Exploring the Impact of Fatigue on Work Ability of People with Rheumatic Diseases. - PDF Document

Presentation Transcript

  1. Exploring the Impact of Fatigue on Work Ability of People with Rheumatic Diseases. 1

  2. Exploring the Impact of Fatigue on Work Ability of People with Rheumatic Diseases. Discipline of Occupational Therapy, Trinity College, Dublin and the Rheumatology Department, St. James’ Hospital, Dublin. 2

  3. Exploring the impact of fatigue on work ability of people with Rheumatic Diseases. Edited by: Deirdre Connolly, Lynn O’Toole, Clodagh Fitzpatrick, Michele Doran, Finbar O’Shea. To be cited as: Connolly D, O’Toole L, Fitzpatrick C, Doran M, O’Shea F. (2016) Exploring the impact of fatigue on work ability of people with Rheumatic Diseases. Trinity College, Dublin, and the National Disability Authority, Ireland. Discipline of Occupational Therapy, Trinity Centre for Health Sciences, St. James’ Hospital, James’ Street, Dublin 8 Ireland Tel: +353 1 8963210 Email: connoldm@tcd.ie Rheumatology Department St. James’ Hospital, James’ Street, Dublin 8 Ireland January 2016 3

  4. Acknowledgments The research team wishes to acknowledge and thank The National Disability Authority for funding this project under its Research Promotion Grants Scheme, which in 2014 had the theme of employment. All those attending the weekly Rheumatology clinics in St. James’ Hospital, Dublin, who completed the questionnaires, and to all the people who completed the online survey. The participants of the focus groups and interviews who took the time out of work to travel to the Trinity Centre of Health Sciences to share their experiences of working with a rheumatic disease. The staff of the three voluntary organisations, Arthritis Ireland, Scleroderma Ireland and Lupus Group Ireland, who distributed the link to the online survey to their members. 4

  5. Table of Contents Exploring the Impact of Fatigue on Work Ability of People with Rheumatic Diseases. ........................................................................................................................................ 1 Exploring the impact of fatigue on work ability of people with Rheumatic Diseases. .. 3 Acknowledgments .......................................................................................................... 4 List of Tables .................................................................................................................. 9 Executive Summary ...................................................................................................... 11 Conclusion ................................................................................................................ 12 1. Introduction .............................................................................................................. 14 2. Context of the Research ............................................................................................ 15 2.1 Societal and Economic Cost ............................................................................... 15 2.2 Irish employers’ rights and responsibilities ........................................................ 15 3. Literature Review ..................................................................................................... 17 3.1 Introduction ........................................................................................................ 17 3.2 Challenges in work ............................................................................................. 17 3.3 Work disability in rheumatic diseases ................................................................ 18 3.4 Impact of fatigue on work .................................................................................. 19 3.5 Conclusion .......................................................................................................... 19 4. Methodology ............................................................................................................. 20 4.1 Study aims and objectives .................................................................................. 20 4.2 Study design ....................................................................................................... 20 4.3 Data collection .................................................................................................... 20 Sampling ............................................................................................................... 20 Recruitment Process ............................................................................................. 21 4.4 Data collection methods ..................................................................................... 23 Patient Global Assessment of Disease Activity (PtGA) .......................................... 23 5

  6. Multidimensional Fatigue Inventory ........................................................................ 23 Work Role Functioning Questionnaire ..................................................................... 24 Arthritis Work Spill-over ..................................................................................... 24 EQ-5D- 3L ............................................................................................................ 24 4.5 Qualitative Data Collection ................................................................................ 24 4.6 Data analysis ....................................................................................................... 25 Quantitative .......................................................................................................... 25 Qualitative ............................................................................................................ 25 5. Quantitative results ................................................................................................... 27 5.1 Demographic profile ........................................................................................... 27 5.2 Profile of employment ........................................................................................ 29 5.3 Descriptive Analysis of Outcome Measures ...................................................... 29 Disease activity ..................................................................................................... 29 Impact of fatigue ................................................................................................... 29 Work Ability ......................................................................................................... 30 The Worker Role Functioning (WRF) Questionnaire .......................................... 30 The Arthritis Work Spill-over (AWS) Questionnaire .......................................... 30 Impact on quality of life ....................................................................................... 30 5.4 Correlations ........................................................................................................ 31 Disease activity and relationships with Fatigue, Work and Quality of Life ........ 32 Relationships between fatigue and disease activity, work functioning and quality of life ..................................................................................................................... 33 5.5 Are there differences across severity, age, gender or other variables in disease activity, work role functioning, arthritis-work spill-over and quality of life? ......... 34 Differences based on fatigue severity ....................................................................... 35 Differences in variables based on demographic characteristics ............................... 35 6

  7. Impact of age ........................................................................................................ 35 Impact of Gender .................................................................................................. 37 Impact of education .............................................................................................. 38 Impact of co-morbidity ......................................................................................... 39 Impact of disease duration ........................................................................................ 42 5.6 Differences in variables based on employment characteristics .......................... 44 Working hours ...................................................................................................... 44 Job Type ................................................................................................................... 45 Job Sector ................................................................................................................. 46 5.7 Summary of Quantitative Results ....................................................................... 48 Relationships between variables ........................................................................... 48 Significant differences between variables ............................................................ 48 Demographic differences between fatigue, work and quality of life ....................... 49 Work Characteristics ................................................................................................ 49 6. Qualitative Findings ................................................................................................. 51 6.1 Descriptions of fatigue........................................................................................ 54 Patterns of fatigue ................................................................................................. 54 Factors that increase fatigue ................................................................................. 54 6.2 Impact of fatigue on daily function .................................................................... 55 Impact of fatigue on cognition ............................................................................. 55 Impact of fatigue on mood ................................................................................... 56 Impact of fatigue on physical abilities.................................................................. 56 6.3 Fatigue management strategies ........................................................................... 57 6.4 Disclosure ........................................................................................................... 58 6.5 Recommendations for managing fatigue in work .............................................. 59 7

  8. 6.6 Summary of Qualitative findings ....................................................................... 60 7. Discussion ................................................................................................................. 61 Disclosure ................................................................................................................. 61 Disease activity ......................................................................................................... 62 Fatigue and work ...................................................................................................... 63 Patterns of fatigue and factors that increase fatigue ................................................. 63 Impact of fatigue on work ........................................................................................ 64 Fatigue management strategies ................................................................................. 65 Risk factors for impaired work functioning ............................................................. 66 Demographic characteristics ................................................................................ 66 Work characteristics ............................................................................................. 68 8. Conclusions .............................................................................................................. 69 Bibliography ................................................................................................................. 70 Appendix 1: Definition of terms ................................................................................... 78 Appendix 2: Questionnaire pack (quantitative phase) ................................................. 83 Appendix 3: Focus group and interview schedule (qualitative phase) ....................... 100 8

  9. List of Tables Table 1: Abbreviation list ............................................................................................. 10 Table 2: Comparison of St. James’ and Survey monkey group demographics ............ 22 Table 3: Age and gender demographic variables ......................................................... 27 Table 4: Demographic variables ................................................................................... 27 Table 5: Type of rheumatic disease breakdown ........................................................... 28 Table 6: Disease duration of rheumatic disease ........................................................... 28 Table 7: Employment Details ....................................................................................... 29 Table 8: Mean fatigue scores according to the MFI ..................................................... 30 Table 9: Overall mean worker role functioning scores ................................................ 30 Table 10: EQ-5D descriptive systems breakdown ....................................................... 31 Table 11: Correlations: Disease Activity and outcome measures ................................ 33 Table 12: Correlations: Fatigue and outcome measures .............................................. 34 Table 13: Differences in disease, work and quality of life outcome measures between high and low fatigue ..................................................................................................... 35 Table 14: Differences in disease activity and fatigue levels between age categories .. 36 Table 15: Differences in Work and Quality of Life between Age categories .............. 36 Table 16: Differences in disease activity and fatigue levels between genders ............ 37 Table 17: Differences in work and quality of life between genders ............................ 37 Table 18: Differences in disease activity and fatigue for levels of education .............. 38 Table 19: Differences in work and quality of life between levels of education ........... 39 Table 20: Differences in disease activity and fatigue levels in co-morbidity .............. 41 Table 21: Differences in work and quality of life in co-morbidity .............................. 41 Table 22: Differences in disease activity and fatigue levels for disease duration: ANOVA ........................................................................................................................ 42 Table 23: Differences in work functioning and quality of life for disease duration: ANOVA ........................................................................................................................ 43 Table 24: Disease Duration and outcome measures ANOVA (post hoc test) ............. 43 Table 25: Differences in disease activity and fatigue levels between different working hours ............................................................................................................................. 44 Table 26: Differences in work and quality of life between different working hours ... 45 Table 27: Differences in disease activity and fatigue levels for job type: ANOVA .... 45 Table 28: Differences in work and quality of life for job type: ANOVA .................... 46 Table 29: Job Type and outcome measures ANOVA (post hoc test) .......................... 46 Table 30: Differences in disease activity and fatigue levels for job sector .................. 47 Table 31: Differences in work and quality of life for job sector .................................. 47 Table 32: Demographic, fatigue, work and quality of life profile of focus group and interview participants ................................................................................................... 52 9

