REHABILITATION FOR PATIENTS WITH BURNOUT

Therese Stenlund

Department of Public Health and Clinical Medicine, Occupational and Environmental Medicine, Umeå University 901 87 Umeå Umeå 2009

Copyright©Therese Stenlund New series No. 1237 ISSN: 0346-6612 ISBN: 978-91-7264-711-4 Printed by: Arkitektkopia Umeå, Sverige 2009

Grant me the serenity To accept the things I cannot change, Courage to change the things I can, And the wisdom to know the difference

Reinhold Niebuhr 1926

CONTENTS

ABSTRACT...... 7 SVENSK SAMMANFATTNING ...... 9 ABBREVIATIONS ...... 11 ORIGINAL PAPERS ...... 12 INTRODUCTION...... 13 Prevalence of burnout...... 13 Definition of burnout...... 14 Causes of burnout...... 15 Measurements of burnout...... 16 Work capacity and burnout...... 17 Rehabilitation...... 17 AIMS OF THE THESIS ...... 23 Specific aims ...... 23 METHODS ...... 24 Settings and participants ...... 24 Data sources ...... 25 REST ...... 25 QIST ...... 28 MONICA ...... 29 Ethical approval...... 30 Procedure...... 30 REST ...... 30 QIST ...... 30 MONICA ...... 31 Data ...... 31 Psychological variables, health-related quality of life and perceived relaxation ...... 31 Burnout ...... 31 Anxiety, depression and obsessive-compulsive symptoms...... 31 Stress behaviour ...... 32 Fatigue ...... 32 Self-esteem ...... 32 Health-related quality of life...... 32 Perceived relaxation...... 32 Work characteristics and work-related stress...... 33 Social support ...... 33 Sick leave...... 34 Physical measurements ...... 34 Balance...... 34 Physical capacity and activity ...... 34 Interview transcriptions ...... 35 Rehabilitation...... 35 Cognitively-oriented Behavioural Rehabilitation (CBR)...... 35 Qigong...... 35 Basic Care...... 36 Statistics and analysis ...... 36 Statistics...... 36 Grounded theory...... 37 RESULTS ...... 38 Description of burnout patients in relation to gender and a general population...... 38 Rehabilitation programs...... 38 Dropouts...... 38 Compliance ...... 38 Complementary and conversational treatment ...... 38 Effects of rehabilitation...... 39 DISCUSSION ...... 46 Description of burnout patients in relation to gender and a general population...... 46 Effects of rehabilitation ...... 47 Work capacity and burnout...... 49 Methodological considerations ...... 50 Clinical implications and further research...... 52 CONCLUSIONS...... 55 ACKNOWLEDGEMENTS ...... 56 REFERENCES...... 58

ABSTRACT

Stress-related diseases and burnout have increased in Sweden during the last decades. In 2006, the most common diagnoses for new cases of sickness compensation were mental and behavioural disorders in both women and men. In spite of the large group of people seeking care for and on long-term sickness absence due to stress-related diseases and burnout, there is no agreement on which treatment they should be offered. The overall aim of this thesis was to describe patients on long- term sick leave because of burnout and to evaluate rehabilitation programs for this patient group. Two patient samples were recruited from the Stress Clinic at the University Hospital in Umeå, Sweden: REST (Rehabilitation for stress- related disease and burnout; n=136) and QIST (Qigong for stress-related disease and burnout; n=82). A general population sample was from the 2004 Northern Sweden MONICA survey (n=573). Patients in REST were randomised into a 1-year rehabilitation program to either program A (Cognitively-oriented Behavioural Rehabilitation (CBR) and Qigong), or to program B (Qigong alone). In Paper I, baseline data were compared with data from the MONICA sample. In paper II, programs A and B were compared regarding effects on psychological variables and sick leave rates, and in Paper III, 18 patients from program A and B were interviewed to explore subjective experiences of the rehabilitation programs. Patients in QIST were allocated to an intervention with Qigong twice a week for 12 weeks or a control group. Psychological and physical measurements were assessed in QIST. Data were collected by questionnaires, physical measurements, the register on sick leave, and interviews. Patients with burnout reported a more restricted social network and higher work demands than the general population. In relation to women from a general population, women with burnout more often worked “with people”, reported high job strain, a more sedentary work situation and less emotional support. A per-protocol analysis showed no significant differences in treatment effect between program A and B in REST or between the intervention and control group in QIST. All groups improved significantly over time with reduced levels of burnout, anxiety, depression, and fatigue. In REST, lower scores on obsessive-compulsive symptoms, stress behaviour, and sick leave rates were found in both programs and in QIST both groups increased dynamic balance and physical capacity. In an intention-to-treat analysis, patients in program A in REST had significantly fewer obsessive-compulsive symptoms, and larger effect sizes in stress behaviour and obsessive-compulsive symptoms compared to patients in program B. Patients in both REST programs perceived that the 1-year rehabilitation program gave them specific tools to use in secondary prevention. They also emphasised that the good encounters, affirmation and group cohesiveness they perceived during the

7 rehabilitation was a necessary basis for initiation of a behavioural change leading to recovery. In conclusion, compared to a general population, patients with burnout perceived more demands at work and less social support. Lack of emotional support seemed to be more associated with burnout among women. There were no differences in effect between CBR and Qigong compared to Qigong alone, or between a 12 week Qigong intervention compared to a control condition. Improvements were found in all groups in the rehabilitation programs. CBR combined with Qigong have some advantages compared to Qigong alone. An environment with good encounters and affirmation of the patients was experiences as important by the patients and group rehabilitation had advantages as recognition and support from the group. Early rehabilitation measures are important to prevent long-term sickness absence. In future rehabilitation programs it might be necessary to have a more individualized approach and choose treatments preferred by the patient.

Keywords: Burnout, social support, work demands, population surveillance, epidemiology, randomised controlled trials, cognitive therapy, sick leave, mind-body and relaxation techniques, anxiety, depression, grounded-theory analysis.

8 SVENSK SAMMANFATTNING

Stressrelaterade sjukdomar och ”utbrändhet” har ökat i Sverige de senaste årtionden. Psykiska sjukdomar och syndrom var den största gruppen bland nybeviljade sjukersättningar/aktivitetsersättningar hos både män och kvinnor under 2006. Trots att många söker hjälp för och är sjukskrivna på grund av stressrelaterade sjukdomar och ”utbrändhet” finns det ännu ingen samlad evidens om vilken rehabilitering som bör erbjudas. Det övergripande syftet med denna avhandling var att beskriva patienter, som är långtidssjukskrivna på grund av ”utbrändhet” samt att utvärdera olika rehabiliteringsprogram för denna patientgrupp. Två patientgrupper rekryterades från Stresskliniken, Norrlands Universitetssjukhus, Umeå, Sverige; REST (Rehabilitering vid stress och utbrändhet) (n=136) och QIST (Qigong vid stress och utmattningssyndrom) (n=82) samt en grupp ur befolkningen i samma område; 2004 års MONICA- studie i Norra Sverige (n=573). Patienterna i REST randomiserades till en 1-årig rehabilitering, antingen enligt program A) Kognitivt beteendeinriktad rehabilitering (KBR) och Qigong eller program B) enbart Qigong. I studie I jämfördes baslinjedata från REST med MONICA populationen. I studie II jämfördes program A och B med avseende på psykologiska variabler och sjukfrånvaro och i studie III intervjuades 18 patienter från program A och B för att fånga deras upplevelser av den 1-åriga rehabiliteringen. Patienterna i QIST randomiserades till en interventionsgrupp med Qigong två gånger/vecka i 12 veckor eller till en kontrollgrupp. Effekter av Qigong på både psykologiska och fysiska variabler studerades. Data insamlades med frågeformulär, genom fysiska tester, ur försäkringskassans register över sjukfrånvaro och genom enskilda intervjuer. I jämförelse med befolkningen rapporterade patienter med ”utbrändhet” att de hade ett mera begränsat socialt nätverk och upplevde högre krav i arbetet. Kvinnor med ”utbrändhet” arbetade i större utsträckning i människorelaterade yrken, upplevde arbetet mer psykiskt belastande, hade ett mera stillasittande arbete och upplevde ett sämre emotionellt stöd jämfört med kvinnor i befolkningen. I en per-protokoll analys kunde ingen signifikant skillnad i behandlingseffekt påvisas mellan program A och B i REST eller mellan intervention eller kontrollgruppen i QIST. Alla grupper förbättrades över tid med minskade nivåer av utbrändhet, ångest, depression, och trötthet. Dessutom förbättrades båda programmen i REST med minskade nivåer av tvångssymtom, stressbeteende och grad av sjukskrivning och båda grupperna i QIST förbättrades med ökad dynamisk balans och fysisk kapacitet. I en intention-to treat analys påvisades en signifikant skillnad i REST där patienter i program A hade färre tvångssymtom, samt större effect-size i stressbeteende och tvångssymtom i jämförelse med patienter i program B. Patienterna i båda programmen i REST upplevde att den 1- åriga rehabiliteringen gav dem redskap, som de kunde använda för sitt

9 tillfrisknande. De poängterade också att det positiva bemötandet och den bekräftelse de mötte både av personal och gruppmedlemmar hade stor betydelse för att initiera en beteendeförändring. Sammanfattningsvis skiljde sig patienter med ”utbrändhet” från en befolkningsgrupp genom att de upplevde högre krav i arbetet och lägre socialt stöd. Brist på emotionellt stöd hade större påverkan på kvinnor med ”utbrändhet”. Ingen skillnad i behandlingseffekt kunde påvisas mellan KBR och Qigong och enbart Qigong, ej heller mellan 12 veckors Qigong jämfört med en kontrollgrupp. Alla grupper i de olika rehabiliteringsprogrammen förbättrades över tid. KBR i kombination med Qigong hade vissa fördelar i jämförelse med enbart Qigong. En miljö med gott bemötande och bekräftelse upplevde patienterna som viktig och rehabilitering i grupp hade fördelar som igenkännande och stöd i gruppen. Tidiga rehabiliteringsinsatser är viktiga för att förebygga långtidssjukskrivning. I framtiden kan det vara viktigt att individualisera rehabiliteringen för att tillgodose patientens behov.

10 ABBREVIATIONS

AVAT Availability of Attachment AVSI Availability of Social Integration CBI Copenhagen Burnout Inventory CBR Cognitively-oriented Behavioural Rehabilitation CBT Cognitive Behavioural Therapy CIS Checklist Individual Strength questionnaire CPRS-S-A Self-Rating Scale for Affective Syndromes DSM Diagnostic and Statistical Manual of Mental Disorders ELSS Everyday Life Stress Scale HAD Hospital Anxiety and Depression scale HRQL Health-Related Quality of Life ICD-10 International Classification of Diseases, Tenth Revision ISSI Interview Schedule for Social Interaction MBI Maslach Burnout Inventory MCS Mental Component Summary score MONICA Multinational MONItoring of Trends and Determinants in CArdiovascular disease OLBI Oldenburg Burnout Inventory PCS Physical Component Summary score QIST QIgong for Stress-related disease and burnout RCT Randomised Controlled Trials REST REhabilitation for Stress-related disease and burnout SCQ Self-Concept Questionnaire SF-36 Short Form 36 item health survey SMBQ Shirom-Melamed Burnout Questionnaire S-PARI Swedish version of the Physical Assessment Scale s-USM self-reported exhaustion syndrome Vo2max Maximal oxygen uptake WHO World Health Organization

11 ORIGINAL PAPERS

This thesis is based on the following papers, which will be referred to by their Roman numerals I-IV:

I Stenlund T, Ahlgren C, Lindahl B, Burell G, Knutsson A, Stegmayr B, Slunga Birgander L. Patients with burnout in relation to gender and a general population. Scandinavian Journal of Public Health, 2007; 35: 516-523.

II Stenlund T, Ahlgren C, Lindahl B, Burell G, Steinholtz K, Edlund C, Nilsson L, Knutsson A, Slunga Birgander L. Cognitively oriented behavioral rehabilitation in combination with Qigong for patients on long-term sick leave because of burnout: REST – a randomized clinical trial. International Journal of Behavioral Medicine, Published online, 16 January 2009.

III Fjellman-Wiklund A, Stenlund T, Steinholtz K, Ahlgren C. To take charge – Patients’ experiences in a rehabilitation program for burnout. Manuscript

IV Stenlund T, Slunga Birgander L, Lindahl B, Nilsson L, Ahlgren C. The effects of Qigong in patients with burnout: a randomized controlled trial. Submitted

Article I. © 2007 by SAGE Publications Ltd Article II. Reproduced by courtesy of ISBM and Springer

12 Introduction ______

INTRODUCTION

Despite an increasing amount of research and articles on burnout published there is no consensus on how to define burnout, if burnout is work-related, or how burnout should be measured (1).

Prevalence of burnout The prevalence of burnout has been studied in the general population (2, 3), the working population (2, 4, 5, 6) and in specific workplaces (7-9). In a Swedish population-based study, burnout was measured with Pines Burnout Measure (BM). Burnout was found in 7 % 0f the general population and in 5.7 % of the individuals who were working (2). Using the Maslach Burnout Inventory- General Survey (MBI-GS), 25.2 % of employees were found to suffer from mild burnout and 2.4 % from severe burnout in a Finnish study (4). Lindblom et al. (5) studied working individuals in Sweden and burnout was measured with the MBI-GS. Burnout groups were created using cut-off points based on the 25th and 75th percentiles for each of the three dimensions. About 19 % of the individuals belonged to a low level group of burnout and about 18 % to a high level group of burnout (5). Another recent study from Sweden consisting of working individuals also measured burnout with MBI-GS, but only used the emotional exhaustion subscale. Individuals who scored higher than the 75th percentile were classified as having emotional exhaustion symptoms. This was found in 15.7 % of the men and 21.8 % of the women (6). Soares et al. (3) found that 21 % of the women in a population study experienced high burnout when burnout was defined as ≥ 4 points on the Shirom-Melamed Burnout Questionnaire (SMBQ). Prevalence of burnout in human services professionals in the Netherlands was estimated to 3-16 % based on cut-off points of the MBI (7). Of 63 white-collar female employees in Sweden, 43 were allocated to a high level burnout group based on ≥ 3.75 points on the SMBQ (8). Peterson et al. (9) studied burnout in healthcare workers in Sweden. Burnout was measured with the Oldenburg Burnout Inventory (OLBI) and 33.6 % were allocated to have both high exhaustion and high disengagement. Compared to other occupations on the labour market, burnout is more common in social workers, physicians, nurses and especially among teachers (2, 7). Burnout is reported to be more common among both younger (3, 10, 11) and older workers (5, 11). Some studies reported that having a high level of burnout was associated with being a woman (5, 6, 10, 12). Women have been found to report higher scores on exhaustion and men on cynicism (10, 13). One reason for this could be confounding with occupation because more women work in human services (10, 14). Men who are single seem to experience higher burnout than married men (10, 13). Women who were divorced or single reported higher burnout scores than married women (3).

