Minor long-term effects 3-4 years after the ReDO™ intervention for women with stress-related disorders: A focus on sick leave rate,everyday occupations and well-being

Abstract

BACKGROUND:

The Redesigning Daily Occupations (ReDO™) work rehabilitation method has been found effective, compared with care as usual (CAU), for women with stress-related disorders.

OBJECTIVE:

To conduct a long-term follow-up of former ReDO™ and CAU participants with respect to sick leave, well-being and everyday occupations 3-4 years after completed work rehabilitation.

METHODS:

Forty-two women in each group participated. An index day was decided to estimate sick-leave rate, retrieved from register data. Fifty-five women also participated in a telephone interview addressing well-being, everyday occupations and life events.

RESULTS:

Both groups had reduced their sick-leave rate further, but no difference between the groups was established. The ReDO™ women perceived a better balance in the work domain of everyday occupations, whereas the CAU group reported more over-occupation. No differences were found on well-being. The groups had experienced similar important life events, affecting the women’s work and private lives. Previous stress and anxiety predicted sick leave at the long-term follow-up.

CONCLUSIONS:

Although the ReDO™ intervention had speeded up return to work in the immediate follow-up perspective, the CAU had caught up in the longer term. Still, the ReDO™ women exhibited better balance in the work domain.

Keywords

Work rehabilitation employment female long-term outcomes Trial registration ClinicalTrial.gov NCT01234961. Registered September 2007.

1 Background

Stress-related disorders are a common cause for sick leave in many western countries [1 –4], and the negative consequences are manifold for the affected individuals, their families, workplaces and society at large [5]. There is an economic impact on all these levels, but for the individual him/herself there are also losses in terms of life roles, social capital, quality of life and self-esteem [6]. Stress-related disorders are a subgroup among mental disorders, which constitute the most prevalent reason for sick leave in Sweden [7]. Sometimes the terms; common mental disorders [8, 9] and minor mental disorders [10] are used, and these typically include diagnoses such as mild to moderate depression and adjustment disorders [2, 10].

Different rehabilitation interventions have been developed to enable individuals’ return to work. In Sweden the local Social insurance agencies have the responsibility for the coordination of work rehabilitation, although the employer has the main responsibility for implementing work rehabilitation for those who are in work [11].

Return to work after a longer period of sick leave is a complex process that is contingent on many issues. Factors that facilitate this process are an appropriate intervention [12 –15], believing that returning to work will be possible [16, 17], and an understanding and supporting attitudes among the relevant authorities and professionals [18 –20].

About two thirds of those who are on long-term sick leave (>60 days) in Sweden are women [21]. It is thus urgent to develop feasible interventions that facilitate return to work for women, such as the Redesigning Daily Occupations (ReDO™) [2, 22]. The ReDO™ is a lifestyle intervention that builds upon principles from occupational therapy and occupational science, stating that an individual’s pattern of daily occupations can promote health if that pattern is personally satisfying and on par with the individual’s capacities and preferences [22, 23]. This rehabilitation program was implemented in primary health care during 2007–2010. A social insurance officer who coordinated all pertinent rehabilitation alternatives assessed those who were eligible for the ReDO™ intervention. Besides female gender, having an employment, currently being on sick leave due to a stress-related disorder, and being in need for a lifestyle intervention such as the ReDO™ were the inclusion criteria. The intervention is a group-based rehabilitation program where the women meet twice a week for ten weeks. This is followed by a six-week individual work-practice period at the woman’s regular place of work. The findings from the ReDO™ project showed that the return-to-work rate, which was the main outcome measure, was higher in the ReDO™ group compared to “care as usual”, CAU [2]. The main alternative, among the interventions labeled as CAU, was support from the social insurance officers in combination with measures taken by the employer, in accordance with Swedish routines. The ReDO™ program was also effective in terms of greater reduction in sick leave and greater improvement in self-esteem [2] and quality of life [24] compared to the CAU group. Furthermore, those who participated in ReDO™ increased their satisfaction with everyday occupations more during the course of the ReDO™ [25]. The follow-up period in this project was one year, but in intervention projects that target return to work a longer follow-up would be desirable and reveal whether effects remain over time. Another intervention aimed at facilitating return to work showed to be successful in a 1.5-year perspective, but no long-term benefits (at 2.5 years) were indicated [26].

It seems vital to address long-term outcomes of interventions aimed at return to work to learn more about their effectiveness. The aim of this study was thus to conduct a long-term follow-up of the ReDO™ intervention for women with stress-related disorders and look at their situation with respect to sick leave, well-being and everyday occupations 3-4 years after having participated in that work-oriented rehabilitation. The term occupation, as in everyday occupation, is here used to represent all the everyday activities a person is involved in, including paid work, pastimes, household work, caring for others, socializing and the like.

The research questions were:

Was there any difference in sick-leave rate between the intervention group and the control group at the long-term follow-up?

Did the groups differ regarding perceptions of well-being and everyday occupations?

Were perceptions of well-being and everyday occupations related to currently being in work or not?

Which, if any, previous factors pertaining to stress, anxiety and depression could explain currently being in work?

Which life events had the women encountered since they had participated in work-oriented rehabilitation?

Was the importance of those life events related with currently being in work or not?

