2011 John Wiley & Sons A ⁄ S

Acta Neurol Scand 2012: 125: 248–253 DOI: 10.1111/j.1600-0404.2011.01557.x

ACTA NEUROLOGICA SCANDINAVICA

Post-stroke fatigue and return to work: a 2-year follow-up Andersen G, Christensen D, Kirkevold M, Johnsen SP. Post-stroke fatigue and return to work: a 2-year follow-up. Acta Neurol Scand: 2012: 125: 248–253.  2011 John Wiley & Sons A ⁄ S. Background – Post-stroke fatigue may affect the ability to return to work but quantitative studies are lacking. Method – We included 83 first-ever stroke patients <60 years and employed either full-time (n = 77) or part-time (n = 6) at baseline. The patients were recruited from stroke units at Aarhus University Hospital between 2003 and 2005 and were followed for 2 years. Fatigue was assessed by the Multidimensional Fatigue Inventory. Pathological fatigue was defined as a score ‡12 on the General Fatigue dimension. Return to paid work was defined as working at least 10 h per week. Data were analyzed using multivariable logistic regression. Results – A total of 58% of patients had returned to paid work after 2 years. The adjusted Odds Ratio (OR) for returning to paid work was 0.39 (95% confidence interval (CI) 0.16–1.08) for patients with a General Fatigue score ‡12 at baseline. Persisting pathological fatigue after 2 years of follow-up was associated with a lower chance of returning to paid work [adjusted OR 0.29 (95% CI 0.11–0.74)]. Higher scores of General Fatigue at follow-up also correlated negatively with the chance of returning to paid work when analyzing fatigue on a continuous scale (adjusted OR 0.87, 95% CI 0.80–0.94 for each point increase in General Fatigue). Conclusions – Post-stroke fatigue appears to be an independent determinant of not being able to resume paid work following stroke.

Introduction

Return to work is a key goal in recovery following stroke in young patients but is often problematic because adjustments to the work situation may be necessary, even in cases where there is no obvious disability (1–6). The socioeconomic burden caused by disability among patients with stroke is substantial for the individual patient, relatives, and society, respectively, because of changes in income and lost productivity. The overall chance of returning to paid work was 44% in a recent systematic review (7); however, the chance varied widely in the studies possibly reflecting differences in stroke settings and selection criteria, small sample sizes, varying completeness of follow-up as well as differing definitions of employment. The rehabilitation process is probably insufficient in both duration and scope in preparing patients for resuming employment, and little is known about 248

G. Andersen1, D. Christensen1, M. Kirkevold2, S. P. Johnsen3 1

Department of Neurology, Aarhus University Hospital, Aarhus C, Denmark; 2Department of Nursing Science, University of Aarhus, Aarhus C, Denmark; 3Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus N, Denmark

Key words: fatigue; socioeconomic outcome; stroke; work S. P. Johnsen, MD, PhD, Department of Clinical Epidemiology, Aarhus University Hospital, Olof Palmes All 43-45, DK-8200 Aarhus N, Denmark Tel.: +45 8942 4803 Fax: +45 8942 4801 e-mail: [email protected] Accepted for publication May 26, 2011

barriers that hinder employment. Most studies have focused on physical barriers (e.g., motor weakness, cognition, and language) as well as social barriers (e.g., education, marital status, job context), while mental barriers and psychiatric morbidity are often overlooked. Factors associated with return to work seem to be less severe stroke and good functional outcome at discharge, absence of post-stroke psychiatric morbidity, high income, white-collar employment, positive personal attitude, and positive attitude of co-workers (2, 5, 6, 8– 10). In contrast, no consistent associations have been found for factors like age, sex, stroke location, marital status, or race. Fatigue affects most stroke patients in the acute phase and is also a frequent long-lasting sequelae in young stroke patients (11–14) that limits a variety of aspects of everyday life. The prevalence of fatigue among post-stroke patients has been estimated to be 40–74% (15). Fatigue may be a possible key factor that prevents return to paid

Post-stroke fatigue and return to work work in otherwise recovered stroke patients, and lack of effective treatment strategies could therefore constitute an important possible barrier for return to work. Although post-stroke fatigue has been identified in a qualitative study as a possible barrier for return to work (8), no existing quantitative studies have, to our knowledge, reported on the long-term impact of post-stroke fatigue on return to work. We therefore conducted a cohort study with long-term follow-up to examine whether fatigue is independently associated with the chance of returning to paid work.