13 Introduction ______

Definition of burnout Burnout was first described by Herbert J. Freudenberger (15) in 1974. Freudenberger worked as an unpaid psychiatrist in the alternative health care clinic and described burnout based on his own and other volunteer workers’ experiences. He stated that burnout was accompanied by a feeling of exhaustion and fatigue in combination with different physical symptoms (for example headaches, gastrointestinal disturbances, sleeplessness, and shortness of breath). Burnout people also looked, acted and seemed depressed. Freudenberger also reported that people who were prone to burnout worked too much, too long and too intensely, they had a need to achieve, and often performed monotonous work (15). About the same time, Christina Maslach (16) studied about 200 professionals in the health and social services by collecting questionnaires and performing interviews. She found that the professionals were often unable to cope with emotional stress at work and burnout was a result (16). Maslach et al. (17) defined burnout as a work-related syndrome consisting of three dimensions that occurred among individuals who work with people: emotional exhaustion, depersonalization, and reduced personal accomplishment. Emotional exhaustion refers to feelings of depleted emotional resources. Depersonalization refers to negative, cynical attitudes about clients. Finally, reduced personal accomplishment refers to a tendency to evaluate one’s own work effort negatively, with feelings of low self-esteem in the work situation. Pines and Aronson (18) define burnout more broadly as a state of physical, emotional and mental exhaustion caused by long periods in emotionally demanding situations. Burnout is characterized by physical depletion and fatigue, feelings of helplessness and hopelessness, emotional drain, development of negative attitudes to one’s self, work, others and life. Schaufeli and Enzmann (7) formulated a working definition of burnout that is characterised by the core indicator exhaustion which is accompanied by four general symptoms: distress, a sense of reduced effectiveness, decreased motivation, and dysfunctional attitudes and behaviours at work. Burnout is thought to develop gradually over time in individuals who do not suffer from psychopathology. Schaufeli and Greenglass (14) later defined burnout as a state of physical, emotional and mental exhaustion resulting from long-term involvement in emotionally demanding work situations. Shirom (19) characterized burnout as a chronic depletion of an individual’s energetic resources, including physical fatigue, emotional exhaustion and cognitive weariness. Burnout is a reaction to long-time exposure to stress at work. In Sweden, the concept of burnout has been replaced by a new diagnosis called the exhaustion syndrome by the Swedish health care and social welfare system. In 2003, the Swedish National Board of Health and Welfare created diagnostic criteria (20) (Table I) for the exhaustion syndrome (ICD-10, F43.8A) (21) according to ICD and DSM criteria. The

14 Introduction ______

exhaustion syndrome is characterized by mental exhaustion and reduced endurance as a result of identifiable stress factors for at least six months. The stress factors can be identified both in and outside of work situations.

Table I. Diagnostic criteria for exhaustion syndrome from the National Board of Health and Welfare in Sweden. All capital letters must be fulfilled for the diagnosis. A Physical and psychological exhaustion for at least two weeks. The symptoms should be developed as a consequence of one or several stressors during at least six months. B Evident lack of mental energy in the form of reduced initiative, reduced endurance, or an extended time for recovery after mental stress. C At least four of the following symptoms almost every day during a two week period: 1 Difficulties with concentration or memory 2 Reduced ability to handle demands or to work with time pressure 3 Emotionally unstable 4 Disturbed sleep 5 Physical weariness 6 Physical symptoms as pain, chest pain, palpitations, digestive complaints, dizziness or sound hypersensitivity D Symptoms should cause clinical suffering or reduced capacity at work, in social life or in other important respects. E The condition is not caused by substances or somatic disease. F If criteria are fulfilled for depressive episode, dysthymia or anxiety disorder, the exhaustion syndrome will be reported as a secondary diagnosis.

Causes of burnout In initial studies, burnout was described as a work-related syndrome in human services and health care (17). Later it was also reported from work beyond human services and in all types of occupations (7, 10, 19). Schaufeli and Taris (22) pointed out that burnout also can be applied to work-like activities such as those of students, artists or volunteers. High demands, low control and low social support at work are work environment factors that have been associated with burnout (5, 6, 8, 10). There are different theories or models that increase the understanding of the relationship between work-related stress and burnout. Examples of such models are the demand-control model (23), the effort-reward imbalance (ERI) model (24), and the job demands-resources (JD-R) model (25). The demand-control model consists of two dimensions: job decision latitude and psychological demands. A high job strain situation is defined as a high level of psychological demands combined with a low level of decision latitude. Individuals who are exposed to high job strain have an increased risk for psychological strain and mental illness (23, 26, 27). Work-related social support has been added to the demand-control model (28), and further improves the model’s ability to predict illness.

15 Introduction ______

The ERI assesses “high cost/low gain” at work and if there is an imbalance between high effort spent and low reward received there will be an increased risk for poor health (24, 27, 29). The JD-R model consists of the two categories: job demands and job resources. An imbalance with high job demands and insufficient job resources implies a stress situation and increases risk of developing burnout (25). Job resources are found to buffer against the burnout risk from high job demands (14, 30). Effects of burnout have also been found in workmates. Female co-workers were torn between a desire to help their workmate and a necessity to manage their own work (31). There are different views of whether burnout is work-related or not. Evidence shows that burnout can occur outside work in connection with life events like marriage (32) and in mothers of severely ill children (33). Struggling to fulfil responsibilities towards work, family and other peoples are experiences reported by many patients during the time preceding burnout (34). Middle aged women with a double burden of high job strain and a large amount of unpaid work have an increased risk for poor self-reported health (35), and for emotional exhaustion when more time is spent on childcare (36). In comparison with a general population, patients with burnout report a more restricted social network and women report less emotional support (37). Personality traits have also been identified as important factors for burnout. Low levels of hardiness, an external locus of control, an avoidant or passive coping style, poor self-esteem, a sensitive type, type-A behaviour and neuroticism are qualities suggested to make the individual more sensitive to strain and to psychological distress (10, 19).

Measurements of burnout The most frequently used measurement in scientific studies on burnout is the Maslach Burnout Inventory (MBI). The MBI assesses the three core dimensions of emotional exhaustion, depersonalization, and reduced personal accomplishment. Three different versions of the MBI are used: MBI-Human Services Survey (MBI-HSS), MBI-Educators Survey (MBI-ES), and MBI- General Survey (MBI-GS) (10, 17). Pines et al. (38) constructed the Burnout Measure (BM) which is a reliable and reasonably valid instrument and often used as a one- dimensional questionnaire (7). The BM can also be divided in three subscales: demoralisation, exhaustion, and loss of motive. High correlation has been found between the emotional exhaustion subscale of the MBI and the BM three subscales (2). In this thesis, the Shirom-Melamed Burnout Questionnaire (SMBQ) was used in all papers. The SMBQ contains 22 items, each rated on a 7- point scale. The SMBQ comprises the subscales emotional and physical exhaustion, tension, listlessness, and cognitive weariness. An overall index was computed as the mean of all items. A higher score indicates a higher level of burnout (39-41). The SMBQ has been shown to correlate highly with the emotional exhaustion subscale of the MBI (r = .77) and with the BM (r = .87) (8).

16 Introduction ______

The Oldenburg Burnout Inventory (OLBI) includes the two dimensions exhaustion and disengagement (25). Validity and reliability have been confirmed of the OLBI (42). The Copenhagen Burnout Inventory (CBI) consists of three sub- dimensions: personal burnout, work-related burnout, and client-related burnout, and can be measured in various domains. The CBI is reported to have acceptable reliability and validity (43). According to the criteria (20) (Table I) for the exhaustion syndrome (ICD-10, F43.8A) (21) in Sweden a new questionnaire has been constructed called the self-reported exhaustion syndrome (s-USM) (44), that is currently being evaluated. To summarize, there are many instruments that measure burnout. Many researchers agree that the central aspect of burnout is emotional exhaustion (1, 19, 36), or the two dimensions exhaustion and depersonalization/disengagement (22, 42). The lack of agreement on whether burnout should be regarded as work-related or a general phenomenon remains. Depending on which opinion the researcher uses, different instruments are employed to measure burnout. However, researchers have pointed out the importance of finding a unanimous and clear definition of burnout in order to understand how to measure burnout (1, 45).

Work capacity and burnout In Sweden mental and behavioural disorders as the cause of new sickness compensations increased from 28 % to 40% between 2002 and 2006. These disorders were also the most common diagnosis in both women and men seeking sickness compensation in 2006 (46). In the Fourth European Working Conditions Survey (2005) 22 % of the EU workers reported fatigue and/or stress symptoms (47). In a 2008 survey on work-related disorders in Sweden, the most common disorders for women were stress or other types of mental strain (10.6 %), followed by strenuous working postures (7.7 %). For men, strenuous working postures were the most common disorders (6.3 %) and stress and other types of mental strain the second common reason (6.1 %) (48).

Rehabilitation Most research on rehabilitation of individuals suffering from burnout has been done among employees who remain at work. There are randomised controlled trials (RCT) using different measures, e.g. interventions for mindfulness-based stress reduction (49), comparison between cognitive behavioural therapy and yoga (50), web-based stress management (51), and peer-support groups (52), and all shown to have positive effect. However, there is still insufficient evidence to determine which type of rehabilitation should be offered to patients on sick leave because of burnout. Table II shows an overview of the different RCT for patients on sick leave due to stress-related illness or burnout.

17 Introduction ______

Cognitive behavioural therapy (CBT) is a method commonly used in the treatment of various psychological and psychiatric disorders. Historically, CBT has evolved through three phases. The first phase started in 1920 with the investigation of classic and operant conditioning in animals and humans, i.e. the basics in learning. Based on the principles derived from this research, behaviour therapy (BT) was developed. In 1970 the second phase put cognition in focus and cognitive therapy (CT) was developed by Aaron T. Beck and Albert Ellis. They found that psychological/psychiatric disturbance was a result of both faulty cognition and faulty cognitive processing. Since the early 1980s, BT and CT have been integrated into CBT (53, 54). In a CBT approach the patient is helped or guided toward awareness of thoughts, feelings, the consequent behaviour, and the accompanying physiological response (54). Originally, CBT was performed individually, but group CBT has also been shown effective (55). CBT continues to move forward with new interventions, including dialectical behaviour therapy (DBT), acceptance and commitment therapy (ACT), mindfulness-based stress reduction (MBSR), and mindfulness-based cognitive therapy (MBCT) (54). Interest in mind-body therapies have increased in the Western world. Traditional Chinese medicine (TCM) consists of several therapies that aim to promote health and healing based on balancing qi according to the theory of yin and yang and the five elements (wood, fire, soil, metal, water). Qigong is one therapy originating from TCM and can easily be performed by anyone. The word Qigong is composed of qi, the vital energy of the body, and gong, the practice or skill of working with qi (56). There are several forms and styles of Qigong but the basic components are concentration, relaxation, mind exercises, breathing exercises, body posture, and slow movements. These components could affect and balancing qi. Qigong can also be divided into static/passive and dynamic/active qigong. Static Qigong is mostly performed in sitting or lying posture with focus on an internal body training approach. Dynamic Qigong is mostly performed in a standing position with slow movements (57). RCT have evaluated Qigong for different diagnoses with varying results and also used different control conditions (Table III).

18 Introduction ______

Liggande Table II del 1

19 Introduction ______

Liggande Table II del 2

20 Introduction ______

Liggande Table III del 1

21 Introduction ______

Liggande Table III del 2

22 Aims ______

AIMS OF THE THESIS

The overall aim of this thesis is to describe patients on long-term sick leave because of burnout and to evaluate rehabilitation programs for this patient group.

Specific aims

• To compare patients with burnout to a general population sample with respect to physical, psychosocial, and work variables. Paper I

• To evaluate gender differences in patients with burnout. Papers I & II.

• To compare two rehabilitation programs, one consisting of Cognitively-oriented Behavioural Rehabilitation (CBR) and Qigong, and another consisting of Qigong alone, with regard to effects on psychological variables and sick leave rates. Paper II.

• To illuminate burnout patients subjective experiences from participating in two rehabilitation programs. Paper III.

• To evaluate the effectiveness of a 12-week Qigong program compared with a control condition. Paper IV.

23 Methods ______

METHODS

Settings and participants All participants in this thesis came from the county of Västerbotten, located in the north of Sweden. The population in Västerbotten is nearly 260 000 people and almost half of them (112 000) live in Umeå (the Birch City). The mean age of the population in Umeå is 38 years and about 50% have a higher education. This is due to the presence of Umeå University which in 2007 had almost 29 000 students, 60% of whom were women. The labour market in Umeå is almost equally divided between the public and the private sector (58). This thesis is based on three samples. Two patient samples, REST (rehabilitation for stress-related disease and burnout) (Papers I-III) and QIST (Qigong for stress-related disease and burnout) (Paper IV), were recruited from the Stress Clinic at the University Hospital in Umeå. A general population sample was recruited from the 2004 Northern Sweden MONICA survey (Paper I) (Figure 1). Baseline characteristics for the samples in Paper I-IV are summarized in Table IV.

Paper I REST 136 patients

Paper II QIST 82 patients Paper III 18 patients MONICA 573 participants

Paper IV

Figure 1. Source of the samples analysed in Papers I-IV. Two patient samples were from the Rehabilitation for stress-related disease and burnout (REST) and Qigong for stress-related disease and burnout (QIST) and one general population sample was from the Multinational Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA) survey.