2 Methods

The long-term follow-up was an extension of a quasi-experimental project evaluating the ReDO™ intervention [2, 24]. The project, which also included a qualitative study [27], initially comprised four measurement points; prior to the intervention (= baseline), after completed intervention, and follow-ups 6 and 12 months after completed intervention. The current long-term follow-up, based on both quantitative and qualitative data, was made 2-3 years after the 12-month follow-up, thus 3-4 years after completion of the intervention. The institutional research ethics committee at Lund University approved the study, Reg. No. 2011/718. All procedures followed were in accordance with the ethical standards of the responsible institutional committee and with the 1964 Helsinki Declaration and its later amendments. Informed consent was obtained from all patients for being included in the study.

2.1 The ReDO™ intervention

The manualized ReDO™ intervention is described in detail elsewhere [28]. It is a group-based, three-part program encompassing 16 weeks. The first part (weeks 1–6) addresses hindrances for a balanced and health-promoting pattern of daily occupations, such as too little time for leisure and unequal distribution between family members with respect to household chores. The second part (weeks 6–10) focuses on obstacles for return to work, such as unclear work duties and a heavy work load. During this total of ten weeks, the group meets each week for two 2.5-hour sessions when problems with everyday occupations are in focus and strategies for how to solve them are developed. Each group had three to five participants. In between group meetings, the participants test these strategies and attempt to shape a satisfying pattern of everyday occupations for themselves. The next-coming session includes a follow-up of those endeavors, but also the identification of new problems, when relevant, as well as alternative strategies and solutions. The third part (weeks 11–16) constitutes a work-practice period. The women preferably do their work practice at their regular workplace; otherwise they are offered another workplace that can support their process towards returning to work. Two booster sessions are held during this third and last part of the program to provide feedback and support.

Four occupational therapists with specific training in the ReDO™ method worked two by two and led a total of ten groups that were included in the project reported here.

2.2 The CAU interventions

The CAU received by the comparison group varied from only regular follow-up meetings with the social insurance officer to also including a structured rehabilitation program. Work rehabilitation, mindfulness training, cognitive behavioral therapy and physiotherapy were examples of interventions that were offered in addition to the follow-up meetings with the social insurance officer. About half of the participants in the CAU group received some kind of additional rehabilitation besides these meetings. Only few participants received more than one type of rehabilitation, such as physiotherapy and cognitive behavioral therapy.

2.3 Data collection

2.3.1 Register data

The main outcome variable targeted in this long-term follow-up was percentage of time off sick calculated as the proportion of the woman’s normal working hours. Thus, a woman normally working 40 hours per week but now working 32 hours was off sick for 25%. These data were retrieved through the register of the Social Insurance Offices and a social insurance officer anonymized and coded the data before they were delivered to the research team. The register data were retrieved on a specific date (1 September 2012) for all participants, and the time from having completed one’s rehabilitation to the index day varied between 3-4 years.

2.3.2 Data collection at the long-term follow-up

A research assistant with vast experience from working with people on sick leave, as well as from working in research projects, performed the rest of the data collection through telephone interviews based on structured questionnaires. The questionnaires were first sent to the interviewees so that the questions were known to them prior to the phone call. The interviewee also had the questions before her during the interview to avoid mishearing. The interview was based on five questionnaires with together 63 structured questions and took 30–45 minutes. The questionnaires were as follows:

Self-mastery assesses whether people perceive they influence and control their own life situation or whether other circumstances have the major impact. This was estimated by Pearlin’s Mastery Scale [29], which has seven items and uses a response scale with four alternatives from agree to disagree. The Swedish version was used, found to be valid and reliable and to represent a logical continuum of the measured construct according to the Rasch measurement model [30].

Self-esteem was assessed by the self-report instrument Rosenberg’s Self Esteem scale [31, 32]. It includes ten items targeting positive and negative aspects of self-esteem. A response scale with the alternatives yes/no/I do not know, proposed by Oliver and colleagues [33], was used for this study. The final score ranges from –1 to 1, and a higher rating signifies better self-esteem. Good internal consistency and test-retest reliability have been demonstrated [34].

A background questionnaire asking for socio-demographic information (age, civil status, having children, having friends, education, type of work) and causes of any sick leave was administered. There was also a question regarding any life events that significantly influenced the woman’s everyday life and had occurred since completion of the last interview (the one-year follow-up). Up to four events could be listed, and each of them was then rated according to significance on a Likert-type scale that varied from 1 = very little importance to 5 = great importance.

The Satisfaction with Daily Occupations and Occupational Balance (SDO-OB) assessment was used, which is an interview-based questionnaire addressing the respondent’s satisfaction with everyday occupations and perceived balance in relation to one’s everyday occupations [35]. It consists of 13 items covering the occupational areas of work, leisure, home chores and self-care. The respondent first states whether or not (s)he presently performs the targeted occupation. This gives yes/no answers and the number of affirmative responses are calculated into an activity level score. The respondent then rates his/her satisfaction with the occupation on a seven-point scale where 1 = worst possible situation and 7 = best possible situation. Summarizing these ratings renders an occupational satisfaction score. In relation to each occupational area, the respondent also estimates whether (s)he does too little, just enough or too much of that type of occupation. There are five items of this type, one for each occupational area and one general question. A five-point response scale is used, ranging from –2 (way too little) to 2 (way too much). Zero denotes just enough. These ratings are not summarized, but result in a five-facet profile of occupational balance in terms of perception of time allocated to various occupational domains [35, 36]. The original SDO was without the balance items and has shown to be valid and reliable across target groups and cultures [37 –39]. The balance items have shown initial concurrent, discriminant and criterion validity [35].

