Exploring return to work barriers through the lens of model of human occupation. The NOW WHAT project

Abstract

Background

The challenges of returning to work after sickness absence demands a wide conceptual understanding of what hinders the employee’s work participation. Thus, there is a need to know more about self-perceived barriers for Return to Work (RTW).

Aim

This study aimed to investigate RTW barriers experienced by employees on long-term sick leave, through the lens of the Model of Human Occupation (MOHO).

Material and methods

The study was a large-scale qualitative interview study (n = 85) using semi-structured telephone interviews. Eligible participants had received sick leave benefits for between 6 months and 1.5 years. The data were analysed with quantitative and qualitative content analysis. A deductive approach using the MOHO concepts guided the analysis process.

Results

The study generated 941 coded meaning units describing barriers for RTW, of which we were able to code 895 within the framework of MOHO. In the person-specific concepts, performance capacity barriers were most often described (n = 303), followed by volitional barriers (n = 165) and barriers related to habituation (n = 66). Barriers related to the environmental components amounted to 361. Barriers in the occupational environment was dominant (n = 214).

Conclusion

Experienced barriers related to both environmental components and person-specific concepts.

Significance

The habituational and volitional perspective on barriers can contribute to the identification and communication of performance capacity-related barriers not previously identified.

Keywords:

• Absenteeism
• conceptual model
• content analysis
• employee
• employment
• sick leave
• sustainable working life
• vocational rehabilitation
• work disability
• work participation

Introduction

Sickness absence often has a multifactorial origin and different societal, occupational and individual factors may interplay and serve as risk factors [1,2]. Work characteristics such as higher physical work demands [2], poor quality of leadership [3,4], lack of social support [3,4] and exposure to harassing behaviour [5] are risk factors associated with sickness absence. However, the general population does not seem to be affected randomly by sickness absence. Sociodemographic factors such as older age [2,6–10], female gender [1,8], and lower educational- and income level [2,10] are consistent prognostic factors for long term sick leave (LTSL) across a multitude of medical diagnoses.

The financial burden imposed by LTSL on society is substantial, with an annual cost of approximately 180 billion NOK in Norway [11]. Aside from being a significant economic challenge, LTSL is widely acknowledged as a major risk factor for permanent exclusion from the labour market and dependence on disability pension [12]. Work is considered the main occupation for many adults, and work participation can impact health through its latent qualities [13]. In occupational therapy practice work is regarded one of the main domains to facilitate participation in for adults [14]. Considering the extensive individual and societal implications, its crucial to acquire a deeper understanding of the obstacles faced by individuals on sick leave that hinder their timely and safe return to work.

While sociodemographic risk factors related to sickness absence and return to work (RTW) are well studied, less is known regarding the self-perceived barriers for RTW that also address environmental and occupational factors. Previous qualitative research on self-perceived barriers for RTW among sickness absentees seems to be limited to specific diagnostic groups, such as chronic pain patients [15,16], cancer survivors [17,18], stroke survivors [17], persons diagnosed with common mental disorders [19] and patients with cardiovascular diseases [20].

A meta-ethnography on challenges for RTW in chronic pain patients identified three main categories for RTW-obstacles: managing pain, managing work relationships, and making work adjustments [16]. Employers’ over-estimation of work ability—especially with regard to invisible symptoms, such as fatigue—were perceived as barriers for RTW in a systematic review of cancer survivors [18]. Another review found positive attitudes towards sick leave among significant others in the life of the absentee with a chronic disease to be a major barrier for RTW [21]. A systematic review of vocational interventions found disruption of the worker role and its associated habits over time to be a barrier for RTW [22]. This can be explained in part by a lack of confidence due to prolonged work absenteeism, which was reported as a major psychological barrier for RTW in a qualitative study by Grant et.al. [15]. As shown above, most studies exploring barriers to RTW have so far been inductive, and few conceptual models that might guide understanding and communication of these crucial aspects have been used.

