complaints. On this ground, assessing work ability and following required RTW actions were perceived as more urgent in mental than in physical cases. Despite these perceptions, in practice, the assessment of work ability seemed to impair the RTW process in mental cases (but not in physical ones), and the (non-)uptake of RTW actions appeared to have similar results in both mental and physical cases.


CONCLUSIONS
With these outcomes, the effectiveness of a differential approach is questioned, and the relevance of a bidirectional dialog on work ability and a phased RTW plan is highlighted, regardless of the absence cause. Our study also demonstrates how policymakers need to strike a balance between obligatory and permissive legislation to better involve workplaces in RTW issues. Implications for rehabilitation Both physically and mentally sick-listed employees could benefit from a bidirectional dialog on work ability as well as from a phased RTW plan. A greater role for employers in the RTW process should be accompanied with a support for sick-listed employees, in both physical and mental sickness absence cases. Dutch and Danish RTW legislation could be improved by carefully balancing obligatory and permissive rules and regulations to involve workplaces in RTW matters.


Introduction
During the past two decades, the main purpose of many European welfare states in case of sickness absence has changed from providing benefits, to activating sick-listed employees to return to work (RTW) early. [1,2] These "activation policies" regard work as a better form of welfare than passive benefit receipt, [3] and therefore promote an early RTW of sick-listed employees before they have reached full recovery. [4] Hence, the focus is no longer on determining the inabilities of sick-listed employees, but on discovering their remaining ability to work, despite their illness. [5] The emphasis on work has led to an increasingly important role for employers (and workplaces) in the RTW process, which is based on the belief that they are well positioned to judge what work their employees can still perform as well as the required work(place) adjustments. [1] In doing so, employers are expected to be able to reduce the economic burden of sickness absence to society. [6] Consequently, the RTW process has become the domain of a multiplicity of workplace stakeholders, such as sick-listed employees, immediate supervisors, Human Resource (HR) managers, occupational health physicians (OHPs), unions, and coworkers. [4,7,8] Recently, researchers are catching up on their understanding of how workplace actors actually experience the RTW process, thereby revealing how workplace relationsespecially those between sicklisted employees and employers (represented by supervisors and HR managers)affect this process. Several studies demonstrated that employers appear to have the upper hand in decisions about RTW issues, for example about work(place) adjustments or the timing and speed of the RTW. [9][10][11][12] In making these decisions, employers seem to base their approach, inter alia, on their perceptions of the sick-listed employee's image, attitude, personality and openness about the illness and the RTW. [13,14] Moreover, employers' actions appear influenced by the value and the replaceability of the sick-listed employee, and by the presence of goodwill and trust. [4,6,[15][16][17][18][19] Finally, research has pointed to the potentially facilitating role of OHPs in the RTW process, [4] the stimulating role of unions on employers taking proactive RTW measures, [8,20] and to the relevance of coworkers in the phase before, during and after the RTW. [21,22] As such, these studies highlight the significance (and complexity) of workplace relations and multidisciplinary collaboration in the RTW process.
However, to date, empirical research that explicitly compares stakeholders' RTW experiences in cases of physical versus mental health conditions is scarce. [4,23] This is because scholars have either addressed RTW experiences in relation to a specific symptom group, such as musculoskeletal disorders, [7,14] cancer, [13,17] common mental disorders [19,24] and depression, [20] or have not distinguished between the two health conditions in their findings. [6,9,11,12,15,16,18] As a result, the limited available research has proven inconclusive so far: While some literature reviews suggest the existence of similarities between RTW experiences in physical (here, musculoskeletal) and mental health conditions, [23] such as the importance of work adjustments, [4,25] qualitative studies noted differences in RTW experiences between both health conditions. For instance, with regard to early contact, Tjulin et al. [26] found that supervisors and coworkers felt that the timing of the RTW should vary between physical and mental illnesses, and Hoefsmit et al. [10] observed that in mental cases, supervisors and employees tended to not have early contact, compared to physical cases. The inconclusiveness of evidence leaves a significant gap in the existing literature, since understanding how the RTW process can be tailored (or might not need to be tailored) to the needs of the physically or mentally sick-listed employee may promote an earlier RTW, and in so doing create a "win-win situation" [17,26] for all parties involved.
In this study, we therefore aim (1) to explore if and why the RTW experiences of various workplace stakeholders differ between physical and mental health conditions, and (2) to understand the consequences of potentially different experiences for the RTW process in both health conditions. Since legislation has been shown to affect workplace stakeholders' RTW behavior, [10,27] for instance by giving actors their "teeth" to act, [28] this research examined actual RTW processes in two countries to take this influence into account: the Netherlands and Denmark. These countries are viewed as frontrunners in the "activation" of sick-listed employees to RTW early, [29,30] yet differ regarding the degree to which workplaces are given a statutory role in this process. [31] The RTW legislation of the two countries is further explained and compared below.