The Occupational Value with pre-defined items (OVal-pd) instrument [40, 41] was used to address the value associated with one’s everyday occupations. It has 18 items, addressing three dimensions of occupational value; concrete, symbolic and self-reward value [42]. Each item is formulated as a statement, starting with “In the past four weeks, I have done occupations that... ” and then followed by a specific value experience such as “... that were fun or playful”. This example represents self-reward value; things people do simply because they are enjoyable. Items addressing concrete value concern actions that produce tangible outcomes, such as a nice product or a reimbursement, whereas symbolic value has to do with belongingness and feeling part of a culture. Ratings are performed using a four-point scale ranging from 1 (not at all) to 4 (very often). The OVal-pd has shown to be valid, reliable in terms of internal consistency and cross-culturally robust [41 , 44].

2.3.3 Previously collected data

This study was also based on data collection regarding perceived stress, anxiety and depression at the follow-up performed one year after completed work rehabilitation (ReDO™ or CAU). These data were collected two to three years before the index day (1 September 2012) by the research assistant who also performed the long-term follow-up data around the index day. The following instruments were used at the one-year follow-up:

The Perceived Stress Scale (PSS) is a self-report questionnaire that assesses the degree to which people perceive their lives as unpredictable, uncontrollable and overloaded [45]. The PSS has 14 items, which are rated in relation to how frequently the respondent has perceived the targeted type of stress. A five-point response scale is used that ranged from never (0) to very often (4). The Swedish version was used, which has shown adequate properties regarding factor structure, reliability and construct validity [46, 47]. A higher score indicated more perceived stress.

The Hospital Anxiety and Depression Scale (HADS) [48], developed to identify anxiety and depression in samples without serious psychiatric illness, was also used. HADS is a self-report questionnaire with 14 items that are rated on a four-point response scale. Higher scores designate states of more severe anxiety and depression. Separate scores for anxiety and depression were used for the current study. The HADS is extensively used and performs well as an identifier of anxiety disorders and depression [49].

2.4 Participants

The project comprised of 84 women, 42 who received the ReDO™ intervention and 42 who were matched on a number of variables; number of children, married/unmarried, education level, housing situation (villa/flat), reason for being off sick at baseline and number of months off sick at baseline. Register data at the long-term follow-up were obtained for all 84 participants, but only 24 women (57%) in the ReDO™ group and 31 (74%) in the CAU were reached and agreed to participate in the follow-up telephone interviews. This did not constitute a statistically significant group difference in participation rate (p = 0.108). Analyses between participants in the follow-up interview and non-participants indicated that, in the CAU group, the participants were significantly older than the non-participants (see Table 1). Furthermore, in the ReDO™ group those who were interviewed consisted of fewer Managers and professionals than the non-participants. No other difference was found between interview completers and non-completers (p-values ranging between 0.155 and 0.895).

Table 1
Baseline characteristics of the women participating in the long-term follow-up; data for the total sample (N= 84) and the interviewees (N= 55)

The total sample (N= 84) The interviewees (N= 55)

Characteristic ReDO group (n = 42) CAU group (n = 42) ReDO group (n = 24) CAU group (n = 31)

Age; mean (SD)¹⁾ 45 (19) 46 (9) 43 (8) 40 (9)

Married or living with a partner; n (%) 30 (71%) 27 (64%) 15 (63%) 21 (68%)

Number of children; mean (SD) 2.4 (1.4) 2 (1) 2.5 (1.1) 2 (0.8)

Type of occupation^2,3)

Managers and professionals 15 (36%) 21 (50%) 4 (17%) 17 (55%)

Technicians and associate professionals 13 (31%) 6 (14%) 10 (42%) 4 (13%)

Clerical support, service and sales workers 13 (31%) 15 (36%) 9 (38%) 10 (32%)

Plant and machine operators 1 (2%) – 1 (4%) –

Having a university degree; n (%) 16 (40%) 21 (51%) 7 (30%) 16 (52%)

Percentage of baseline sick leave; mean (SD) 92 (18) 83 (26) 90 (21) 78 (30)

Sick leave (months) before baseline; mean (SD) 13 (20) 10 (10) 11 (20) 11 (11)

First diagnosis

Depression 19 (45%) 23 (54%) 12 (50%) 15 (48%)

Stress/Exhaustion 20 (48%) 17 (41%) 10 (42%) 14 (45%)

Physical main diagnosis 3 (7%) 2 (5%) 2 (8%) 2 (7%)

	The total sample (N= 84)	The interviewees (N= 55)
Age; mean (SD)¹⁾	45 (19)	46 (9)	43 (8)	40 (9)
Married or living with a partner; n (%)	30 (71%)	27 (64%)	15 (63%)	21 (68%)
Number of children; mean (SD)	2.4 (1.4)	2 (1)	2.5 (1.1)	2 (0.8)
Type of occupation^2,3)
Managers and professionals	15 (36%)	21 (50%)	4 (17%)	17 (55%)
Technicians and associate professionals	13 (31%)	6 (14%)	10 (42%)	4 (13%)
Clerical support, service and sales workers	13 (31%)	15 (36%)	9 (38%)	10 (32%)
Plant and machine operators	1 (2%)	–	1 (4%)	–
Having a university degree; n (%)	16 (40%)	21 (51%)	7 (30%)	16 (52%)
Percentage of baseline sick leave; mean (SD)	92 (18)	83 (26)	90 (21)	78 (30)
Sick leave (months) before baseline; mean (SD)	13 (20)	10 (10)	11 (20)	11 (11)
First diagnosis
Depression	19 (45%)	23 (54%)	12 (50%)	15 (48%)
Stress/Exhaustion	20 (48%)	17 (41%)	10 (42%)	14 (45%)
Physical main diagnosis	3 (7%)	2 (5%)	2 (8%)	2 (7%)

Note. Statistically significant differences are indicated in bold and further explicated below. ¹⁾The CAU women, who participated in the interview at the long-term follow-up, were younger than the dropouts (p = 0.020). ²⁾The ReDO™ women, who participated in the interview at the long-term follow-up, comprised fewer Managers and professionals than the dropouts (p = 0.026). ³⁾At the long-term follow-up the interviewees in ReDO™ group comprised of fewer Managers and professionals and more Technicians and associate professionals compared to the CAU group (p = 0.012).