Current knowledge on RTW supports the appropriateness of a biopsychosocial model, including attempts to understand the RTW processes for sick-listed employees [23,24]. One such model, Kielhofner’s Model of Human Occupation (MOHO) [25–27], is designed to explain human occupation and might therefore be useful for the study of RTW barriers (i.e. factors having an impact on work). MOHO conceptualises human’s occupational propensities as comprising volition, habituation and performance capacity. Volition refers to what a person finds interesting or important, as well as what the persons believe they are able to do [25–27]. Habituation is comprised of the roles persons engage in, as well as the acquired tendency to perform activities in a given pattern [25–27]. Performance capacity is the person’s ability to do things. This capacity is given by the underlying physical and mental capacities as well as the corresponding individual experience [25–27]. These three concepts are interrelated and integrated in the person, constantly influenced by the environment in which the occupation is being performed. Kielhofner and colleagues [25–27] argued that the key to understanding work disability lies in examining the intersection of environmental components and employee volition, habituation and performance capacity. Within the occupational rehabilitation area, MOHO-based assessment instruments are is extensively researched [28–31]. Hence, scientific evidence exists for the relevance of several constructs within the model for the present population. A research study from 2013 concluded that MOHO was the only occupational therapy model that explicitly mentioned RTW and could be a promising model for facilitating RTW in their population of breast cancer survivors [32]. Furthermore, MOHO has been used in vocational rehabilitation interventions [22,33]. We choose the MOHO since it is the most widely used and most researched occupational therapy practice models, and it is used both in occupational rehabilitation research and in occupational therapy practice [29, 34]. And thus, the model seems promising for identifying barriers for RTW among employees on LTSL.

There is a need to gain more understanding of perceived barriers among those on sick leave and a need for research across diverse diagnostic groups. This knowledge could inform occupational therapy practitioners and other professionals working in occupational health services. Therefore, in this study we aimed to explore the first-person perspective, through the lens of MOHO, by asking: What do employees on long-term sick leave experience as barriers for returning to work?

Material and methods

Design

The study was designed as a large-scale qualitative telephone interview study, using inductive semi-structured telephone interviews with employees on long-term sick leave (n = 85). This type of design, which involves a large number of informants, seeks to uncover breadth of experience [35–38] to reach adequate information power [39]. We used a combination of qualitative and quantitative content analysis to identify the barriers in the transcripts [40–42]. We aimed to comply with the Standards for Reporting Qualitative Research (SRQR) guidelines when reporting on methods and findings [43].

Setting/context

This study is a part of the NOW WHAT project. In the NOW WHAT project, researchers are seeking to innovate how RTW services are designed, organised, and delivered. This research was conducted in Norway, a social welfare state. Sickness absence rates are high in Norway [44]. The most frequent diagnosis used on sickness certificates in Norway are musculoskeletal disorders, followed by common mental disorders [44]. Norwegian employees are entitled to receive sick leave benefits for up to a year, with full payment limited to 530,000 NOK. After 1-year, social insurance proceeds with an assessment to determine work (dis)ability. This allowance is called Work Assessment Allowance (AAP) and can be granted for a maximum duration of 3 years at a time. The social insurance is funded by taxes.

Participants

Employees were eligible to participate if they had received sick leave benefits for a longer period (>6 months and <1.5 years) at the time of recruitment for at least 50% of their employed work hours. The participants were recruited from one social insurance (NAV) county in Norway. Eligible participants were invited via a short telephone text message (SMS) to sign up for the interview study. The SMS was sent by the Labour and Welfare directorate. To be able to contact and process the participants, the invitation text was sent in two bulks. This gave the researchers time to follow up, in a timely fashion, on the participants who signed up for the study.

Participants’ sociodemographic characteristics

Table 1 presents the participants’ characteristics. Fifty females and thirty-five males participated in the study. Their age ranged from 23–68 years, with a median age of 51 years. The participants were employed in a broad range of industries. Table 1 organises their employment sector by the United Nations International Standard Industrial Classification of All Economic Activities (ISIC). Twenty-six of the participants had leadership responsibilities.