Subjects and methods Study sample

The patients were recruited from three acute stroke units at Aarhus University Hospital between 1 March 2003 and 28 February 2005, and all baseline data were prospectively registered in a national stroke database as part of the National Indicator Project (16). All patients had an acute Computer Tomografi or Magnetic Resonance Imaging scan upon admission. The study was part of a larger prospective study concerning fatigue in a cohort of 165 first-ever stroke patients (15). We included all patients aged <60 years who were full-time (n = 77) or part-time (n = 6) employed at baseline in the present follow-up study. Thus, the patient sample in this study included 83 first-ever stroke patients diagnosed according to the WHO criteria (acute focal or global cerebral deficit with symptoms lasting more than 24 h of presumed cerebrovascular etiology). Patients with subarachnoidal hemorrhage were not included in the study. Stroke patients were treated according to national clinical guidelines, which included thrombolysis when relevant (17). Presumed post-stroke depression and pathological crying were detected early and prevented or treated with use of antidepressants (18). If needed, patients were referred directly from the acute stroke ward to rehabilitation in a dedicated rehabilitation hospital specialized in stroke rehabilitation and with focus on post-stroke employment (19). In general, patients were advised to resume paid work, if possible parttime, approximately 3 months after the stroke incident. The social security system in Denmark provides full financial compensation to patients during a run-in period with a maximum length of 6 months. Decisions on whether patients should be granted social benefits including sick-leave, early retirement and so-called flex-job (a shielded work-

ing condition with financial compensation) are made by the social authorities. Patients were included in the study upon giving their consent and were scheduled for follow-up visits at 10 days, 3 months, 1 year, and 2 years after stroke onset. Data were collected in person by one of the authors (DC) at the stroke units, rehabilitation units, the neurological outpatient clinic, or in the patientÕs home. Data on fatigue

Fatigue was assessed using the self-rating Multidimensional Fatigue Inventory (MFI-20). MFI-20 is based on different modes of expressing fatigue and encompasses the following dimensions: General Fatigue includes overall feelings of being tired; Physical Fatigue refers to the physical sensations related to fatigue; Mental Fatigue includes deficits in cognitive functioning; Reduced Motivation and Reduced Activity refer to lack of motivation to do anything useful. Each dimension consists of four items with a response rate from one to five. Scores can range from a minimum of four to a maximum of twenty. A high score on fatigue items indicates a high level of fatigue, while a high score on nonfatigue items indicates a low level of fatigue. The score on the General Fatigue dimension may serve as an overall indicator of fatigue (20). In this study, we used a cutoff score ‡12 on the General Fatigue dimension as an indicator of clinically significant pathological fatigue (16). Data on return to work

At baseline, the work status was registered in the database: employed vs unemployed, student, pensioner, not known. In Denmark, it is possible to retire and receive a social pension beginning at 60, 65, 68, or 70 years of age. We chose the following definition of return to paid work: Working (or studying) at least 10 h. per week (25% of full-time work) at a workplace on the open market (21). Changes from baseline were registered at follow-up visits at one and 2 years. Data on patient characteristics

Data on sociodemographic and clinical characteristics of the stroke patients were collected during the in-person contacts. Sociodemographic data in addition to employment status included age and sex. Clinical variables included lesion location, Scandinavian Stroke Scale (SSS) score at admission, and follow-up (22), Barthel Index-100 (BI) (23), Symbol Digit Modalities Test (SDMT) (24), 249