24 Methods ______

Table IV. Baseline characteristics for the study samples used in this thesis. REST REST QIST MONICA Variable (n=136) (n=18) (n=82) (n=573) Sex (female/male), n 96/40 10/8 68/14 283/290 Age, mean (SD) 41.6 (7.4) 42.4 (7.4) 44.3 (9.1) 40.7 (8.5) Shirom-Melamed Burnout Questionnaire, mean (SD) 5.7 (0.7) 5.5 (0.6) 5.6 (0.7) 3.1 (1.0) Education, n (%) Primary/secondary school 68 (50) 13 (72) 41 (50) 360 (64) University 68 (50) 5 (28) 41 (50) 202 (36) Family situation, n (%) Living alone 12 (9) 1 (5.5) 9 (11) 83 (15) Living with another adult 20 (15) 1 (5.5) 19 (23) 128 (23) Living with another adult and children 87 (64) 12 (67) 45 (55) 302 (55) Single parent 17 (12) 4 (22) 9 (11) 38 (7) Type of work, n (%) With people 80 (59) 7 (39) 52 (63) 232 (43) With things 17 (12) 5 (28) 9 (11) 181 (34) With data 39 (29) 6 (33) 21 (26) 126 (23) Physical activity, n (%) < 30 min/day 45 (37) 6 (35) 28 (35) 104 (18) ≥ 30 min/day 77 (63) 11 (65) 53 (65) 467 (82) Physical exercise, n (%) ≤ 2 hours/week 99 (82) 13 (76) 69 (88) 442 (77) > 2 hours/week 21 (18) 4 (24) 9 (12) 129 (23)

Data sources REST A total of 259 patients were consecutively screened for participation in the randomised controlled trial, REhabilitation for Stress-related disease and burnout (REST) at the Stress Clinic, University Hospital of Umeå between December 2002 until December 2004. Inclusion criteria for the patients in the REST study were: age 25-55 years; sick leave for burnout ≥ 25% of working hours during the previous 3-24 months; and an average score ≥ 4.6 on the Shirom-Melamed Burnout Questionnaire (SMBQ) (39-41). Exclusion criteria were: other diseases that could result in fatigue and/or stress-related symptoms; other diseases that could be a reason for future sick leave; other diseases or treatments that could interfere with active participation; post-traumatic stress disorders; unemployment for more than two years; known abuse of alcohol or drugs; speech and language difficulties; need for individual therapy; or participation in other intervention studies. One hundred fifty-eight patients met the inclusion criteria. Of these, 22 patients declined to participate due to a personal reason, long travel distance, or because they had recovered and returned to work. There were no significant differences in age, sex or SMBQ score between patients

25 Methods ______who declined and those who participated in the study. The final study sample consisted of 136 patients. In Paper III a sub-sample of 59 patients from REST who had completed the 1-year rehabilitation were asked to participate in an interview study. Seventeen accepted and were interviewed. One patient who had not completed the 1-year rehabilitation was also interviewed. This resulted in a total of 18 patients interviewed. A flowchart of the patients throughout the REST study is presented in Figure 2.

26 Methods ______

Assessed for eligibility (n=259) (187 women/72 men)

Excluded (n=123) Did not meet inclusion criteria (n=101) (76 women/25 men) Refused to participate (n=22) (15 women/7 men)

Randomised (n=136) (96 women/40 men)

Allocated to rehabilitation program A (n=67) Allocated to rehabilitation program B (n=69) (49 women/18 men) (47 women/22 men) Received allocated intervention (n=62) Received allocated intervention (n=64) Did not receive allocated intervention (n=5) Did not receive allocated intervention (n=5) 1 difficulty attending study because of work 1 long travel distance 2 long travel distance 1 personal reason 2 personal reason 1 moved out of the area 2 disappointed because of randomisation

Analyzed 6 months (n=60) (44 women/16 men) Analyzed 6 months (n=50) (36 women/14 men) Discontinued intervention (n=2) Discontinued intervention (n=14) 1 personal reason 4 difficulty attending study because of work 1 did not give any reason 1 long travel distance 3 felt too ill to continue 2 wanted to seek another treatment 1 personal reason 3 did not give any reason

Analyzed at 1-year rehabilitation (n=58) Analyzed at 1-year rehabilitation (n=49) (42 women/16 men) (36 women/13 men) Discontinued intervention (n=2) Discontinued intervention (n=1) 1 drug problem 1 personal reason 1wanted to seek another treatment

Analyzed at 6 month follow-up (n=53) Analyzed at 6 month follow-up (n=36) (39 women/14 men) (27 women/9 men)

Analyzed at 12 month follow-up (n=54) Analyzed at 12 month follow-up (n=39) (41 women/13 men) (27 women/12 men)

Interview (n=10) Interview (n=8) (6 women/4 men) (4 women/4 men)

Figure 2. Flowchart of the patients throughout the REST study. In Paper I-II, all 136 patients are analysed. In Paper III, 18 patients are analyzed.

27 Methods ______

QIST A total of 290 patients were consecutively screened for participation in the randomised controlled trial, QIgong for Stress-related disease and burnout (QIST), at the Stress Clinic, University Hospital of Umeå between January 2005 and December 2006. Inclusion criteria for the patients in the QIST study were: diagnosis of burnout and exhaustion syndrome according to the Swedish criteria (20); aged 25-65 years; and an average score ≥ 4.0 on the Shirom-Melamed Burnout Questionnaire (SMBQ) (39-41). Exclusion criteria were known abuse of alcohol or drugs, or participation in other intervention studies. One hundred twenty-eight patients met the inclusion criteria. Of these, 46 patients declined to participate. There were no significant differences in age, sex or SMBQ score between patients who participated and those who declined. The final study sample consisted of 82 patients. A flowchart of the patients throughout the QIST study is presented in Figure 3.

28 Methods ______

Assessed for eligibility (n=290) (217 women/73 men)

Excluded (n=208) Did not meet inclusion criteria (n=162) (114 women/48 men) Refused to participate (n=46) (35 women/11 men)

Randomised (n=82) (68 women/14 men)

Allocated to intervention group (n=41) Allocated to control group (n=41) (34 women/7 men) (34 women/7 men) Received allocated intervention (n=38) Received allocated control condition (n=38) Did not receive allocated intervention (n=3) Did not receive allocated control condition (n=3) 3 difficulty attending study because of work 1 difficulty attending study because of work 1 personal reason 1 did not give any reason

Analyzed at 4 weeks (n= 38) Analyzed at 4 weeks (n=38) (32 women/6 men) (32 women/6 men)

Analyzed at 8 weeks (n=34) Analyzed at 8 weeks (n=37) (30 women/4 men) (31 women/6 men) Discontinued intervention (n=4) Discontinued control condition (n=1) 2 personal reason 1 started another treatment 1 underwent surgery 1 felt too ill to continue

Analyzed at 12 weeks (n=33) Analyzed at 12 weeks (n=35) (29 women/4 men) (29 women/6 men) Discontinued intervention (n=1) Discontinued control condition (n=2) 1 did not give any reason 2 did not give any reason

Figure 3. Flowchart of the patients throughout the QIST study.

MONICA The Multinational MONItoring of Trends and Determinants in CArdiovascular Disease (MONICA) project was initiated in 1982 by the World Health Organization (WHO) to assess trends in cardiovascular mortality in relation to known risk factors. MONICA included 38 populations in 26 countries and the Northern Sweden MONICA survey was included from 1986. The MONICA survey was performed as a repeated cross-sectional survey in 1986, 1990, 1994, 1999, and 2004 in the counties of Västerbotten and Norrbotten (59). In this thesis a sample of 573 participants from Västerbotten County in the 2004 Northern Sweden MONICA survey were used as an age-adjusted reference population (25-55 years).

29 Methods ______

Ethical approval All studies in this thesis were approved by the Research Ethics Committee at the Medical Faculty of Umeå University. All patients received verbal and written information regarding the respective study. The patients were informed that participation was voluntary, they could withdraw at any time without stating a reason, and they were promised confidentiality. The patients who agreed to participate gave written informed consent.

Procedure REST Most of the patients in REST were referred to the Stress Clinic by general practitioners. Before inclusion, all patients underwent medical and psychological examination at the Stress Clinic. They also answered a comprehensive questionnaire that was mailed to them before the first visit. Patients who met the inclusion criteria received verbal and written information regarding the study from a research nurse. After baseline assessments the patients were randomised to either program A (Cognitively-oriented Behavioural Rehabilitation (CBR) and Qigong) or program B (Qigong alone). The randomisation procedure was performed by a research nurse who opened opaque sealed envelopes. In total, 10 groups participated in program A and seven groups in program B. In Paper I, baseline data were compared with data from the MONICA sample. In Paper II, patients in program A and B were compared at baseline, after six months and at the end of the 1-year rehabilitation as well as at six and 12 month follow-ups after the 1-year rehabilitation. At the end of the 1-year rehabilitation program patients from nine groups were asked to participate in an interview study (Paper III) with the purpose of assessing their experiences in the programs. The goal was to interview two patients in each group. In five groups more than two patients volunteered and participation was decided by drawing lots. One patient who had withdrawn from the study was also interviewed. All interviews were performed at the Stress Clinic by two interviewers who were not involved in the rehabilitation program.

QIST Most of the patients in QIST were referred to the Stress Clinic by general practitioners. Before inclusion, all patients underwent medical and psychological examination at the Stress Clinic. The patients received verbal and written information regarding the study if they met the inclusion criteria. Patients who agreed to participate in the study were randomly allocated to either an intervention group (Qigong twice a week during 12 weeks in combination with basic care) or a control condition (basic care only). In total, there were 4 groups in the intervention group

30 Methods ______and 4 groups in the control group. The patients completed questionnaires and performed physical tests at baseline and after the 12 week intervention period.

MONICA The 2004 Northern Sweden MONICA project randomly selected 250 women and 250 men in each of five age groups (24-34, 35-44, 45-54, 55- 64, and 65-74 years) from an updated national database. Selected individuals were invited to a health examination at the nearest healthcare centre. Measurements in the MONICA project were questionnaires, physiological measurements and event registration of acute stroke and acute myocardial infarction. The survey procedure and assessments are described elsewhere (59).

Data Questionnaires, register data on sick leave, physical measurements, and interview transcription used in this thesis are described below:

Psychological variables, health-related quality of life and perceived relaxation Burnout In all papers burnout was measured using the Shirom-Melamed Burnout Questionnaire (SMBQ) (39-41). The SMBQ contains 22 items and each item is rated on a seven-point scale graded from 1 (almost never) to 7 (almost always). The SMBQ is comprised of four subscales: emotional and physical exhaustion (eight items); tension (four items); listlessness (four items); and cognitive weariness (six items). An overall index with the four subscales was computed as the mean of all items. Higher scores indicate a higher level of burnout.

Anxiety, depression and obsessive-compulsive symptoms In Paper II, anxiety, depression and obsessive-compulsive symptoms were measured using the Self-Rating Scale for Affective Syndromes (CPRS-S-A) (60, 61). The CPRS-S-A contains 19 items, each rated on a seven-point scale (0-3 with half-point steps). The patients rated the degree of symptoms in the previous three days and the three subscales were calculated separately (anxiety 0-27 points; depression 0-27 points; obsessive-compulsive symptoms 0-24 points). In Paper IV, anxiety and depression was measured using the Hospital Anxiety and Depression scale (HAD) (62). HAD contains 14 items where both subscales contains seven items each, scored on a four-point scale. The patients rated the degree of symptoms in the previous week and each

31 Methods ______subscale was calculated separately (0-21 points). Higher scores indicate more symptoms.

Stress behaviour Stress behaviour was assessed in Paper II by the self-rated questionnaire Everyday Life Stress Scale (ELSS) (63). ELSS contains 20 statements that refer to stress behaviours in everyday life situations. Each item was scored on a four-point scale and a total summary score (0-60 points) was calculated. Higher scores indicate more self-rated stress behaviours.

Fatigue In Paper II and IV, fatigue was measured using the Checklist Individual Strength questionnaire (CIS) (64). CIS contains 20 items, each rated on a seven-point scale graded from 1 (“Yes, that is true”) to 7 (“No, that is not true”). The CIS is composed of four dimensions of fatigue: subjective experience of fatigue; concentration; motivation; and physical activity level. The patients estimated their fatigue during the previous two weeks and a total score (20-140 points) was calculated. Higher scores indicate a worse condition.

Self-esteem In Paper IV, self-esteem was measured by the Self-Concept Questionnaire (SCQ) (65). SCQ contains 30 items and is based on seven components of self-esteem: significance; worthiness; appearance and social acceptability; competence; resilience and determination; control over personal destiny; and the value of existence. The patients estimated their self-esteem on an eight-point scale (0 = “completely disagree”) to (7 = “completely agree”), and these were summed to a total score (0-210 points). Higher scores indicate more self-esteem.

Health-related quality of life In Paper IV, Health-Related Quality of Life (HRQL) was assessed by the Swedish version of the SF-36. The SF-36 is a generic quality of life instrument and consists of 36 items that measure eight physical and mental dimensions of health: “physical functioning”; “role limitations due to physical problems”; “bodily pain”; “general health perceptions”; “vitality”; “social functioning”; “role limitations due to emotional problems”; and “mental health”. Scores on each scale ranged from 0 (worst possible health state) to 100 (best possible health state). The eight dimensions were divided into the physical component summary score (PCS) and the mental component summary score (MCS) (66, 67).

Perceived relaxation The Relaxation Inventory is a self-report measure used to assess relaxation and consists of three subscales: Physiological Tension Scale;

32 Methods ______

Physical Assessment Scale; and Cognitive Tension Scale (68). In Paper IV, degree of relaxation was measured with four items from the Swedish version of the Physical Assessment Scale (S-PARI) (69). The patients estimated their current state of relaxation on a five-point scale (1= “not at all”) to (5= “very good”), and these were summed to a total score (4-20 points). Higher scores indicate a more relaxed condition.

Work characteristics and work-related stress Work characteristics were measured as type of work (“with people”, “with things”, or “with data”) (70) (Paper I-IV), overtime work (Paper I) and work-related stress (Paper I). Work-related stress was measured by the demand-control model (23). In the Swedish version there are five questions dealing with psychological demands and six questions dealing with decision latitude. In the demand dimension there were two questions in the MONICA study and one question in the REST study that did not correspond to the Swedish version. A factor analysis resulted in five questions in the demand dimension that were used when analyzing gender differences among burnout patients and four questions when analyzing differences between patients and MONICA population. The patients and the MONICA population estimated their working conditions on a four-point scale. When analyzing gender differences among burnout patients a summary index was calculated separately for demand (5-20 points) and control (0-24 points), where higher scores indicate high demands or high control. When analyzing differences between patients and the MONICA population, a dichotomization point was set so that high demands were defined as more than the 75th percentile and low control as less than the 25th percentile. High job strain was defined as having both high demands and low control.