Table 1. Characteristics of the participants.

Education level	N-participants
Primary/lower secondary school	4
Upper secondary/High school	32
University/college ≥4 years	31
University/college 5< years	18
Area of living
Urban	61
Rural	24
Employment sector
Public sector	30
Private sector	47
Private sector, with public duties	7
Missing info	1
Industry
Human health and social work activities	18
Mining and quarrying	18
Education	13
Transportation and storage	6
Wholesale and retail trade	6
Construction	4
Information and communication	3
Manufacturing	3
Administrative and support service activities	3
Financial and insurance activities	2
Agriculture, forestry and fishing	1
Electricity; gas, steam and air conditioning supply	1
Water supply; sewerage, waste management and remediation activities	1
Public administration and defence; compulsory social security	1
Professional, scientific and technical activities	1
Not elsewhere classified	1
Missing info	3

The participants’ initial sick leave diagnoses according to ICPC-2 disease classification are presented in Table 2. The two major groups were psychological diagnosis (29%) and musculoskeletal diagnosis (28%), followed by general and unspecified diagnosis (10%).

Table 2. The participants initial sick leave diagnosis.

Sick leave diagnosis classified by ICPC-2 Disease classification	N-participants
General and unspecified	10
Blood, blood forming organs and immune mechanism	1
Digestive	0
Eye	0
Ear	1
Cardiovascular	6
Musculoskeletal	23
Neurological	8
Psychological	24
Respiratory	1
Skin	0
Endocrine/metabolic and nutritional	3
Urological	1
Pregnancy, childbearing, family planning	0
Female genital	3
Male genital	0
Social problems	2

Data collection

To guide the interviews, one senior researcher developed and tested an interview guide consisting of two parts. To obtain sociodemographic and personal information, the first part was structured as a questionnaire. The second part was a semi-structured interview guide consisting of the open-ended questions. “What are the real obstacles or hinderances you experience that are preventing you or making it difficult for you to return to work?” This question was followed by verification of content and follow-up questions that encouraged them to remember all relevant barriers, such as “Can you think of any other obstacles or barriers, at work/at home/in contact with health care services, that make it difficult for you to resume work?”.

The interviews were conducted by telephone from spring 2018 to spring 2020. The interviews had a mean duration of 14.34 min (min. 8.14 to max. 34.07 min). Eleven research collaborators who participated in the NOW WHAT project as master’s students or research assistants participated in the data collection; all had a health or social science background: occupational therapists (n = 7), nurses (n = 2), a chiropractor (n = 1) and a social worker (n = 1). They took part in the testing of procedures and data collection. The interviewers had a diversity of previous experience related to working with employees on sick leave, from extensive experience to little experience. Before the data collection, the interviewers were trained in the interview procedure; their interview progress and potential challenges were closely followed up on during the data collection period.

Data analysis

We used a deductive qualitative content analysis combined with quantitative content analysis [41,42,45]. The data were coded with a manifest deductive approach, guided by the MOHO [25,26]. To apply our deductive data analysis strategy, one senior researcher and one PhD candidate constructed a coding frame in NVivo based on the main components of MOHO. Once the first version of the coding frame was ready, the senior researcher and the PhD candidate separately tried the frame in a selected number of interviews (n = 5). The two researchers compared their coding, discussed possible challenges, and considered how the frame could be improved. The frame was modified and adjusted according to issues revealed in the discussion. The coding frame is provided in Supplementary Appendix A. The coding proceeded with the two researchers coding 10 interviews, followed by a meeting to ensure that the coding frame worked across the two coders and had descriptions that could help them code in a way that was as similar as possible.