  10. Table 1: Abbreviation list ANOVA AWS AS EQ5D-VAS MFI or MFI-20 MFIGF N= NDA NVIVO Analysis of variance Arthritis Work Spill-over Ankylosing Spondylitis EQ-5D Visual Analogue Scale Multidimensional Fatigue Inventory MFI General Fatigue Subscale Number of participants National Disability Authority Nudist Vivo version 10 (NVivo) software analysis for qualitative data Osteoarthritis p-value or probability value Polymyalgia Rheumatica Psoriatic Arthritis Patient Global Assessment of Disease Activity OA p PMR PsA PtGA r RA SLE SSc SPSS ±SD Statistical Significance WRF WRFWSD WRFOD WRFPD WRFMD WRFSD Rho or pearson product-moment correlation coefficient Rheumatoid Arthritis Systemic Lupus Erythematosus Systemic sclerosis Statistical Package for Social Sciences ±Standard Deviation p-value Work Role Functioning Questionnaire Work Role Functioning Work Scheduling Demands Work Role Functioning Output Demands Work Role Functioning Physical Demands Work Role Functioning Mental Demands Work Role Functioning Social Demands 10

  11. Executive Summary Rheumatic diseases such as arthritis are one of the leading costs of work disability throughout the western world. In Ireland, the cost of lost productive time for those with Rheumatoid Arthritis has been estimated at €1.6 billion (Arthritis Ireland, 2008). The Employment Equality Acts 1998-2011 outline the legal obligations employers have to reasonably accommodate employees with a disability. Fatigue is reported in up to 90% of people with rheumatic diseases and therefore must be considered in any study of work disability and productivity (Norheim et al., 2011). This study, which was funded by the National Disability Authority, was carried out to explore the impact of fatigue in people with a rheumatic disease. The specific aims of the study were to: Examine the impact of fatigue on work-related activities for those with rheumatic diseases. Examine the impact of fatigue on the interrelationship between arthritis and work. Explore people’s experiences of the impact of fatigue on their work ability Investigate the range of strategies used by people in the work place to manage their fatigue The majority of people in the study have Rheumatoid Arthritis, are working fulltime in non-manual jobs, in the private sector or self-employed. The average level of disease activity was 4.66 out of a maximum of 10. Those with higher levels of disease activity also have higher levels of fatigue, have more difficulties in work and lower quality of life. This indicates that disease activity interferes with many aspects of a person’s life. Age was related to work performance with the younger participants having significantly more difficulties in work. Perhaps older participants have reduced their work hours, changed jobs or have developed strategies manage the demands of their work. This indicates the importance of early interventions to assist those with a newly diagnosed rheumatic disease to manage work effectively. The majority of respondents were female. On comparison with the male participants, the female participants had significantly higher levels of fatigue, significantly more difficulty managing the majority of their work demands, and reported lower quality of life. Those with third level education reported significantly more difficulty with the mental and social aspects of their work than those who achieved up to second level education, however the reasons for this are unclear. Work characteristics have an impact on study participants’ disease, fatigue and quality of life. Those working part-time have significantly higher disease activity than their full time counterparts. They also have significantly higher levels of fatigue in the majority of fatigue domains. Although not significant, they had more difficulty meeting the demands associated with their work. The exception to this was that full- time workers have significantly more difficulty with the physical demands of work. Further studies are required examining other characteristics of part-time workers that might explain these differences. 11

  12. The majority of study participants have severe fatigue with physical fatigue being the most problematic. In the qualitative phase some participants identified a pattern to their fatigue while others reported no discernible pattern and described their fatigue as unpredictable. This unpredictability makes fatigue more difficult to manage. Those with severe fatigue have significantly more difficulty fulfilling the demands of their work, their arthritis impacts more severely on their work and vice versa, and they have a lower quality of life. This demonstrates the debilitating nature of fatigue for people with rheumatic diseases and how it impacts on a range of work activities. In the qualitative phase of the study, participants discussed how fatigue also causes cognitive difficulties in work which leads to emotional problems. There are a range of fatigue management strategies that could be implemented in the workplace including prioritising work activities, pacing activities across the day and the week, making ergonomic changes to work stations and taking short breaks throughout the day. The participants in this study also recommended that employers and co-workers are educated on fatigue in rheumatic diseases and how it differs from regular fatigue. Health professionals have a role to play in providing such education. Just under half of participants reported moderate or extreme anxiety/depression. This could be related to many factors such as having a diagnosis of a chronic disease; experiencing pain and fatigue, or having difficulty balancing the management of the symptoms of their disease with meeting the demands of their work. Some of the participants in the qualitative phase of the study discussed difficulty telling their employers about their diagnosis as they feared this would impact on the security of their jobs or promotional opportunities, although such discrimination would be illegal under the Employment Equality Acts. It is vital that employers appreciate the importance of creating an environment that facilitates those with rheumatic diseases to feel safe in disclosing their diagnosis. Otherwise this could ultimately impact on employees’ mental health which will increase absenteeism in the workplace. In this study 40% of participants reported having multi-morbidity. That is they have at least one other chronic disease in addition to their rheumatic disease. In comparison to those without any other chronic condition, those with multi-morbidity have significantly higher levels of fatigue, significantly more difficulty in fulfilling work-related demands, their arthritis impacts more severely on their work; and they have significantly lower quality of life. These findings demonstrate the consequences of multi-morbidity within and outside the workplace. This indicates the need for early identification of people with multi- morbidity as they appear to be at higher risk for work-related difficulties. Conclusion Fatigue is a pervasive symptom of rheumatic diseases and it impacts on many elements of work performance. Those with severe fatigue have significantly more 12

  13. difficulty in work than those without severe fatigue and they also have a lower quality of life. Employers therefore need to accommodate employees with severe fatigue to enable them to meet the demands of their work. Interventions are required early in the disease trajectory to assist people to effectively manage fatigue related to Rheumatic diseases. 13

  14. 1. Introduction Rheumatic diseases are common in the general population. They are painful conditions usually caused by inflammation, swelling, and pain in the joints or muscles. Examples of rheumatic diseases include Osteoarthritis, Rheumatoid Arthritis, Ankylosing Spondylitis and Systemic Lupus Erythematosus. Some rheumatic diseases like osteoarthritis occur as result of ‘wear and tear’ to the joints. Other rheumatic diseases, such as Rheumatoid Arthritis, are believed to be caused by an autoimmune reaction whereby the immune system attacks the linings of joints, causing joint pain and swelling. Due to the range of symptoms in rheumatic diseases, all aspects of a person’s life can be affected. Rheumatic diseases are one of the main causes of physical disability, contribute to societal and economic costs and lead to loss of productivity in the workplace (Bevan et al, 2009). They are also one of the leading costs of work disability throughout the western world. In Ireland, the cost of lost productive time for those with Rheumatoid Arthritis was estimated at €1.6 billion (Barlow et al., 2001). Barlow et al., (2001) explored preventing work disability in people with rheumatologic diseases and reported that anxiety, depression and negative mood increased in those not fulfilling their full employment potential. Higher levels of pain and depression have been found in people with rheumatologic diseases when they are experiencing instability in employment (Barlow et al., 2001). A European wide ‘Fit for Work’ study investigated the impact of musculoskeletal disorders (MSD), including Rheumatoid Arthritis(RA) and Ankylosing Spondylitis (AS), on work loss of Irish workers and estimated that MSD’s cost in the region of €750m (Bevan, Magee and Quadrello, 2009). They also estimated that unemployment rates for people with AS are three times higher than the general population. Research has found that work gives people a sense of identity, a role in society and a sense of independence (Boonen et al., 2001). Work provides purpose to a person’s daily activities, gives people the opportunity to be productive, provides financial support to themselves and their families and provides regular social interactions. In addition, work can be a distraction from health problems and can even give a person the chance for regular physical exercise that helps minimise symptoms of the condition (Gignac et al., 2014). Therefore, when people are unemployed or they are finding it difficult to participate at work, their psychological well-being may be affected and research has found that depression and poor self-esteem are associated with unemployment (Dooley et al., 2000). Fatigue is reported to affect up to 90% of people with rheumatic diseases and must be considered in any study of work disability and productivity (Norheim et al., 2011). An improved understanding of the connection between a person’s life, work, rheumatic disease and fatigue, can assist in identifying individuals at risk of difficulty with maintaining employment. This information will also help health professionals to make recommendations for clients on how to manage their rheumatic diseases in the workplace (Gignac et al., 2014). Gignac et al., (2014) identified research priorities for 14

  15. rheumatic diseases. These included a need for further exploration of factors and symptoms that impact on work performance in the workplace and that limit and/or improve work outcomes. 2. Context of the Research 2.1 Societal and Economic Cost Rheumatic diseases are one of the main causes of work disability in people of working age in Ireland. Each year in Ireland seven million working days are lost due to symptoms of rheumatic diseases. This results in a €750 million cost to the economy (Bevan et al., 2009). Work disability has been found to be more prominent in those with Rheumatic diseases than in the general population (Barrett et al., 2000). For example Boonen et al. (2001), found that people with Ankylosing Spondylitis (AS) are three times more likely to withdraw from work than the general population. Fit for Work is a pan-European project which has a particular focus on supporting people with musculo skeletal disorders, including arthritis and other rheumatic conditions to get back to and continue in employment.1 2.2 Irish employers’ rights and responsibilities The Employment Equality Acts (1998-2011) apply to all employees (full, part-time and temporary) in the public and private sector. The key legal provisions are: Employers must not treat employees less favourably in employment, training or promotion as a result of their disability. Dismissal is not permitted unless an employee cannot meet the essential duties of the job By law a person with a disability is considered fully competent and capable of undertaking any duties, if the person would be fully competent and capable when reasonably accommodated by the employer Employers must take ‘appropriate measures’ to meet the needs of people with disabilities in the workplace, except if in doing so, a ‘disproportionate burden’ is imposed on the employer. ‘Appropriate measures’ are helpful and practicable actions employers should choose to suit workers with a disability (Equality Authority, 2011). The overall objectives of appropriate measures are to ensure the person with a disability is regarded equally as other employees when applying for work, in work and when applying for promotions or training. Some examples of appropriate measures are: 1 http://www.fitforworkeurope.eu/research.htm 15