Methods
To explore workplace stakeholders' RTW experiences in physical versus mental health conditions, we used a comparative multiple-case study design. This design is particularly convenient to perform a multi-stakeholder analysis, especially in unexplored research areas. [32] The study was conducted from December 2012 to August 2014.

Research setting
This research was carried out in four non-profit hospitals, equally divided over the Netherlands and Denmark. Workplace actors in both countries have obtained a statutory role in activating sicklisted employees to RTW early, albeit to a different extent. Legislation in the Netherlands stipulates that the RTW process is the shared responsibility of employers and employees; by contrast, since legislation in Denmark places the responsibility to initiate the RTW process with municipalities, less policy initiatives are focused on workplace actors, [15,31] who are the focus of this study. As shown in Table 1, the Dutch RTW legislation for workplaces is characterized by extensive, obligatory "must rules" coupled with sanctions, [27] whereas noncompulsory "may rules" (without sanctions) are characteristic of the Danish legislation. [31] However, despite this difference, the RTW legislation contains similar "activation" components in both countries, such as an analysis of the remaining work ability of sick-listed employees (i.e. the problem analysis in the Netherlands and the possibility attest in Denmark), a RTW plan, and meetings between sick-listed employees and their employers. Including these two countries with their different policies enables a better understanding of the influence of legislation on the behavior of workplace actors in the RTW process.

Sample and procedure
We included a total of 21 cases of long-term sickness absence, defined as absence for six weeks or longer, in this research (i.e. five to six cases per hospital in both the Netherlands and Denmark). These cases were purposefully selected to achieve diversity with regard to the type of the health condition (i.e. physical or mental) as well as the job and ward of the sick-listed employee, in order to include as many different workplace actors as possible. This has led to a sample consisting of 13 cases of physical illnesses (e.g. musculoskeletal disorders, neurological disorders and cancer) and eight cases of mental illnesses (common mental disorders, such as stress, burnout and depression). On average, the employees were absent from work for 9.2 months. They were all employed on different wards in the hospitals, or else had different supervisors. The final data set consisted of 61 semi-structured interviews with the workplace actors involved in the 21 sickness absence cases. This means that in the Dutch cases, the sick-listed employee, the supervisor, the HR manager, and the OHP were included, andwhere relevantother work and health professionals. In the Danish cases, the sick-listed employee and the supervisor were interviewed as the main workplace actors involved, and in each hospital an HR manager and a union representative were interviewed. Additionally, in one Danish hospital a social worker was included. An overview of the interviewees and the characteristics of each sickness absence case is provided in Table 2; it should be noted that the difference in the number of interviews per country is related to variations in national and organizational policies (e.g. the involvement of the OHP in the RTW process is obligatory in the Netherlands but not in Denmark).
The interviews were semi-structured and guided by a topic list for both physical and mental cases. Topics entailed: (1) the interviewees' current tasks and professional background, (2) their views on the cause and solution for the sickness absence, (3) their perceptions of their own role in the RTW process, (4) their perspectives on the ideal and actual roles of the other workplace actors involved, and on (5) (the effectiveness of) the RTW actions that are taken, like adjustments of the work(place) or working hours. The interviews took between 40 to 60 minutes, with a few exceptions (e.g. they could last two hours when interviewees were involved in multiple cases). They were recorded (with permission) and transcribed verbatim, leading to a total of 728 pages of transcripts.