The process of data analysis is shown in Table 3. We coded the manifest content of the interview transcripts. After identifying meaning units representing barriers, we categorised them, when possible, deductively within the categories of MOHO, and gave the code a short descriptive inductive title, as shown in Table 3. During the process the inductive descriptive titles were collapsed into groups with the same or similar content. For instance, two inductively described categories such as “Difficulties walking” and “Trouble with standing” were collapsed into one inductive code named “Walking and Standing”. In Table 4 and 5, the deductive categories from the codebook and the inductive descriptive codes can be discerned.

Table 3. The process of deductive content analysis.

Meaning unit	MOHO – main category (deductive)	Name of subcategory (inductive)
“I have difficulties walking (…)”	Person specific concepts →Performance capacity →→Motor skills	Walking and standing
“So… So, my disease makes my memory very impaired”	Person specific concepts →Performance capacity →→Process skills	Memory
“It’s my vision. My eyesight, which is my main barrier”	Person specific concepts →Performance capacity →→Communication and interaction skills	Eyesight
“I need to take painkillers (…)”	Person specific concepts →Habituation →→Habits	Routine medication
“For someone my age, I struggle to keep up, you know”	Person specific concepts →Volition →→Sense of capacity	Sense of capacity: Age
“My ageing parents are ill, (…) and caring for them”	Environmental components →Social Environment →→Social relationships	Care burden (outside of work)
“The social insurance systems structure and inability to see the individual and accommodate the individual”	Environmental components →Occupational environment →→Health and welfare systems	Social insurance
“Yes, it’s the layout of the vehicle. The design of the vehicle, it’s so constrained”	Environmental components →Physical environment →→Room	Physical spaces at work

Table 4. Overview of the results of RTW barriers.

MOHO concepts and components	N- coded units describing barriers	% of total units describing barriers
Person specific concepts	534	60
→Volition	165	18,5
→Habitation	66	7,5
→Performance capacity	303	34
Environmental components	361	40
Physical environment	59	6
Social environment	88	10
Occupational environment	214	24
Total number of coded units	895	100

Table 5. Overview of the results from the quantitative content analysis: person specific concepts.

Person specific components	Units per system (% of total)	N-units per main concepts	N-units per subcategory
Volitional system	165 (31)
^#Values		35
^#Personal causation:		119
Self-efficacy			28
Sense of capacity: general			28
Sense of capacity: age-related			13
Sense of capacity: chronic disease			25
Sense of capacity: recovery			25
^#Interests		11
Habituational system	66 (12)
^#Habits:		52
Rest and sleep			24
Routine medication			6
Work			5
Stress			9
Habits general/other			7
^#Internalised roles		14
Role loss and role-lessness			8
Worker role			6
Performance capacity system	303 (57)
^#Motor skills:		67
Walking and standing			32
Physical strength and perseverance			13
Upper extremities			14
Back, neck and torso			8
^#Process skills		45
Concentration			13
Memory			10
Mental energy			6
Orientation in time and space			2
Other			14
^#Communication and interaction skills:		9
Vision			4
Hearing			3
Verbal and non-verbal communication			2
^#Subjective experience of performance capacity:		154
Lack of energy or fatigue			64
Pain			31
Poor health			55
Other			4
^#Occupational performance in ADL		28
Total	534 (100)	544	470

Primary system within person specific concepts displayed in bold with total number units and percentage of total units in brackets. ^#Refers to a main concept, deductively derived from MOHO. Inductively named subcategories in plain text, no superscript.

The meaning units which we were not able to categorise deductively were coded to an open code named “Unable to code in relation to MOHO concepts”, for example functions of body structures such as hypoglycaemia. Following the analysis of all material, we counted the frequency of meaning units describing barriers in their respective code. The software programme NVivo 12 Pro (QSR International) was used to manage the data and as a platform for collaboration when coding the data.

Ethics

The study was presented for the Regional Committee for Medical and Healthcare Research (REK) in Norway and was deemed outside of their jurisdiction. The study is approved by the Norwegian Centre for Research Data (NSD), with reference number 57078. The participants were informed of the aim of the study and provided information on confidentiality. They gave written consent and were informed that their participation was voluntary and would not affect the services they received. Upon obtaining permission, we recorded the telephone interviews and transcribed them verbatim. The transcribed records were anonymized.