Andersen et al. Charlson Comorbidity Index (25), new onset depression, and antidepressant treatment. Lesion location was characterized first as hemispheric or bilateral, subsequently as Total Anterior Circulation (TAC), Partial Anterior Circulation (PAC), Lacunar (affecting deep perforating arteries (lacunar stroke)), or Posterior Circulation (POC) (26). SDMT was used to assess change in cognitive functioning (attention and speed) over time. The SDMT consists of a sheet of paper with geometric figures and numbers which the patient is requested to match within 90 s. The score consists of a total score, summing up all the correct matches (24). The Charlson comorbidity index score, which covers 19 major disease categories, was computed for each patient based on all discharge diagnoses recorded prior to the hospitalization for stroke. Data were obtained from the National Registry of Patients which contains data on all discharges from non-psychiatric hospitals in Denmark since 1977 (27). Depression was assessed using the Major Depression Inventory from which depression can be diagnosed according to the International Classification of Diseases 10th revision (ICD-10) (28). The patients were systematically asked to perform the SDMT after MFI-20 and the Major Depression Inventory. The study was approved by the local ethics committee, and all participating patients provided informed consent (J no. 20030012). Statistical analysis

Dichotomous data are presented as percentages. Continuous variables are presented as medians (25th–75th percentiles). Crude and adjusted odds ratios (OR) with 95% confidence intervals (CI) were computed for the association between the General Fatigue score and the chance of returning to work after 2 years of follow-up using logistic regression. General Fatigue score at baseline and after one and 2 years of follow-up were considered both as continuous and dichotomous variables. The latter were defined as a General Fatigue score ‡12. Covariates associated with the chance of returning to work after 2 years of follow-up in univariable logistic regression analyses (Wald test P < 0.10) were included in the multivariable logistic regression models. All analyses were conducted using R statistical software (version 2.8.1) (http://www.r-project.org/). Results

At 1-year follow-up 44 of 83 first-ever stroke patients (53%) had resumed work (32 full-time, 10 250

part-time, and two patients in ÔflexÕ-job, 27 were still on sick leave, eight had been granted a pension or other social benefits, and in four patients information was missing. At 2-year follow-up 48 (58%) had resumed work (27 full-time, 9 part-time, and 12 in ÔflexÕ-job), 16 had been granted a pension or other social benefits, three were retired, and in 10 patients, information was missing. From 1-year till 2-year follow-up one patient in part-time employment changed to full-time employment while one patient working part-time went back on sick leave. One patient on sick leave changed to part-time employment, and an additional eight patients on sick leave changed to a ÔflexÕ-job. The descriptive baseline characteristics of the 83 employed patients at stroke onset are shown in Table 1. Data are shown for all patients and according to job status at 2-year follow-up. No patients developed depression during follow-up according to the Major Depression Inventory; however, antidepressant treatment was initiated for 36% of the patients to prevent or treat symptoms of post-stroke depression or pathological crying during follow-up. Post-stroke fatigue was associated with a lower chance of returning to paid work during follow-up as shown in Table 2 although not all associations reached statistical significance. The crude ORs for returning to paid work were 0.51 (95% CI 0.18– 1.39) and 0.39 (95% CI 0.16–1.08), respectively after one and 2 years of follow-up for patients with a General Fatigue score ‡12 at baseline. Only marginal changes in the associations were observed when adjusting for covariates, which were associated with the chance of returning to work in univariable analyses, including stroke severity at admission, BI at discharge, and SDTM score at follow-up, resulting in adjusted ORs of 0.53 (95% CI 0.17–1.64) after 1 year and 0.38 (95% CI 0.15– 1.09) after 2 years, respectively. The adjusted OR for returning to paid work was also low in patients with persisting pathological General Fatigue at follow-up at one (adjusted OR 0.55, 95% CI 0.18– 1.65) and 2 years (adjusted OR 0.29, 95% CI 0.11– 0.74). Higher scores on General Fatigue at baseline and during follow-up also correlated negatively with the chance of returning to paid work when including General Fatigue as a continuous variable, although again not all associations were statistical significant. Hence, an increase of one point in the General Fatigue score at baseline equaled an adjusted OR of returning to work of 0.85 (95% CI 0.70–1.05) after 2 years of follow-up, whereas an increase in one point at 2 years of follow-up equalled an adjusted OR of 0.84 (0.76– 0.92).