Data analysis
Data analysis was done according to open, axial and selective coding techniques, [33] supported by the use of MAXQDA, a software program for qualitative data analysis. First, the interview transcripts were coded line-by-line following the topics of our topic list. We also coded for instances where workplace stakeholders made explicit distinctions between (the RTW process in case of) physical and mental illnesses. This open coding resulted in a list of categories, which we subsequently structured and elaborated using Strauss and Corbin's [33] organizational scheme or "paradigm". The paradigm enables researchers to code around the axis of categories, by relating them to subcategories that define the conditions, (inter)actions and consequences leading to or flowing from these categories. Finally, in selective coding, we were able to distinguish two core categories (or phenomena) as the main themes in our comparison of workplace stakeholders' RTW experiences between physical and mental health conditions: (1) conflicting interpretations of work ability, and (2) different perceptions of required RTW actions. Figure 1 illustrates our data coding structure and shows how the interplay between different perceptions of physical and mental illnesses and the RTW legislation in the Netherlands and Denmark play a role in shaping our two main themes, and the resulting (inter)actions in and consequences for the RTW process.

Ethical considerations
Although approval by an ethical committee for this type of study is not required in the Netherlands and Denmark, all procedures followed were in accordance with the ethical standards laid down in the 1964 Helsinki declaration and its later amendments. The sickness absence cases were selected by our contact person within the hospitals, which was an HR manager, a manager on occupational health issues, or one of the work and health professionals. Potential participants were given written or oral information about the research. Only when the sick-listed employees gave their consent to participate, the other actors involved in the cases were asked for their participation. Informed consent was obtained from all individual participants included in the study. Moreover, at the start of each interview, participants were assured of anonymity and confidentiality in handling the data.

Findings
In comparing stakeholders' RTW experiences between physical and mental health conditions, we will present the findings following the two main themes that have been revealed during data analysis: (1) conflicting interpretations of work ability, and (2) different perceptions of required RTW actions. The conditions, (inter)actions and consequences leading to or resulting from each theme ( Figure 1) are woven into our storyline. Since the first theme mainly occurred in the Netherlands, we take the Dutch cases as a starting-point and compare these with the Danish ones, revealing how the assessment of work ability works out differently for mental and physical cases. By contrast, the second theme is presented by comparing the Danish cases with the Dutch ones, and demonstrates how the (non-)uptake of actions has similar implications for (inter)actions and consequences in mental and physical cases.

Conflicting interpretations of work ability
The workplace stakeholders in our study seemed to have different perceptions of physical and mental illnesses in relation to managing the RTW process. In both countries, they argued howin their experiencephysical cases are mostly "easy" to manage because of the overall visibility of the health complaints and the general predictability of their course and duration, whereas mental cases are seen as "difficult" due to the lack of these characteristics: Physical complaints are much easier for everybody. There is a clear time frame and that's it, and it's much more accepted by everyone (Supervisor NL-5; emphasis added) It would be much easier if it were a broken leg or a cancer disease. When it's a depression or another kind of mental illness, it's very difficult to manage (Supervisor DK-2; emphasis added) Because of the invisibility of mental health complaints and apparent doubts about the credibility of the diagnosis (are they "really ill"?), mentally sick-listed employees seemingly have to work harder to prove the existence of their illness and to convince others of their (in)ability to work than physically sick-listed employees. This is illustrated by the following examples:

