Skip Navigation


Rheumatology Advance Access originally published online on August 18, 2009
Rheumatology 2009 48(10):1283-1289; doi:10.1093/rheumatology/kep239
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
48/10/1283    most recent
kep239v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Anis, A.
Right arrow Articles by Sato, R.
PubMed
Right arrow PubMed Citation
Right arrow Articles by Anis, A.
Right arrow Articles by Sato, R.
Related Collections
Right arrow Pharmacology
Right arrow Rheumatoid Arthritis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2009. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

The effect of etanercept on work productivity in patients with early active rheumatoid arthritis: results from the COMET study

Aslam Anis1,2, Wei Zhang2, Paul Emery3, Huiying Sun2, Amitabh Singh4, Bruce Freundlich5 and Reiko Sato4

1School of Population and Public Health, University of British Columbia, 2Centre for Health Evaluation and Outcome Sciences, St Paul's Hospital, Vancouver, Canada, 3Academic Unit of Musculoskeletal Disease, Leeds University, Leeds, UK, 4Global Health Outcomes Assessment and 5Global Medical Affairs, Wyeth, Collegeville, PA, USA.

Correspondence to: Aslam Anis, Centre for Health Evaluation and Outcome Sciences, 620-1081 Burrard Street, Vancouver, BC V6Z 1Y6, Canada. E-mail: aslam.anis{at}ubc.ca


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Acknowledgements
 References
 
Objectives. To compare the impact of the combination of etanercept (ETN) and MTX with MTX alone on work productivity among MTX-naïve patients with active early RA over a 12-month period.

Methods. The COMET (COmbination of Methotrexate and ETanercept) trial was a 2-year double-blind randomized clinical trial. Absenteeism during the first year was measured and it included: (i) number of missed workdays; (ii) reduced working time; and (iii) number of stopped workdays. Each absenteeism measure was estimated using a mixed model, and their variations were estimated by bootstrapping. As a sensitivity analysis, the lost workdays due to presenteeism (reduced performance at work) was also estimated.

Results. Two hundred and five patients [MTX (n = 100) vs ETN + MTX (n = 105)], who were working full time or part time at baseline and had at least one follow-up observation, were included in the analysis. Compared with the MTX group, the ETN + MTX group had a maximum of 37 fewer missed workdays or at minimum 22 fewer missed workdays. The associated productivity gain equalled £2586 and £1555, respectively. When additionally accounting for presenteeism, the total improvement could be as high as 42 (95% CI 16, 69) fewer lost workdays representing a productivity gain of £2968.

Conclusions. Our results demonstrated that early treatment with ETN + MTX led to a significant attenuation of absenteeism among patients with early active RA. These productivity gains represent benefit beyond the traditional measures of clinical and radiographic improvements. Further research to simultaneously measure both absenteeism and presenteeism is warranted.

KEY WORDS: Rheumatoid arthritis, Absenteeism, Presenteeism, Etanercept, Work productivity


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Acknowledgements
 References
 
RA is the most common form of inflammatory arthritis [1] with a prevalence rate of ~1% and an annual incidence of 3 per 10 000 adults [2]. Patients with RA are often absent from work, decrease their routine work hours or suffer from job loss, all of which contribute to productivity losses. Many RA patients are also unable to work at their full potential while at work, known as ‘presenteeism’ [3, 4]. Studies have shown that the associated costs are substantial [4–6]. One recent study [5] found that productivity losses accounted for 32% of the total annual costs per RA patient in Europe, exceeding other cost components such as medical costs (21%), drug costs (14%), non-medical costs (14%) or informal care costs (19%).

Recent studies have demonstrated that disease-related absence from work, reduction in work hours and job loss can occur even during the very early phase of RA. A systematic review of cohort studies on productivity losses due to RA showed that the time between RA onset until 50% probability of being permanently work disabled varied from 4.5 to 22 years [3]. Merkesdal et al. [7] found that within the first 3 years of RA, there was an average of 82 days of sick leave per person-year and 26% of the patients lost work because of RA. Eberhardt et al. [8] investigated the work disability rates over 15 years in an early RA cohort, where the mean duration of RA was 11 months at baseline. The work disability rate increased from 28% at baseline to 35% after 5 years. Among a total of 148 patients, 47 patients reduced working hours by changing working status from full time to part time. Thus, work productivity loss is a serious and common problem even for patients with very early RA. When RA patients were asked to prioritize as to which dimension of health they would like to see improvement, almost one-third of patients cited health status related to work as an important priority [9].