Results

Frequency of meaning units describing RTW-barriers

The analysis generated 941 coded meaning units describing RTW barriers. We were able to deductively code 895 meaning units within the framework of MOHO across 15 main concepts and components and a total of 45 subcategories. The remaining 46 meaning units were all barriers stated as the name of the disease, specific symptoms or diagnosis itself. We identified a mean of ten meaning units describing barriers per interview transcription (min. 1, max. 25). All participants, except for two, reported barriers related to both the person-specific concepts and the environmental components. Table 4 provides an overview of the results from the quantitative content analysis. The results from the quantitative analysis will be reported first, and the results from the qualitative analysis will follow.

Frequency of meaning units in the person-specific concepts

Within the person-specific concepts, the concept of performance capacity was the dominant category, containing 60% of the barriers. Furthermore, 154 units of coding related to aspects affecting the subjective experience of performance capacity that permeated the three skills categories. This included fatigue or lack of energy (64 units), a general subjective experience of having a body in poor health (55 units) and pain (31 units). A total of 121 meaning units representing barriers were linked to motor, process, and communication and interaction skills. Volitional barriers (165 units) accounted for 31% of all person-specific concepts, and were related to personal causation (119 units), values (35 units) and interest (11 units). Barriers in the personal causation code was related to self-efficacy (28 units), and one’s sense of capacity (91 units). Barriers related to the habituational system (66 units) were mostly tied to habits (52 units) and to a lesser extent expressed as related to internalised roles (14 units).

Frequency of meaning units in the environmental components

Barriers related to the environmental component were divided into categories related to the physical environment (59 units), social environment (88 units) and occupational environment (214 units). Reported barriers in the social environment related mainly to social relationships, such as conflicts and harassment at the workplace (28 units), difficult relationship with supervisor (25 units) and care burden outside of work (28 units). The barriers related to the occupational environment were dominated by work demand barriers (48 units), work hours and time requirements (27 units) and barriers related to the health and welfare system (86 units).

Tables 5 and 6 provide an overview of the extended results of the identified barriers related to the MOHO concepts and the corresponding inductive subcategories.

Table 6. Overview of the results from the quantitative content analysis: Environmental components.

Environmental components	Units per system (% of total)	N-units per main concepts	N-units per subcategory
Social environment	88 (24,4)
^#Social interactions		6
High social demands			2
Low/No social demands			4
^#Social relationships:		82
Harassment, bullying, conflict			28
Careburden (outside of work)			28
Supervisor relationship and/or leadership			25
Social work environment unspecified			1
Physical environment	59 (16,3)	59
Space, room			12
Working conditions with few possibilities for sitting or resting			22
Modifications of work and equipment			25
Occupational environment	214 (59,3)
^#Work		109
Presence of adequate occupations			17
Work demands			48
Work hours and time requirement’s			27
Work transitions and reorganising			17
^#Health and welfare services		86
The organisation and availability of the social insurance			40
The availability and access to medical treatment			22
Clinical treatment (cooperation with professional)			24
^#Transportation (timing, schedule, system)		10
^#Political and societal environment		9
Total	361 (100)	361	342

Primary system within environmental components displayed in bold with total number units and percentage of total units in brackets. ^#Refers to a main component deductively derived from MOHO. Inductively named subcategories in plain text, no superscript.

Qualitative description of deductive content analysis

Person-specific concepts

Frequently reported barriers were related to the corresponding subjective experience of performance capacity that extend throughout the three skills categories, including fatigue or lack of energy, pain, and the general subjective experience of having a body in poor health. One participant said, “My employer was really good at facilitating, so it has nothing to do with that. It’s just that I am ill.” This quotation, which is representative of several participants, illustrates the subjective experience of being in such poor health that one is unable to work, despite modifications/efforts made. Furthermore, the meaning units describing barriers related to motor skills, most were related to gross motor skills, hereafter the torso and upper and lower extremities, involving difficulties such as moving and lifting. Only one participant reported struggling with fine motor skills.