Starting-point: the Netherlands
In the Dutch cases, the assessment of the sick-listed employee's work ability by the OHP (the statutory problem analysis, see Table 1) is indeed perceived as more urgent to manage the RTW process in mental than in physical cases. As one HR manager, for instance, explained: With mentally sick-listed employees, we often need the advice of the OHP.
[ … ] In case of something physical, [like] a broken leg, it's clear: six weeks using crutches, no physical load. But in case of a burnout?
[ … ] In those cases, the OHP and the occupational health nurse are the two players we often consult (HR manager NL-6/7) However, in the mental cases in our research, the "subjectivity" of mental illnesses led to conflicting interpretations of the employee's work ability between the OHP and the sick-listed employee. That is to say, in their assessment of the remaining ability to work, the two OHPs included in our study seemed to doubt the severity of the illness (e.g. "people who are diagnosed with a burnout have nothing more than an adjustment disorder, are mildly overworked", NL1-5) and therefore decided that working while having a mental illness is not necessarily harmful for the sick-listed employee's recovery. Although the mentally sick-listed employees argued that they wanted to RTW as soon as possible as well, they did not share the OHP's interpretation of a full ability to work ("In the beginning, I really wasn't able to work", NL-4), and described that they needed a (slower) phased RTW.
However, upon closer inspection of the data, the conflicting interpretations per se did not seem to cause frictions between OHPs and sick-listed employees, but rather the way in which these assessments of work ability were communicated (see, for instance, the above quote about burnout). As one of the work and health professionals described, "the OHP [of this hospital] can be confrontational, and I think that's okay, but I always say, 'it's the tone Figure 1. Data coding structure. [33,43] that makes the music'" (Vocational expert NL-6/9/10). Indeed, the mentally sick-listed employees argued that the OHPs' directness in stating that they were able to work without having listened to their story first, appeared to "cut the ground from under their feet" (NL-5/7). As a result of not acknowledging the employees' illnesses, according to the supervisors and HR managers, the employees were set back in their recovery and the RTW process took longer than might have been necessary. An illustration of such a mental case is provided in Table 3.
By contrast, these conflicting interpretations of work ability did not occur in the physical cases, where the sick-listed employees appreciated the directness and critical view of the two OHPs in our study.
Comparing with Denmark Although disagreements about work ability (especially in relation to the speed of the RTW) occurred in Danish mental cases as well; here, the RTW process appeared ameliorated by the inclusion of a union representative and/or a municipality councilor, who served as the sick-listed employee's "backup". That is to say, in mental cases, the union representative and the municipality councilor acted first and foremost as the employee's support (their "second pair of ears and eyes", UR DK-2). This occurred either "front stage", by reminding the workplace actors of the rights of sick-listed employees and signaling when rules are not followed in joint meetings, or "back stage", by explaining the employer's expectations towards sick-listed employees in conversations with the employees only. Moreover, union representatives and municipality councilors seemed able to counterbalance a supervisor's push for a (too) quick RTW, as "the little voice of morality [ … ] maintaining social responsibility" (UR DK-2). To illustrate, one employee described the role of the municipality councilor as follows: He [the municipality councilor] did a great job and he was on my side, he was helping me and he was not pushing me back to work [ … ]. That really made a difference (Employee DK-2, common mental disorder) Another sick-listed employee confirmed the union representative's own role description as being "a catalyst in the process" (UR DK-1): The union representative and the municipality councilor were not involved or had a less significant role in physical sickness absence cases. Only in one case (DK-11), where the RTW process did not match the department head's personal experiences with a cancer illness, the union representative had a similar role as in the mental cases.