Table 1 indicates further socio-demographic and sickness-related information regarding both the sample as a whole (for which only register data were available) and those who were reached for the interview. Regarding the sample as a whole, there was no difference between the ReDO™ group and the CAU group on any of the variables. Among those who participated in the follow-up telephone interview fewer women in the ReDO™ group and more in the CAU were Managers and professionals, whereas more ReDO™ women and fewer CAU women than expected were found among Clerical support, service and sales workers (p = 0.012). No other group differences among interviewees were found, p-values ranging between 0.095 and 0.946.

2.5 Data analyses

Non-parametric statistics were seen as appropriate since the questionnaires used produced ordinal data. Differences between groups were tested with chi² analyses (for categorical data) or the Mann-Whitney U-test (for ordinal data). These analyses concerned the ReDO™ group versus the CAU in the first place. In order to address the research question whether previous levels of stress, anxiety and depression could explain currently being in work, groups were also formed based on dichotomization of the variables of stress, anxiety and depression as measured at the one-year follow-up. Differences between two measurement points were analyzed by the Wilcoxon test for related samples.

Open-ended responses regarding life events were categorized according to manifest, quantitative content analysis as described by Krippendorff [50], i.e., responses with similar contents were grouped in categories and counted.

The software applied was the IBM SPSS version 21. The level for statistical significance was set at p = 0.05.

3 Results

3.1 Sick leave rate based on register data

As indicated in Table 2, there was no difference between the groups regarding sick leave at the follow-up 3-4 years after completed rehabilitation. Four of the ReDO™ women (N= 42) and three of the CAU women (N= 42) were on sick leave on the index day. The percentage of sick leave at the long-term follow-up, in terms of percentage of normal working hours, did not either reveal any difference between the groups. When looking back at the preceding two years, 16 in the ReDO™ group and 13 in the CAU group had been on sick leave. Comparing number of days off sick during that period, in terms of equivalents of full-time off-sick days, indicated no statistically significant difference between the groups.

Table 2
Sick-leave for the total sample at the long-term follow-up

ReDO group (n = 42) CAU group (n = 42) P-value

Number of women on sick leave on the index day; n (%) 4 (9.5%) 3 (7.1%) 1.0

Off sick as percentage of normal hours on the index day; mean (SD) 7.7 (25) 6 (22.6) 0.702

Number of women on sick leave at any time during the two-year period preceding the follow-up; n (%) 16 (45%) 13 (31%) 0.178

Number of days off sick during the two years preceding the follow-up; mean (SD) 65 (128) 43 (126) 0.109

	ReDO group (n = 42)	CAU group (n = 42)	P-value
Number of women on sick leave on the index day; n (%)	4 (9.5%)	3 (7.1%)	1.0
Off sick as percentage of normal hours on the index day; mean (SD)	7.7 (25)	6 (22.6)	0.702
Number of women on sick leave at any time during the two-year period preceding the follow-up; n (%)	16 (45%)	13 (31%)	0.178
Number of days off sick during the two years preceding the follow-up; mean (SD)	65 (128)	43 (126)	0.109

When analyzing the difference between sick-leave rate at the one-year follow-up and the long-term follow-up in the ReDO™ and the CAU groups separately, a significant decrease was found for both groups (p = 0.002 for the ReDO™ group and p = 0.01 for the CAU group).

3.2 Everyday occupations, well-being and significant life events

Analyses based on the 55 women who participated at the long-term follow-up indicated no differences between the ReDO™ and the CAU groups regarding self-mastery, self-esteem, satisfaction with daily occupations or occupational value. Regarding occupational balance, there was a tendency that the ReDO™ group scored lower than the CAU on balance in the work domain. Frequency distribution indicated that the ReDO™ women more often than average responded that they were in balance whereas the CAU women more often answered that they worked too much or far too much. No group difference was found for any of the other occupational balance items.

Further calculations were based on the follow-up sample as a whole. Women who were not currently on sick leave scored higher on self-mastery and self-esteem than those who were (p = 0.001 in both cases). Those who were not currently on sick leave also valued their everyday occupations more at the long-term follow-up (p = 0.016). The group difference with respect to satisfaction with everyday occupations was just above statistical significance (p = 0.056), but the tendency was in the positive direction for those who were not on sick leave.

Those who belonged to the group with a high level of stress at the one-year follow-up showed a higher degree (in percentage) of current sick leave, compared to those who belonged to the low group (p = 0.005). The same was found for level of anxiety at the one-year follow-up (p = 0.036), whereas level of depression at that time was not related with current degree of sick leave (p = 0.250).

