Return to Work: Promoting Health and Productivity in Workers With Common Mental Disorders

Common mental disorders, such as adjustment, anxiety and depressive disorders are highly prevalent in the working population and are associated with impaired work functioning and high sick leave rates. For mental health disorders, several established treatments exist, such as Cognitive Behavior Therapy (CBT), pharmacotherapy, and physical activity. However, less evidence is available on which treatments that increase an individual's ability to return to work (RTW) when he/she has a common mental disorder. In particular, the effectiveness of a psychotherapeutic intervention for RTW is largely unknown even though these types of interventions are common and are recommended by the National Board of Health and Welfare for common mental disorders in Sweden. The few studies in which psychotherapeutic interventions (mostly CBT) have been evaluated indicate that these were equally or less effective in enhancing RTW compared to other interventions. In the Swedish rehabilitation guarantee, CBT-treatments are subsided based on the assumption that improved health status will contribute to earlier RTW. However, the results from the first evaluations of the rehabilitation guarantee point to the contrary. RTW was actually delayed for CBT for common mental disorders compared to treatment as usual (TAU) (5).

Traditionally, CBT manuals have been oriented towards reducing symptoms. Likewise, outcome measures generally consist of symptom-based scales. In a recent study, a specific RTW CBT-manual was developed including targeting return to a work context, resulting in earlier RTW. Self-reported mental health symptoms were reduced to a similar extent as in regular CBT. This implies a room for improvement in the CBT-manuals through orienting the treatment towards specific areas of functioning, without a loss of symptom specific improvement.

Even though CBT has proven effective for several mental conditions, little is known about why the interventions lead to change or how the change came about. This is especially true when it comes to RTW. There are various treatment intents with varying results, however, little or nothing is known beyond subjective reasoning about the active processes in treatment, mediators that might lead to reduced sickness absence. Moderators refer to characteristics that influence the direction or magnitude of the relation between the intervention and outcome. There are several studies discussing what factors might predict sickness absence. However, to help us understand how a treatment works, for whom it works and under which conditions, a more specific focus on mediators as well as moderators within a study for reducing sick leave is needed.

One construct that would appear useful in understanding and facilitating RTW when CBT is implemented is self-efficacy. In short, self-efficacy is the belief that an individual has in his/her capacity to perform a specific behavior successfully. When applied to RTW, people with low self-efficacy would believe that they might fail to fulfill their work demands or work role. These efficacy cognitions are expected to be prominently present among those with mental health problems, as mental disorders often erode a positive self-concept by the very nature of the disorder. Lagerveld et al have developed a self-efficacy questionnaire specifically oriented to capture self-efficacy expectations regarding RTW and return-to-work self-efficacy for sick listed employees with mental health problem. RTW-SE has been proven a robust predictor of actual return to work, however if it also serves as a mediator of change remains to be explored.

Another intervention model with some support for increased RTW is the inclusion of a workplace intervention (WI) in a rehabilitation program. In a Dutch RCT, Blonk et al compared CBT performed by trained therapists with treatment by "labor experts" who had had a brief instruction in CBT principles, with controls. They found a significantly better RTW in the labor expert group compared to CBT, which did not differ from the controls. In a Swedish study, a manualized WI was evaluated and found significantly better compared to (non-randomized) controls.

A third method that is interesting in relation to RTW due to its primary focus on improved function rather than symptom reduction is the CBT-method, Acceptance and Commitment Therapy (ACT). According to the theoretical frame of ACT, psychological discomfort is the result of experiential avoidance. Experiential avoidance implies attempts of avoiding painful thoughts, emotions and physical sensations even when these attempts counteracts effective behavior in terms of living in accordance with personal values. High levels of experiential avoidance have been associated with psychopathology including depression, anxiety and low quality of life. The clinical goal in ACT is to reduce experiential avoidance and instead enhance psychological flexibility which implies the development of a broader and more flexible behavior in accordance with personal values, also in the presence of negative stimuli. Several studies have shown that a general measure of this psychological process predicts a wide-range of work-related outcomes, from mental health and work attitudes to job performance and absence rates. A side effect of living a more functional life is often that self-reported psychological symptoms decline. According to ACT theory, psychological flexibility would be the key process of change, however there is no data available in the present to support this statement in relation to RTW.

The duration of sick leave is also influenced by many other factors such as socio-demographic, medical, work-related and organizational factors, as well as by factors in the health-care and legislative systems. Diagnosis such as burnout, depression and anxiety disorders are associated with an increased risk of long-term sickness absence. Previous sickness absence increases the risk for long-term sickness absence also after adjustment for socio-demographic factors and self-reported health status. This may imply that short-term sickness absence has social and health-related consequences beyond the effects of the ill health it reflects. However, these findings may be further explored by taking severity of diagnosis into account, and need to be evaluated in a comprehensive study.