Recently, clinical trials have demonstrated that aggressive treatment in early RA leads to positive work-related outcomes. Initial aggressive treatment of RA with a combination of DMARDs relative to therapy with a single DMARD significantly reduced the cumulative duration of sick leave and RA-related disability [10]. Other studies have reported that early intervention with a combination of biologic therapy with MTX significantly improved the employment potential of early RA patients [11–15] and reduced absent workdays [11, 13, 14]. However, these studies focused on measuring the impact of combination therapy on job loss, employability and absent workdays of patients with early RA. The consequent impact on overall productivity has yet to be established.

In an early study examining employment outcomes and biologic therapy, Yelin et al. [16] reported an association between use of etanercept (ETN) and a higher employment rate and greater working hours compared with an observational cohort. Given that these results were observed in patients with long-standing RA, the question of whether productivity benefits would also accrue to patients with early active RA receiving ETN remained unanswered. Therefore, the objective of this study was to evaluate the impact of the combination of ETN and MTX relative to MTX alone on work productivity among MTX-naïve patients with active early RA over a 12-month period.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Acknowledgements
 References
 
Study design
The COMET (COmbination of Methotrexate and ETanercept) trial was a 24-month, double-blind, two-period, multi-centre, randomized clinical trial [17, 18]. Both Periods 1 and 2 were 12 months in duration. In the first year, namely Period 1, a total of 542 patients were randomized equally to two treatment groups: a combination of ETN and MTX vs MTX alone. The main inclusion criteria were: (i) subject was >=18 years of age; (ii) subject had met the ACR criteria for the diagnosis of RA; (iii) disease duration was between 3 months and 2 years; and (iv) subject had a disease activity score (28 joints) (DAS28) >=3.2 and at least one of the following two criteria: ESR >=28 mm/h or CPR level >=20 mg/l. As a sub-analysis of COMET, this article compares the impact of the combination of ETN and MTX with MTX alone on work productivity among MTX-naïve patients with active early RA over the first 12 months. The study complied with the Declaration of Helsinki, and received institutional review board approval and regulatory review prior to site initiation. All subjects provided written informed consent.

Absenteeism
The primary outcome of this study was absenteeism during the first 12 months among patients who were working at baseline. The information on work status at baseline and absenteeism during the trial was collected by a self-reported questionnaire. At baseline, patients were asked ‘Which best describes your work situation right now?’. Patients were able to choose among responses including ‘I am working full time’, ‘I am a student’, ‘I am unemployed but currently looking for work’, ‘I am retired’, ‘I am working part time’, ‘I am receiving unemployment benefits’, ‘I am a homemaker and not looking for outside work’, ‘I work as volunteer’, ‘I am unable to work due to health problems or injury’ and ‘Other work situation’. During the trial, patients were asked to specify, since their last visit, ‘the number of days missed from work because of health’, ‘the number of hours they cut down from routine work because of health’, and whether they stopped working because of health. Absenteeism comprised three components: (i) number of missed workdays; (ii) reduced working time; and (iii) number of stopped workdays. The number of missed workdays indicates, within the 1-year study period, how many days patients did not work because of health while they were still employed. The number of stopped workdays indicates the duration that patients were unemployed because of health within the first year. The outcome was measured at four follow-up visits in the COMET trial: Weeks 12, 24, 36 and 52. Patients working full time or part time at baseline and having absenteeism data available on at least one of the four follow-up visits were included in the analysis. The recall period for absenteeism at Weeks 12 and 24 was for the prior 4-week period, and at Weeks 36 and 52 was for the prior 8-week period.

Patients reported reduced working time in hours and these were transformed into days by assuming an 8-h workday for full time and 4-h workday for part time workers. We assumed that the total number of workdays per week was 5 days and that the total number of workdays per year was 208 days, which was calculated by the average annual hours actually worked per UK worker in 2007 (1670 h) divided by eight [19]. Since the exact date of work stoppage/resumption was not collected, this was estimated according to two scenarios. Scenario I assumed that once a patient reported stopping working, he or she would not restart working during the remaining portion of the year. Total absenteeism was then calculated by summing up the three components. Alternatively, Scenario II assumed that patients could resume working after a period of work stoppage. The date of work stoppage would be at the middle time point between when patients reported ‘not stopped’ and when they reported ‘stopped’. The maximum of the three components was counted as total absenteeism. Accordingly, Scenario I provides a maximum estimate of absenteeism and Scenario II provides the minimum estimate. Under either scenario, total absenteeism could not exceed 208 workdays.