Social support Social support was assessed in Paper I by using the instruments Availability of Attachment (AVAT) and Availability of Social Integration (AVSI) from a condensed version of the Interview Schedule for Social Interaction (ISSI) (71). The two subscales consist of six items each. In paper I six items in AVAT was used but in AVSI only two items were available. AVAT estimates aspects of emotional support and AVSI assesses network. The patients estimated their emotional support (0-6 points) and network (0-2 points). When analyzing differences between patients and the MONICA population, a dichotomization point was set so that low emotional support was defined as less than the 25th percentile and low network support was less than the 25th percentile.

33 Methods ______

Sick leave Number of sick leave days before randomization was collected from medical journals. The patient’s sick leave rate at baseline, after the 1-year rehabilitation, and at 12 month follow-up were collected from registers at the Social Insurance Agency and used in Paper II. Sick leave included sickness benefit, activity and sickness compensation, and rehabilitation compensation. In Sweden, the first 14 days of sick leave are paid by the employer. Thus, only sick-leave periods exceeding 14 day are included in the analyses.

Physical measurements Balance Static balance was assessed by standing on one leg without shoes and with eyes open. The lifted leg had contact with the calf of the supporting leg. The arms were hanging along the sides with the thumbs in contact with the thighs and at the same time the patient was turning the head from side to side at a self-selected speed. The time to perform the test without losing the balance was registered in seconds and a maximum test time was set to 120 seconds (72). The best of two trials was used in the analyses. Static balance was assessed in both REST and QIST but only reported in Paper IV. Dynamic balance was assessed in QIST (Paper IV) by walking 20 steps on a one cm thick line with shoes on, heel to toe, with eyes open. The patient performed one trial and time in seconds as well as number of steps outside the line was recorded (73).

Physical capacity and activity The Two-kilometre (km) Walk Test (74) was used to assess physical capacity. The patients walked 10 laps on a 200 m athletic track indoors. An indoor track was used to avoid different weather conditions that could affect the results. Before the trial the patients got verbal instructions to “walk the distance as fast as you can without risking your health”. Patients’ heart rate at the end of the walk, walking time, age, sex, and body mass index (BMI) were recorded and used in a prediction equation (75) to predict maximal oxygen uptake (VO2max). Due to limitations in the prediction equation only walking time is presented. The Two-km Walk Test is a valid test for moderately fit men and women (76), and obese adults (77). Physical capacity was assessed in both REST and QIST but only reported in Paper IV. Physical activity in everyday life was assessed in QIST (Paper IV) by pedometer use (SILVA Pedometer Plus). The pedometer recorded total daily step counts during two days. The patients wore the pedometer from getting up in the morning until to going to bed and an average number of steps for the two days was recorded.

34 Methods ______

Interview transcriptions Semi structured face to face interviews were performed. The interviews lasted 45 to 90 minutes, were tape-recorded, and transcribed verbatim. An interview guide was used that included themes to be covered. The themes were how the patient perceived her/his own rehabilitation process, how different parts of the programs were experienced, the meaning of the group and the personnel at the Clinic. The interview ended with the patient reflecting on the future and how she/he should manage without support from the program. The themes were slightly adjusted between the interviews so that information in an interview could be further explored in a following interview.

Rehabilitation Cognitively-oriented Behavioural Rehabilitation (CBR) The CBR program evaluated in Papers II and III was developed from a stress management program for women with coronary heart disease (78). The program was modified to suit patients with burnout by adding discussion components of affect awareness (79), sleep, and work rehabilitation. The CBR program consisted of 30 group sessions, with 20 meetings the first six months and 10 meetings the last six months. Each group session was three hours long with a 30 minute break in the middle. A group leader especially trained in CBR led the group consisting of six to nine patients. There were at least two patients of the same sex in each group. Every session started with 10 minutes of seated relaxation followed by a specific theme. The CBR program consisted of five key components: 1) education (for example, stress reactions, sleep, affect, medication, the importance of rest in order to recover); 2) awareness of reactions and “self-talk”; 3) development of behavioural/cognitive/emotional skills; 4) spiritual issues and life values; and 5) preparation for return to work. The CBR group reassembled at three, six, and 12 months after the 1-year rehabilitation. During the 1-year rehabilitation there were three group meetings for spouses and/or relatives.

Qigong Qigong was evaluated in Papers II-IV. In the REST study, Qigong was performed in 1-hour sessions, once a week for one year. In the QIST study, Qigong was performed in 1-hour sessions, twice a week for 12 weeks. The Qigong groups were gender-mixed with 8-16 patients in each. A physiotherapist specially trained in Qigong by © TaiChi Akademien AB, Stockholm, Sweden led the groups. The same Qigong exercises were used in both studies and the program consisted of three parts: 1) warm-

35 Methods ______up movements; 2) basic movements to affect body awareness, balance and coordination, breathing and muscular tension; and 3) relaxation and mindfulness meditation with self-performed body massage at the end. The basic movements visualize different animals like “the crane stretches itself”, “the rhinoceros looks at the stars” and “the shambling bear”. In the mindfulness meditation the patients were asked to think of a positive memory and re-establish the situation with their senses. During the Qigong program background music was used and the movements were mostly performed standing. Patients who preferred could perform the movements in a sitting or lying position. A printed summary of the Qigong movements was offered to patients who wanted to perform additional Qigong at home.

Basic Care All patients in the REST and the QIST study took part in basic care at the Stress Clinic. Basic care included follow-up visits to the physician who handled medication prescriptions and sick leave certificates. The physician also gave general advice concerning recovery, sleep, routines in daily life and physical activity. Work rehabilitation support was also offered to the patients. The concept of work rehabilitation support used in Papers II and III consisted of a rehabilitation meeting at the end of the program in which the patient, the responsible physician, the employer and the officer at the Social Insurance Agency participated.

Statistics and analysis Statistics All statistical analyses were performed in SPSS© version 13.0 (Paper I) or 14.0 (Paper II-IV) (SPSS Inc., Chicago, IL, USA) and a p-value less than 0.05 was considered significant. Power analyses were performed in Papers II and IV. Differences between groups were analyzed using Pearson’s chi-squared test (categorical variables) and independent two sample t-test (continuous variables). To test if there were any differences between women and men in comparison with a general population in emotional support and jobstrain, Breslow-Day test with Tarone’s adjustment were used and only reported in the cover of the thesis. In analyses of intervention effects, repeated measures analyses of variance (Paper II), effect size (80) (Paper II), and non-parametric statistics (Wilcoxon Signed Ranks Test and Mann-Whitney Test) (Paper IV) were used. Per-protocol analyses (patients who completed the intervention period) and intention-to-treat analyses (all randomised patients) were performed in Papers II and IV. An imputation procedure was used to adjust for missing responses in single items. Missing responses were replaced with the median response for the group and this occurred in 0.3- 0.4 percent (Paper I), 0.2-0.6 (Paper II), and 0.2-0.7 (Paper IV) of the items. Missing responses in the intention-to-treat analyses (Papers II and

36 Methods ______

IV) were replaced with the mean value change between the measurement points. Mean value change was calculated separately for the randomisation groups as well as for women and men.

Grounded theory Grounded Theory method (81) of constant comparison was used in Paper III to describe the patients’ experiences of the REST program. Transcribed interviews were first analysed by open coding, followed by a mutual comparison and a final negotiated outcome (81, 82). The next step was to perform selective coding where open codes with similar elements were compiled into categories (82). Further, there was a process to find axes between the categories and to identify a core category. Finally, a theoretic model of the rehabilitation process for patients with burnout was developed.

37 Results ______

RESULTS

Description of burnout patients in relation to gender and a general population Significant differences were found when comparing the patients with burnout with the general population. Patients with burnout reported a higher rate of unemployment, a more restricted social network and higher work demands. In addition, women with burnout reported less emotional support, a more sedentary work situation, higher job strain and worked to a greater extent “with people” than women in the general population. Men with burnout were more often living in families compared with men in a general population (Paper I). When comparing women and men with burnout some significant gender differences were found. Women with burnout reported a higher degree of sleep disturbances, lower job control, more unpaid work and they worked to a greater extent “with people”. Men with burnout perceived a more restricted social network and reported more overtime work (Paper I). Further analysis displayed that compared to men with burnout, women with burnout have higher odds of less emotional support in comparison with a general population sample. This difference was not found for job strain.

Rehabilitation programs Dropouts A total of 29 patients withdrew from the REST study (Paper II) and 14 patients from the QIST study. There were no significant differences in SMBQ score, sex or age between patients who completed the intervention period and the dropouts. In Paper III, one patient withdrew before the interview.

Compliance In the REST study (Paper II) the mean attendance in program A was 25 ± 3.1 of the 30 CBR group sessions and 21.7 ± 7.5 of the 35 Qigong sessions. Program B attendance was 21.6 ± 7.9 of the 35 Qigong sessions. In the QIST study, patients in the intervention group attended 15.2 ± 5.3 of the 24 Qigong sessions.

Complementary and conversational treatment Some patients in REST (Paper II) (26 % program A and 44 % program B) and in QIST (21 % intervention group and 20 % control group) reported having complementary treatment outside the Stress Clinic during the intervention period. These included massage,

38 Results ______acupuncture and different treatments of body awareness. Patients in REST (Paper II) (16 % program A and 44 % program B) and in QIST (30 % intervention group and 33 % control group) also reported having individual or group conversational therapy outside the Stress Clinic during the intervention period. The additional conversational therapy was used significantly more often in program B compared with program A in the REST study (Paper II).

Effects of rehabilitation Per-protocol analysis There was no significant time by group interaction or group effect between program A and B in REST (Paper II) and no significant between- group differences between the intervention and control group in QIST. All groups improved significantly over time with reduced levels of burnout, anxiety, depression and fatigue. In REST (Paper II) both programs reported a significant reduction of stress behaviour and obsessive-compulsive symptoms. In QIST, both groups improved significantly over time with increased dynamic balance and physical capacity. Values of the psychological variables, HRQL, perceived relaxation and physical measurements at baseline and after the intervention in REST (Paper II) and QIST are presented in Table V.

39 Results ______

Liggande Table V

40 Results ______

The time course of burnout (SMBQ) in the REST study (Paper II) showed an improvement during the intervention period and follow-ups in both programs (Figure 4). The time course for the other psychological variables was almost the same. There was no significant difference between the programs in REST (Paper II) with regard to the effect size of the psychological variables (Figure 5).

6,00

5,50

5,00

Estimated Marginal Means Marginal Estimated 4,50

4,00

Baseline 6 month one-year 6 month 12 month rehabilitation follow-up follow-up Figure 4. Results of the SMBQ, from the analyses of variance with repeated measures in the REST study (Paper II). Rehabilitation program A Rehabilitation program B

2

1,6

1,2 Rehabilitation program A 0,8 Rehabilitation

values size effect Mean program B 0,4

0

CIS ELSS SMBQ

Anxiety Obsession Depression Figure 5. Effect size values in the per-protocol analyses in the REST study (Paper II) after 1-year rehabilitation for Shirom-Melamed Burnout Questionnaire (SMBQ), Everyday Life Stress Scale (ELSS), Checklist Individual Strength questionnaire (CIS), Self-Rating Scale for Affective Syndromes (CPRS-S-A) depression, anxiety and obsessive-compulsive symptoms in rehabilitation program A (n=58) and rehabilitation program B (n=49). Effect size values: < 0.2 = no effect, 0.2-0.5 = small effect, 0.5-0.8 = medium effect, > 0.8 = large effect. 41 Results ______

When comparing women and men in REST (Paper II), a significant time by sex interaction in burnout (SMBQ) (Figure 6) and fatigue were found in program A, and women improved significantly compared to men. That interaction was not found between women and men in program B (Figure 7). At 12 month follow-up, mean score for burnout in program A was significantly lower for women (3.8) compared to men (4.8). In addition, significantly more women in program A (78%) than men (31%) reported an SMBQ score < 4.6 at 12 month follow-up.

5,50

5,00

4,50 Estimated Marginal Means

4,00

3,50 Baseline 6 months one-year 6 month 12 month rehabilitation follow-up follow-up SMBQ

Figure 6. Results of the SMBQ in the REST study (Paper II) program A, from the analyses of variance with repeated measures. Men Women

42 Results ______

6,00

5,50

5,00

Estimated Marginal Means 4,50

4,00

Baseline 6 month one-year 6 month 12 month rehabilitation follow-up follow-up smbq

Figure 7. Results of the SMBQ in the REST study (Paper II) program B, from the analyses of variance with repeated measures. Men Women

A sub-analysis was performed of patients in REST (Paper II) (n=73) who had no additional conversational therapy outside the Stress Clinic during the intervention period. A significant time by group interaction was found in depression, anxiety and obsessive-compulsive symptoms, with patients in program A significantly improved.

Intention-to-treat analysis In the REST study (Paper II) there was a significant time by group interaction in obsessive-compulsive symptoms, with patients in program A significantly improved. Patients in program A had also significantly larger effect sizes in stress behaviour and obsessive-compulsive symptoms compared to patients in program B. There were no significant between-group differences between the intervention and control group in the QIST study.

Sick leave There were no significant differences between patients in program A and B in the REST study (Paper II) regarding sick leave rates at baseline, after the intervention period or at 12 month follow-up. Patients in both programs improved significantly over time with decreased sick leave rates. At the 12 month follow-up, 39 % of the patients in program A and 37 % of the patients in program B remained on full time sick leave. There were significant association between a SMBQ score ≥ 4.6 and sick leave, and between length of sick leave before randomisation and future sick leave. Sixty-one percent of the patients who were on full time sick leave at 12 month follow-up had an average SMBQ score ≥ 4.6, and 88 % of the

43 Results ______patients who were on full time sick leave at 12 month follow-up had been on sick leave for more than six months at randomisation.

Experiences from the rehabilitation programs in the REST study The patients’ experiences of the 1-year rehabilitation programs in the REST study (Paper III) were that good encounters with affirmation and group cohesiveness were necessary base for rehabilitation. After time in the rehabilitation program the patients’ started to reflect upon what had contributed to their burnout. In this process the patients experienced support and affirmation from the group and the clinic personnel. The patients developed insight into the importance of balancing work, family and one’s own recovery and understood the need to take responsibility for their lives again. With the help of tools learned and practiced in the rehabilitation programs, the intellectual insights became embodied by practising to listen to body signals, setting boundaries and finding new ways of behaving. The way patients in the respective groups obtained insight and embodied it differed. Patients in program A primarily used cognitive tools learned in group discussions. In contrast, patients in program B used physical and psychological experiences from the Qigong movements. At the end of the rehabilitation process the patients had thoughts about the future that included both hope and fear. The process during the 1-year rehabilitation programs in the REST-study, as described by the patients was used to develop a theoretical model (Figure 8) (Paper III).