Based on the above summary on current knowledge, the primary aims of the present study is to evaluate the efficacy of two different interventions, ACT and a workplace intervention (WPI), both as standalone treatments and combined, and compare these to TAU. The second aim is to examine different variables that moderate program efficacy and mediates change in RTW for the different interventions. The third aim is to evaluate cost-effectiveness.

General research questions in this project includes

1. What is the efficacy and cost-effectiveness of two different rehabilitation models, one based on psychotherapy, ACT, and the other on workplace-interventions, WPI , when these are offered as standalone interventions and in combination for patients with adjustment, anxiety and depressive disorders?
2. Examination of mediators and moderators that might explain differences in RTW for individuals with common mental disorders participating in rehabilitation interventions?
3. Does actual use of prescribed selective serotonin re-uptake inhibitors (SSRI) or Serotonin-norepinephrine reuptake inhibitors (SNRI) increase RTW compared to non-use of prescribed SSRI or SNRI among patients with adjustment, anxiety and depressive disorders? Methods The study is an RCT with repeated measures and a mixed design. Measures are taken at pre- and post-treatment and at 6, 12, 24 and 60 months follow-up. Participants are recruited from the National Insurance Office (NIO) starting in February 2013. Individuals ages 18-60 living in the county of Stockholm with at least 50% employment rate applying for sick pay due to common mental disorders are consecutively invited to an diagnostic interview and assessment during the inclusion period. Individuals that meet the inclusion criteria and gives informed consent are invited to participate. A power analysis based on a study on the WPI by Björn Karlsson gives that in order to get 80 % power, 72 participants is required in each group. 320 participants will be randomized to one of 4 groups.

The interventions:

1. ACT; The ACT intervention consists of 6 manual-based face-to-face sessions and internet-based homework modules. The manual is based on the six core processes in the ACT-model: acceptance, mindfulness, defusion, self as context, values and committed action.
2. WPI; This interventions aims at the facilitation of dialogue between the participant and the workplace through a series of steps consisting of individual interviews with the participant and his/her nearest supervisor and a so called "convergence dialogue meeting" in order to agree upon short- and long-term solutions.
3. ACT + WPI; In this group, the participants will collaterally receive the ACT and the WPI intervention.
4. Control group; Treatment as usual (TAU) which means that the participant continues in ordinary health care and does not receive interventions other than the initial assessment.

Independent variables:

1. Consists of the within-group variable Time with five measurement points: pre-, and post treatment, 6, 12, 24 and 60 months.
2. Consists of the between-groups variable Treatment condition with four conditions as described above.
3. Consists of the between-groups variable Use or non-use of prescribed SSRI or SNRI reported by the patients in a questionnaire.

Dependent variables Primary outcome measure is RTW based on register data from the NIO, self-reported data regarding short-term absence (periods of less than 14 days) and self-reported work ability according to the Work Ability Index (WAI).

Secondary outcome measures are symptom severity and general function. Depression is measured with MADRS-S (Montgomery Åsberg Depression Rating Scale) and HAD - subscale Depression (Hospital Anxiety and Depression Scale), anxiety with HAD - subscale Anxiety, Burnout symptoms with Karolinska Exhaustion Disorder Scale (KEDS). General function as measured by the Work and Social Adjustment Scale and general satisfaction with life with the Satisfaction with life scale (SWLS).

Measures of mediators will be taken in connection with treatment sessions and include: the Return To Work Self Efficacy scale (RTW-SE), The work-related acceptance and action questionnaire (WAAQ) and Bull´s Eye Values Survey (BEVS).

Measures of moderators include: demographic data, sick days in the past five years, pre scores on self-rated symptoms (MADRS-S, HAD, KEDS).

Statistical analyses Efficacy will be assessed with Mixed models for repeated measures. Mixed regression models are a powerful way to conduct an intent-to-treat analysis. Unlike normal analysis of variance/analysis of covariance, these models use all available data from all participants and take into account the obtained outcome and missing data. Mediational analyses will be based on Baron \& Kenny´s requirements for testing mediation. Cost-effectiveness are based on a combination of cost-effectiveness analyses and cost-utility-analyses according to the Manual for Assessment of Cost-effectiveness-analysis in REHSAM based on contemporary guidelines in health economic science practice.

This research provides an opportunity to better understand the process of RTW for individuals on sick-absence due to common mental disorders. A theoretical understanding of the mechanisms behind treatment change, which may differ for different subgroups, is important to ultimately be able to maximize or change treatments for individuals on sick-leave. This study might significantly contribute to answer real everyday questions that clinicians face as well as policy makers and agencies working with the task to increase RTW and promote public health