Post-stroke fatigue and return to work Table 1 Descriptive baseline characteristics of employed patients with stroke. Presented for all patients and according to employment status after 2 years of follow-up All patients (n = 83)

Characteristics Age* Gender (%) Women Men Living arrangement (%) Living alone Cohabiting Scandinavian Stroke Scale* Aphasia (%) Barthel Index* Oxford Community Stroke Classification (%) LAC PAC POC TAC Unknown Charlson comorbidity index (%) None Low High Symbol Digit Modalities Test* MFI-20: general fatigue ‡12 (%)

Returned to work (n = 48)

Did not return to work (n = 25)

Died or lost to follow-up (n = 10)

53.8 (45.4, 58.2)

54.1 (45.4, 58.4)

55.1 (49.9, 58.0)

46.2 (41.7, 53.2)

48 52

46 54

48 52

60 40

27 73 58 (53, 58) 13 100 (95, 100)

27 73 58 (55, 58) 12 100 (100, 100)

16 57 16 8 3

19 54 17 4 6

76 20 4 35.00 (23.75, 44.75) 57

28 72 58 (43, 58) 16 100 (60, 100) 16 64 12 8 0

79 17 4 39.00 (28.50, 48.50) 46

72 28 0 27.00 (20.00, 35.00) 64

20 80 57 (30, 58) 10 100 (15, 100) 0 50 20 30 0 70 10 20 35.00 (10.75, 52.25) 90

LAC, lacunar stroke; PAC, partial anterior circulation stroke; POC, posterior circulation stroke; TAC, total anterior circulation stroke; MFI-20, Multidimensional Fatigue Inventory. *Median (lower quartile, upper quartile).

Table 2 General fatigue and return to work among patients with stroke after 2 years of follow-up. Adjusted Odds ratios (OR) with 95% confidence intervals (CI) of chance of return to work according to General fatigue score at baseline and after 2 years of follow-up

General General General General

fatigue fatigue fatigue fatigue

score >12 (baseline) score continuous (baseline) score >12 (follow-up) continuous (follow-up)

1 year

2 years

Adjusted OR (95% CI)

Adjusted OR (95% CI)

0.53 0.70 0.55 0.58

(0.17–1.64) (0.31–1.57) (0.18–1.65) (0.24–1.39)

0.38 0.85 0.29 0.84

(0.15–1.09) (0.70–1.05) (0.11–0.74) (0.76–0.92)

*Adjusted for Scandinavian Stroke Scale score (at time of admission), Symbol Digit Modalities, and Barthel Index (at time of discharge). At 1 year also adjusted for aphasia.

Discussion

This study was undertaken at a time with a historic low rate of unemployment in Denmark. Efforts were therefore made on behalf of the social health care system to support adjustments in paid work for stroke patients if needed post-stroke, to maintain employment for as many patients as possible. Although the time period was unique in this respect, our follow-up study on paid work after first stroke in patients aged <60 years at stroke onset showed that no more than 41% of patients returned to full-time paid employment after 1 year. This number decreased to 37% after 2 years.

Furthermore, 13% had resumed part-time paid work after 1 year and 12% after 2 years. Poststroke fatigue was in our study for the first time identified as a strong independent determinant of the chance of returning to paid work, an ultimate goal in rehabilitation for young stroke patients. The conditions for resuming paid work in our stroke cohort were in general good. Rehabilitation was undertaken in a dedicated hospital setting for those patients who needed in-hospital treatment, the level of social security was high, and extensive efforts were made to help stroke patients back to paid work under shielded conditions (typically sick leave for 3 months followed by part-time employment or ÔflexÕ-job as long as needed ⁄ 6 months). The proportion of stroke patients returning to work from one till 2 years after stroke did not increase in our study. However, the proportion of patients who were full- or part-time employed after 1 year was comparable with findings 5– 15 years after stroke in another Danish cohort admitted in the early 1990s (3). Three Scandinavian studies with long-term follow-up have previously reported a low proportion of stroke patients returning to work (3, 11, 21); however, an increasing proportion of patients may return to work even more than 5 years post-stroke (3, 11, 21). In the present study, it seemed that more patients reached their final employment status earlier than previously reported, but the propor251