  16. Adjust premises and equipment such as ergonomic seating and equipment at a desk space Flexible working hours Allocate work activities to suit workers with disabilities Provide training and other resources that might help recovery or transition back to work. ‘Disproportionate burden’ is unreasonable pressure, either financially or due to the size of the business, on employers who are trying to carry out ‘reasonable accommodations’. However, before employers can claim ‘disproportionate burden’ they must enquire about obtaining public funding or grants which may provide the extra assistance needed to make the accommodations possible (Equality Authority, 2011). The National Disability Authority in Ireland has published good practice guidelines for employers on how to retain and support employees with disabilities in the workplace. They suggest that employers have clear written policy to include the following reasonable accommodations: Provide early intervention to help employees with disability remain in employment after absence from work Keep in touch when an employee is absent from work Provide employees with an assessment of work ability before return to work Make a plan for returning to work Discuss with employees what supports they need put in place before their return to work Give employees an option for phased return to work Discuss with employee what they would like to tell their colleagues about their condition Inform and educate line managers on appropriate supports their colleague now needs Encourage persons with a disability to continue to pursue career goals Monitor work performance against career goals Offer redeployment to a suitable vacancy when someone is unable to do their job anymore. Rheumatic diseases can present a number of challenges for those in paid employment. There are many symptoms of the disease including pain and fatigue that contribute to these challenges. However, employers are legally obliged to reasonably accommodate workers who experience these challenges. Therefore, this study aims to explore work- related challenges related to symptoms of Rheumatic diseases and the reasonable accommodations that people with Rheumatic diseases need to get from employers to maintain productivity levels in the workplace. 16

  17. 3. Literature Review 3.1 Introduction This brief review of the literature discusses the impact of Rheumatic diseases on the ability to work with a specific focus on the impact of fatigue on work ability. The term Rheumatic diseases covers over 100 various conditions such as Rheumatoid Arthritis (RA), and Osteoarthritis (OA) and other auto-immune conditions such as Ankylosing Spondylitis (AS), and Spondyloarthropies, Systemic Lupus Erythematosus (SLE), Fibromyalgia and Systemic Sclerosis (SSc) (Sangha, 2000). In Ireland approximately 915,000 people have a Rheumatic disease with RA, OA and Fibromyalgia being the most common (Arthritis Ireland, 2013). A worldwide report on prevalence of arthritis and rheumatism found the prevalence of OA to be between 8-16.4%, RA between 1-6%, Ankylosing Spondylitis to be between 0.1-0.5% and autoimmune conditions such as SLE SSc and Sjögren’s to be between 0.1-0.5% (Wong et al., 2010). Dadoun et al. (2014) carried out a study on 813 people of working age with newly diagnosed Rheumatoid arthritis (RA). They found that work productivity is affected negatively in the first three years after diagnosis. Also identified from this study was that absence from work, along with poor physical and mental health, can result in reduced work productivity. A study by Barlow et al, (2001), with 133 people with Ankylosing Spondylitis (AS) found that work loss and disability can lead to loss of identity, lower or depressed mood and lower self-esteem. 3.2 Challenges in work Research has found that Rheumatic diseases are one of the most common chronic conditions to affect a person’s ability to remain in paid employment (Burton et al., 2006). Work disability and loss are related to the challenges that people with Rheumatic diseases have in the workplace. These challenges can be related to three areas: Firstly, some symptoms associated with Rheumatic diseases such as fatigue and pain can pose issues for people in work with regards to productivity levels and absenteeism. Secondly, challenges occur related to being fully able to participate in work activities due to the symptoms. Some people experience particular challenges within their work environments and issues with other colleagues. Finally, challenges occur in relation to the emotional impact of work disability on the person with the Rheumatic disease (Lacaille et al., 2007, Gignac et al., 2011). A study by De Croon et al. (2005) on work ability of 78 employees with early RA found that predictors of low work ability include fatigue, use of manual strength at work, and a lack of support, autonomy and participation in decision making. Another qualitative study by Lacille et al, (2007) identified fatigue as the main symptom of arthritis that caused the most difficulty at work. Participants identified a lack of awareness of colleagues and employers of fatigue due to it being an invisible symptom. This made it more difficult for people with arthritis to disclose information about fatigue and identify accommodations they may need. Therefore it was 17

  18. recommended that fatigue in the workplace needs to be addressed through medical and employment interventions such as medications that can help reduce fatigue levels and education on self-management strategies to cope with fatigue in work. 3.3 Work disability in rheumatic diseases Work disability is prevalent in those with rheumatic diseases and has substantial costs from a personal, societal and economic view (Bevan et al., 2009).Work disability in Rheumatic diseases is a multidimensional concept that encompasses more than just employment, it includes: reduction in employed hours, loss of prospects of being promoted, more frequent use of sick leave, increased employment changes and early retirement (Allaire et al., 1996). Learner et al., (2002) found that Osteoarthritis is the leading cause of work disability in adults (Lerner et al., 2002). Similarly, research into work disability in RA has found that people with RA are more likely to discontinue work than the overall population (Barrett et al., 2000). A 10-year longitudinal study of 1,235 people with early RA reported that pain and fatigue (low vitality) predicted work loss (McWilliams et al., 2014). Tillett et al., (2015) conducted a large multicentre UK study into factors that influence work disability in people with Psoriatic Arthritis (PsA). Three hundred and eighteen participants were assessed in the study of which 26% were unemployed. They identified that work disability was correlated with older age, disease duration of 2 to 5 years and poorer physical ability. Previous research also investigates how fatigue impacts on work disability in people with PsA (Wallenius et al., 2009). A similar study carried out by Barlow et al., (2001), into work disability in Ankylosing Spondylitis (AS) found that fatigue was one of the main challenges contributing to work disability for people with AS. The impact of fatigue on work for the study participants also filtered into other areas of their lives such as home, family and leisure activities. This shows that fatigue in the workplace is a multi-dimensional issue affecting a variety of areas in people’s lives. Bevan, McGee and Quadrello (2009) carried out a study in Ireland as part of the European Fit for Work study examining the impact of musculoskeletal disorders (MSD) on work ability. MSDs include regional pain such as back pain, joint and limb pain. They also included people with Rheumatoid Arthritis (RA) and Ankylosing Spondylitis (AS). They reported that MSDs can impact on endurance, cognitive abilities and mobility in work and that this can severely affect a person’s ability to remain in work. Loss of work results in financial and psychological difficulties for people. Bevan, Magee and Quadrello (2009) therefore recommend a biopsychosocial model to support those with MSDs to remain in the workforce. Baker and Pope (2009) identified that 32.4% of those with Systemic Lupus Erythematosus are work disabled due to psychosocial difficulties and symptoms of disease such as fatigue. They recommended further research into the impact of work disability on the person and possible interventions to help in the workplace (Baker and Pope, 2009). Previous research into work ability in women with Fibromyalgia and Systemic Sclerosis has found that greater work ability was associated with better ability to perform daily activities, increased satisfaction with ability to engage in 18

  19. activities, improved well-being and health (Sandqvist et al., 2008). Research in Trinity College Dublin found that work is an important and valued activity for people with SSc and that fatigue was one of the main issues that affected work ability along with effects on skin, pain and breathlessness (Mowlds et al., 2013). Therefore, further research is required to clarify barriers and facilitators to remaining in employment and the accommodations made by employers for people with a range of Rheumatic diseases. 3.4 Impact of fatigue on work Fatigue is part of the inflammatory or disease process of the majority of rheumatic diseases (Norheim et al., 2011). Fatigue is one of the main predictors of challenges to work in rheumatic diseases and also is a factor in limiting people while in work (Gignac et al., 2006, Arthritis Ireland, 2008). Gignac et al. (2014), found that up to now there has been little research carried out into possible interventions related to the impact of fatigue on work and challenges to activity participation in work. Therefore, there is a gap in the knowledge on challenges people with Rheumatic diseases may have in managing fatigue in the workplace and if fatigue is a reason for giving up work temporarily or permanently. Even though OA is one of the most commonly diagnosed rheumatic diseases, recent research has found that fatigue in OA is not regularly evaluated in practice or in research (Power et al., 2008). A Canadian study by Gignac et al, (2014), on 352 participants with a rheumatic disease found that fatigue can make it more difficult for a person to manage work, a family and social life balance. However, previous research has found that most people with RA do not discuss the impact of their fatigue with their health professionals because they either just accept it as a consequence of their RA or they feel it’s not addressed by health professionals (Repping-Wuts et al., 2008). Similar findings on communication of fatigue during medical appointments was found by Feldthusen et al. (2013) which left patients taking charge of managing their fatigue themselves. Therefore, there is a need to identify what specific challenges people with rheumatic diseases experiencing fatigue have in the work place, what they are currently doing to accommodate these challenges and what areas they would like strategies to help with work ability. 3.5 Conclusion Rheumatic diseases impact on work ability, performance in work and contribute to loss of work. This is mainly due to the symptoms of the condition, with fatigue emerging as a symptom that impacts on all areas of life but particularly work. However, how exactly and what areas in particular people are having difficulties with are not yet defined. Therefore, research is needed to understand how different symptoms of rheumatic diseases particularly fatigue, affect work ability and performance. 19