Different perceptions of required RTW actions
According to the Dutch and Danish workplace stakeholders in this study (except for the sick-listed employees), the RTW process requires different actions based on the absence cause. While physical cases are perceived as in no need of strictly following official guidelines, because of their (generally) clearer and more predictable trajectories; mental cases are considered as requiring frequent contacts and a gradual build-up of hours and tasks according to a RTW plan, due to the subjectivity of mental health complaints and their varying trajectories and durations. As two workplace stakeholders explained: People who are ill for mental reasons create such a barrier [to RTW] if they don't stay in touch with the workplace, [ … ] so you need to make sure that you keep in contact, even if it's only to have a cup of coffee or to go for a short walk (HR manager NL-9) You make the small [RTW] steps and have more frequent meetings with someone with a mental illness than you have in case of a bad knee or a bone break (Supervisor DK-5) Starting-point: Denmark. Despite these perceptions of the need for a different RTW approach according to the cause of the absence, in the Danish cases this distinction was not observed in practice. RTW plans were only made in three mental cases; in the other two mental cases these plans were only drawn up after the involvement of the union representative or the municipality councilor. By contrast, RTW plans were still made in three of the six physical cases (all concerning cancer illnesses), althoughas we have shown aboveworkplace actors described them as less necessary in physical situations.
What is more, the analysis revealed the importance of having and holding on to an appropriate, phased RTW plan, irrespective of the absence cause. Namely, the RTW process appeared to proceed problematically in cases of mental as well as physical health conditions when no RTW plan was made (i.e. no phased RTW took place), when the plan did not match the sick-listed employee's needs, or when supervisors deviated from established plans. In these situations, conflicts occurred between the sick-listed employee and the supervisor over the speed of the RTW; in physical cases especially when the RTW took longer than the expected duration. To illustrate these findings, Table 4 gives examples of how both physically and mentally sick-listed employees in these problematic Danish cases described the importance of a RTW plan.
Comparing with the Netherlands. In the Dutch cases, RTW plans were made in all of the physical and mental cases (with two exceptions where the physical illness was at such an advanced stage that RTW was not yet or no longer possible). Although RTW [ … ] Of course you want an employee to return to work as soon as possible, [ … ] but in this case he gave such a blunt advise: "you have to return or else you'll lose your job" (Supervisor NL-5) The OHP is very firm in his judgment of mental health problems [ … ]. And that may backfire, so that employees don't want to return to work (HR manager NL-4/5) plans are developed for each individual case separately, increasing the amount of working hours appeared to be prioritized over building up regular tasks, as one OHP said: "the most important is returning to normal work routines (the hours and days you work) and then we accept a temporary reduction in the difficulty of tasks" (NL [1][2][3][4][5]. This means that most sick-listed employees increased their working hours by one hour every two to four weeks, job duties were reduced to less complex tasks, and irregular shifts were temporarily canceled. Interestingly, all of the mentally sick-listed employees started their RTW by visiting the workplace several times to have a cup of coffee, in order to keep in touch with their colleagues and to lower the threshold to RTW. Despite a perceived need to distinguish RTW actions according to the cause of the absence, both physically and mentally sicklisted employees in our Dutch cases appreciated the RTW plan as a support and guidance in returning to work. In particular, these employees were satisfied with how their needs were considered in establishing the RTW plan, as they mentioned, "what has contributed to the RTW was the time and space that I was given" (NL-4, mental), "they give me all the room that I need to RTW" (NL-6, physical) and "my manager gives me carte blanche" (NL-8, physical). Even in two of the three mental cases where the RTW process was at first impaired by conflicting interpretations of work ability, as we saw above, the sick-listed employee and the supervisor managed to make a plan that got the RTW process back on track in a positive way. As one of the supervisors explained, this meant "partly taking on the [OHP's] advice and partly going our own way" (NL-7). Next to having and holding on to a RTW plan, basing these plans on the needs of physically and mentally sick-listed employees thus seems important for an unproblematic process of returning to work.

Discussion
The aim of our study was to explore if and why the RTW experiences of various workplace stakeholders in the Netherlands and Denmark differ between physical and mental health conditions, and to understand the consequences of potentially different experiences for the RTW process in both health conditions. The data revealed the existence of a discrepancy between perceptions of a required distinction between physical and mental cases on the one hand, and the reality of having to make this distinction on the other. On this ground, this study highlights the importance of (1) involving sick-listed employees in a bidirectional dialog on work ability, (2) establishing an adequate, phased RTW plan, regardless of the absence cause, and (3) striking a balance between "must rules" and "may rules" in RTW legislation aimed at involving workplaces. We discuss these key findings below.