In all 38 of the 55 women, 17 (71%) in the ReDO™ group and 21 (68%) in the CAU group, reported that something significantly affecting their everyday life had occurred since the one-year follow-up. Twenty-two (13 and 8, respectively) also reported a second event. Only seven in the ReDO™ group and four in the CAU group reported a third event and nobody reported a fourth. Similar events were reported in both groups. A manifest content analysis indicated that a job-related event was the most common one. Eighteen (33%) in the total group of 55 women had experienced changing, losing and/or getting a new job. Fourteen (25%) reported that illness, either their own or that of a close relative, had affected their everyday life, whereas 13 (24%) reported that someone close to them had died. Serious problems in close relationships, such as a divorce or a lasting family conflict, were reported by eight (15%). Fourteen events concerned a variety of both positive and negative experiences, such as getting a child or a grandchild, children moving out or in, the husband losing his job, or learning a new skill. A few of those 14 responses were too imprecise to categorize (such as “a friend”). Table 4 summarizes the categories of events per group and the participants’ ratings of their significance. As shown there, the groups did not differ on any aspect of reported life events. Both groups rated the importance of the life events around 4 on a five-step scale.

The significance of the events was not of importance to currently being off sick (p = 0.919 for the first reported event and p = 0.545 for the second).

Table 3
Ratings of well-being and everyday occupations among those who were interviewed at the long-term follow-up

ReDO group (n = 24) CAU group (n = 31) P-value

Self-mastery; mean (SD) 4 (9.5%) 3 (7.1%) 1.0

Self-esteem; mean (SD) 7.7 (25) 6 (22.6) 0.702

Satisfaction with daily occupations; mean (SD) 16 (45%) 13 (31%) 0.178

Occupational value; mean (SD) 65 (128) 43 (126) 0.109

Occupational balance; mean (SD) per occupational area¹⁾

Work 0.3 (0.7) 0.6 (0.9) 0.05

Leisure –0.7 (0.8) –0.5 (0.9) 0.435

Home chores 0.4 (0.9) 0.3 (0.8) 0.486

Self-care –0.4 (0.7) –0.4 (0.6) 0.693

General occupational balance 0.1 (0.8) 0.4 (0.6) 0.153

	ReDO group (n = 24)	CAU group (n = 31)	P-value
Self-mastery; mean (SD)	4 (9.5%)	3 (7.1%)	1.0
Self-esteem; mean (SD)	7.7 (25)	6 (22.6)	0.702
Satisfaction with daily occupations; mean (SD)	16 (45%)	13 (31%)	0.178
Occupational value; mean (SD)	65 (128)	43 (126)	0.109
Occupational balance; mean (SD) per occupational area¹⁾
Work	0.3 (0.7)	0.6 (0.9)	0.05
Leisure	–0.7 (0.8)	–0.5 (0.9)	0.435
Home chores	0.4 (0.9)	0.3 (0.8)	0.486
Self-care	–0.4 (0.7)	–0.4 (0.6)	0.693
General occupational balance	0.1 (0.8)	0.4 (0.6)	0.153

¹⁾Zero indicates balance; a value <0 indicates being under-occupied; a value >0 indicates being over-occupied.

Table 4

Significant life events since the one-year follow-up, as reported by the interviewees at the long-term follow-up

	ReDO group (n = 24)	CAU group (n = 31)	P-value
Work-related event; n (%)	9 (38%)	9 (29%)	0.457
Illness, own or a near person; n (%)	8 (33%)	6 (19%)	0.217
Death in one’s own family; n (%)	6 (25%)	7 (23%)	0.774
Problematic relationships; n (%)	4 (17%)	4 (13%)	0.657
Other; n (%)	8 (33%)	6 (19%)	0.228
Significance rating of the first mentioned event; mean (SD)	4.1 (1.2)	4 (0.9)	0.523
Significance rating of the second mentioned event; mean (SD)	4.2 (1.1)	3.9 (1.2)	0.647

4 Discussion

The lower one-year follow-up sick-leave rate in the ReDO™ group, compared to those who had received work rehabilitation according to CAU [2], was not maintained at the long-term follow-up. Thus, although the ReDO™ seemed to have speeded up return to work in the immediate follow-up perspective [2], the CAU had caught up at this long-term follow-up. However, the ReDO™ women perceived they had a better balanced working life. The CAU women were more often over-occupied with work tasks. This may be a lasting result of the ReDO™ intervention, the aim of which was that the participants should learn how to observe and analyze their own situation and develop strategies for how to maintain a balanced pattern of daily occupations [28]. These strategies were obviously not effective enough to maintain a group difference regarding sick leave at the long-term follow-up. This lack of a lasting effect on sick leave is similar to that found by Karlson and colleagues [26], and it is an open question whether booster sessions during a follow-up period could have prolonged the effect of the ReDO™ intervention. Nevertheless, very few in both the ReDO™ group and the CAU group were off sick on the index day and both groups had decreased their sick leave rate from the one-year to the long-term follow-up. The principal value of the ReDO™ seems to have been to accelerate the return to work process immediately after long-term sick leave [2].