44 Results ______

CHOICE OF TRACK

Faith and hope

HOW CAN I BE THE Body and mind HOW CAN I BE THE ONE I WANT TO BE? awareness ONE I WANT TO BE? Cognitive strategies Physical activity Q Q I C I G B G TAKE CHARGE O R O N N G G

GET TO KNOW MYSELF GET TO KNOW MYSELF Self-esteem Balancing between Physical and psychological Self-confidence energy and relaxation experiences Empowerment

GOOD ENCOUNTERS AFFIRMATION GROUP COHESIVENESS

Figure 8. Theoretical model constructed from burnout patients’ experiences from a 1-year rehabilitation program in the REST study (Paper III). Sub-categories, categories and the core category “Take Charge” are presented.

45 Discussion ______

DISCUSSION

The overall aim of this thesis was to describe patients on long-term sick leave because of burnout and to evaluate rehabilitation programs for this patient group. The findings indicate that patients with burnout differ from a general population with regard to perceived social support and work-related factors. There were no significant differences in effect in a per-protocol analysis of CBR and Qigong compared to Qigong alone (REST), or between a Qigong intervention compared to a control condition (QIST). Improvements were found in all rehabilitation program groups. The patients in programs with CBR and Qigong or Qigong alone experienced beneficial improvements that resulted in an ability to take charge of their lives.

Description of burnout patients in relation to gender and a general population Both female and male patients in the REST study reported high levels of burnout that were confirmed by clinical and psychological measurements. Reported levels of burnout were even higher than other patient samples with burnout (8, 83). More female patients with burnout were included in our studies which are similar to study populations in most studies on rehabilitation for burnout (83, 84). This could reflect that more women are affected by burnout or could also be due to women being more inclined to consult doctors when sick. In population studies women scored higher on exhaustion compared to men (10, 13), but in REST there was no significant difference in levels of burnout at baseline between women and men. One reason for this could be that burnout measured with the SMBQ was used as inclusion criteria. Differences between women and men in the study were mostly the same as those found in Swedish society, i.e., the women more often worked with “people” (85), worked fewer hours in paid work, and had more hours in domestic work compared to the men (86). An interesting finding was that male patients with burnout more often lived in families compared to men in the general population. This is contradictory to earlier studies where being married is found to be protective for men (10, 13). Maybe men who are living in families are supported to consult a doctor, while men who are living alone tries to manage on their own. However, there were few men in the study and a large number of comparisons were made, so association by coincidence can not be ruled out. The burnout women in our study reported lower job control than the men. This has previously been found in a European multicentre study on gender differences in perceived job stress (87). Patients in the REST study reported higher work demands compared with the general population. The REST study was cross-sectional and no conclusions regarding cause and effect can be drawn. However, a recent prospective

46 Discussion ______study of a Swedish working population sample found that low control, lack of social support, downsizing and especially demands at work were associated with emotional exhaustion (6). Gender differences in the prospective study showed that high demands and low control were more common in women. Furthermore, lack of support from fellow workers predicted emotional exhaustion symptoms in women, whereas lack of support from superiors increased the risk for men (6). Thus, social support is an important factor in relation to burnout. Patients with burnout in the REST study perceived a more restricted network compared to the general population and female patients experienced less emotional support compared to women in the general population. Our data are cross-sectional but the association is in accordance with findings in the longitudinal study (6). Associations between poor social support and burnout is found in other studies as well (5, 8). Social support at work is also emphasized as important for work return in women on sickness absence due to work related strain (88). In this thesis, social support was not measured specifically at work but defined in more general terms to include family, friends, co-workers or others, but the association seems to be similar. There were significant gender differences in social support among patients with burnout. Men reported a more restricted network compared to women. This could be an effect of that caring for relations is a larger part of the female identity, but also due to differences in work situations, as men in the study reported more overtime work, which reduces time for social contacts. The same relation between overtime work and restricted social networks was found in women in a study by Verdonk et al. (89) who found mental health problems and social isolation among young highly educated women, with high rate of overtime work. In our study lack of emotional support seemed to be more associated with burnout among women. This raise questions on whether women and men are in need of different social support as well as if the social support provided by women and men differ. Kawachi and Berkman (90) have shown that women have more emotionally intimate relationships than men and therefore suffer more from others’ problems. This could explain the higher risk for burnout found in women “working with people” but could also mean that women, when they are sick, experience guilt because they are not able to help others as they used to. Men, on the other hand, could be the receivers of support provided by women.

Effects of rehabilitation All groups in the rehabilitation programs (REST and QIST) showed positive effects in psychological variables over time. The same pattern has been seen in other rehabilitation studies of sickness absent patients with stress-related illness or burnout (83, 84, 91-93). Explanations to the lack of differences between the rehabilitation programs could be viewed in several ways. There could be a natural recovery course in burnout, and to identify additional effects from

47 Discussion ______rehabilitation programs might require larger study populations. Also, the instruments used in the evaluations might not have been sensitive enough. Another explanation could be that all groups reported high frequencies of complementary and conversational treatment outside the Stress Clinic during the intervention periods. Almost half of the patients in program B in the REST study reported having conversational therapy outside the program. This could have levelled out the differences between the programs in REST, but could not explain the lack of difference between the groups in QIST. Instead, low compliance in the intervention group in QIST could have made the intervention less effective. Similarities between the rehabilitation programs must also be taken into account. In the third developmental phase, CBT is influenced by eastern philosophy (54) from which Qigong also emerged. CBR used in our study is not equivalent with CBT but is based on the same principles. In both CBR and Qigong the patients learned to be mindful and to be ‘here and now’, but in different ways as the patients confirmed in the interviews. During CBR, patients learned to use cognitive tools to be mindful and also practiced awareness daily with different types of homework, e.g., awareness in watching nature or eating with combination with breathing. In Qigong, patients received mindfulness through slow, focused movements and breathing. Granath et al. (50) found similar results when they evaluated CBT and yoga for individuals with stress-related problems. Both groups improved with no differences between the two interventions (50). Other similarities between the rehabilitation programs could be the “common factors”. Common factors are those such as expectations for improvement, warmth and attention, understanding, encouragement in the treatment setting, and are not specific to any particular technique (94). The results from the interview study (REST) confirmed that there were common factors that influenced both the CBR and the Qigong groups. Good encounters, affirmation, and support from professionals and other patients were expressed as being very important for the informants and a necessary basis to initiate behavioural change. All programs were group based. Meeting other patients in same situation decreased their feeling of being insane. This has previously been confirmed in focus groups discussions by women on sick leave due to work related strain (88). Support groups for women with breast cancer have also been shown beneficial for recovery (95). Meta-analyses are conducted to determine the effect of interventions that aimed to reduce work-related stress. Interventions on the individual level are reported to be stronger (96, 97), and this is particularly true for cognitive-behaviour approaches (97, 98). A meta-analysis recommended that cognitive-behaviour programs should not be combined with other treatments because of reduced effects (98). If this is the case, the effect of CBR seen in our study might have been stronger if it was not combined with Qigong. As many patients also searched for additional treatment outside the Stress Clinic, three or more treatments could have been evaluated at the same time.

48 Discussion ______

The CBR groups in the REST study were gender-mixed and contained at least two men in each group, but there was always a majority of women. Women in these groups had more favourable results than men in burnout and fatigue outcomes. Similar results are reported by Jensen et al. (99) who found that women with non-specific spinal pain had greater improvement in health-related quality of life than men in a gender-mixed CBT rehabilitation programs. Contradictory findings on the role of gender for treatment effects on depression from individual and group CBT are reported by Watson and Nathan (100). They found equivalent benefits in women and men but they studied gender-separate groups (100). In cardiac rehabilitation, gender-mixed stress management programs with a majority of men have been common. This situation was supposed to be unfavourable for women. Therefore a separate stress management program for women with coronary heart disease was developed (78) and found to be beneficial for women’s psychological well- being (63). A more even distribution of women and men in our groups or separate groups for women and men might have affected the results. No gender differences were found in relation to qigong. Neither in the rehabilitation programme B in REST or in the QIST study (stratified analyses not shown in the manuscript). Gender differences which emerged in the interview study in REST were in most cases the same as between women and men in the Swedish society in general.

Work capacity and burnout Burnout was first described in work settings among individuals still at work. Today more research is published on burnout in patients on sick leave. Sickness absence in Sweden has varied widely over time and this has been especially true during the last two decades (101). Change in working life from physical to mental work and blending of work and private life are discussed as some reasons for increased sickness absences (102). General changes in legislation and the trade cycle also are important factors in explaining sickness absence in Sweden. Sickness absence increases when unemployment rates decrease and conversely when unemployment rates increase, sickness absences decrease (101). There is no sharp distinction between sick and not sick in relation to stress-related and mood disorders. Similar symptoms have been given different diagnoses throughout history. The diagnosis neurasthenia with mental fatigue was developed in 1869, the concept of stress in 1940, chronic fatigue syndrome with decreased immune defence in 1980, and finally burnout syndrome in the 2000 (103). This shows that individuals perceived similar stress-related symptoms regardless of the diagnoses. Among individuals with stress-related illness, sickness absence from work for more than one year reduced their ability to return to the labour market (104). Early intervention could be an important factor in increasing work return as was shown by van der Klink et al. (92). In the REST study, the average sickness absence rate before randomisation was almost one year, which probably affected the work return rate. This

49 Discussion ______interpretation is supported by the association between length of sick leave before randomisation and future sick leave. Most of the patients who were on fulltime sick leave at 12 months follow-up had been on sick leave for more than six months at randomisation. Another factor found to increase work return rate is to increase the focus on workplace rehabilitation support during the intervention (91). In the REST study, work rehabilitation support often consisted of a single rehabilitation meeting. It could be important for the patients to integrate newly adopted behaviours in the work place by structured activating plans and a coordinated rehabilitation between responsible parties. It might be important for some patients to have the support from the rehabilitation team in work place changes and/or to try to work in a different setting. This was suggested by some patients in the REST interview study. Another hindrance for work return could be domestic strain. Domestic strain is a hindrance in rehabilitation for long-term sick listed women with musculoskeletal disorders (105). In the REST study, female patients reported more household work, but there was no difference in work return rate between women and men.

Methodological considerations At the time when this thesis’ studies were performed, there was no structured rehabilitation for patients with burnout in health care. Many patients struggled to come to the Stress Clinic and this could have led to the selection of more motivated patients. The patient samples in our study had a high education level, which could reflect the labor market in Umeå, and raise the question of whether our results are generalizable. There are few studies on the effect of education on the outcome of CBR and Qigong, but in a study by Watson and Nathan (100) both women and men of different levels of education progressed during individual or group CBT. Thus, a reasonable assumption is that educational level does not have an impact on CBR treatment on patients with burnout. Paper I used a cross-sectional design that compared patients with a general population. Therefore no conclusion regarding causality could be drawn. Despite this, a picture of differences was found between the two populations. Paper II and IV consist of RCT that are the foundation of evidence-based research. Advantages of RCT are that participants are assigned randomly to different groups to protect against selection bias (106). However, disadvantages with RCT for patients on long-term sick leave should be discussed. Despite randomization, all potential variables that could affect rehabilitation might not be randomly distributed. Some patients may have greater benefits from rehabilitation in an early stage while others who have been on sick leave for a long time may need different rehabilitation measures such as individual CBT. Furthermore, patients in both REST and QIST reported having complementary and conversational treatment outside the Stress Clinic during the intervention period and this probably affected our results. This is a potential problem in both shorter (Paper IV) and longer (Paper II) RCT intervention

50 Discussion ______periods, where patients suffering from severe exhaustion tend to search for additional treatments (93). Under these circumstances it is difficult to find suitable control conditions. To have a control condition without any treatment for patients with severe and lengthy diseases could be considered as unethical. Another disadvantage with a RCT design is that patients could have a positive or negative attitude toward a particular treatment (107). Our and others (93) experiences are that most of the patients allocated to a program with less intervention were disappointed. On the other hand, some patients were uncomfortable with CBR because of talking in a group and were more content with only Qigong (Paper III). Some patients were disappointed to be allotted to program B and dropped out. This could create a problem with statistical power and this was confirmed in REST where an intention-to treat analysis found differences between programs. From a clinical perspective, RCT might not be the optimal research method for evaluation of rehabilitation programs where different patients might need different combinations of measures. Stress management intervention programs are based mainly on psychological variables. Often cognitive and relaxation interventions are evaluated with psychological measurements, while exercise interventions are measured with physiological measurements. In these circumstances it is difficult to compare the different interventions (98). A broader perspective concerning measurements was taken into consideration when rehabilitation programmes were evaluated in this thesis. Psychological variables, HRQL, perceived relaxation, and physical measurements were assessed. Since different instruments has been used to measure burnout there are problems in the estimation and comparison of the prevalence of burnout. There are also problems when different cut-off scores, often based on quartile splits, are used to define burnout. Cox and Tisserand (1) discuss the problem of whether burnout is a continuous condition or a dichotomized state. One of the main issues for burnout researchers should be to define cut-off scores and diagnostic criteria for burnout in the clinical setting (1). In this thesis, the diagnostic criteria for the Swedish exhaustion syndrome is only used in Paper IV because the REST study started before the diagnostic criteria for exhaustion syndrome was developed. The diagnosis for burnout in the REST study was used as defined by Shaufeli and Enzmann (7). They formulated a definition of burnout that is applicable to the workplace. Because of this a questionnaire measuring burnout in the work domain such as the MBI would be used. We decided not to use the MBI because some of the questions on depersonalization (such as “I don’t really care what happens to some recipients”) do not fit our culture. Other researchers in Denmark have given this as a reason not to use the MBI (43). The REST study and the Stress Clinic started during the same time period. We learned from Swedish colleagues in the same field who had positive experiences with the SMBQ. Therefore we also chose the SMBQ for our evaluations. Our experiences in working with the SMBQ are that it could be answered

51 Discussion ______easily and used by patients in different settings. There was only one question that was not always answered and the question was “I feel that I don’t have the strength to go to work in the morning”. Patients who have been on long-term sick leave have difficulty with this question. Since diagnostic criteria for the Swedish exhaustion syndrome are now established and stress factors can be identified in both work and general settings, a suitable questionnaire should be used. Kristensen et al. (43) formulated a valid and reliable instrument to measure burnout, namely the CBI with the three sub-dimensions: personal burnout, work-related burnout, and client-related burnout. The CBI can be used both inside and outside the work place and this would satisfy the Swedish diagnosis of the exhaustion syndrome. In Sweden a new measurement of burnout according to the new diagnosis criteria, the s-USM, have been formulated and are being validated (44). These measurements need also to be evaluated in the clinical setting where important information of their usefulness can be obtained. In this thesis anxiety and depression were measured with the CPRS- S-A (Paper II) and the HAD scale (Paper IV). We changed to the HAD scale because it was an easily self-reported instrument that was beginning to be used in the clinical health care setting. The HAD scale is also useful in screening for clinically significant anxiety and depression (62). The State-Trait Anxiety Inventory (STAI) (108) would have been interesting to use in this thesis to understand if the anxiety consists of a more state (temporary) or trait (stable) condition. State Anxiety scores have previously been reported to decrease significantly after 30 minutes of Qigong in comparison with a non-active control group (109). Before inclusion in our studies, all patients underwent medical and psychological examination to determine if the patients fulfilled the inclusion criteria. Post-traumatic stress disorders (PTSD) was one of the exclusion criteria in the REST study but a specific questionnaire measuring PTSD was not used. This was a limitation since PTSD and/or attention deficit hyperactivity disorder (ADHD) are reported to be under- diagnosed among long-term sick leave patients with burnout (110).