Presenteeism and total work productivity loss
Previous studies have shown that presenteeism reduces work productivity by more than absenteeism and is the largest component of work productivity loss [3, 4]. In a recent study on productivity loss due to arthritis, the largest component of overall productivity loss was presenteeism, accounting for 41% of the total productivity loss. This exceeded productivity loss due to job loss/change (37%), decreased work hours (12%) and missed workdays (10%) [4]. Therefore, presenting the productivity loss due to absenteeism alone would be an underestimate. Given that presenteeism was not measured in the COMET trial, we incorporated the mapping algorithm published recently on the association between measures of disease severity and presenteeism among patients with OA and RA [20]. In this study by Zhang et al., presenteeism was not found to differ between OA and RA patients. The percent of work productivity loss (while at work) due to presenteeism was according to two instruments: (i) the Work Productivity and Activity Impairment Questionnaire (WPAI) [21, 22] and (ii) the Work Limitations Questionnaire (WLQ) [23, 24]. Patients’ functional disability level was measured according to the HAQ [25, 26]. The association between categorical HAQ score (by quartile: <0.25, 0.25–0.75, 0.75–1.25 and >=1.25) and the percent of work productivity loss at work was estimated by regression analysis.

Since our presenteeism estimate was not directly measured in the COMET trial, but imputed, we included it only as part of a sensitivity analysis around the base case estimate of absenteeism. To do so, we first estimated the percent of productivity loss while at work in the COMET study by HAQ score. That is, using the association between percent of productivity loss at work measured by WPAI and WLQ and categorical HAQ score in Zhang et al.'s study [20], we mapped the corresponding estimates of percent of productivity loss at work in the COMET study based on the measured HAQ score. Secondly, lost workdays due to presenteeism were then estimated by multiplying the percent of productivity loss at work by the actual workdays of COMET patients in the first year. The actual workdays of COMET patients were calculated as the total possible number of workdays (208 days) minus the total absenteeism. Thirdly, total work productivity loss was calculated by summing up the total absenteeism and the lost workdays due to presenteeism. Table 1 provides a detailed explanation of the estimation procedure for each component of work productivity loss.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Components of work productivity loss measure

 
Trade-off between absenteeism and presenteeism
The relationship between absenteeism and presenteeism is given by the equation, P = (208 - A)x p, where P is lost workdays due to presenteeism, A is absenteeism, p is the percent of work productivity loss while at work and 208 is the assumed total number of possible workdays. As the equation shows, the larger the A is, the smaller the P will be and that there exists a trade-off between A and P. Intuitively, presenteeism only occurs when people are at work and, hence, is inversely related to absenteeism.

Work productivity cost
Adopting a societal perspective, the human capital approach was used to estimate work productivity costs that were calculated by multiplying work productivity loss (days) by an average daily pay rate for all patients according to British labour statistics. British national statistics reported median/mean weekly pay by work status (full time vs part time), sex and age group in 2007 [27]. We first calculated the weighted median/mean daily pay in three ways: (i) weighting wage only by the distribution of work status in COMET data (median/mean = £74.25/£89.87); (ii) weighting by the distribution of work status and sex in COMET data (£70.66/£84.70); and (iii) weighting by the distribution of work status, sex and age group in COMET data (£75.18/£89.47). The daily pay was assumed to be weekly pay divided by five. The three weighted median/mean daily pay estimates were similar, whereas the median estimates were somewhat lower than mean estimates. Therefore, the lowest median daily pay, which was weighted by work status and sex (£70.66), was used to estimate work productivity costs.

Statistical analysis
To estimate total absenteeism according to its three components as measured in COMET, we first used a mixed model for longitudinal data, which included intercept as a random effect. A mixed model was chosen since it accounts for within-person correlations that occur with repeated measurements and acknowledges that each patient has a different number of missed work days. The model can be expressed as: Yij = (β0 + b0i) + β 1Treatmentij + β 2Weekij+ β 3TreatmentijxWeekij + {epsilon}ij; where Y is one of the three components, i indicates individual patient, j indicates the j-th observations, b is the random effect for intercept and the variable week can be 16, 24, 36 and 52.