  20. 4. Methodology 4.1 Study aims and objectives The main aim of the study was to explore the perceptions of people with rheumatic diseases of the impact of fatigue on work ability. The study objectives were to: Examine the impact of fatigue on work-related activities for those with Rheumatic diseases. Examine the impact of fatigue on the interrelationship between arthritis and work. Explore people’s experiences of the impact of fatigue on their work ability. Investigate the range of strategies used by people in work to manage their fatigue. 4.2 Study design A sequential exploratory mixed methodology approach was used in this study (Creswell and Plano Clark, 2011). Quantitative data were gathered using self-report questionnaires followed by semi-structured interviews and focus groups to generate qualitative data. The data from both approaches were analysed separately (Corcoran, 2006). 4.3 Data collection Sampling Two recruitment methods were utilised in this study. The primary method of recruitment was through weekly rheumatology clinics in St. James’s Hospital. Participants were also recruited through the distribution of an online survey via arthritis-related voluntary organisations. The online survey was used in order to attempt a wide geographical representation beyond the catchment area of St. James’s Hospital which typically deals with people in the greater Dublin area. The inclusion criteria were that participants: had a definite diagnosis of a Rheumatic disease were between the ages of 18 and 65 years of age were currently in paid employment. In relation to recruitment in the weekly rheumatology clinics, every person who attended the clinic and met the first two inclusion criteria were provided with a Participant Information Leaflet explaining the purpose of the study and what was involved in participating in the study. Those who agreed to participate in the study then approached the researchers who were present at the clinics and were provided with the questionnaires. On completion of the questionnaires, respondents were 20

  21. invited to participate in a focus group or interview which were held on a different day and arranged to suit respondents’ schedules. Participants were also recruited through the distribution of an online survey by voluntary organisations for those with rheumatic diseases. The following organisations distributed the survey to their members: Arthritis Ireland Scleroderma Ireland Lupus group Ireland. The diagnosis of all participants who attended the outpatient rheumatology clinics was confirmed by a chart audit, however it was not possible to confirm a diagnosis for those who were recruited via the online survey as all diagnoses for these participants were self-reported. Recruitment Process The recruitment process took place through two sources. The first was through St. James’s Hospital. Recruitment here took place over a 15 week period. During this time, 814 people attended the weekly clinics in St James’s Hospital. Of these 543 met the age and diagnosis criteria. Of these, 196 completed questionnaires and were currently working or had worked in the past 24 months giving a 36% response rate. The second source of recruitment was through online questionnaires advertised through voluntary organisations for rheumatic diseases. One hundred and one participants accessed the online questionnaire, of these 86 completed the questionnaire. As the survey was distributed online via voluntary organisations it is not possible to determine the numbers of individuals the survey was distributed to in order to calculate a response rate. Table 2 shows a demographic comparison of the two groups. 21

  22. Table 2: Comparison of St. James’ and online survey group (a) demographics Recruitment group Online Survey St. James’ Hospital Age 18-30 31-40 41-50 51-60 61-67 Gender Male Female Duration of disease Up to 5 years 6-10 years 10 years + Did not say Work hours Full-time Part-time Did not say Job type Non-manual work Mixed work Manual work Did not say No. % No. % 12 50 55 60 19 6% 26% 28% 31% 10% 17 20% 32 37% 27 31% 8 2 5 81 94% 36 42% 21 24% 29 34% 0 53 62% 33 38% 0 59 69% 25 29% 2 0 9% 2% 88 108 44% 56% 6% 76 56 61 3 39% 29% 31% 2% 0% 131 52 13 67% 27% 7% 0% 80 71 36 9 41% 36% 18% 5% 2% 0% (b)Mean scores on different instruments Mean 12.5 Instrument Multi-dimensional Fatigue Inventory (MFI) - General fatigue Work Role Functioning (WRF) Arthritis Work Spill- over (AWS) EQ-5D-VAS SD ±4.1 Mean 16.5 SD ±3.2 ±17.9 63.8 ±19.2 45.3 ±6.1 ±5.3 17.3 22.0 ±18.5 43.4 ±19.0 63.7 Overall 282 people completed questionnaires. Of these 234 participants were currently working and 48 had been working in the last 24 months but were not currently working. 22

  23. In terms of the qualitative phase, 62 respondents from the St. James’ weekly rheumatology clinics provided contact details to participate in a focus group. Of these, contact was made with thirty-one people, with 11 individuals agreeing to participate. Ten people attended four focus groups and one person took part in an individual interview. Forty-seven respondents of the online survey provided contact details. Contact was made with fourteen of these respondents. Of these, seven individuals agreed to participate with three people attending a focus group and four people participated in an individual interview. In total, 18 people took part in the qualitative phase of the study. 4.4 Data collection methods Five questionnaires were used to collect quantitative data: A short demographic questionnaire designed by the research team Patient Global Assessment of Disease Activity (PtGA) Multidimensional Fatigue Inventory (MFI) Work Role Functioning (WRF) Arthritis Work Spill-over (AWS) EQ-5D-3L. See Appendix 2 for a copy of these measures. Patient Global Assessment of Disease Activity (PtGA) The Patient Global assessment of disease activity (PtGA) is a single item scale that was originally designed for use with people with RA (Anderson et al., 2011). It was designed to measure a summary of how the disease affects the patient at a point in time. In this study the NRS was used ranging on a scale of 0 to 10. The higher the score, the higher the disease activity (Lassere et al., 2001). The PtGA has been found to have good test-re test reliability, validity and response to changes (Lassere et al., 2001). Multidimensional Fatigue Inventory The Multidimensional Fatigue Inventory (MFI-20) (Smets et al., 1995) is a 20 item self-report questionnaire that measures different aspects of fatigue over the past seven days. The MFI contains five domains: general fatigue, physical fatigue, reduced activity, reduced motivation and mental fatigue. Scores range from a minimum of 4 to a maximum of 20 for each domain. Higher scores indicate more fatigue. The MFI-20 has been used in previous research into fatigue in Rheumatic diseases and has been found to be a reliable and valid measure (Da Costa et al., 2006, Thombs et al., 2008, Van Tubergen et al., 2002, Reeves et al., 2005). A score of 13 or above in the general fatigue category indicates severe fatigue (Reeves et al., 2005). 23

  24. Work Role Functioning Questionnaire The Work Role Functioning (WRF) (Amick et al., 2004), questionnaire is a self-report questionnaire which measures the limitations a person experiences in work due to their health condition. It is a 27-item questionnaire that is divided into 5 subscales: work scheduling demands, output demands, physical demands, mental demands and social demands. Items are scored on a 5-point scale 1 (difficult none of the time) to 5 (difficult all of time) with each subscale being scored separately. A score is calculated by adding the response of each subscale, getting an average score and multiplying the score by 25 to get an overall percentage from 0% (have no problems meeting demands of the job) to 100% (always have problems meeting the demands of the job). This single summated score is calculated if a person is not missing 20% of the items. The WRF questionnaire has been used in previous research in people with Rheumatic diseases and is reliable and valid when used with other measures of work ability and impact of health conditions on work (Roy et al., 2011). Arthritis Work Spill-over The Arthritis Work Spill-over (AWS) is a 6-item self-report questionnaire which measures the degree to which the demands of arthritis impedes work performance and the degree to which work impedes the management of arthritis. The questionnaire is scored on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree) (Gignac et al., 2006).Total scores range from 6-30 with higher scores indicating higher spill-over, or that arthritis is negatively affecting work between work and arthritis and vice versa (Gignac et al., 2006, Gignac et al., 2007). EQ-5D- 3L The EQ-5D-3L is a self-report measure of health status developed by the EuroQol group. It is a non-disease specific assessment for use in describing and valuing health related quality of life (Rabin and Charro, 2001). There are two parts to the measure: the descriptive category (EQ-5D descriptive system) and the Visual Analogue Scale (EQ-VAS). The descriptive category contains five elements: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression with each containing three levels: no problems, some problems and extreme problems. Participants mark the statement which is most relevant to their health state on the day. The VAS section of the measure is a vertical scale ranging from 0 ‘worst imaginable health state’ to 100 ‘best imaginable health state’. VAS is considered a valid and reliable measure for a range of chronic conditions (EuroQolGroup, 1990). 4.5 Qualitative Data Collection Focus groups and semi-structured individual interviews were the qualitative methods used in this study. Both methods were based on pragmatic concerns in order to recruit as many participants as possible for the qualitative phase of the study. Individual interviews were offered to participants unable or unwilling to attend a focus group. Semi-structured individual interviews collect detailed accounts of participants’ thoughts, attitudes, beliefs, and knowledge pertaining to a particular phenomenon and allow interviewers flexibility in questioning to further explore areas of interest 24