Clinical implications and further research Burnout is troublesome for both women and men. Questions could be raised of how prevention and/or a rehabilitation program should be performed for this patient group. An early rehabilitation is important in order to shorten the sick leave period and result in earlier return to work (92, 104). A multimodal rehabilitation is preferable. A team of different professionals would be best suited to adjust multimodal rehabilitation to individual needs. This should include a base of good encounters, affirmation, and support. In future evaluations of rehabilitation programs it is important to find other evaluation methods than RCT’s, as rehabilitation should emanate from the individual patient’s needs and motivation.

52 Discussion ______

Timing of the rehabilitation is important and should be based on the patient’s health status and motivation. Rehabilitation gives the patients’ a structure to their daily lives (54). The rehabilitation could include support for the patients to acquire daily routines such as exercise, recovery, and eating and sleeping routines. These are reported to be important for women with sickness absence due to work related strain to regain control over their own situations and to master life as a whole (88). Rehabilitation could be performed individually, as part of a group, or a combination of both, depending on what is appropriate for the patient (111). Further research should investigate if group rehabilitation for patients with burnout would be more effective if there is a more even gender distribution versus separate groups for women and men. The patients need support for mental insights, to find new ways to behave in stress related situations, and for balancing activity and recovery, which could be done with both CBR and Qigong. Since a Qigong intervention twice a week for 12 weeks had no additional effect as compared to basic care it would be interesting to study if a more intense physical activity would be more efficient. Cognitive impairments have previously been found in patients with chronic burnout syndrome (112, 113), and burnout patients are often depressed. Physical activity and particularly aerobic fitness training have positive effects on brain function and cognition (114), and physical activity is also effective in reducing depression (115). From this perspective it would be interesting to study early rehabilitation with physical activity and prescription of preferred physical activity (FaR) for patients with burnout. An uncontrolled study found that after six months, FaR in primary health care changed the patients’ behaviours to a more physically active lifestyle and an increased quality of life (116). In 1974, Freudenberger (15) reported advice on the prevention of burnout. Since burnout is an emotional and mental exhaustion, the physical activity should not consist of meditation or yoga. He suggested that the type of physical exercise should be more active and cause the individual to become physically tired. The physical exercise should be individualised to increase the patient’s motivation for a sustainable physically active lifestyle. Patients who have been exhausted for a long time maybe need support to increase their physical activity level. Other patients who have high level of physical activity due to anxiety may need to decrease the activity level and perform exercises that are more focused on body awareness and mindfulness. In an early multimodal rehabilitation, work rehabilitation support should be included. A structured activating plan and coordinated rehabilitation should be set up by the responsible parties, i.e. the officer at the Social Insurance Agency, the employer, the responsible physician, and the patient. Some patients maybe need support to change their work. However, most patients could go back to their ordinary work with support at the work place. Individual adjustments at the work place might be necessary as secondary prevention and gender aspects need be taken into consideration. Since domestic strain is reported to be a

53 Discussion ______hindrance in rehabilitation for women (105) a more even distribution of unpaid work between women and men would be desirable. The employer could also adjust the work setting with preventive strategies for the women. For example, the opportunity to perform physical exercise might be difficult for women outside work because of household responsibilities. In some Swedish male dominated occupations, i.e. police, fireman, and military, physical activity is part of the work time. This is not obvious in female dominated work, i.e. health care. Another important factor in work rehabilitation is to have continuous contact with the supervisors and work-mates. This has previously been shown to be important for women with sickness absence due to work related strain. Support from the employer and work-mates increased the women’s self- confidence and encouraged them to return to work (88). This is important for both women and men.

54 Conclusion ______

CONCLUSIONS

• Compared to a general population, patients with burnout perceived more demands at work and less social support. The difference is more pronounced among women.

• Gender differences among burnout patients are mostly the same as those seen in society as a whole, but female patients are more vulnerable due to lack of emotional support.

In rehabilitation for patients on long-term sick leave because of burnout

• Patients improve in psychological variables and sick leave rates from both a program with CBR combined with Qigong and one with Qigong alone, and there were no differences between the programs.

• CBR combined with Qigong has advantages compared to Qigong alone when drop outs are included as well as when patients with additional conversational therapy are excluded from the analyses.

• CBR combined with Qigong seems to suit women better than men.

• Patients experienced tools learned in both CBR combined with Qigong and Qigong alone as helpful in recovery and for taking charge over their lives.

• Qigong twice a week for 12 weeks has no additional effect as compared to basic care.

• Good encounters with affirmation and support are important as a basis for behavioural change.

• Group rehabilitation has advantages as recognition and support from the group.

• Early rehabilitation measures are important to prevent long-term sickness absence.

55 Acknowledgements ______

ACKNOWLEDGEMENTS

Despite some troublesome life events during my years as a PhD student, I have mostly positive memories of this time. I have given this some thought and found that my social support must be a strong reason for this. I would like to thank all of my friends who have supported me during these years and I would particularly like to express my sincere gratitude to:

Christina Ahlgren, my main supervisor, for sharing your knowledge, and for always supporting and encouraging me. You are my model both as a physiotherapist and as a person. You have been all that I could have asked for as a supervisor.

Lisbeth Slunga Birgander, my co-supervisor and employer, for your support, advice and brilliant knowledge in this research field. It has been a privilege to have the opportunity to perform research in the clinic. You have been all that I could ask for of an employer.

Bernt Lindahl, my co-supervisor, for your support and comments on my work and your perpetually positive attitude.

Patients from both REST and QIST, for you patiently answered questionnaires, performed physical tests, and participated in the rehabilitation programs.

Co-authors Ann-Christine Fjellman-Wiklund, Gunilla Burell, Katarina Steinholtz, Curt Edlund, Leif Nilsson, Anders Knutsson, Birgitta Stegmayr. For all your specialist knowledge, support and help with putting our results to paper.

Leif Nilsson, for your help with my never-ending statistical questions. Evidently, no questions are too stupid for you.

Anne Nafziger, for your excellent and quick language revisions.

Monika, for your support with the layout of my thesis.

My colleagues and friends, at the Division of Occupational and Environmental Medicine at Umeå University, the Stress Clinic and the Occupational and Environmental Medicine Clinic at the University Hospital of Umeå, for encouraging discussions on general as well as scientific topics, and for sharing many laughs during lunchtimes and coffee breaks. It is never boring to go to the work. A special thanks to my colleagues at the Stress Clinic, for my exciting journey with you starting with the Stress Clinic in 2002 with the REST project, and for the opportunity to participate in your team work.

56 Acknowledgements ______

Fellow PhD students in the Division of Occupational and Environmental Medicine at Umeå University, for sharing experiences through hard work and pleasant meetings.

Sofia Norlund and Anita Pettersson Strömbäck, my dear roommates during these years, for your friendship and emotional support during sunny and rainy days.

Katarina Steinholtz, my dear friend and colleague, for your friendship and support during all the years we have worked together.

Ullis and Esse, my dear friends, because you are always there for me and for all our pleasant times together.

Patrik and Camilla, for your emotional and practical support.

Annika, my dear friend and training mate, for your friendship and support and for all the wonderful experiences we have together.

Lotta, my Sunday’s running mate, for all our conversations and your positive attitude.

My family, och speciellt min mamma, jag uppskattar din trygghet och förnöjsamhet. Tack också för hjälp med barnen när jag har varit bortrest.

Emanuel and Elina, my wonderful children, because you remind me of what is important in life. I love you with all my heart.

This thesis was supported by grants from the Swedish Council for Working Life and Social Research, the Swedish Research Council, the Swedish Social Insurance Agency, the Faculty of Medicine, Umeå University, and the Västerbotten County Council.

57 References ______

REFERENCES

1. Cox T, Tisserand M, Taris T (Editorial). The conceptualization and measurements of burnout: Questions and directions. Work Stress 2005;19(3):187-191. 2. Hallsten L, Bellaagh K, Gustafsson K. Utbränning i Sverige en populationsstudie [Burnout in Sweden a population study]. Arbete och Hälsa 2002:6. Stockholm: National Institute for Working Life; (in Swedish); 2002. 3. Soares J, Grossi G, Sundin Ö. Burnout among women: associations with demographic/socio-economic, work, life-style and health factors. Arch Womens Ment Health 2007;10:61-71. 4. Ahola K, Honkonen T, Isometsä E, Kalimo R, Nykyri E, Aromaa A, et al. The relationship between job-related burnout and depressive disorders - results from the Finnish Health 2000 Study. J Affect Disord 2005;88:55-62. 5. Lindblom KM, Linton SJ, Fedeli C, Bryngelsson IL. Burnout in the working population: relations to psychosocial work factors. Int J.Behav Med 2006;13(1):51-59. 6. Magnusson Hansson LL, Theorell T, Oxenstierna G, Hyde M, Westerlund H. Demand, control and social climate as predictors of emotional exhaustion symptoms in working Swedish men and women. Scand J Public Health 2008;36:737-743. 7. Schaufeli W, Enzmann D. The burnout companion to study & practice: a critical analysis. London: Taylor & Francis; 1998. 8. Grossi G, Perski A, Evengård.B, Blomkvist V, Orth-Gomér K. Physiological correlates of burnout among women. J Psychosom Res 2003;55:309-316. 9. Peterson U, Demerouti E, Bergström G, Samuelsson M, Åsberg M, Nygren Å. Burnout and physical and mental health among Swedish healthcare workers. J Adv Nurs 2008;62(1):84-95. 10. Maslach C, Schaufeli WB, Leiter MP. Job Burnout. Annu Rev Psychol 2001;52:397-422. 11. Ahola K, Honkonen T, Virtanen M, Aromaa A, Lönnqvist J. Burnout in relation to age in the adult working population. J Occup Health 2008;50:362-365. 12. Ahola K, Toppinen-Tanner S, Huuhtanen P, Koskinen A, Väänänen A. Occupational burnout and chronic work disability: An eight-year cohort study on pensioning among Finnish forest industry workers. J Affect Disord (2008), doi: 10.1016/j.jad.2008.09.021. 13. Ahola K, Honkonen T, Isometsä E, Kalimo R, Nykyri E, Koskinen S, et al. Burnout in the general population. Soc Psych Psych Epid 2006;41:11-17. 14. Schaufeli WB, Greenglass ER. Introduction to special issue on burnout and health. Psychology and Health 2001;16:501-510. 15. Freudenberger HJ. Staff Burn-Out. J Soc Iss 1974;30(1):159-165.

58 References ______

16. Maslach C. Burned-out. Human Behavior 1976;5:16-22. 17. Maslach C, Jackson S, Leiter M. Maslach Burnout Inventory Manual. Third ed. Palo Alto, California: Consulting Psychologists; 1996. 18. Pines A, Aronson E. Combatting Burnout. Child Youth Serv Rev 1983;5:263-275. 19. Shirom A. Burnout in work organizations. In: Cooper CL and Robertson I, ed., International Review of Industrial and Organizational Psychology 1989 (pp. 25-48). New York: John Wiley & Sons Ltd. 20. National Board of Health and Welfare. Utmattningssyndrom. Stressrelaterad psykisk ohälsa. (Exhaustion syndrome. Stress- related mental poor health) (in Swedish), Bjurner & Bruno AB, Stockholm; 2003. 21. National Board of Health and Welfare. Ändringar i och tillägg till klassifikation av sjukdomar och hälsoproblem 1997 (KSH97) alfabetisk förteckning. (Changes in and supplement to classifications of diseases and health problems 1997 - an alphabetic list.) (in Swedish), Stockholm; 2005. 22. Schaufeli WB, Taris TW. The conceptualization and measurements of burnout: Common ground and worlds apart. Work Stress 2005;19(3):256-262. 23. Karasek R, Theorell T. Healthy work: Stress, productivity and the reconstruction of working life. New York: Basic Books; 1990. 24. Siegrist J. Adverse health effects of high-effort/low-reward conditions. J Occup Health Psychol 1996;1(1):27-41. 25. Demerouti E, Nachreiner F, Bakker AB, Schaufeli WB. The job demands-resources model of burnout. J Appl Psychol 2001;86(3):499-512. 26. Theorell T, Karasek RA. Current issues relating to psychosocial job strain and cardiovascular disease research. J Occup Health Psychol 1996;1(1):9-26. 27. Stansfeld S, Candy B. Psychosocial work environment and mental health - a meta-analytic review. Scand J Work Environ Health 2006;32(6 ):443-462. 28. Johnson JV, Hall EM. Job strain, work place social support, and cardiovascular disease: a cross-sectional study of a random sample of the Swedish working population. Am J Public Health 1988;78(10):1336-1342. 29. Siegrist J, Starke D, Chandola T, Godin I, Marmot M, Niedhammer I, et al. The measurement of effort-reward imbalance at work: European comparisons. Soc Sci Med 2004;58:1483-1499. 30. Bakker AB, Demerouti E, Euwema MC. Job resources buffer the impact of job demands on burnout. J Occup Health Psychol 2005;10(2):170-180.