Andersen et al. tion might increase even further with longer follow-up. Our data suggest that an early, focused effort including advice from professionals on how to return to work at a slower pace, together with financial support from social security systems may increase the ability to adjust to a reduced working capacity and thereby increase the number of fulltime paid workers within the first year after first stroke. The focus on paid work and working capacity in rehabilitation settings is seemingly important, as also stressed by others (8, 29), and it is of importance to note that returning to work is a long, time-consuming process which necessitates continuous adjustments (19). In our setting, in which the vast majority of patients were independently performing activities of daily living at time of discharge, general fatigue was an important determinant for the patientsÕ ability to cope sufficiently and manage their jobs as previously. This association was independent of physical disability or cognitive deficits which we controlled for in the multivariable analyses. In the present cohort of young first-ever strokes the functional outcome in general was very good, but still less than half the patients were able to resume paid work after 2 years without adjustments. Antidepressant treatment was commonly used in our study population, and most patients were probably well-treated because no patients had a significant concomitant depression during followup according to the Major Depression Inventory. Antidepressant medication did not appear to influence the level of fatigue in our population (15); however, others have found a close correlation between the post-stroke depression and the chance of resuming paid work (2). There is good evidence that antidepressant treatment is effective for relieving depressive symptoms, but antidepressants do not seem effective in treating fatigue (30). Post-stroke fatigue therefore needs more attention and needs to be explored in more detail to find ways to cope with or intervene with the problem. The size of our study was modest and not all presented reached statistical significance, and some caution is therefore warranted when interpreting the findings. However, it is noteworthy that all point estimates consistently indicated that increased fatigue was associated with a lower risk of returning to work. Return to work after stroke recovery may contribute to an improvement of life satisfaction and self-esteem as well as securing financial independence. We suggest a stronger focus on fatigue in the process of rehabilitation and followup of young stroke patients and propose that 252

programs should be set up to help stroke patients better cope with fatigue, as fatigue appears to be a key factor with a negative impact on working capacity. Acknowledgements Supported by grants from the Aarhus University Hospital Initiative, the Department of Nursing Science, University of Aarhus, and the Danish Medical Research Council.

Conflicts of interest We are not aware of any relevant conflicts of interest.

References 1. Alaszewski A, Alaszewski H, Potter J, Penhale B. Working after a stroke: survivors¢ experiences and perceptions of barriers to and facilitators of the return to paid employment. Disabil Rehabil 2007;29:1858–69. 2. Glozier N, Maree L, Hackett ML, Parag V, Anderson CS. The influence of psychiatric morbidity on return to paid work after stroke in younger adults. Stroke 2008;39:1526– 32. 3. Teasdale TW, Engberg AW. Psychosocial consequences of stroke: a long-term population-based follow-up. Brain Inj 2005;19:1049–58. 4. Busch MA, Coshall C, Heuschmann PU, Mckevitt C, Wolfe CD. Sociodemographic differences in return to work after stroke: the South London Stroke Register (SLSR). J Neurol Neurosurg Psychiatry 2009;80:888–93. 5. Lindstro¨m B, Ro¨ding J, Sundelin G. Positive attitudes and preserved high level of motor performance are important factors for return to work in younger persons after stroke: a national survey. J Rehabil Med 2009;41:714–8. 6. Saeki S, Toyonaga T. Determinants of early return to work after first stroke in Japan. J Rehabil Med 2010;42:254–8. 7. Daniel K, Wolfe CD, Busch MA, Mckevitt C. What are the social consequences of stroke for working-aged adults? A systematic review. Stroke 2009;40:e431–40. 8. Medin J, Barajas J, Ekberg K. Stroke patientsÕ experiences of return to work. Disabil Rehabil 2006;28:51–60. 9. Appelros P. Prevalence and predictors of pain and fatigue after stroke: a population-based study. Int J Rehabil Res 2006;29:329–33. 10. Bendz M. The first year of rehabilitation after a strokefrom two perspectives. Scand J Caring Sci 2003;7:215–22. 11. Naess H, Nyland HI, Thomassen L, Aarseth J, Myhr KM. Long-term outcome of cerebral infarction in young adults. Acta Neurol Scand 2004;110:107–12. 12. Naess H, Nyland HI, Thomassen L, Aarseth J, Myhr KM. Fatigue at long-term follow-up in young adults with cerebral infarction. Cerebrovasc Dis 2005;20:245–50. 13. Choi-Kwon S, Han SW, Kwon SU, Kim JS. Poststroke fatigue: characteristics and related factors. Cerebrovasc Dis 2005;19:84–90. 14. Schepers VP, Visser-Meily AM, Ketelaar M, Lindeman E. Poststroke fatigue: course and its relation to personal and stroke-related factors. Arch Phys Med Rehabil 2006;87:184–8. 15. Christensen D, Johnsen SP, Watt T, Harder I, Kirkevold M, Andersen G. Dimensions of post-stroke fatigue: a twoyear follow-up study. Cerebrovasc Dis 2008;26:134–41.