  25. (Dillaway et al., 2006). Focus group interviews are a data collection method where participants express their views by interacting in a group discussion of the issues (Davidson et al., 2010, Carter and Henderson, 2005). Focus group interviews are useful for evaluating health care interventions and explaining quantitative findings as they provide both an individual and collective perspective (Liamputtong, 2010). Focus groups have been proposed as a particularly useful method of data collection when taking a qualitative descriptive approach and as counterpart to quantitative research to obtain a broad range of information about events (Sandelowski, 2000). The purpose of both the interviews and focus groups was to explore participants’ perceptions of fatigue and how it impacts on work and their ability to maintain paid employment. Another aspect explored was the issue of disclosure to employers and the accommodations, if any, provided/offered if they did disclose. The advantage of using a combination of these methods is enhancement of data completeness, as each method reveals different parts of the phenomenon of interest (complementary views) and contributes to a more comprehensive understanding (expanding the breadth and/or depth of the findings). For example, individual interviews may be used to explore personal experiences, whereas focus groups may be used to examine opinions and beliefs about the phenomenon (Lambert and Loiselle, 2008). 4.6 Data analysis Quantitative Data analysis was carried out using the Statistical Package for Social Science (SPSS) version 20. Initial descriptive statistics using frequencies, means and standard deviations were calculated. Pearson-Product Moment correlations were used to explore relationships between variables. Independent t-tests and one-way between groups analysis of variances were used to assess differences between the scores of sub categories of participants. Qualitative Both the interviews and focus groups were transcribed verbatim, then these transcripts were reviewed and were initially pre-coded manually, highlighting and marking quotes that stood out as significant and worthy of attention (Layder, 1998). Coding is the practice of grouping and labelling ideas and views so that the ideas generated then reflect wider perceptions. Once the codes are grouped they can form themes and viewpoints. These are then the findings of the qualitative method that provide responses to the research questions (Corcoran, 2006). In this study constant comparative analysis coding was used to analyse the qualitative interviews (Glaser and Strauss, 1967). Following this provisional coding the transcripts were entered into Nudist Vivo version 10 (NVivo) software for analysis of qualitative data. Patterns in the codes were identified by re-reading the transcripts and then grouping together similar codes and descriptions into new categories (Stein et al., 2013). These new categories were then compared to one another and discussed by the researchers and given theme labels (Coffey and Atkinson, 1996). 25

  26. 26

  27. 5. Quantitative results Overall 282 people completed questionnaires. Of these 234 participants were currently working and 48 were not currently working but had worked in the previous 24 months. However, not all participants completed all measures fully, therefore there are incidences of missing data. The demographic profile of those currently working (n=234) is presented along with the results of descriptive analysis, correlations between variables and finally differences in variables based on demographic and work characteristics. 5.1 Demographic profile The majority of those working were women and aged between 41 and 50 years of age (table 3). Table 4 shows the demographic details of participants. Table 3: Age and gender demographic variables Frequency (n) Percent (%) Age 18-30 31-40 41-50 51-60 61-67 Total Gender Male Female Total 28 67 71 54 14 234 12 29 30 23 6 100 69 165 234 30 70 100 Table 4: Demographic variables Frequency (n) Percent (%) Co-morbid condition One or more None Did not say Total Marital Status Single Married Separated/divorced Widowed Did not say 91 137 39 59 3 100 6 234 77 120 26 33 52 11 3 1 8 3 27

  28. Total Living Situation Living alone Living with someone Did not say Total Education level Up to Second level College/ University Did not say Total 234 100 Frequency (n) Percent (%) 42 188 18.0 80 4 2 234 100 92 137 39 59 2 100 5 234 Table 5 and 6 present the main types of rheumatic diseases of participants and disease duration. The largest single groups were those with Rheumatoid Arthritis and those who had their condition for less than five years. Table 5: Type of rheumatic disease breakdown Rheumatic disease type Rheumatoid Arthritis Psoriatic Arthritis Osteoarthritis Systemic Lupus Erythematosus Ankylosing Spondylitis Fibromyalgia Systemic Sclerosis Gout Polymyalgia Rheumatica Other Multiple Rheumatic conditions Frequency (n) Percentage (%) 100 24 43 10 4 8 6 5 0.4 9 19 14 12 1 8 2 16 28 3 1 7 12 Table 6: Duration of rheumatic disease Disease Duration (n=234) Up to 5 years 6-10 years 10 years or more Did not say Frequency (n) Percentage (%) 91 65 76 2 39 28 32 1 28

  29. 5.2 Profile of employment Participants provided information on the type of employment in which they are engaged. The majority of respondents are in full-time employment (70%) and in non- manual work (51%). Non manual labour consists of administrative, managerial, supervisory, office work and other professionals for example teacher. The disclosure question and job sector were both added to the questionnaire later in the data collection process and asked of just over half the sample (55%). Of these, three quarters (76%) had disclosed their rheumatic disease to their employer. Table 7: Employment Details Description of employment 2 (n=226) Non-Manual Mixed non-manual and manual Manual Employment hours (n-234) Full-time Part-time Job Sector (n=153) Private Company/Self-Employed Public Service Disclosure (n = 157) Yes No Not applicable (Self-employed) No. % 120 80 26 53 35 11 163 71 70 30 82 71 56 46 119 36 2 76 23 1 5.3 Descriptive Analysis of Outcome Measures Disease activity Participants were asked to rate their disease activity over the past week on a scale of visual analogue scale of 0 (no disease activity) to 10 (severe disease activity). The disease activity mean score was 4.7 (SD±2.6), with 225 people answering this question. Impact of fatigue The fatigue categories of the Multidimensional Fatigue Inventory (MFI) are: general fatigue, physical fatigue, reduced activity, reduced motivation and mental fatigue. Scores show that ‘General Fatigue’ has the highest mean score. General fatigue 2Explanation of employment description terms: - Non-manual work in this study covers administrative, managerial, supervisory office or other professional work such as a teacher. - Mixed work in this study covers occupations such as sales and service occupations such as waitress, personal care attendant; patients care nurse, nurses’ aide, or driver. - Manual work consists of such as carpenter, roofer or loader. 29

  30. includes global/broad statements about fatigue and was designed to encompass both physical and psychological aspects of fatigue and provides a general indicator of fatigue (Reeves et al., 2005). Table 8: Mean fatigue scores according to the MFI (range 4-20) (High scores indicate greater fatigue) General Fatigue 13.8 Physical Fatigue 12.5 Reduced Activity 10.1 Reduced Motivation Mental Fatigue 10.3 n=220 Mean Score Standard deviation 10.2 SD±4.2 SD±4.3 SD±4.2 SD±3.5 SD±4.2 Work Ability This section presents information on participants’ work ability as measured by the Work Role Functioning Questionnaire and the Arthritis Work Spill-over questionnaire. The Worker Role Functioning (WRF) Questionnaire The WRF questionnaire measures a person’s interpretation of how they are functioning in the workplace and has five subsections (Table 9). A high score indicates more difficulty. Participants’ mean scores indicated that participants perceived most difficulty in the physical demands of work, closely followed by difficulty managing work scheduling demands. Table 9: Overall mean worker role functioning scores (range 0-100) (High scores show greater difficulties) Total Work Scheduling Demands Output demands Physical Demands Mental Demands Social n=212 Demands Mean scores SD 49.7 55.4 47.1 56.7 46.6 37.6 ±.19.0 ±.25.8 ±.22.1 ±.27.9 ±.22.5 ±.17.7 The Arthritis Work Spill-over (AWS) Questionnaire The AWS (score range 6-30) measures the degree to which the demands of arthritis effects work performance and the degree to which work effects the management of arthritis, where higher scores indicate more negative outcomes. The mean score was 18.3 (SD±.6.1), for 220 participants completing this measure. This indicates that there is a moderate level of spill-over between participants’ work and their arthritis. Impact on quality of life The EQ-5D assessed quality of life, under five headings (Table 10). The most frequently-occurring situations were some problems with mobility (53%), no problems with self-care (77%), some problems with usual activity (56%), moderate pain/discomfort (67%) and not anxious/depressed (52%). The mean EQ-VAS score, ranging from 0 for extreme difficulty up to 100, completed by 210 participants, was 59.5 (SD±.19.2). 30

  31. Table 10: EQ-5D descriptive findings No. % EQ-5D Descriptive (n=223) Mobility No problems Some problems Confined to bed Missing Self-care No problems Some problems Unable to perform Self-care activities Missing Usual Activity No problems Some problems Unable to perform usual activities Missing Pain/Discomfort No pain/discomfort Moderate pain/discomfort Extreme pain/discomfort Missing Anxiety/ Depression Not anxious/depressed Moderately anxious/depressed Extremely anxious/depressed Missing 96 125 2 11 41 53 1 5 179 44 0 11 77 19 0 5 86 131 6 11 37 56 3 5 36 157 30 11 15 67 13 5 121 92 10 11 Mean 52 39 4 5 Standard deviation ±.19.2 EQ-5D- VAS (0-100) (n=210) 59.5 5.4 Correlations This section examines the relationships between: Disease activity and the variables of fatigue, work functioning, arthritis work spill- over and quality of life Fatigue and the variables of disease activity, work functioning, arthritis work spill- over and quality of life The relationships between variables were investigated using Pearson product-moment correlation coefficients. Preliminary analyses were performed to ensure no violation of the assumptions of normality, linearity and homoscedasticity. The strength of the correlations can be interpreted as follows (Cohen 1988, Pallant 2010) 31