59 References ______

31. Ericsson-Lidman E, Norberg A, Strandberg G. Meanings of being a female co-worker to a person developing burnout. Scand J Caring Sci 2007;21:155-162. 32. Pines A. Marriage burnout: a new conceptual framework for working with couples. Psychother Priv Pract 1987;5:31-44. 33. Lindahl Norberg A. Burnout in mothers and fathers of children surviving brain tumour. J Clin Psychol Med Settings 2007;14:130- 137. 34. Ekstedt M, Fagerberg I. Lived experiences of the time preceding burnout. J Adv Nurs 2005;49(1):59-67. 35. Mellner C, Krantz G, Lundberg U. Symptom reporting and self- rated health among women in mid-life: the role of work characteristics and family responsibilities. Int J Behav Med 2006;13(1):1-7. 36. Bekker MHJ, Croon MA, Bressers B. Childcare involvement, job characteristics, gender and work attitudes as predictors of emotional exhaustion and sickness absence. Work Stress 2005;19(3):221-237. 37. Stenlund T, Ahlgren C, Lindahl B, Burell G, Knutsson A, Stegmayr B, et al. Patients with burnout in relation to gender and a general population. Scand J Public Health 2007;35:516-523. 38. Pines A, Aronson E, Kafry D. Burnout: From tedium to personal growth. New York: The Free Press; 1981. 39. Kushnir T, Melamed S. The Gulf War and its impact on burnout and well-being of working civilians. Psychol Med 1992;22:987- 995. 40. Melamed S, Kushnir T, Shirom A. Burnout and risk factors for cardiovascular diseases. Behav Med 1992;18:53-60. 41. Melamed S, Ugarten U, Shirom A, Kahana L, Lerman Y, Froom P. Chronic burnout, somatic arousal and elevated salivary cortisol levels. J Psychosom Res 1999;46:591-598. 42. Halbesleben JRB, Demerouti E. The construct validity of an alternative measure of burnout: Investigating the English translation of the Oldenburg Burnout Inventory. Work Stress 2005;19(3):208-220. 43. Kristensen TS, Borritz M, Villadsen E, Christensen KB. The Copenhagen Burnout Inventory: A new tool for the assessment of burnout. Work Stress 2005;19(3):192-207. 44. jr. Ahlborg G, Ljung T, Swan G, Glise K, Jonsdottir I, Hadzibajramovic E, et al. Stressrelaterad ohälsa bland anställda vid Västra Götalandsregionen och Försäkringskassan i Västra Götalands län. Delrapport 1 - enkätundersökning i maj-juni 2004. [Stress-related ill health among employees in the county council and the regional Social Insurance Agency in Västra Götaland. Report nr. 1 - survey between may and june 2004] (in Swedish), Göteborg; 2006. 45. Shirom A. Reflections on the study of burnout. Work Stress 2005;19(3):263-270.

60 References ______

46. Forsakringskassan.se [homepage on the Internet]. Statistik. Nybeviljade sjukersättningar/aktivitetsersättningar. Fördelning på län och diagnos, 2006.(Statistics. New compensations. Divided at county and diagnoses, 2006). Report No. 2007:2 (in Swedish), Stockholm: Swedish Social Insurence Agency; 2007. Available from: http://www.forsakringskassan.se/. 47. Parent-Thirion A, Macias E, Hurley J, Vermeylen G. Fourth European Working Conditions Survey. European foundation for the improvement of living and working conditions, Dublin 2007. 48. Arbetsorsakade besvär 2008. Arbetsmiljöstatistik Rapport 2008:5. (Work-related disorders 2008. Work environment statistics, report 2008:5). Arbetsmiljöverket. Available from: http://www.av.se/. 49. Cohen-Katz J, Wiley SD, Capuano T, Baker DM, Shapiro S. The effects of mindfulness-based stress reduction on nurse stress and burnout. A quantitative and qualitative study. Holist Nurs Pract 2004;18(6):302-308. 50. Granath J, Ingvarsson S, von Thiele U, Lundberg U. Stress management: a randomized study of cognitive behavioural therapy and Yoga. Cogn Behav Ther 2006;35(1):3-10. 51. Hasson D, Anderberg U, Theorell T, Arnetz B. Psychophysiological effects of a web-based stress management system: a prospective, randomized controlled intervention study of IT and media workers. BMC Public Health 2005;5(78):1-14. 52. Peterson U, Bergström G, Samuelsson M, Åsberg M, Nygren Å. Reflecting peer-support groups in the prevention of stress and burnout: randomized controlled trial. J Adv Nurs 2008;63(5):506-516. 53. Shawe-Taylor M, Rigby J. Cognitive behaviour therapy: its evolution and basic principles. Royal Society for the Promotion of Health Journal 1999;119(4):244-246. 54. Kåver A. KBT i utveckling. En introduktion till kognitiv beteendeterapi. [The development of Cognitive Behavioural Therapy. An introduction to cognitive behavioural therapy] (in Swedish). Stockholm: Natur och Kultur; 2006. 55. Sanders D, Wills F. Cognitive Therapy. An introduction. Second ed. SAGE Publications; London; 2005. 56. Sancier K. Medical applications of Qigong. Altern Ther 1996;2(1):40-46. 57. Tsang H, Cheung L, Lak D. Qigong as a psychosocial intervention for depressed elderly with chronic physical illnesses. Int J Geriatr Psychiatry 2002;17:1146-1154. 58. Umeå Index 2008, statistics. Available from: http://www.umea.se/. 59. Stegmayr B, Lundberg V, Asplund K. The events registration and survey procedures in the Northern Sweden MONICA Project. Scand J Public Health 2003;31(Suppl. 61):9-17.

61 References ______

60. Mattila-Evenden M, Svanborg P, Gustavsson P, Åsberg M. Determinants of self-rating and expert rating concordance in psychiatric out-patients, using the affective subscales of the CPRS. Acta Psychiatr Scand 1996;94:386-396. 61. Svanborg P, Åsberg M. A new self-rating scale for depression and anxiety states based on the Comprehensive Psychopathological Rating Scale. Acta Psychiatr Scand 1994;89:21-28. 62. Zigmond A, Snaith R. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 1983;67:361-370. 63. Claesson M, Slunga Birgander L, Lindahl B, Nasic S, Åström M, Asplund K, et al. Women's hearts - stress management for women with ischemic heart disease. J Cardiopulm Rehabil 2005;25:93- 102. 64. Beurskens A, Bultmann U, Kant I, Vercoulen J, Bleijenberg G, Swaen G. Fatigue among working people: validity of a questionnaire measure. Occup Environ Med 2000;57:353-357. 65. Robson P. Development of a new self-report questionnaire to measure self esteem. Psychol Med 1989;19:513-518. 66. Sullivan M, Karlsson J, Ware Jr J. The Swedish SF-36 Health Survey I. Evaluation of data quality, scaling assumptions, reliability and construct validity across general populations in Sweden. Soc Sci Med 1995;41(10):1349-1358. 67. Persson L, Karlsson J, Bengtsson C, Steen B, Sullivan M. The Swedish SF-36 Health Survey II. Evaluation of Clinical Validity: Results from population studies of elderly and women in Gothenborg. J Clin Epidemiol 1998;51(11):1095-1103. 68. Crist DA, Rickard HC, Prentice-Dunn S, Barker HR. The Relaxation Inventory: Self-Report Scales of Relaxation Training Effects. J Pers Assess 1989;53(4):716-726. 69. Lindgren B, Sundin Ö. S-PARI. Den svenska versionen av the Physical Assessment Scale of the Relaxation Inventory (PARI). (The Swedish version of the Physical Assessment Scale of the Relaxation Inventory) (in Swedish). Nordisk Fysioterapi 1997;1:114-120. 70. Kohn M, Schooler C. Work and personality: an inquiry into the impact of social stratification. Norwood, NJ: Ablex; 1983. 71. Undén A, Orth-Gomér K. Development of a social support instrument for use in population surveys. Soc Sci Med 1989;29:1387-1392. 72. Barnekow-Bergkvist M, Hedberg G, Janlert U, Jansson E. Development of muscular endurance and strength from adolescence to adulthood and level of physical capacity in men and women at the age of 34 years. Scand J Med Sci Sports 1996;6:145-155. 73. Bränström H, Malmgren-Olsson E, Barnekow-Bergkvist M. Balance performance in patients with whiplash associated disorders and patients with prolonged musculoskeletal disorders. Adv Physiother 2001;3:120-127.

62 References ______

74. Oja P, Laukkanen R, Pasanen M, Tyry T, Vuori I. A 2-km walking test for assessing the cardiorespiratory fitness of healthy adults. Int J Sports Med 1991;12:356-362. 75. Oja P, Mänttäri A, Pokki T, Kukkonen-Harjala K, Laukkanen R, Malmberg J, et al. Tester's guide. UKK walk test. Tampere, Finland; 2001. 76. Laukkanen R, Oja P, Pasanen M, Vuori I. Criterion validity of a two-kilometer walking test for predicting the maximal oxygen uptake of moderately to highly active middle-aged adults. Scand J Med Sci Sports 1993;3:267-272. 77. Laukkanen R, Oja P, Pasanen M, Vuori I. Validity of a two kilometre walking test for estimating maximal aerobic power in overweight adults. I J Obes 1992;16:263-268. 78. Burell G, Granlund B. Women's hearts need special treatment. Int J Behav Med 2002;9:228-242. 79. Bergdahl J, Larsson A, Nilsson LG, Riklund Åhlström K, Nyberg L. Treatment of chronic stress in employees: Subjective, cognitive and neural correlates. Scand J Psychol 2005;46:395-402. 80. Dawson-Saunders B, Trapp RG. Basic & Clinical Biostatistics. East Norwalk, Connecticut: Appleton & Lange: 1994. 81. Glaser BG, Strauss AL. The discovery of grounded theory - strategies for qualitative research. Chicago; Aldine: 1967. 82. Dahlgren L, Emmelin M, Winkvist A. Qualitative methodology for international public health. Department of Public Health and Clinical Medicine: Umeå University; 2004. 83. Heiden M, Lyskov E, Nakata M, Sahlin K, Sahlin T, Barnekow- Bergkvist M. Evaluation of cognitive-behavioural training and physical activity for patients with stress-related illness: a randomized controlled study. J Rehabil Med 2007;39:366-373. 84. Brattberg G. Internet-based rehabilitation for individuals with chronic pain and burnout: a randomized trial. I J Rehabil Res 2006;29:221-227. 85. Yrkesregistret med yrkesstatistik 2006. Statistiska Centralbyrån [The Swedish occupational register with statistic 2006, Statistics Sweden]. 86. Berntsson L, Lundberg U, Krantz G. Gender differences in work- home interplay and symptom perception among Swedish white- collar employees. J Epidemiol Community Health 2006;60:1070- 1076. 87. de Smet P, Sans S, Dramaix M, Boulenguez C, de Backer G, Ferrario M, et al. Gender and regional differences in perceived job stress across Europe. Eur J Public Health 2005;15(5):536-545. 88. Holmgren K, Dahlin Ivanoff S. Women on sickness absence - views of possibilities and obstacles for returning to work. A focus group study. Disabil Rehabil 2004;26(4):213-222. 89. Verdonk P, de Rijk A, Klinge I, de Vries A. Sickness absence as an interactive process: Gender experiences of young, highly educated

63 References ______

women with mental health problems. Patient Educ Couns 2008;73:300-306. 90. Kawachi I, Berkman LF. Social ties and mental health. J Urban Health 2001;78(3):458-467. 91. Blonk R, Brenninkmeijer V, Lagerveld S, Houtman I. Return to work: A comparison of two cognitive behavioural interventions in cases of work-related psychological complaints among the self- employed. Work Stress 2006;20(2):129-144. 92. van der Klink J, Blonk R, Schene A, van Dijk F. Reducing long term sickness absence by an activating intervention in adjustment disorders: a cluster randomised controlled design. Occup Environ Med 2003;60:429-437. 93. de Vente W, Kamphuis J, Emmelkamp P, Blonk R. Individual and group cognitive-behavioral treatment for work-related stress complaints and sickness absence: a randomized controlled trial. J Occup Health Psychol 2008;13(3):214-231. 94. Lambert M. Bergin and Garfield´s handbook of psychotherapy and behavioral change. Fifth Ed. New York: John Wiley & Sons, Inc.; 2004. 95. Pinheiro CPO, da Silva RM, Mamede MV, Fernandes AFC. Participating in a support group: Experience lived by women with breast cancer. Rev Latino-am Enfermagerm 2008;16(4):733-738. 96. Mimura C, Griffiths P. The effectiveness of current approaches to workplace stress management in the nursing profession: an evidence based literature review. Occup Environ Med 2003;60:10-15. 97. van der Klink J, Blonk R, Schene A, van Dijk F. The benefits of interventions for work-related stress. Am J Public Health 2001;91(2):270-276. 98. Richardsson K, Rothstein H. Effects of occupational stress management intervention programs: a meta-analysis. J Occup Health Psychol 2008;13(1):69-93. 99. Jensen IB, Bergström G, Ljungquist T, Bodin L, Nygren ÅL. A randomized controlled component analysis of a behavioral medicine rehabilitation program for chronic spinal pain: are the effects dependent on gender? Pain 2001;91:65-78. 100. Watson HJ, Nathan PR. Role of gender in depressive disorder outcome for individual and group cognitive-behavioral treatment. J Clin Psychol 2008;64(12):1323-1337. 101. Lidwall U, Marklund S, Skogman Thoursie P. Utvecklingen av sjukfrånvaron i Sverige. [Development of the sickness absence in Sweden] (in Swedish) In: Gustafsson RÅ, Lundberg I (eds). Arbetsliv och hälsa 2004 (pp. 173-193) Stockholm: Arbetslivsinstitutet; 2004. 102. Kompier MAJ. New systems of work organization and workers' health. Scand J Work Environ Health 2006;32(6, special issue):421-430.