Barriers related to volition included barriers connected to values, interests, and personal causation such as one’s sense of self-efficacy and one’s sense of capacity. Barriers related to values were expressed as personal convictions related to personal involvement in the RTW process: “No, no, right now my focus is not on returning to work.” Some participants expressed their values as anticipation of RTW outcome, or as decisions based on previous experiences: “I have fought my way back to work several times before. But this time … when I got sick listed, now there’s no chance, it is just… It’s a no go.” Several participants started their responses with, “It is important for me to work, but…”, thus expressing the participants’ experience that working is meaningful and a valued occupation. A few participants expressed barriers related to interest. This was articulated either as only finding enjoyment in their current work tasks, or as an unfulfilled interest or desire to attempt new work tasks.

Sense of capacity was expressed as a barrier related to older age, and hence the sense of changed capacity. However, most of the meaning units reflected uncertainty regarding one’s sense of capacity. The following quotation highlights the barrier of uncertainty regarding capacity and the unpredictability of capacity:

I can get up in the morning and really not be in such good shape when I get up, and then it may improve during the day. But that is actually not so common. How I feel varies terribly, terribly, and I never, never know how the day is going to be. It is very frustrating, and that is what is stopping me from going back to work.

Barriers related to habituation were expressed as both necessary and undesirable habits. Necessary habits were barriers related to the need for prescription drugs, the consumption of which was incompatible with work. Participants expressed undesirable habits as habits related to the need for frequent rest, sleep, difficulties adjusting to shift work/shift hours and new work roles, and not being able to find work that was compatible with the individual’s desired or internalised role. ‘I am a carpenter, and as long as I can’t be that, it’s difficult for me’. This quote features the perspective of roles and role loss in relation to experienced barriers.

Environmental components

Barriers related to the environmental component were categorised into groups related to the physical environment, social environment and occupational environment. Physical environment barriers were experienced as obstacles related to physical modifications to the work and workspace, or general accessibility issues at the workplace. One participant articulated a physical environment barrier by stating ‘It is very difficult with the stairs, and the distances at (name of worksite)’. Some of those interviewed expressed a desire to be able to test different modifications before committing to alterations. They assumed that doing so could reduce uncertainty regarding the functionality of the modifications that could enable a return to work.

Barriers related to the social environment revolved around the social environment at work, such as social relationships with supervisors and colleagues, harassment, conflict, social interaction as part of work tasks and the experienced burden of caretaking for family members with extensive care needs. The following quote is about experiencing a conflict at work, as one of several barriers for RTW, ‘(…) and also, I had a conflict at the workplace’.

The reported barriers related to the component of occupational environment encompassed the availability of adequate work tasks, the political dimension and barriers related to the health and welfare systems. These included barriers with presence of, and availability of, adequate treatment and healthcare services, as well as barriers related to difficulties navigating and interacting with the social insurance system/NAV. The following quote illustrates the barriers related to adequate healthcare and the barrier of the time element: “The other thing is that the medical examinations take such a long time. I have had more than two years of examinations and they still haven’t figured anything out.”

Furthermore, the accessibility of appropriate and adequate work tasks was a frequently reported RTW barrier. Several participants pointed to a gap between their work capacity and the demands of their work, such as responsibility, physical presence, co-operation and customer contact. Work hours and work tempo presented another barrier. These barriers referred to high speed/tempo, time pressure, difficulties maintaining a full-time position, challenges in accommodating work hours for varying work capacity and a lack of flexibility regarding the number of hours required. A few participants stated that they found their work undemanding, and therefore boring.