We calculated the average 1-year estimates of the three components and the total absenteeism for the two treatment groups (ETN + MTX vs MTX) and their difference. Next, since the distribution of the absenteeism data was highly skewed, bootstrapping of 10 000 samples was used to help measure the variations in all the three component estimates and estimate the 95% CIs. Simultaneously, a proportion of each HAQ category (<0.25, 0.25–0.75, 0.75–1.25 and >=1.25) during the first year was generated from each of the bootstrapped sample. Finally, Monte Carlo simulation method was used to estimate the distribution of the percent of work productivity loss at work, the lost workdays due to presenteeism and the total work productivity loss in COMET study by taking account of the uncertainty around the HAQ coefficients estimated from the Zhang et al. [20] study, the HAQ estimates from COMET study and the actual workdays of COMET patients. The simulation was repeated 10 000 times. For each outcome, the 95% CI (percentile interval from bootstrapped samples or from Monte Carlo simulations) was calculated. Fisher's exact test was used to compare the probability of stopping work for the first time by each follow-up visit between treatment groups.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Acknowledgements
 References
 
Among the total 542 randomized patients, 214 (39.5%) were working full time or part time at baseline. Among those patients who were working at baseline, nine were excluded from our analysis because they did not have any follow-up observation on absenteeism. Thus, a total of 205 patients who were working at baseline [MTX (n = 100) vs ETN + MTX (n = 105)] were included in our analysis. The mean age was 45 years and 69% were females (Table 2). At baseline, the mean RA duration was 8.7 months. Patients had high disease activity and pain, moderate-to-severe functional disability and poor quality of life. Most patients (73%) were working full time at baseline. Patients in the two treatment groups had similar demographics and baseline disease characteristics. Table 2 also presents the demographics and baseline disease characteristics of the remaining 328 randomized patients who were not working full time or part time at baseline (n = 311) or who did not report their employment status (n = 17). Compared with these 328 patients, the patients who were working at baseline were younger and had slightly lower disease activity and better physical function and quality of life.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Demographics and baseline disease characteristics

 
Absenteeism
For total absenteeism and each of its components, the number of lost workdays in patients treated with ETN + MTX was approximately half of that of the patients receiving MTX (Table 3). The subjects in ETN + MTX group missed 18 (95% CI 2, 34) fewer workdays in the first year compared with subjects in MTX group (ETN + MTX: 14 days vs MTX: 32 days). Under Scenario I, the total annual absenteeism was 29 workdays for the ETN + MTX group compared with 66 workdays for the MTX group. This corresponded to 37 fewer workdays lost (95% CI 6, 68) equalling £2586 productivity gain for the ETN + MTX group. Under Scenario II, 22 workdays were lost for the ETN + MTX group vs 44 workdays in the MTX groups, resulting in 22 fewer workdays lost (95% CI 2, 43) or a productivity gain of £1555 for the ETN + MTX group.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Absenteeism during 52 weeks

 
Presenteeism
As a sensitivity analysis, we estimated the percent of work productivity loss at work and the lost workdays due to presenteeism by treatment groups, scenarios (Scenario I vs II) and presenteeism instruments (WPAI vs WLQ) (Table 4). Regardless of instrument, the percentage of work productivity loss at work among patients treated with MTX was significantly higher than that among patients treated with combination therapy. But the magnitudes of the estimates varied widely according to the instrument. The difference in the percentage of work productivity loss at work between the two treatment groups was –7.5 (95% CI –11.2, –4.2) according to the WPAI vs –1.4 (95% CI –2.1, –0.7) according to the WLQ. Therefore, the lost workdays due to presenteeism and the total work productivity loss were correspondingly different. Recall that absenteeism estimated under Scenario I was greater than that under Scenario II (Table 3). Due to the inverse relationship between absenteeism and presenteeism, the lost workdays due to presenteeism estimated under Scenario I were, therefore, lower than that for Scenario II within each treatment group.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Presenteeism and total work productivity loss during 52 weeks

 
Total work productivity loss
As a part of sensitivity analysis, the total work productivity loss (the sum of absenteeism and presenteeism) was calculated (Table 4). Since absenteeism was only partially traded off by the lost workdays due to presenteeism, the total work productivity loss under Scenario I was still higher than the loss under Scenario II in each treatment group. When the WPAI was used to estimate the percentage work productivity loss at work, total work productivity loss was 42 workdays or £2968 less in ETN + MTX treatment group than MTX group under Scenario I, or 31 workdays or £2212 less under Scenario II. When WLQ was used to estimate the percentage work productivity loss at work, total work productivity loss was 37 workdays or £2607 less in the ETN + MTX group than MTX group under Scenario I, or 23 workdays or £1646 less under Scenario II.

Probability of stopping working
Prior to Weeks 12, 24, 36 and 52, the probabilities of the first time work stoppage indicated by patients in MTX treatment group were significantly higher than those in ETN + MTX group (Table 5). At the end of the first year, 24% of the patients in MTX group vs 8.6% in ETN + MTX group stopped working at least once.