Bidirectional dialogs on work ability
A first difference in actors' RTW experiences between physical and mental health conditions revealed in the assessment of the sicklisted employees' remaining work ability, which was perceived as more urgent in mental than in physical cases. The need to prove a mental illness confirms how doubts regarding the credibility of the diagnosis are specific to common mental disorders. [19] However, despite this perceived need for evidence, in practice, the assessment of work ability appeared counterproductive in the mental cases in our study: conflicting interpretations of work ability (initially) impaired the RTW process. With one exception, these divergent interpretations did not occur in physical cases.
In our research, conflicts about an employee's remaining work ability seemed to stem from a lack of (sufficient) two-way communication between the employee and the OHP (in the Dutch cases) or between the employee and his or her supervisor (in the Danish cases). We found that the mentally sick-listed employees in our study were not given an equal voice, although including employees in assessing their work ability is described as a best practice to ensure active participation in the RTW process. [23] The lack of voice created resistance in these employees towards an early RTW (they felt like "the ground was cut from under their feet") and caused them to put their foot down, which delayed the RTW process. These findings are in line with earlier research showing that, on a more general level, organizational decisions fail more often when employees are not consulted. [34] It thus seems that the RTW process can benefit from bidirectional communication.
In fact, bidirectional communication among stakeholders has been suggested as a potential (yet neglected) facilitator of a successful RTW. [35] Reflecting on the Danish cases in our study where frictions existed between the supervisor and the sick-listed employee, the union representative and/or the municipality councilor could ensure bidirectional communication between both actors. After their involvement, the RTW process got back on track, as employees were given a say in determining the RTW approach based on their work ability. Indeed, chances of returning to work appear significantly higher when a RTW plan is made together with the sick-listed employee, compared to when no plan is made or when it is done by the OHP or the employer solely. [36] This underscores the importance of effective communication between workplace stakeholders.
Although a greater role for employers in the RTW process is increasingly advocated, for instance by the OECD, [1] our study suggests that this should not occur without including support for the sick-listed employee, especially in mental but also in some physical cases. This implies that the role of the union or the municipality could be further strengthened in Denmark (e.g. active involvement from the outset, in all cases), while it is not yet clear who should take this supporting role in the Netherlands. Up until The importance of a phased RTW plan The discrepancy between perceptions and reality of having to make a distinction between physical and mental cases furthermore revealed regarding the implementation of RTW actions. Whereas workplace actors perceived the need to follow RTW procedures in "difficult" mental cases and not in "easy" physical ones, the findings showed that the RTW process could benefit from adequate implementation of a phased RTW plan, irrespective of the cause of the absence. The most important issue seemed to be that the RTW process (in terms of pace and timing) is based on the employee's needs, echoing Franche et al. [8] who described that "in the optimal self-organized return to work, the worker is typically asked by the employer what s/he needs". Basing actions on the needs of sick-listed employees is also recognized as a best practice in managing the RTW for employees with musculoskeletal disorders as well as for those with common mental disorders. [23] This does not necessarily mean moving away from an activation approach, since virtually all the (physically and mentally) sick-listed employees in our study expressed a desire to RTW early. The finding of the importance of a phased RTW plan is in line with quantitative evidence in the Netherlands and Denmark showing that sick-listed employees who partially RTW have a higher chance of a full RTW compared to those who do not RTW partially. [36,37] However, while a phased RTW appeared only effective for physically sick-listed and not for mentally sick-listed employees in Denmark, [38] our study of actors' experiences revealed how both groups of (Dutch as well as Danish) employees valued the structure, guidance and support provided by an adequate, phased RTW plan. This contradiction might be explained by the observation that all the mentally sick-listed employees in our research mentioned the desire to RTW early (i.e. before having reached full recovery), which has been shown to reduce the time to RTW in mental cases when a phased RTW plan is made. [39] In conclusion, while workplace stakeholders perceived the necessity to treat physically and mentally sick-listed employees differently, in practice, the effectiveness of making this distinction appeared questionable, since both physically and mentally sicklisted employees seemed to benefit from having a phased RTW plan that is based on their needs.