Post-stroke fatigue and return to work 16. Mainz J, Krog BR, Bjornshave B, Bartels P. Nationwide continuous quality improvement using clinical indicators: the Danish National Indicator Project. Int J Qual Health Care 2004;16(Suppl 1):i45–50. 17. Referenceprogram for behandling af patienter med apopleksi. Danish Board of Health. Copenhagen. http://www.sst.dk/ publ/Publ2006/CEMTV/SfR/Apopl_refprg.pdf (accessed 9 June 2011) [in Danish]. 18. Andersen G, Vestergaard K, Lauritzen L. Effective treatment of poststroke depression with the selective serotonin reuptake inhibitor citalopram. Stroke 1994;25:1099–104. 19. Blicher JU, Jensen CB, Westh J, Hellemann I. Housing and work outcomes one year after severe brain injury. An interview-based follow-up study. Ugeskr Laeger 2007;169:228–31. 20. Smets EMA, Garssen B, Bonke B, Dehaes JCJM. The multidimensional fatigue inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res 1995;39:315–25. 21. Hofgren C, Bjo¨rkdahl A, Esbjo¨rnsson E, Stibrant-Sunnerhagen K. Recovery after stroke: cognition, ADL function and return to work. Acta Neurol Scand 2007;115:73–80. 22. Jorgensen HS, Nakayama H, Raaschou HO, Vive-Larsen J, Stoier M, Olsen TS. Outcome and time course of recovery in stroke. Part II: time course of recovery. The Copenhagen Stroke Study. Arch Phys Med Rehabil 1995;76:406–12.

23. Mahoney FI, Barthel DW. Functional evaluation: the BARTHEL Index. Md State Med J 1965;14:61–5. 24. Smith A. Symbol digit modalities test: manual. Los Angeles: Western Psychological Services, 1982. 25. Charlson ME, Pompei P, Ales KL, Mackenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373–83. 26. Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet 1991;337:1521–6. 27. Andersen TF, Madsen M, Jorgensen J, Mellemkjoer L, Olsen JH. The Danish National Hospital Register. A valuable source of data for modern health sciences. Dan Med Bull 1999;46:263–8. 28. Bech P, Wermuth L. Applicability and validity of the major depression inventory in patients with ParkinsonÕs disease. Nord J Psychiatry 1998;52:305–9. 29. Van De Port IGL, Van Den Bos GA, Voorendt M, Kwakkel G, Lindeman E. Identification of risk factors related to perceived unmet demands in patients with chronic stroke. Disabil Rehabil 2007;29:841–6. 30. Choi-Kwon S, Choi J, Kwon SU, Kang DW, Kim JS. Fluoxetine is not effective in the treatment of poststroke fatigue: a double-blind, placebo-controlled study. Cerebrovasc Dis 2007;23:103–8.

253