View this table:
[in this window]
[in a new window]

  		TABLE 5. Probability of stopping work for the first time by each follow-up visit

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Acknowledgements
 References
 
We compared the impact of treatment with the combination of ETN and MTX to MTX alone on work productivity among MTX-naïve patients with active early RA over 12 months. Our findings show that work productivity loss was significantly lower among patients in the ETN combination treatment group compared with the MTX alone group. Our analysis assumes a societal perspective and, therefore, the improvements in patients’ ability to work translate into a production gain for society. Thus, from a societal perspective, the cost of treatment with ETN is partially offset by productivity gains.

This study confirmed that for patients with early RA, work loss is rapid in patients with highly active disease. At the end of the first year of the COMET study, 24% of the patients in the MTX group and 8.6% in ETN + MTX group stopped working at least once. The fact that half of the first work stoppage occurred in the first 3 months indicated the high risk of work disability for patients with very early RA and the urgency for aggressive treatment. There were significantly fewer work stoppages in the ETN + MTX group compared with the MTX group as early as Week 12. Numerous trials comparing the efficacy of ETN alone or in combination with MTX vs MTX alone have demonstrated that ETN had a more rapid rate of improvement with significant differences observed as early as Week 2 [17, 18, 28–30]. This early onset of treatment effect is likely to have contributed to fewer work stoppages in the ETM + MTX group.

The productivity loss attributable to absenteeism was substantial. The mean number of missed workdays was half in the ETN + MTX group (14 days) compared with that of the MTX group (32 days), despite the high dose of MTX monotherapy. When the number of missed workdays in the MTX group (32 days) is considered in context of 208 possible work days per year in the UK, this amounts to ~15% of the work days or >1 month of work time.

Recently, three trials were conducted to compare the impact of combination biologic therapy with MTX vs MTX monotherapy on work productivity in terms of absenteeism and employment status among MTX-naïve patients with early RA [11, 13, 14]. The biologic therapies investigated were infliximab in the ASPIRE trial [14], and adalimumab in the PREMIER [11] and a trial among the Yorkshire Early Arthritis registry patients [13]. However, the results from these trials were not comparable, mainly because the definition and measurement of work disability and productivity widely differed across studies. For example, simple endpoints such as employment status or job loss were defined differently. The ASPIRE trial measured employability (feeling well enough to be able to work); the PREMIER measured positive change (retain job or gain job) and negative change (lose job or never gain job) of employment status; the trial in the Yorkshire Early Arthritis registry patients measured a combination of job loss and/or imminent job loss; and the COMET trial measured work stoppage. In terms of absenteeism, the COMET trial assessed missed workdays due to health, whereas the PREMIER study assessed missed workdays due to RA. Despite the different ways of defining the outcomes, the results from these trials suggest that compared with MTX monotherapy, combination of a biologic with MTX was more likely to reduce the number of missed workdays and prevent job loss or work disability among patients with early RA. Data from a US observational study also suggest that for patients with shorter disease duration (<11 years), anti-TNF use had a protective effect on primary work disability and RA-attributable work disability [31].

Contrary to the above, an observational study by Wolfe et al. [32] did not discern a positive effect of anti-TNF therapy on the risk of work disability. They found that anti-TNF therapy was not associated with social security disability, but was associated with an increased risk of self-reported disability (odds ratio = 1.6) after adjustment for all RA-related factors. However, according to the editorial accompanying the publication, the study had limitations that prevent one from concluding that anti-TNF therapy does not reduce work disability and unemployment in RA [33]. Given that observational studies can yield different results from randomized controlled trials such as ours and other trials [11, 13, 14], future observational studies and clinical trials are necessary to further evaluate the impact of biologics on work productivity.

Despite the growing number of clinical trials examining the impact of biologic agents and work productivity, most of these trials did not show whether combination therapy could reduce lost days due to presenteeism. Our study is the first attempt to map lost workdays due to presenteeism according to Zhang et al. [20]. In Zhang's study, the WPAI and WLQ, which are validated presenteeism instruments [34], were used to measure the time loss due to presenteeism. According to these two instruments, there seems to be a trend in favour of combination therapy. However, previous studies have shown that the time loss estimates due to presenteeism vary widely according to measurement instrument [35–38], but which instrument gives a valid estimate of time loss due to presenteeism is not clear. Therefore, the lost workdays due to presenteeism estimated in our study were not considered as the primary outcome but as additional supportive information.

Our study has several limitations. First, in the COMET trial, patients were asked about their absenteeism ‘since last visit’, which did not give a clear definition of the recall period. In this study, we assumed that the last visit was the last clinical visit and, thus, the recall period was the prior 4 weeks for visits at Weeks 12 and 24 and prior 8 weeks for visits at Weeks 36 and 52. A maximum of 5 days per week was used to limit the total absenteeism per patient.