"Must rules" versus "may rules"
This research is one in a few that has explored the RTW process cross-nationally [e.g. 27] and included two countries with varying RTW legislation for workplaces, ranging from strong "must rules" in the Netherlands [27] to weak "may rules" in Denmark. [31] While similar perceptions of the differences between (the RTW process for) physical and mental cases were observed, workplace actors' behavior differed between the two countries, with varying consequences for the RTW process. In our Dutch cases, and according to the legislation, RTW plans were made (where possible), but the obligatory problem analysis has led to the imposition of an activation approach on (mentally) sick-listed employeesalbeit with a counterproductive effect. In our Danish cases, RTW plans were either not drawn up or not followed through in a majority of cases, but union representatives and municipality councilors appeared successful in getting the RTW process back on track by being the "second pair of ears and eyes" for (especially mentally) sick-listed employees.
Hence, while "must rules" coupled with sanctions (as in the Netherlands) may increase the probability that RTW actions are taken, our explorative study suggests that these rules can also be used to push sick-listed employees to RTW quickly without considering their needs first. Previous research also showed how employers in the Netherlands used the legislation to force meetings with the sick-listed employee out of distrust regarding the employee's (in)ability to work. [10] These observations nuance earlier quantitative findings suggesting that the Dutch legislation promotes the RTW of long-term sick-listed employees more than the Danish legislation, [40] since statistics remain rather silent on how workplace actors actually experience the RTW process. [41] It may be assumed that negative experiences, such as feelings of being pushed, impair (the sustainability of) the RTW and damage the employment relationship.
Moreover, similar to an earlier study by Stochkendahl et al., [15] this research found that a reliance on "may rules" does not mean that no actions are taken in Denmark. In almost half of the cases, supervisors took the responsibility to draw up a RTW plan, while Danish labor agreements ensured that wages are paid after the mandatory wage payment period of 30 days (to four to nine weeks for blue-collar workers and to one year for white-collar workers). [42] Moreover, while employers in Denmark are allowed to dismiss a sick-listed employee, only four of the eleven Danish employees in this study were dismissed after an average of eight months of absence, and they were given the notice period as a last chance to RTW. Based on a study comparing employers' compliance with "must rules" and "may rules" in Dutch and Belgian RTW legislation, [27] a possible explanation for conformity in Denmark may be that some actors considered these "may rules" as useful or matching their own goals.
The above suggests that neither an overreliance on "must rules" and sanctions, nor the use of "may rules" only should be seen as the holy grail in involving workplaces in RTW issues. [27,40] To quote Van Raak et al., [27] "[a] combination of "must rules" and "may rules" may be more productive, on condition that the latter correspond with internalized rules, with the goals of agents and with their ideas about useful and applicable rules". Hence, it seems that policymakers need to take different stakeholders and their interests into account, when designing policies that aim to involve workplaces but that simultaneously should protect and empower (sick-listed) employees.

Limitations and recommendations for future research
The strengths of our study include its comparison of physical and mental cases in natural settings involving a diverse set of workplace stakeholders, and its link with national legislation in two countries. This has enabled a more in-depth insight into the RTW process and the experiences of a multitude of workplace actors therewith. Although our research comprised an explorative study with a limited number of cases, the use of multiple data sources in terms of interviews with various workplace actors (sick-listed employees, supervisors, HR managers, union representatives, OHPs, and other work and health professionals) helped to triangulate the findings by illuminating the RTW process from different angles.
Nevertheless, due to the explorative nature of our study, further research is needed that compares RTW experiences between physical and mental health conditions, in order to verify and extend our findings. For example, while the two OHPs in our study had fairly strong opinions about mental illnesses and the possibility to RTW, other perspectives may exist among OHPs as well. Moreover, it would be worthwhile to include external stakeholders, such as medical doctors and psychologists, to investigate how their role might affect the process of, and experiences with, returning to work in physical and mental cases. Finally, since our research was conducted in four hospitals, its generalizability to other settings has to be further investigated. It is unclear whether workplace stakeholders in sectors other than healthcare will experience the RTW process differently, for instance due to their lack of medical knowledge. Replicating our study in different sectors would therefore be an interesting avenue for future research.

Conclusion
The legislation aimed at the "activation" of sick-listed employees in the Netherlands and Denmark does not distinguish between causes of absence; rather, it provides general procedures to manage the RTW process, regardless of the illness. Yet, by comparing the experiences of various workplace stakeholders with the RTW process in actual settings, this study shows how cases of physical and mental health conditions are perceived and treated differently. At the same time, the findings question the effectiveness of this differential approach, as it appears that the "dis-able bodied" as well as the "dis-able minded" in the two countries could benefit from a treatment (and especially a phased RTW plan) that is achieved through bidirectional dialog, considering the needs of individual employees without necessarily moving away from an activation approach.

Disclosure statement
The authors report no conflicts of interest.