  32. Range .10 to .29 – weak Range .30 to .49 – moderate Range .50 to 1.0 – strong Disease activity and relationships with Fatigue, Work and Quality of Life The relationship between disease activity and the following measures were investigated: Multidimensional Fatigue Inventory Work Role Functioning Questionnaire Arthritis Work Spill-over EQ-VAS Statistically significant relationships are marked with an asterisk in Table 11 below. Significant positive correlations were found between disease activity and all domains of fatigue (general fatigue, physical fatigue, reduced activity, reduced motivation and mental fatigue), indicating that as disease activity increases so does fatigue. Disease activity was also significantly related to categories of Work Role Functioning (except managing the social demands of work) indicating that as disease activity increases difficulties in work functioning also increase. Another significantly positive relationship was found between disease activity and the AWS, indicating that arthritis- work spill-over increases when disease activity increases. A significant negative correlation was also found between disease activity and quality of life indicating that those with higher disease activity have a lower health-related quality of life. 32

  33. Table 11: Correlations: Disease Activity and outcome measures Variables Disease Activity n=225 R p-value Multi-dimensional Fatigue Inventory (MFI) (n=220) MFI General MFI Physical MFI Reduced Activity MFI Reduced Motivation MFI Mental Work Role Function (WRF) (n=212) WRF Total WRF Work Scheduling Demands WRF Output Demands WRF Physical Demands WRF Mental Demands WRF Social Demands AWS (n=220) EQ-VAS (n=210) .543 .448 .322 .406 .414 .000* .000* .000* .000* .000* .347 .373 .190 .430 .171 .126 .440 -.487 .000* .000* .006** .000* .014** .072 .000* .000* **Significant at p≤0.01** indicating a 99% chance of the relationship being true *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Relationships between fatigue and (i) disease activity, (ii) work functioning and (iii) quality of life Relationships were examined between the Multidimensional Fatigue Inventory and the following variables: Work Role Functioning Arthritis Work Spill-over EQ-VAS Overall significant positive correlations, of varying strength, were found between all domains of fatigue as measured by the MFI and all aspects of work role functioning, as measured by the Work Role Functioning questionnaire. This indicates that higher levels of fatigue are associated with higher levels of difficulty in all aspects of work role functioning. Significant positive correlations, of varying strength, were also found between all MFI domains of fatigue and arthritis work spill-over. This indicates that higher levels of self-reported fatigue, across a number of domains, are associated with the extent to which arthritis impacts on work and vice versa.Negative correlations were found between all domains of the MFI and self-rated health as measured by the EQ-VAS. This indicates that higher levels of different types of fatigue are associated with lower levels of self-rated health as measured by the EQ-VAS. Significant relationships are marked with an asterisk in Table 12. 33

  34. Table 12: Correlations: Fatigue and outcome measures (n=220) MFI MFI MFI MFI MFI Multi-dimensional Fatigue Inventory (MFI) Work Role Functioning (WRF) WRF Total WRF Work Scheduling Demands WRF Output Demands WRF Physical Demands WRF Mental Demands WRF Social Demands AWS (n=220) EQ-VAS (n=210) General Fatigue Physical Fatigue Reduced Activity Reduced Motivation Mental Fatigue R R R R R .48** .45** .53** .52** .48** .43** .41** .36** .57** .52** .36** .31** .42** .29** .45** .49** .48** .48** .41** .35** .36** .37** .37** .31** .58** .16* .27** .16* .19* .34** .45** -.59** .41** -.50** .54** -.57** .32** -.46** .41** -.42** **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true 5.5 Are there differences across severity, age, gender or other variables in disease activity, work role functioning, arthritis-work spill-over and quality of life? This next section of the report presents the results of a statistical procedure, called the independent samples t-test, to ascertain if there are significant differences in disease activity, fatigue, work role functioning, arthritis work spill over and self-rated health between subgroups of: severity of fatigue age gender education levels co-morbidity duration of disease The tables that follow present the average (mean) values for the subgroups on the various tests for these factors, along with a measure of how dispersed the values for the subgroup are around that average (standard deviation). The final column of these 34

  35. tables, the P value, is a measure of whether the groups are genuinely different or not – if it is below 0.05, the groups concerned show genuinely different characteristics that are highly unlikely to be due to chance. Differences based on fatigue severity Reeves et al., (2005) define severe fatigue as a score of greater than or equal to 13 on the MFI general fatigue subscale, and we have used this to divide the sample into severe and less severe fatigue groups. Table 13 presents the results. There were significant differences in scores for those with higher and lower levels of fatigue in all variables except for work role functioning – social demands. This indicates that those with severe levels of fatigue have higher disease activity, more difficulty in work functioning, higher levels of arthritis work spill-over, and lower self-rated health than those with less severe levels of fatigue. Table 13: Differences in disease, work functioning and quality of life outcome measures between severe and non-severe fatigue Outcome Measure Disease Activity Fatigue Level Less severe Severe Less severe Severe n Mean for sub-group Std P value deviation 88 126 82 122 3.0 5.9 40.1 56.5 ±2.0 ±2.3 ±16.4 ±17.7 .000** Work Role Functioning (WRF) Total WRF Work Scheduling Demands WRF Output Demands .000** Less severe Severe 82 122 41.7 64.8 ±20.7 ±24.1 .000** Less severe Severe Less severe Severe Less severe Severe Less severe Severe Less severe Severe Less severe Severe 82 122 82 122 82 122 82 122 85 123 82 119 38.0 53.6 43.0 66.0 40.2 52.3 35.3 39.3 15.2 20.8 71.1 52.2 ±16.7 ±22.4 ±23.7 ±26.9 ±19.8 ±22.2 ±18.4 ±17.2 ±5.8 ±5.0 ±17.0 ±16.6 .001** WRF Physical Demands .000** WRF Mental Demands .000** WRF Social Demands .117 AWS .000** EQ5D-VAS .000** **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Differences in variables based on demographic characteristics Impact of age Independent samples t-tests were conducted to compare disease activity, fatigue, work role functioning, arthritis work spill-over and self-rated health for younger and older respondents. There were no significant differences in disease activity levels, fatigue 35

  36. levels, arthritis work spill-over and self-rated health. There were however, significant differences between younger and older respondents for overall work role functioning, and the mental and social demands of work, with younger participants having more difficulty in these areas (see Table 14 and 15). Table 14: Differences in disease activity and fatigue levels between age categories (younger up to 50, older 51+) Outcome Measure Age N Mean for sub-group Standard deviation P value Disease Activity Younger Older 163 62 4.5 5.2 ±2.6 ±2.5 .056 Multi-dimensional Fatigue Inventory (MFI) MFI General Fatigue MFI Physical Fatigue Younger Older 164 56 14.0 13.2 ±4.1 ±4.4 .208 Younger Older 164 56 12.4 12.6 ±4.2 ±4.7 .823 MFI Reduced Activity Younger Older 164 56 9.9 10.4 ±4.1 ±4.7 .490 MFI Reduced Motivation Younger Older 164 56 10.1 10.3 ±3.4 ±3.8 .717 MFI Mental Fatigue Younger 164 56 10.5 9.9 ±4.1 ±4.4 .373 Older **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Table 15: Differences in Work and Quality of Life between Age categories (younger up to 50, older 51+) Outcome measure Age N Mean for sub- Standard deviation ±18.8 ±19.0 P value group 51.5 44.9 Work Role Functioning (WRF) Total WRF Work Scheduling Demands WRF Output Demands Younger Younger Older 153 59 .023* Younger Older 153 59 57.5 50.1 ±25.0 ±27.4 .063 153 59 48.6 43.2 ±21.9 ±22.3 .115 Older WRF Physical Demands Younger Older 153 59 57.4 55.0 ±27.4 ±29.5 .584 WRF Mental Younger 153 49.4 ±22.7 .003* 36

  37. Outcome measure Age N Mean for sub- Standard deviation ±20.3 P value group 39.3 Older 59 WRF Social Demands Younger 153 59 39.6 32.3 ±18.6 ±13.6 .006* Older AWS Younger Older 157 63 18.8 17.2 ±5.9 ±6.5 .079 EQ5D-VAS Younger Older 153 57 59.2 60.1 ±19.7 ±17.7 .769 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Impact of Gender Women had significantly higher levels of general, physical and mental fatigue, in comparison to men (See Table 16 and 17). Women also had significantly more difficulty in all categories of work role functioning than men, except for meeting the social demands of work where no difference was found. Females were also found to have significantly lower self-rated health than males. Table 16: Differences in disease activity and fatigue levels between genders Outcome Measure Disease Activity Gender N Mean of sub-groups Standard deviation P value Male Female 63 162 4.6 5.7 ±3.0 ±2.4 .817 MFI General Fatigue Male Female 64 156 12.3 14.4 ±4.0 ±4.1 .000** MFI Physical Fatigue Male Female 63 162 11.1 13.0 ±4.2 ±4.3 .003* MFI Reduced Activity Male Female 63 162 10.0 10.2 ±4.3 ±4.2 .302 MFI Reduced Motivation Male Female 63 162 9.7 10.4 ±3.3 ±3.6 .163 MFI Mental Fatigue Male Female 63 162 9.1 10.8 ±4.2 ±4.1 .007* **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010). Table 17: Differences in work and quality of life between genders Outcome Measure Gender (n) N Mean of group Standard deviation P value 37