Secondly, the data on patients’ normal workweek and occupation were not collected. We assumed a 5-day workweek and 208 days per year and used the national median daily wage as the proxy of the cost for 1 day of work loss, which potentially reduces the variability.

Thirdly, since in COMET trial the exact time of work stoppage/resumption was not collected, we could not obtain a precise estimate on the number of stopped workdays and thus total absenteeism. However, we estimated the possible maximum and minimum stopped workdays by assuming that: (i) once a patient reported stopping work, he or she would not restart working and (ii) patients could resume working after a period of work stoppage. Actual total absenteeism is probably somewhere between our maximum and minimum estimate. Fourthly, the study did not collect information on education and job characteristics, which de Buck et al. [39] found to be associated with work productivity loss. Nonetheless, given that our study was a randomized controlled trial, these unmeasured characteristics are likely to be similar between the two groups.

Finally, it should be noted that the study questionnaire asked patients about loss of work productivity that was health related and not specifically RA related. Although Young et al. [40] have distinguished work disability due to RA from that due to other reasons, such as age and personal or social reasons, RA plus comorbidity and RA plus other reasons, within the context of patient self-reports in a clinical trial, it may be more appropriate to use the concept of health-related work productivity loss. Distinguishing reasons of work productivity loss in such a detail as in Young et al.'s study [40], requires patients to be able to distinguish between the impact of comorbidities, side-effects and/or toxicities. For example, if a patient was absent from work due to RA treatment drug-related toxicity, the missed workdays might not be correctly attributed to RA. Productivity loss resulting from RA-related fatigue may not also be attributed to RA in some patients. By asking for work productivity loss due to health, this implicitly excludes productivity loss due to age, personal or social reasons while still capturing all positive and negative potential treatment effects. Nonetheless, even with an adequate sample size and a randomized controlled design, our findings may still be influenced by outliers reporting unrelated work productivity loss (e.g. other illnesses, car accidents).

In conclusion, our results show that early treatment with ETN + MTX led to an attenuation of absenteeism in terms of sick days and work stoppage among patients with early active RA. From a societal perspective, the associated productivity gain attributable to ETN was substantial. These productivity gains represent benefit beyond the traditional measures of clinical and radiographic improvements. Further research is warranted to examine the robustness of association between early treatment and work-related outcomes. Our sensitivity analysis suggests that fewer lost workdays due to presenteeism can also be attributed to treatment with ETN + MTX. However, in order to obtain a direct estimate of time loss due to presenteeism, further studies, explicitly measuring presenteeism, should be conducted.

Formula


   Acknowledgements
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
References
 
The authors wish to thank the investigators, their staff and the subjects involved in the trial. W.Z. is a recipient of a Canadian Institutes of Health Research Doctoral Research Award in the Area of Public Health Research and a Canadian Arthritis Network Graduate Award.

Funding: This work was supported by Wyeth.

Disclosure statement: R.S. is an employee of Wyeth who manufactures etanercept. A.A. has received research funding and/or honoraria from Wyeth, Schering Plough and Abbott, all of whom manufacture biologics. W.Z. and H.S. are partially supported by Wyeth. A.S. is an employee of Wyeth and owns stocks/shares in Wyeth. P.E. is a consultant for and has received honoraria from Wyeth. B.F. is a Vice President of Medical Affairs at Wyeth and owns stocks/shares in Wyeth.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
 References
 