64 References ______

103. Johannisson K. När otillräckligheten byter namn. Ett idéhistoriskt perspektiv på den nya ohälsan (in Swedish) In: Ekman R, Arnetz B. Stress: molekylerna, individen, organisationen, samhället (pp. 13-29) Stockholm: Liber AB; 2002. 104. Engström LG, Janson S. Stress-related sickness absence and return to labour market. Disabil Rehabil 2007;29(5):411-416. 105. Östlund G, Cedersund E, Hensing G, Alexandersson K. Domestic strain: a hindrance in rehabilitation? Scand J Caring Sci 2004;18:49-56. 106. Kunz R, Vist GE, Oxman AD. Randomisation to protect against selection bias in health care trials (Review). Cochrane Library 2008;4. 107. TenHave TR, Coyne J, Salzer M, Katz I. Research to improve the quality of care for depression: alternatives to the simple randomized clinical trial. Gen Hosp Psych 2003;25:115-123. 108. Spielberger CD. Assessment of state and trait anxiety: conceptual and methodological issues. Southern Psychologist 1985;2:6-16. 109. Johansson M, Hassmén P, Jouper J. Acute effects of Qigong exercise on mood and anxiety. Int J Stress Manage 2008;15(2):199-207. 110. Brattberg G. PTSD and ADHD: underlying factors in many cases of burnout. Stress and Health 2006b;22:305-313. 111. Almén N. Stress- och utmattningsproblem. Kognitiva och beteendeterapeutiska metoder. [Stress and exhaustion disorder. Cognitive and behavioural approaches] (in Swedish). Studentlitteratur; 2007. 112. Sandström A, Nyström Rhodin I, Lundberg M, Olsson T, Nyberg L. Impaired cognitive performance in patients with chronic burnout syndrome. Biol Psychol 2005;69:271-279. 113. Öhman L, Nordin S, Bergdahl J, Slunga Birgander L, Stigsdotter Neely A. Cognitive function in outpatients with perceived chronic stress. Scand J Work Environ Health 2007;33(3):223-232. 114. Hillman CH, Erickson KI, Kramer AF. Be smart, exercise your heart: exercise effects on brain and cognition. Nat Rev Neurosci 2008;9(1):58-65. 115. Harris AHS, Cronkite R, Moos R. Physical activity, exercise coping, and depression in a 10-year cohort study of depressed patients. J Affect Disord 2006;93:79-85. 116. Kallings LV, Leijon M, Hellénius ML, Ståhle A. Physical activity on prescription in primary health care: a follow-up of physical activity level and quality of life. Scand J Med Sci Sports 2008;18:154-161. 117. Stenlund T, Ahlgren C, Lindahl B, Burell G, Steinholtz K, Edlund C, et al. Cognitively oriented behavioral rehabilitation in combination with Qigong for patients on long-term sick leave because of burnout: REST - a randomized clinical trial. Int J Behav Med, published online, 16 January 2009.

65 References ______

118. Tsang H, Mok C, Yeung A, Chan S. The effect of Qigong on general and psychosocial health of elderly with chronic physical illnesses: a randomized clinical trial. Int J Geriatr Psychiatry 2003;18:441- 449. 119. Mannerkorpi K, Arndorw M. Efficacy and feasibility of a combination of body awareness therapy and qigong in patients with fibromyalgia: a pilot study. J Rehabil Med 2004;36:279-281. 120. Wenneberg S, Gunnarsson LG, Ahlström G. Using a novel exercise programme for patients with muscular dystrophy. Part II: a quantitative study. Disabil Rehabil 2004;26(10):595-602. 121. Stenlund T, Lindström B, Granlund M, Burell G. Cardiac rehabilitation for the elderly: Qigong and group discussions. Eur J Cardiovasc Prev Rehabil 2005;12:5-11. 122. Cheung B, Lo J, Fong D, Chan M, Wong S, Wong V, et al. Randomised controlled trial of qigong in the treatment of mild essential hypertension. J Hum Hypertens 2005; 19:697-704. 123. Tsang H, Fung K, Chan A, Lee G, Chan F. Effect of a qigong exercise programme on elderly with depression. Int J Geriatr Psychiatry 2006;21:890-897. 124. Lansinger B, Larsson E, Persson L, Carlsson J. Qigong and exercise therapy in patients with long-term neck pain. SPINE 2007;32(22):2415-2422.

66 Dissertations ______

Dissertations written by physiotherapists, Umeå University 1989–2009 Birgitta Bergman Being a physiotherapist - Professional role, utilization of time and vocational strategies. Umeå University Medical Dissertations, New Series no 251, 1989 (Department of Physical Medicine and Rehabilitation)

Inger Wadell Influences from peripheral sense organs on primary and secondary spindle afferents via gamma-motoneurones - A feedback mechanism for motor control and regulation of muscle stiffness. Umeå University Medical Dissertations, New Series no 307, 1991 (Department of Physiology)

Jessica Elert The pattern of activation and relaxation during fatiguing isocinetic contractions in subjects with and without muscle pain. Arbete och hälsa 1991:47. Diss. 1992 (sammanfattning). (Departments of Clinical Physiology, National Institute of Occupational Health and Physical Medicine and Rehabilitation)

Gunnevi Sundelin Electromyography of shoulder muscles - The effects of pauses, drafts and repetitive work cycles. Arbete och Hälsa, 1992:16. Diss. (sammanfattning). (Departments of Anatomy, National Institute of Occupational Health, Division of Work and Environmental Physiology, Divisions of Occupational Medicine and Applied Work Physiology and Occupational Medicine)

Birgit Rösblad Visual and proprioceptive control of arm movement - Studies of development and dysfunction. Diss. (sammanfattning) 1994 (Department of Paediatrics)

Charlotte Häger-Ross To grip and not to slip - Sensorimotor mechanisms during reactive control of grasp stability. Umeå University Medical Dissertations, New Series no 429, 1995 (Department of Physiology)

Lars Nyberg Falls in the frail elderly – Incidence, characterics and prediction with special reference to patients with stroke and hip fractures. Umeå Medical Dissertations, New Series no 483, 1996 (Department of Geriatric Medicine)

Margareta Barnekow-Bergkvist Physical capacity, physical activity and health -A population based fitness study of adolescents with an 18-year follow-up. Umeå University Medical Dissertations, New Series no 494, 1997 (Departments of Physiology and Technology, National Institute for Working Life and Epidemiology and Public Health)

Britta Lindström Knee muscle function in healthy persons and patients with upper motor neurone syndrome. Umeå University Medical Dissertations, New Series no 505, 1997 (Departments of Physical Medicine and Rehabilitation and Clinical Neuroscience)

Monica Mattsson Body Awareness - applications in physiotherapy. Umeå University Medical Dissertations, New Series no 543, 1998 (Departments of Psychiatry and Family Medicine)

67 Dissertations ______

Hildur Kalman The structure of knowing. Existential trust as an epistemological category. Umeå studies in the humanities, 145, 1999 (Department of Philosophy and Linguistics)

Lillemor Lundin-Olsson Prediction and prevention of falls among elderly people in residential care. Umeå University Medical Dissertations, New Series no 671, 2000 (Departments of Community Medicine and Rehabilitation, Physiotherapy and Geriatric Medicine)

Christina Ahlgren Aspects of rehabilitation – with focus on women with trapezius myalgia Umeå University Medical Dissertations, New Series no 715, 2001 (Department of Public Health and Clinical Medicine, Occupational Medicine)

Ann Öhman Profession on the move - changing conditions and gendered development in physiotherapy. Umeå University Medical Dissertations, New series No 730, 2001 (Departments of Community Medicine and Rehabilitation, Physiotherapy and Public Health and Clinical Medicine, Epidemiology)

Hamayun Zafar Integrated jaw and neck function in man: studies of mandibular and head-neck movements during jaw opening-closing tasks. (Department of Odontology, Clinical Oral Physiology and Centre for Musculoskeletal Research, National Institute for Working Life, Umeå)

Kerstin Söderman The female soccer player – Injury pattern, risk factors and intervention. Umeå University Medical Dissertations, New series no 735, 2001 (Departments of Surgical and Perioperative Sciences, Sports Medicine, and Community Medicine and Rehabilitation, Physiotherapy)

Lena Grönblom-Lundström Rehabilitation in light of different theories of health. Outcome for patients with low-back complaints – a theoretical discussion. Umeå University Medical Dissertations, New series no 760, 2001 (Departments of Public Health and Clinical Medicine, Epidemiology, and Community Medicine and Rehabilitation, Social Medicine)

Kerstin Waling Pain in women with work-related trapezius myalgia. Intervention effects and variability. Umeå University Medical Dissertations, New series no 762, 2001 (Departments of Public Health and Clinical Medicine, Occupational Medicine, and Community Medicine and Rehabilitation, Physiotherapy)

Eva-Britt Malmgren-Olsson Health problems and treatment effects in patients with non-specific musculoskeletal disorders. A comparitson between Body Awareness Therapy, Feldenkrais and Individual Physiotherapy. Umeå University Medical Dissertations, New series no 774, 2002 (Department of Community Medicine and Rehabilitation, Physiotherapy and Department of Psychology)

Jane Jensen Fall and injury prevention in older people living in residential care facilities. Umeå University Medical Dissertations, New series no 812, 2003 (Department of Community Medicine and Rehabilitation, Physiotherapy and Geriatric Medicine)

Anncristine Fjellman-Wiklund Musicianship and teaching. Aspects of musculoskeletal disorders, physical and psychosocial work factors in musicians with focus on music teachers. Umeå University Medical Dissertations, New series no 825, 2003 (Department of Community Medicine and Rehabilitation, Physiotherapy)

68 Dissertations ______

Börje Rehn Musculoskeletal disorders and whole-body vibration exposure among professional drivers of all-terrain vehicles. Umeå University Medical Dissertations, New series no 852, 2004 (Department of Public Health and Clinical Medicine, Occupational Medicine)

Martin Björklund Effects on repetitive work on proprioception and of stretching on sensory mechanisms. Implications for work-related neorumuscular disorders. Umeå University Medical Dissertations, New series no 877, 2004 (Departments of Surgical and Perioperative Sciences, Sports Medicine Unit, Umeå University, The Center for Musculoskeletal Reseach, University of Gävle, Umeå, and Alfta Forskningsstiftelse, Alfta)

Karin Wadell Physical training in patients with chronic obstructive pulmonary disease – COPD. Umeå University Medical Dissertations, New series no 917, 2004 (Departments of Community Medicine and Rehabilitation, Physiotherapy; Public Health and Clinical Medicine, Respiratory Medicine and Allergy, Surgical and Perioperative Sciences, Sports Medicine)

Peter Michaelson Sensorimotor characteristics in chronic neck pain. Possible pathophysiological mechanisms and implications for rehabilitation. Umeå University Medical Dissertations, New series no 924, 2004 (Departments of Surgical and Perioperative Sciences, Sports Medicine Unit, University of Umeå, Southern Lappland Research Department, Vilhelmina, Centre for Musculoskeletal Research, University of Gävle, Umeå)

Ulrika Aasa Ambulance work. Relationships between occupational demands, individual characteristics and health related outcomes. Umeå University Medical Dissertations, New series no 943, 2005 (Department of Surgical and Perioperative Sciences, Sports Medicine and Surgery, University of Umeå and Centre for Musculoskeletal Research, University of Gävle)

Ann-Katrin Stensdotter Motor Control of the knee. Kinematic and EMG studies of healthy individuals and people with patellofemoral pain. Umeå University Medical Dissertations, New series no 987, 2005 (Departments of Community Medicine and Rehabilitation, Physiotherapy)

Tania Janaudis Ferreira Aspects of muscle function and training with oxygen in patients with chronic obstructive pulmonary disease – COPD. Umeå University Licentiate Thesis, 2005 (Department of Community Medicine and Rehabilitation, Physiotherapy)

Erik Rosendahl Fall prediction and a high-intensity functional exercise programme to improve physical functions and to prevent falls among older people living in residential care facilities. Umeå University Medical Dissertations, New Series No 1024, 2006 (Department of Community Medicine and Rehabilitation, Geriatric Medicine and Physiotherapy)

Michael Stenvall Hip fractures among old people. Their prevalence, consequences and complications and the evaluation of a multi-factorial intervention program designed to prevent falls and injuries and enhance performance of activities of daily living. Umeå University Medical Dissertations, New Series No 1040, 2006 (Department of Community Medicine and Rehabilitation, Geriatric Medicine and Physiotherapy)

Petra von Heideken Wågert Health, physical ability, falls and morale in very old people: the Umeå 85+ Study. Umeå University Medical Dissertations, New Series No 1038, 2006 (Department of Community Medicine and Rehabilitation, Geriatric Medicine and Physiotherapy)

69 Dissertations ______

Gerd Flodgren Effect of low–load repetitive work and mental load on sensitising substances and metabolism in the trapezius muscle. Umeå University Medical Dissertations, New series no 1130, 2007 (Department of Surgical and Perioperative Sciences, Sports Medicine Unit, Umeå University, Centre of Musculoskeletal Research, University of Gävle, Umeå, and the Department of Community Medicine and Rehabilitation, Rehabilitation Medicine, Umeå University, Umeå Sweden)

Staffan Eriksson, Falls in people with dementia. Umeå University Medical Dissertations, New series no 1135, 2007 (Department of Community Medicine and Rehabilitation, Physiotheraphy and Geriatric Medicine, Umeå University, Umeå, Sweden)

Jonas Sandlund, Position-matching and goal-directed reaching acuity of the upper limb in chronic neck pain: Associations to self-rated characteristics. Umeå University Medical Dissertations, New series no 1182, 2008 (Department of Surgical and Perioperative Sciences, Sports Medicine Unit, Umeå University, Centre of Musculoskeletal Research, University of Gävle, Umeå)

Gunilla Larsson Motor function over time in Rett syndrome-loss, difficulties and possibilities. Umeå University Licentiate Thesis, 2008 (Department of Community Medicine and Rehabilitation, Physiotherapy)

Charlotte Åström Effects of vibration on muscles in the neck and upper limbs. With focus on occupational terrain vehicle drivers. Umeå University Medical Dissertations, New series no 1135, 2008 (Departments of Community Medicine and Rehabilitation, Physiotherapy)

Ellinor Nordin Assessment of balance control in relation to fall risk among older people. Umeå University Medical Dissertations, New series no 1198, 2008 (Departments of Community Medicine and Rehabilitation, Physiotherapy)

Jenny Röding Stroke in the younger. Self- reported impact on work situation, cognitive function, physical function and life satisfaction. A national survey. Umeå University Medical Dissertations, New series no 1241, 2009 (Departments of Community Medicine and Rehabilitation, Physiotherapy)

70