Discussion

This study aimed to explore the first-person perspectives of experienced barriers for RTW, asking: What do employees on long-term sick leave experience as barriers for returning to work, according to the terminology from MOHO? Three main findings will be discussed here: (1) 6 out of 10 experienced barriers were related to person specific concepts (volition, habituation and performance capacity); (2) 4 out of 10 RTW barriers belonged to the environmental components, indicating the potential for adjustments in the workplace and/or the RTW services; and (3) a broad range of concepts from the MOHO were necessary to describe each of the employee’s experienced RTW barriers, representing a diversity and complexity of RTW barriers.

The majority of the coded meaning units were related to the concept of the subjective experience of performance capacity, such as lack of energy, pain and the subjective experience of one’s health being poor. The experienced hindering symptoms affecting performance capacity are supported in the research literature concerning prognostic factors related to LTSL [2,20]. Other studies have found that barriers such as pain [15,16], fatigue [16,17], emotional and physical complaints [19] were experienced as having a hindering effect on performance capacity. However, it is possible that the participants in our study overemphasised subjective performance capacity barriers. In Norway, a medical diagnosis or symptom diagnosis is required to qualify for sick leave benefits. The meta linguistics in the society related to sick leave might affect how the sick-listed communicated about the different aspects of their sick leave. Consequently, individuals may have felt compelled to frame their issues related to sick leave in biomedical terms. This societal and personal incentive most likely creates the perception that disclosing biological factors is more legitimate; this can amplify performance capacity barriers and thus might impact the sick-listed communication style [46]. That the participants reported a plethora of barriers related to performance capacity was, therefore, not an unexpected finding.

One in four barriers within the person-specific concept were identified as habituational or volitional barriers. Volitional barriers—expressed as a lack of sense of self-efficacy—are supported in previous research, which has identified a high sense of self-efficacy as a facilitating factor for RTW [2,3], and poor coping strategies as a barrier for RTW [19]. Those barriers in our study identified as being related to values could be linked as adversarial to the facilitating factors labelled as positive attitudes towards RTW in the review by Etuknwa et al. Also connected may be barriers related to certain aspects of volition, such as the experience of unpredictable reliability of work capacity [15], loss of motivation [17,19] and lack of self-confidence [19]. However, to the best of our knowledge, our study has identified RTW barriers related to habits and roles that have not previously been reported. For example, as far as we can tell, research on work modifications aimed at facilitating the integration of necessary habits, which are experienced as a barrier for RTW by employees on LTSL, is currently lacking.

Our results show that four out of ten barriers were related to environmental components, with barriers in the occupational environment being the most dominant. Barriers in the environmental components can be addressed at the workplace, or in working conditions [2,4,19,47]. In our results, barriers related to the physical environment at the workplace were less prevalent than barriers in the occupational environment, such as timing of, availability of and suitability of work/work tasks. Several informants stated that provision of ergonomic gear in the workspace had been offered, yet almost none of the participants talked about modifications of their actual work or work tasks. Effective RTW programs should incorporate several elements, including work modifications [47], as the lack of adequate work modifications acts as a barrier for RTW [2,16,17,19]. In the systematic review by Greidanus et al. [18], both employers’ over-estimation of work ability and under-estimation of work ability were perceived as barriers. Over- and under-estimation of employees’ work abilities can be linked to inadequate work modifications that restrict employees’ access to adequate work tasks [18,19]. It can be a balancing act when deciding what to share about personal health, incorporating the need for privacy whilst sharing enough information with the employer so that adequate modifications and support can be provided [17].

Reported barriers regarding interactions with the social insurance system differ from the negative interactions previously identified, and include issues such as being treated with little respect [48]. In our study, the reported barriers to interaction highlight the difficulties of navigating the welfare system and the lack of individual tailoring. According to Hakvaag et al.’s scoping review, the majority of sick-listed individuals reported positive interactions with the sickness insurance officers in the Scandinavian welfare system [48]. However, negative interactions were also identified, and aligned with some of the barriers we identified in our study, such as being misunderstood or disbelieved [48]. Barriers related to access to sick leave benefits and/or workers’ financial compensation have been reported previously [17,18], though not in our study.