  1. Lee DM, Weinblatt ME. Rheumatoid arthritis [see comment]. Lancet (2001) 358:903–11.[CrossRef][Web of Science][Medline]
  2. Gabriel SE. The epidemiology of rheumatoid arthritis. Rheum Dis Clin North Am (2001) 27:269–81.[CrossRef][Web of Science][Medline]
  3. Burton W, Morrison A, Maclean R, Ruderman E. Systematic review of studies of productivity loss due to rheumatoid arthritis. Occup Med (2006) 56:18–27.[Abstract/Free Full Text]
  4. Li X, Gignac MA, Anis AH. The indirect costs of arthritis resulting from unemployment, reduced performance, and occupational changes while at work. Med Care (2006) 44:304–10.[CrossRef][Web of Science][Medline]
  5. Lundkvist J, Kastang F, Kobelt G. The burden of rheumatoid arthritis and access to treatment: health burden and costs. Eur J Health Econ (2008) 8(Suppl. 2):S49–60.[CrossRef][Web of Science][Medline]
  6. Xie F. The need for standardization: a literature review of indirect costs of rheumatoid arthritis and osteoarthritis. Arthritis Rheum (2008) 59:1027–33.[CrossRef][Web of Science][Medline]
  7. Merkesdal S, Ruof J, Schoffski O, Bernitt K, Zeidler H, Mau W. Indirect medical costs in early rheumatoid arthritis: composition of and changes in indirect costs within the first three years of disease. Arthritis Rheum (2001) 44:528–34.[CrossRef][Web of Science][Medline]
  8. Eberhardt K, Larsson BM, Nived K, Lindqvist E. Work disability in rheumatoid arthritis–development over 15 years and evaluation of predictive factors over time. J Rheumatol (2007) 34:481–7.[Abstract/Free Full Text]
  9. ten Klooster PM, Veehof MM, Taal E, van Riel PL, van de Laar MA. Changes in priorities for improvement in patients with rheumatoid arthritis during 1 year of anti-tumour necrosis factor treatment. Ann Rheum Dis (2007) 66:1485–90.[Abstract/Free Full Text]
  10. Puolakka K, Kautiainen H, Mottonen T, et al. Impact of initial aggressive drug treatment with a combination of disease-modifying antirheumatic drugs on the development of work disability in early rheumatoid arthritis: a five-year randomized follow up trial. Arthritis Rheum (2004) 50:55–62.[CrossRef][Web of Science][Medline]
  11. van Vollenhoven RF, Farraccioli GF, Breedveld FC, Ray S, Cifaldi M. Effect of adalimumab combination therapy on work performance: results from a companion study to PREMIER [abstract]. Arthritis Rheum (2007) 9:S88.[CrossRef]
  12. Kimel M, Cifaldi M, Chen N, Revicki D. Adalimumab plus methotrexate improved SF-36 scores and reduced the effect of rheumatoid arthritis (RA) on work activity for patients with early RA. J Rheumatol (2008) 35:206–15.[Abstract/Free Full Text]
  13. Bejarano V, Quinn M, Conaghan PG, et al. Effect of the early use of the anti-tumor necrosis factor adalimumab on the prevention of job loss in patients with early rheumatoid arthritis. Arthritis Rheum (2008) 59:1467–74.[CrossRef][Web of Science][Medline]
  14. Smolen JS, Han C, van der Heijde D, et al. Infliximab treatment maintains employability in patients with early rheumatoid arthritis. Arthritis Rheum (2006) 54:716–22.[CrossRef][Web of Science][Medline]
  15. Han C, Smolen J, Kavanaugh A, St Clair EW, Baker D, Bala M. Comparison of employability outcomes among patients with early or long-standing rheumatoid arthritis. Arthritis Rheum (2008) 59:510–14.[CrossRef][Web of Science][Medline]
  16. Yelin E, Trupin L, Katz P, Lubeck D, Rush S, Wanke L. Association between etanercept use and employment outcomes among patients with rheumatoid arthritis. Arthritis Rheum (2003) 48:3046–54.[CrossRef][Web of Science][Medline]
  17. Emery P, Breedveld FC, Hall S, et al. Comparison of methotrexate monotherapy with a combination of methotrexate and etanercept in active, early, moderate to severe rheumatoid arthritis (COMET): a randomised, double-blind, parallel treatment trial. Lancet (2008) 372:375–82.[CrossRef][Web of Science][Medline]
  18. Kremer JM. COMET's path, and the new biologicals in rheumatoid arthritis. Lancet (2008) 372:347–8.[CrossRef][Web of Science][Medline]
  19. OECD. Average annual hours actually worked per worker dataset, OECD.Stat, Paris, France. (July 24, 2009, date last accessed). http://stats.oecd.org/wbos/Index.aspx?DatasetCode=ANHRS.
  20. Zhang W, Bansback N, Beaton D, et al. How is reduced performance at work (presenteeism) associated with measures of disease severity in patients with osteoarthritis (OA) and rheumatoid arthritis (RA)? [abstract]. Ann Rheum Dis (2008) 67(Suppl. II):583.
  21. Reilly MC, Zbrozek AS, Dukes EM. The validity and reproducibility of a work productivity and activity impairment instrument. Pharmacoeconomics (1993) 4:353–65.[Web of Science][Medline]