It was obvious that the two intervention groups did not differ on well-being. Those who were off sick on the index day, as calculated on both groups, were however worse off regarding self-mastery, self-esteem and the frequency with which they felt they performed valued occupations. This seems like a logical finding; research has repeatedly shown that the working population rate their quality of life and well-being higher than those who are on sick leave or are unemployed [51 –53]. It has been argued that a good working life promotes well-being and that the benefits of work lie in its potential for promoting physical well-being and in people’s desire for social approval, which they may obtain by working [51]. This indicates that well-being is influenced by qualities of the workplace, but research has also pointed to a relationship in the reversed direction, that the individual’s well-being affects the work; a longitudinal study showed that people with low levels of well-being were more likely to have sickness benefit compensated days during the next-coming year [52]. The present findings also showed that previous stress and anxiety, as reported at the one-year follow-up, predicted sick leave at the long-term follow-up. Depression at the one-year follow-up was not, however, related to sick leave at the follow-up. This is mainly in line with findings from the one-year follow-up of the ReDO™ project, where both anxiety and depression predicted the return to work rate [54], but anxiety played a heavier role in explaining stress [55]. Otherwise depression has attracted the greater attention in the context of stress-related disorders, both as a consequence of a stressful work situation [56] and an aggravating circumstance with respect to return to work [57].

Both groups managed to decrease their sick-leave rate despite considerable turbulence in terms of life events that significantly affected their everyday lives. Job turnover, own illness, illness and deaths in the near environment and difficult relationships were common and affected a majority in both groups. The significance of these events, however, which was rated fairly high in both groups, was not related with sick-leave rate. This indicates the participants had a capacity for handling the events in such a way so as not to affect their work presence.

4.1 Methodological considerations

A weakness of this study was that there was a substantial dropout at the long-term follow-up interview. The circumstances that could explain changes in the participants’ sick-leave rate could not be investigated in the group as a whole, and this weakens the internal validity of that part of the study. On the other hand, the fact that register data were available for the whole sample renders robustness to the hard facts of sick-leave rate.

Some of those who declined to participate stated they did so because they felt they did not have the time. Possibly, those who became non-participants at the long-term follow-up may have had an even more complicated life situation than that described by the interviewees. This was indicated by the fact that a leading position in working life was more common among those in the ReDO™ group who dropped out from the long-term follow-up interview, compared to those who accepted to participate. The same was, however, not found in the CAU group. The tendency to decline to participate among the ReDO™ women with a leading position was possibly part of the strategy they had learnt during the intervention, namely to set limits in order to manage their everyday occupations [28].

The fact that the same research assistant performed the data collection at the one-year follow-up and the long-term follow-up means that any possible interviewer bias would have been minimized. There is always a risk, however, that a telephone interview hampers the exchange of information compared to a personal meeting, which was counteracted by using structured questions and also by sending the questionnaires to the respondents beforehand.

This study was based on a fairly small sample, particularly the interview part, which infers a risk of Type-II errors. Interpretations of non-findings, such as lacking differences between participants and dropouts, should thus be made with caution.

5 Conclusion

As one of very few, this long-term follow-up study has provided data regarding sick leave rate and life situation regarding well-being, everyday occupations and significant life events among women who had previously participated in work rehabilitation due to stress-related disorders. In all, both groups exhibited a positive development with regard to being off sick, and the superior effect found for the ReDO™ intervention at the one-year follow-up was levelled out at this long-term follow-up. This latter group showed, however, a tendency of applying the strategies they had developed during the intervention, as indicated by report of better occupational balance in the work domain compared to the CAU group and a tendency for women in leading positions in working life to decline to participate in the study due to time pressure. The small scale of this study warrants caution when interpreting the findings, but they may generate hypotheses for further research. It would, for example, be valuable to know whether a better balanced pattern of daily occupations would be a long-term effect of a lifestyle intervention such as ReDO™ also in future, larger studies.

Conflict of interest

None to report.

Footnotes

Acknowledgments

Anita Bjerle-Frisk is kindly acknowledged for having performed the data collection. The author declares no conflict of interests. No specific funding was allocated to this long-term follow-up.

References

Åsberg

, Nygren

, Leopardi

, Rylander

, Peterson

, Wilczek

, et al. Novel biochemical markers of psychosocial stress in women. PLoS One 2009;4(1):e3590.

Eklund

, Erlandsson

L-K

. Return to work outcomes of the Redesigning Daily Occupations (ReDO) program for women with stress-related disorders-A comparative study. Women Health 2011;51(7):676–92.

Lander

, Friche

, Tornemand

, Andersen

, Kirkeskov

. Can we enhance the ability to return to work among workers with stress-related disorders? BMC Public Health 2009;9:372.

van Oostrom

, van Mechelen

, Terluin

, de Vet

, Knol

, Anema

. A workplace intervention for sick-listed employees with distress: Results of a randomised controlled trial. Occup Environ Med 2010;67(9):596–602.

Ilmarinen

. Work ability–a comprehensive concept for occupational health research and prevention. Scand J Work Environ Health 2009;35(1):1–5.

Bertilsson

, Petersson

, Ostlund

, Waern

, Hensing

. Capacity to work while depressed and anxious–a phenomenological study. Disabil Rehabil 2013;35(20):1705–11.

SFS-2010:110. Socialförsäkringsbalk [Social Insurance Law] In Swedish Stockholm: Swedish Government; 2014 [cited 2014 14 February]. Available from: http://rkrattsdb.gov.se/SFSdoc/10/100110.PDF

Noordik

, Nieuwenhuijsen

, Varekamp

, van der Klink

, van Dijk

. Exploring the return-to-work process for workers partially returned to work and partially on long-term sick leave due to common mental disorders: A qualitative study. Disabil Rehabil 2011;(17-18):1625–35.

Rebergen

, Bruinvels

, Bos

, van der Beek

, van Mechelen

. Return to work and occupational physicians’ management of common mental health problems–process evaluation of a randomized controlled trial. Scand J Work Environ Health 2010;36(6):488–98.