A broad range of concepts from the MOHO were needed to describe the RTW barriers experienced by the sick-listed employees. This is in line with previous research, which has acknowledged that sick leave can be multifactorial [1,2]. Using the framework of MOHO enabled us to identify nuances within barriers related to both the environmental components and the person-specific concepts. A particular strength of MOHO is that it enables the communication of barriers related to volitional and habituational aspects. This suggests that, from the perspective of the sickness absentee, MOHO is an adequate framework for describing barriers for RTW. This is supported by previous research that has applied MOHO in the context of occupational/vocational rehabilitation [22,32,49,50], and this study has shown that MOHO is adequate for describing RTW barriers experienced by employees on LTSL. One critique, however, is that due to the focus of the model—which aims to describe what drives and shapes human occupation—it does not contain sufficient components to describe all aspects of health. For instance, we were not able to code certain reported symptoms at the level of body structure, such as hypoglycaemia, as barriers within our MOHO-based coding framework. Previous research support that MOHO, with the limitations as mentioned above, seems to be a promising model to use in RTW research and RTW practice [32].

Strengths and limitations

A strength of our study is its large number of participants with diverse characteristics, which has enabled us to examine a broad range of experienced barriers for RTW. This was accomplished through the use of telephone interviews, which allowed us to cover a large geographical area while keeping costs low and avoiding unnecessary environmental impact due to travel. Although telephone interviews are rarely used in qualitative research, they are very common in quantitative surveys [51] and have been found to be well suited for interviewing participants on sensitive subjects [51–53]. The fact that the majority of participants in our study were sick listed with a musculoskeletal or psychological diagnosis, as well as the higher representation of females compared to males, aligns well with general sick leave statistics. Another strength of the study is its use of a theoretical foundation, which helped to enhance the study’s reflexivity. In addition, the use of a conceptual framework in our analysis facilitates transparency of the analysis process. Our study design, with its deductive content analysis, aims to enable the transferability of results to similar welfare systems [54]. Furthermore, the reflexivity and transferability of the study are enhanced by the collaborative efforts of the multiple researchers who conducted the interviews. Even though we acknowledge that the researcher’s role during data collection influences the interview setting, the focused questions of the interview guides ensured that all participants were asked the same (main) questions. A common critique of content analysis is that the need to interpret meaning units can undermine the content analysis. As we have used the definitions from a theoretical framework to define the content of our codes, this disadvantage is less present in our study.

In this qualitative interview study, we aim to gain insight into a diversity of experienced RTW barriers among persons on LTSL. One aspect to consider with our study is the higher proportion of participants with higher education (n = 49). This is contrary to the general LTSL population, as persons with lower education and income levels have a higher risk of being on LTSL [2, 10]. And thus, might have some implications for transferability to other populations.

Implications for future research and practice

To the best of our knowledge, this is the first large-scale interview study to explore the barriers for RTW experienced by long-term sick-listed persons across various diagnoses. It could be interesting to do a similar study in another population to explore the transferability of our results. Expanding the knowledge of the diversity and complexity of the barriers for RTW could be helpful in the development of RTW interventions and in the education of healthcare professionals such as occupational therapists. Our study can contribute to the design and advancement of future RTW services for employees on LTSL. Our findings can inform occupational therapy practice, as it might be possible to improve precision in communication concerning the barriers for RTW from the perspectives of the sick-listed and different stakeholders by applying the terminology from a conceptual framework such as MOHO. This could encourage occupational therapist to apply assessment instrument assessing the volitional aspects of occupational performance, in their vocational rehabilitation practice [28–31]. Furthermore, by specifically applying MOHO, the habituational and volitional perspective on barriers can contribute to the identification and nuances of performance capacity-related barriers, perhaps not previously identified in the RTW services. And thereby making it easier for practitioners to direct MOHO-based interventions.

Acknowledgements

We would like to acknowledge the research assistants and master’s students who participated in the data collection for this study.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The study is funded by the Norwegian Research Council, project number 301937 and the Norwegian Labour and Welfare Directorate.