  22. Reilly Associates Health Outcomes Research. (July 24, 2009, date last accessed). http://www.reillyassociates.net.
  23. Lerner D, Amick BC 3rd, Rogers WH, Malspeis S, Bungay K, Cynn D. The Work Limitations Questionnaire. Med Care (2001) 39:72–85.[CrossRef][Web of Science][Medline]
  24. Lerner D, Reed JI, Massarotti E, Wester LM, Burke TA. The Work Limitations Questionnaire's validity and reliability among patients with osteoarthritis. J Clin Epidemiol (2002) 55:197–208.[CrossRef][Web of Science][Medline]
  25. Fries JF, Spitz P, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis Rheum (1980) 23:137–45.[Web of Science][Medline]
  26. Bruce B, Fries JF. The Stanford Health Assessment Questionnaire: a review of its history, issues, progress, and documentation. J Rheumatol (2003) 30:167–78.[Abstract/Free Full Text]
  27. National Statistics. 2007 annual survey of hours and earnings, National Statistics, London, UK. (July 24, 2009, date last accessed). www.statistics.gov.uk/StatBase/Product.asp?vlnk=13101.
  28. Klareskog L, van der Heijde D, de Jager JP, et al. Therapeutic effect of the combination of etanercept and methotrexate compared with each treatment alone in patients with rheumatoid arthritis: double-blind randomised controlled trial. Lancet (2004) 363:675–81.[CrossRef][Web of Science][Medline]
  29. Bathon JM, Martin RW, Fleischmann RM, et al. A comparison of etanercept and methotrexate in patients with early rheumatoid arthritis. N Engl J Med (2000) 343:1586–93.[Abstract/Free Full Text]
  30. Weinblatt ME, Kremer JM, Bankhurst AD, et al. A trial of etanercept, a recombinant tumor necrosis factor receptor:Fc fusion protein, in patients with rheumatoid arthritis receiving methotrexate. N Engl J Med (1999) 340:253–9.[Abstract/Free Full Text]
  31. Allaire S, Wolfe F, Niu J, Zhang Y, Zhang B, LaValley M. Evaluation of the effect of anti-tumor necrosis factor agent use on rheumatoid arthritis work disability: the jury is still out. Arthritis Rheum (2008) 59:1082–9.[CrossRef][Web of Science][Medline]
  32. Wolfe F, Allaire S, Michaud K. The prevalence and incidence of work disability in rheumatoid arthritis, and the effect of anti-tumor necrosis factor on work disability. J Rheumatol (2007) 34:2211–17.[Abstract/Free Full Text]
  33. Verstappen SM, Jacobs JW, Hyrich KL. Effect of anti-tumor necrosis factor on work disability. J Rheumatol (2007) 34:2126–8.[Free Full Text]
  34. Escorpizo R, Bombardier C, Boonen A, et al. Worker productivity outcome measures in arthritis. J Rheumatol (2007) 34:1372–80.[Abstract/Free Full Text]
  35. Brouwer WB, Koopmanschap MA, Rutten FF. Productivity losses without absence: measurement validation and empirical evidence. Health Policy (1999) 48:13–27.[CrossRef][Web of Science][Medline]
  36. Meerding WJ, Ijzelenberg W, Koopmanschap MA, Severens JL, Burdorf A. Health problems lead to considerable productivity loss at work among workers with high physical load jobs. J Clin Epidemiol (2005) 58:517–23.[CrossRef][Web of Science][Medline]
  37. Ozminkowski RJ, Goetzel RZ, Chang S, Long S. The application of two health and productivity instruments at a large employer. J Occup Environ Med (2004) 46:635–48.[CrossRef][Web of Science][Medline]
  38. Zhang W, Lacaille D, Beaton D, et al. Productivity losses at work among clinical patients with arthritis: comparisons of measurement instruments. In: In: Canadian Arthritis Network's 2007 Annual Scientific Conference. Halifax, NS. Abstract No. 36 and poster.
  39. de Buck PD, de Bock GH, van Dijk F, van den Hout WB, Vandenbroucke JP, Vliet Vlieland TP. Sick leave as a predictor of job loss in patients with chronic arthritis. Int Arch Occup Environ Health (2006) 80:160–70.[CrossRef][Web of Science][Medline]
  40. Young A, Dixey J, Kulinskaya E, et al. Which patients stop working because of rheumatoid arthritis? Results of five years’ follow up in 732 patients from the Early RA Study (ERAS). Ann Rheum Dis (2002) 61:335–40.[Abstract/Free Full Text]
Submitted 22 January 2009; revised version accepted 9 July 2009.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
48/10/1283    most recent
kep239v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Anis, A.
Right arrow Articles by Sato, R.
PubMed
Right arrow PubMed Citation
Right arrow Articles by Anis, A.
Right arrow Articles by Sato, R.
Related Collections
Right arrow Pharmacology
Right arrow Rheumatoid Arthritis
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?