  38. Outcome Measure Work Role Functioning (WRF) Total WRF Work Scheduling Demands WRF Output Demands Gender (n) Male Female N Mean of group 43.5 44.9 Standard deviation ±19.1 ±19.0 P value 63 162 .003* Male Female 63 162 47.3 58.6 ±27.4 ±24.5 .004* Male Female 63 162 41.6 49.3 ±22.5 ±21.7 .024* WRF Physical Demands Male Female 63 162 50.6 59.1 ±29.0 ±27.2 .044* WRF Mental Demands Male Female 63 162 38.4 49.8 ±19.9 ±22.7 .001** WRF Social Demands Male Female 63 162 36.8 37.9 ±20.8 ±16.4 .680 AWS Male Female 63 162 17.2 18.8 ±5.9 ±5.8 .088 EQ5D-VAS Male Female 63 162 63.8 57.7 ±17.5 ±19.6 .035* **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Impact of education Those with education up to second level had significantly higher disease activity levels than those with university level education. Those with university level education had significantly more difficulty in managing the mental and social demands of work role functioning than those with education up to secondary level (Tables 18 and 19). Table 18: Differences in disease activity and fatigue for levels of education Outcome Measure Education Level N Mean of Standard deviation P sub group 5.2 4.3 value Up to 2nd level 3rd level Up to 2nd level 3rd level Up to 2nd level 3rd level Up to 2nd level ±2.7 ±2.5 ±4.2 ±4.1 ±4.6 ±4.2 ±4.3 Disease Activity 88 135 .020* MFI General Fatigue 83 135 13.2 14.2 .078 MFI Physical Fatigue 83 135 12.4 12.6 .757 MFI Reduced Activity 83 9.9 .511 38

  39. Outcome Measure Education Level N Mean of Standard deviation P sub group 10.2 value 3rd level ±4.2 ±3.6 ±3.5 ±4.2 ±4.2 135 MFI Reduced Motivation Up to 2nd level 83 135 10.6 10.0 .264 3rd level Up to 2nd level 3rd level MFI Mental Fatigue 83 135 10.0 10.5 .411 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Table 19: Differences in work and quality of life between levels of education Outcome Measure Education Level Work Role Functioning (WRF) 3rd level Up to 2nd level 3rd level Up to 2nd level 3rd level WRF Physical Demands Up to 2nd level 3rd level Up to 2nd level 3rd level Up to 2nd level 3rd level Up to 2nd level 3rd level Up to 2nd level 3rd level N Mean of sub group Standard deviation P value Up to 2nd level 81 128 47.1 51.6 ±19.5 .096 ±18.6 WRF Work Scheduling Demands 81 128 52.3 57.8 ±28.1 .135 ±24.2 WRF Output Demands 81 128 44.0 49.4 ±22.2 .088 ±22.0 81 128 56.9 57.0 ±29.9 .972 ±27.8 WRF Mental Demands 81 128 42.2 49.5 ±21.5 .021* ±22.8 WRF Social Demands 81 128 34.1 40.1 ±16.4 .018* ±18.2 AWS 86 130 17.4 19.0 ±6.1 .052 ±6.0 EQ5D-VAS 80 126 57.9 60.0 ±17.8 .443 ±20.1 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Impact of co-morbidity The next two tables (Tables 20 and 21) examined if there were differences between those with co-morbidity (multiple condition) in comparison to those with a single rheumatic disease (no co-morbidity). Those with co-morbidity had significantly higher fatigue in all domains except fatigue related reduced motivation. Those with co- morbidity also had significantly higher difficulty in overall work role functioning, managing work scheduling demands and had higher levels of arthritis-work spill-over. 39

  40. Significantly lower levels of self-rated health were found in those with co-morbidity in comparison to those with a single rheumatic disease. 40

  41. Table 20: Differences in disease activity and fatigue levels in co-morbidity Outcome Measure Disease Activity MFI General Fatigue MFI Physical Fatigue MFI Reduced Activity MFI Reduced Motivation MFI Mental Fatigue Co-morbidity N Mean of sub-group Standard P deviation value .003* Yes No Yes No Yes No Yes No Yes No Yes No 89 130 85 129 85 129 85 129 85 129 85 129 5.3 4.2 14.8 13.1 13.4 11.8 11.0 9.3 10.6 9.8 11.3 9.6 ±2.9 ±2.5 ±3.8 ±4.3 ±4.2 ±4.23 ±4.5 ±3.9 ±3.5 ±3.6 ±4.0 ±4.1 .005* .007* .005* .092 .003* **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Table 21: Differences in work and quality of life in co-morbidity Outcome Measure Co- morbidity Yes No Yes No N Mean of sub group Standard deviation ±18.7 ±18.8 ±25.4 ±25.4 P value Work Role Functioning (WRF) WRF Work Scheduling Demands WRF Output Demands WRF Physical Demands WRF Mental Demands WRF Social Demands AWS 85 123 85 123 52.7 47.2 60.5 51.5 .039* .013* Yes No Yes No Yes No Yes No Yes No Yes No 85 123 85 123 85 123 85 123 90 126 83 123 50.4 44.5 59.2 54.9 49.7 44.1 37.6 37.5 19.3 17.5 55.5 62.4 ±23.5 ±20.7 ±26.2 ±29.4 ±21.2 ±22.6 ±16.7 ±18.4 ±6.0 ±6.1 ±19.5 ±18.7 .055 .275 .071 .950 .034* EQ5D-VAS .013* **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) 41

  42. Impact of disease duration There were three categories of disease duration therefore a one way between-groups ANOVA was conducted to examine the impact of disease duration on disease activity, fatigue levels, work role functioning and self-rated health. A statistically significantly difference was found in fatigue-related reduced activity and fatigue-related motivation for the three disease duration groups. Post-hoc comparisons using the Tukey HSD tests were carried out. They indicated that those with disease duration of less than five years were significantly less active as a result of their fatigue than those with longer disease durations. Also it was found that those with disease duration of less than five years were significantly less motivated as a result of their fatigue than those with longer disease durations (Tables 22, 23 and 24). Table 22: Differences in disease activity and fatigue levels for disease duration: ANOVA Scale Under 5 years Mean SD 6-10 years Mean SD 10 years or more Mean (±SD) SD P value Disease Activity MFI General Fatigue MFI Physical Fatigue MFI Reduced Activity MFI Reduced Motivation MFI Mental Fatigue 4.7 ± 2.6 4.4 ± 2.7 4.8 ± 2.6 .669 14.2 ± 3.8 13.3 ± 4.5 13.4 ± 4.3 .466 13.1 ± 4.0 11.7 ± 4.4 12.4 ± 4.7 .143 11.0 ± 4.2 9.0 ± 3.9 9.8 ± 4.5 .012* 11.0 ± 3.8 9.7 ± 3.48 9.6 ± 3.1 .023* 10.5 ± 4.3 10.2 ± 4.2 10.2 ± 4.1 .832 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) 42

  43. Table 23: Differences in work functioning and quality of life for disease duration: ANOVA Scale <5 years Mean 49.2 SD 6-10 years Mean SD 10 years+ Mean SD P value .513 Work Role Functioning (WRF) Work Scheduling Demands Output Demands Physical Demands Mental Demands Social Demands AWS EQ-VAS ±19.1 47.9 ±15.8 51.7 ±21.3 55.7 ±26.54 53.5 ±24.2 56.6 ±26.5 .778 47.4 ±19.8 44.0 ±19.2 49.4 ±19.2 .395 55.4 ±26.7 55.3 ±27.6 59.5 ±29.8 .588 46.8 ±22.1 44.7 ±19.4 48.0 ±25.3 .712 36.8 ±18.3 36.3 ±15.1 39.6 ±19.0 .502 18.9 60.0 ± 6.4 ±18.4 17.6 ±6.3 60.8 ±20.1 18.1 ±5.4 57.7 ±19.5 .400 .627 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Table 24: Disease Duration and outcome measures ANOVA (post hoc test) Dependent Variable MFI Reduced Activity Disease duration (i) Less than 5 years 6-10 years Disease duration (j) 6-10 years 10 years+ Less than 5 years 10 years+ <5 years 6-10 years 6-10 years 10 years+ <5 years 10 years+ <5 years 6-10 years Mean Std. Error .69 .67 .69 P Difference (i-j) 2.0 1.2 -2.0 .073 .040* .073 -.8 -1.2 .8 1.3 1.4 -1.3 .1 -.4 -.1 .73 .67 .73 .58 .56 .58 .60 .68 .74 .984 .040* .984 .073 .040* .073 .984 .040* .984 10 years or more Less than 5 years 6-10 years MFI Reduced Motivation 10 years or more **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) 43

  44. 5.6 Differences in variables based on employment characteristics Differences were analysed between disease activity, fatigue levels, work role functioning, arthritis work spill-over and quality of life in the following employment characteristics: Working hours Job type Job sector Working hours An independent samples t-test was conducted to compare work role functioning, arthritis work spill-over and self-rated health between those working full-time and part-time. Employees working part-time had significantly higher disease activity, general fatigue, physical fatigue, fatigue-related reduced activity and fatigue-related reduced motivation in comparison to those working full-time. Employees working full-time had significantly higher levels of difficulty in meeting the physical demands of work functioning. No differences were found in other aspects of work role functioning or self-rated health (Table 25 and 26). Table 25: Differences in disease activity and fatigue levels between different working hours Outcome Disease Activity Working hours Full- time Part-time N Mean 4.3 5.5 SD P value ±2.6 ±2.3 155 70 .001** MFI General Fatigue Full-time Part-time 153 67 13.4 14.8 ±4.1 ±4.2 .023* MFI Physical Fatigue Full-time Part-time 153 67 12.1 13.4 ±4.3 ±4.4 .030* MFI Reduced Activity Full-time Part-time 153 67 9.7 10.9 ±4.1 ±4.4 .040* MFI Reduced Motivation Full-time Part-time 153 67 9.7 11.3 ±3.6 ±3.2 .001** MFI Mental Fatigue Full-time 153 67 10.1 10.8 ±4.2 ±4.1 .212 Part-time **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) 44