10.

Brouwers

, Tiemens

, Terluin

, Verhaak

. Effectiveness of an intervention to reduce sickness absence in patients with emotional distress or minor mental disorders: A randomized controlled effectiveness trial. Gen Hosp Psychiatry 2006;28(3):223–9.

11.

Swedish Social Insurance Agency. Sjukfrånvarons utveckling. Socialfårsäkringsrapport 2014:12 [The development of sickness absence. Social Insurance Report 2014:12) In Swedish 2014 [16 September]. Available from: http://www.forsakringskassan.se/wps/wcm/connect/c7e1d070-4f90-4c1a-a1aa-ed28dd117e8c/socialforsakringsrapport-2014-12.pdf?MOD=AJPERES

12.

Hillborg

, Svensson

, Danermark

. Towards a working life? Experiences in a rehabilitation process for people with psychiatric disabilities. Scand J Occup Ther 2010;17(2):149–61.

13.

Landstad

, Hedlund

, Wendelborg

, Brataas

. Long-term sick workers experience of professional support for re-integration back to work. Work 2009;32(1):39–48.

14.

Muenchberger

, Kendall

, Grimbeek

, Gee

. Clinical utility of predictors of return-to-work outcome following work-related musculoskeletal injury. J Occup Rehabil 2008;18(2):190–206.

15.

Ahrberg

, Landstad

, Bergroth

, Ekholm

. Desire, longing and vanity: Emotions behind successful return to work for women on long-term sick leave. Work 2010;37(2):167–77.

16.

Hansen

, Edlund

, Henningsson

. Factors relevant to a return to work: A multivariate approach. Work 2006;26(2):179–90.

17.

Nielsen

, Madsen

, Bultmann

, Christensen

, Diderichsen

, Rugulies

. Predictors of return to work in employees sick-listed with mental health problems: Findings from a longitudinal study. Eur J Publ Health 2011;21(6):806–11.

18.

Holmgren

, Dahlin Ivanoff

. Women on sickness absence–views of possibilities and obstacles for returning to work. A focus group study. Disabil Rehabil 2004;26(4):213–22.

19.

Klanghed

, Svensson

, Alexanderson

. Positive encounters with rehabilitation professionals reported by persons with experience of sickness absence. Work 2004;22(3):247–54.

20.

Soderberg

, Jumisko

, Gard

. Clients’ experiences of a work rehabilitation process. Disabil Rehabil 2004;26(7):419–24.

21.

Swedish Social Insurance Agency. Social Insurance in figures 2014 2014. Available from: http://www.forsakringskassan.se/wps/wcm/connect/3107947a-6e12-4dcc-b1a6-e952a289ea95/sfis+2014-e.pdf?MOD=AJPERES

22.

Erlandsson

L-K

, Eklund

. Levels of complexity in patterns of daily occupations: Relationships to women’s well-being. J Occup Sci 2006;13(1):27–36.

23.

Meyer

. The philosophy of occupation therapy. Reprinted from the Archives of Occupational Therapy, Volume 1, 1-10, 1922. Am J Occup Ther 1977;31(10):639–42.

24.

Eklund

, Erlandsson

. Quality of life and client satisfaction as outcomes of the Redesigning Daily Occupations (ReDO) programme for women with stress-related disorders: A comparative study. Work 2013;46(1):51–8.

25.

Eklund

, Erlandsson

. Women’s perceptions of everyday occupations: Outcomes of the Redesigning Daily Occupations (ReDO) programme. Scand J Occup Ther 2014:1–9.

26.

Karlson

, Jonsson

, Osterberg

. Long-term stability of return to work after a workplace-oriented intervention for patients on sick leave for burnout. BMC Public Health 2014;14:821.

27.

Wastberg

, Erlandsson

, Eklund

. Client perceptions of a work rehabilitation programme for women: The Redesigning Daily Occupations (ReDO) project. Scand J Occup Ther 2013;20(2):118–26.

28.

Erlandsson

L-K

. The Redesigning Daily Occupations (ReDO) program: Supporting women with stress-related disorders to return to work-Knowledge base, structure, and content. Occup Ther Ment Health 2013;29(1):85–101.

29.

Pearlin

, Menaghan

, Lieberman

, Mullan

. The stress process. J Health Soc Behav 1981;22(4):337–56.

30.

Eklund

, Erlandsson

, Hagell

. Psychometric properties of a Swedish version of the Pearlin Mastery Scale in people with mental illness and healthy people. Nord J Psychiatry 2012;66(6):380–8.

31.

Sinclair

, Blais

, Gansler

, Sandberg

, Bistis

, LoCicero

. Psychometric properties of the Rosenberg Self-Esteem Scale: Overall and across demographic groups living within the United States. Eval Health Prof 2010;33(1):56–80.

32.

Rosenberg

. Society and the adolescent self image. Princeton, NJ: Princeton University Press; 1965.

33.

Oliver

, Huxley

, Priebe

, Kaiser

. Measuring the quality of life of severely mentally ill people using the Lancashire Quality of Life Profile. Soc Psychiatry Psychiatr Epidemiol 1997;32(2):76–83.

34.

Torrey

, Mueser

, McHugo

, Drake

. Self-esteem as an outcome measure in studies of vocational rehabilitation for adults with severe mental illness. Psychiatr Serv 2000;51(2):229–33.

35.

Eklund

, Argentzell

. Perception of occupational balance by people with mental illness: A new methodology. Scand J Occup Ther 2016;23(4):304–313.