  45. Table 26: Differences in work and quality of life between different working hours Outcome Measure Working hours Full-time Part-time Full-time Part-time Full-time Part-time Full-time Part-time Full-time Part-time Full-time Part-time Full-time Part-time Full-time Part-time N Mean SD P value WRF 142 70 142 70 142 70 142 70 142 70 142 70 150 70 144 66 48.6 ±19.2 51.8 ±18.6 54.9 ±25.3 56.4 ±27.0 46.4 ±22.2 48.5 ±22.1 53.2 ±27.3 63.9 ±28.0 46.3 ±22.4 47.3 ±22.7 38.5 ±18.6 35.8 ±15.7 17.9 19.1 60.8 ±19.9 56.6 ±17.3 .242 WRF Work Scheduling Demands WRF Output Demands WRF Physical Demands WRF Mental Demands WRF Social Demands AWS .685 .524 .008* .768 .313 ±6.0 ±6.1 .187 EQ5D-VAS .144 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Job Type A one way between-groups ANOVA was conducted to examine the impact of job type on disease activity, fatigue levels, work role functioning, arthritis work spill-over and self-rated health. A statistically significantly difference was found in managing the mental demands of work functioning between the three job type groups. Post-hoc comparisons using the Tukey HSD test indicated that non-manual workers had higher difficulty in managing the mental demands of work in comparison to manual workers (Tables 27-29). Table 27: Differences in disease activity and fatigue levels for job type: ANOVA Scale Disease Activity MFI General Fatigue MFI Physical Fatigue MFI Reduced Activity MFI Reduced Motivation Non-manual Mean Mixed SD Mean 4.7 ±2.4 14.3 ±4.3 Manual SD ±2.6 ±3.9 P value Mean 5.6 ±3.2 13.2 ±3.9 SD 4.3 13.5 .184 .288 12.7 ±4.3 12.3 ±4.3 11.8 ±4.5 .591 10.5 ±4.3 9.6 ±4.3 9.9 ±3.9 .409 10.2 ±3.6 10.0 ±3.5 10.8 ±3.2 .606 45

  46. Scale MFI Mental Fatigue Non-manual Mixed Manual P value .252 10.8 ±4.2 9.7 ±4.12 10.4 ±3.8 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Table 28: Differences in work and quality of life for job type: ANOVA Scale Non- Mixed Manual P manual Mean 50.6 ±18.4 57.7 ±24.2 value WRF total WRF Work Scheduling Demands WRF Output Demands WRF Physical Demands WRF Mental Demands WRF Social Demands AWS EQ-VAS SD Mean 49.3 ±20.7 53.3 ±26.8 SD Mean 48.8 ±16.6 52.7 ±29.7 SD .829 .447 48.3 ±22.3 45.5 ±21.6 48.7 ±23.6 .673 54.7 ±26.1 59.5 ±30.9 59.2 ±27.4 .480 49.7 ±22.0 45.4 ±24.5 37.2 ±15.0 .035* 38.9 ±15.6 39.9 ±21.5 35.6 ±14.4 .442 18.2 59.4 ±19.7 ±6.1 18.2 ±6.1 58.4 ±19.4 19.3 63.6 ±16.1 ±6.1 .676 .554 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Table 29: Job Type and outcome measures ANOVA (post hoc test) Dependent Variable WRF Mental Demands Job Type (i) Job Type (j) Mixed Manual Non-manual Manual Mixed Non-manual Mean Std. Error 2.7 4.0 3.4 5.3 5.3 5.0 P Difference (i-j) value .510 .942 .391 .265 .265 .034* Non-manual -3.0 1.3 -4.4 8.2 -8.2 -12.6 Mixed Manual **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Job Sector An independent samples t-test was conducted to compare disease activity, fatigue levels, work role functioning, arthritis work spill over and self-rated health between those working in the public sector and those self-employed or working in the private sector. Those working in public service had significantly higher levels of mental 46

  47. fatigue, difficulty in overall work functioning and managing the physical demands of work (Tables 30 and 31). Table 30: Differences in disease activity and fatigue levels for job sector Outcome Measure Disease Activity Job Sector N Mean SD P value Private or self- employed Public Service Private or self- employed Public Service Private or self- employed Public Service Private or self- employed Public Service Private or self- employed Public Service Private or self- employed Public Service 81 4.7 ±2.6 .795 71 80 4.8 14.1 ±2.5 ±4.3 MFI General Fatigue .159 67 80 15.0 13.0 ±3.8 ±4.3 MFI Physical Fatigue .558 67 80 13.4 9.9 ±4.0 ±4.0 MFI Reduced Activity .132 67 80 11.0 10.1 ±4.2 ±3.6 MFI Reduced Motivation .110 67 80 11.0 10.2 ±3.2 ±4.4 MFI Mental Fatigue .023* 67 11.8 ±3.9 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) Table 31: Differences in work and quality of life for job sector Outcome Measure WRF Job Sector N Mean SD P value Private or self- employed Public Service Private or self- employed Public Service Private or self- employed Public Service Private or self- employed Public Service Private or self- employed 80 49.1 ±20.0 .024* 67 80 56.6 ±18.8 55.3 ±26.7 WRF Work Scheduling Demands WRF Output Demands .082 67 80 62.9 ±24.4 47.2 ±20.9 .245 67 80 51.5 ±23.3 54.6 ±28.8 WRF Physical Demands .014* 67 80 66.4 ±27.3 47.4 ±22.9 WRF Mental Demands .092 47

  48. Outcome Measure Job Sector N Mean SD P value Public Service Private or self- employed Public Service Private or self- employed Public Service Private or self- employed Public Service 67 80 54.2 ±24.0 37.1 ±18.8 WRF Social Demands .076 67 80 42.6 ±17.1 18.0 AWS ±6.2 .088 67 80 19.7 60.9 ±19.6 ±5.9 EQ5D-VAS .133 67 56.1 ±18.4 **Significant at p≤0.01** indicating a 99% chance of the relationship being true. *Significant at p≤0.05 indicating a 95% chance of the relationship being true (Pallant, 2010) 5.7 Summary of Quantitative Results The largest single categories in our sample of those working were female; with a diagnosis of Rheumatoid Arthritis; had had their condition for less than five years; and were between 41-50 years of age. The largest group of respondents were in full-time employment in non-manual type of work. Of those asked about disclosure, three quarters had disclosed their Rheumatic disease to their employer. The group had a mean disease activity score of 4.66. Fatigue levels according to the MFI were highest in the area of general fatigue followed by physical fatigue. The total mean level of difficulty in work role functioning was 49.7% indicating that those working have difficulty with their work activities almost 50% of the time. Physical demands present the highest proportion of difficulty at 56.7%. The Arthritis work spill-over mean score was 18.3. The majority of those currently working identified some problems with mobility, no problems with self-care, some problems with usual activity, moderate pain/discomfort and some difficulty with anxiety and depression. The mean EQ-VAS (0-100) score was 59.5 with a standard deviation of 19.2 Relationships between variables Significant relationships were found between disease activity and fatigue, work role functioning (except managing social demands of work), arthritis work spill-over and health related quality of life. Significant relationships were found between all domains of fatigue and all aspects of work role functioning, arthritis work spill-over and health related quality of life. Significant differences between variables Significant differences were found between those with and without severe fatigue in all variables except for managing the social demands of work 48

  49. There were also significant differences between younger and older respondents in overall work role functioning, in managing the mental and social demands of work, with younger participants having significantly more difficulty. Demographic differences between fatigue, work and quality of life Gender Women had significantly higher levels of general, physical and mental fatigue, in comparison to men. Women also had significantly more difficulty in all aspects of work role functioning than men, except for meeting the social demands of work. Women also reported significantly lower health related quality of life than male participants. Educational attainment Those with education up to second level had significantly higher disease activity levels than those with university-level education. Those with third level education had significantly more difficulty in managing the mental and social demands of their work than those with education up to secondary level. Although those with 3rd level education tended to have higher levels of fatigue, there were no significant differences between the two groups. Co-morbidity Participants with co-morbidities had significantly higher fatigue in all domains of the MFI except fatigue-related motivation than those with no co-morbidity. Those with co-morbidity had a significantly higher proportion of difficulty in overall work role functioning, specifically in managing work scheduling demands. They also reported significantly higher levels of work/arthritis spill-over than those with no co- morbidity. There was also a significantly lower level of health-related quality of life than study respondents with no co-morbidity. Disease duration Statistically significantly differences were found in MFI categories of fatigue-related reduced activity and fatigue-related reduced motivation, between those with a disease duration of less than five years and those with a disease duration of 10 years or more. Work Characteristics Working hours Employees working part-time had significantly higher disease activity, and significantly higher fatigue levels in all domains (except mental fatigue) in comparison to those working full-time. Part time workers scored higher (but not significantly) in all categories of WRF except for social demands. They also reported lower EQ-VAS scores than full-time workers. 49

  50. Employees working full-time had significantly higher levels of difficulty in meeting the physical demands of work than part-time workers. Job type Non-manual workers had higher difficulty (but not significantly) in managing the mental demands required in work functioning in comparison to manual workers. Job sector Participants working in public service had significantly higher levels of mental fatigue, difficulty in overall work functioning and managing the physical demands of work than those in the private sector or self-employed. 50