36.

Wada

, Backman

, Forwell

. Theoretical perspectives of balance and the influence of gender ideologies. J Occup Sci 2010;17(2):92–103.

37.

Eklund

, Backstrom

, Eakman

. Psychometric properties and factor structure of the 13-item satisfaction with daily occupations scale when used with people with mental health problems. Health Qual Life Outc 2014;12(1):7.

38.

Eklund

, Morville

. Psychometric evaluation of the Danish version of Satisfaction with Daily Occupations (SDO). Scand J Occup Ther 2014;21(3):166–71.

39.

Eklund

, Gunnarsson

. Content validity, discriminant validity and responsiveness of the Satisfaction with Daily Occupations (SDO) instrument: A screening tool for people with mental disorders. Br J Occup Ther 2008;71(11):487–95.

40.

Eklund

, Erlandsson

, Persson

. Occupational value among individuals with long-term mental illness. Can J Occup Ther 2003;70(5):276–84.

41.

Eklund

, Erlandsson

, Persson

, Hagell

. Rasch analysis of an instrument for measuring occupational value: Implications for theory and practice. Scand J Occup Ther 2009;16(2):118–28.

42.

Persson

, Erlandsson

L-K

, Eklund

, Iwarsson

. Value dimensions, meaning, and complexity in human occupation-A tentative structure for analysis. Scand J Occup Ther 2001;8(1):7–18.

43.

Cicerali

, Cicerali

, Eklund

. Meanings Turkish people perceive in everyday occupations: Factor structure of Turkish OVal-pd. Int J Psychol 2015;50(1):75–80.

44.

Eakman

, Eklund

. Reliability and structural validity of an assessment of occupational value. Scand J Occup Ther 2011;18:231–40.

45.

Cohen

, Kamarck

, Mermelstein

. A global measure of perceived stress. J Health Soc Behav 1983;24(4):385–96.

46.

Eskin

, Parr

. Introducing a Swedish version of an instrument measuring mental stress. Reports from the Department of Psychology, No 813. Stockholm: Stockholm University; 1996.

47.

Eklund

, Backstrom

, Tuvesson

. Psychometric properties and factor structure of the Swedish version of the Perceived Stress Scale. Nord J Psychiatry 2014;68(7):494–9.

48.

Zigmond

, Snaith

. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67(6):361–70.

49.

Bjelland

, Dahl

, Haug

, Neckelmann

. The validity of the Hospital Anxiety and Depression Scale. An updated literature review. J Psychosom Res 2002;52(2):69–77.

50.

Krippendorff

. Content analysis: An introduction to its methodology. 3rd ed. Los Angeles; London: SAGE; 2013. xiv, 441 p.

51.

van der Meer

. Gender, unemployment and subjective well-being: Why being unemployed is worse for men than for women. Soc Indic Res 2014;115(1):23–44.

52.

Bertilsson

, Vaez

, Waern

, Ahlborg

Jr , Hensing

. A prospective study on self-assessed mental well-being and work capacity as determinants of all-cause sickness absence. J Occup Rehabil 2015;25(1):52–64.

53.

Paul

, Geithner

, Moser

. Latent deprivation among people who are employed, unemployed, or out of the labor force. J Psychol 2009;143(5):477–91.

54.

Eklund

, Wastberg

, Erlandsson

. Work outcomes and their predictors in the Redesigning Daily Occupations (ReDO) rehabilitation programme for women with stress-related disorders. Aust Occup Ther J 2013;60(2):85–92.

55.

Eklund

. Anxiety, depression and stress among women in work rehabilitation for stress-related disorders – The Redesigning Daily Occupations (ReDO) project. Int J Ment Health 2014;42(4):34–47.

56.

Tennant

. Work-related stress and depressive disorders. J Psychosom Res 2001;51(5):697–704.

57.

Brouwers

, Terluin

, Tiemens

, Verhaak

. Predicting return to work in employees sick-listed due to minor mental disorders. J Occup Rehabil 2009;19(4):323–32.

	The total sample (N= 84)		The interviewees (N= 55)
Characteristic	ReDO group (n = 42)	CAU group (n = 42)	ReDO group (n = 24)	CAU group (n = 31)
Age; mean (SD)¹⁾	45 (19)	46 (9)	43 (8)	40 (9)
Married or living with a partner; n (%)	30 (71%)	27 (64%)	15 (63%)	21 (68%)
Number of children; mean (SD)	2.4 (1.4)	2 (1)	2.5 (1.1)	2 (0.8)
Type of occupation^2,3)
Managers and professionals	15 (36%)	21 (50%)	4 (17%)	17 (55%)
Technicians and associate professionals	13 (31%)	6 (14%)	10 (42%)	4 (13%)
Clerical support, service and sales workers	13 (31%)	15 (36%)	9 (38%)	10 (32%)
Plant and machine operators	1 (2%)	–	1 (4%)	–
Having a university degree; n (%)	16 (40%)	21 (51%)	7 (30%)	16 (52%)
Percentage of baseline sick leave; mean (SD)	92 (18)	83 (26)	90 (21)	78 (30)
Sick leave (months) before baseline; mean (SD)	13 (20)	10 (10)	11 (20)	11 (11)
First diagnosis
Depression	19 (45%)	23 (54%)	12 (50%)	15 (48%)
Stress/Exhaustion	20 (48%)	17 (41%)	10 (42%)	14 (45%)
Physical main diagnosis	3 (7%)	2 (5%)	2 (8%)	2 (7%)