Comparison of employability outcomes among patients with early or long-standing rheumatoid arthritis

Authors

  • Chenglong Han,

    Corresponding author
    1. Centocor Research and Development, Malvern, Pennsylvania
    • Centocor Research and Development, 200 Great Valley Parkway, Malvern, PA 19355
    Search for more papers by this author
    • Drs. Han, Baker, and Bala own stock and hold stock options in Johnson & Johnson. Dr. Smolen has received honoraria (less than $10,000 each) from Abbott, Centocor, Roche, Wyeth, Bristol-Meyers Squibb, and Schering-Plough. Dr. St.Clair has received consultant fees (less than $10,000 each) from Centocor, Genentech, Medimmune, and Xoma, and has received research grants from Genentech and Amgen.

  • Josef Smolen,

    1. Medical University of Vienna and Hietzing Hospital, Vienna, Austria
    Search for more papers by this author
    • Drs. Han, Baker, and Bala own stock and hold stock options in Johnson & Johnson. Dr. Smolen has received honoraria (less than $10,000 each) from Abbott, Centocor, Roche, Wyeth, Bristol-Meyers Squibb, and Schering-Plough. Dr. St.Clair has received consultant fees (less than $10,000 each) from Centocor, Genentech, Medimmune, and Xoma, and has received research grants from Genentech and Amgen.

  • Arthur Kavanaugh,

    1. University of California at San Diego
    Search for more papers by this author
  • E. William St.clair,

    1. Duke University Medical Center, Durham, North Carolina
    Search for more papers by this author
    • Drs. Han, Baker, and Bala own stock and hold stock options in Johnson & Johnson. Dr. Smolen has received honoraria (less than $10,000 each) from Abbott, Centocor, Roche, Wyeth, Bristol-Meyers Squibb, and Schering-Plough. Dr. St.Clair has received consultant fees (less than $10,000 each) from Centocor, Genentech, Medimmune, and Xoma, and has received research grants from Genentech and Amgen.

  • Daniel Baker,

    1. Centocor Research and Development, Malvern, Pennsylvania
    Search for more papers by this author
    • Drs. Han, Baker, and Bala own stock and hold stock options in Johnson & Johnson. Dr. Smolen has received honoraria (less than $10,000 each) from Abbott, Centocor, Roche, Wyeth, Bristol-Meyers Squibb, and Schering-Plough. Dr. St.Clair has received consultant fees (less than $10,000 each) from Centocor, Genentech, Medimmune, and Xoma, and has received research grants from Genentech and Amgen.

  • Mohan Bala

    1. Centocor Research and Development, Malvern, Pennsylvania
    Search for more papers by this author
    • Drs. Han, Baker, and Bala own stock and hold stock options in Johnson & Johnson. Dr. Smolen has received honoraria (less than $10,000 each) from Abbott, Centocor, Roche, Wyeth, Bristol-Meyers Squibb, and Schering-Plough. Dr. St.Clair has received consultant fees (less than $10,000 each) from Centocor, Genentech, Medimmune, and Xoma, and has received research grants from Genentech and Amgen.


Abstract

Objective

To compare employability between patients with early and long-standing rheumatoid arthritis (RA) and examine the relationships between improvement in employability and disease stage after adjustment for demographic characteristics, disease activity, physical functioning, and response to therapy.

Methods

We evaluated the employability data from 2 double-blind, randomized, placebo-controlled studies of infliximab plus methotrexate (MTX) in patients with RA. Patients were incomplete responders to MTX in 1 study and had never taken MTX in the other study. Patients age <65 years were categorized as having early RA (≤3 years disease duration) or long-standing RA (>3 years disease duration). Physical functioning was assessed using the Health Assessment Questionnaire (HAQ) and clinical response was determined based on the American College of Rheumatology 20% improvement criteria (ACR20).

Results

Patients with early RA were more likely to be employable at baseline than those with long-standing RA, even after adjusting for baseline HAQ scores. Among patients who were not employable at baseline but achieved an ACR20 response after 1 year of treatment, after adjusting for baseline HAQ score, the patients with early RA who had never taken MTX were 3 times more likely to become employable compared with those with long-standing RA who had an incomplete response to MTX at baseline.

Conclusion

In 2 clinical trials, patients with early RA were more likely to show improved employment outcomes after treatment than those with long-standing RA, suggesting intervention as early as possible in the disease course maximizes an individual patient's employment potential.

INTRODUCTION

The loss of productivity associated with rheumatoid arthritis (RA) disability places significant burden on patients, their families, and society as a whole. This negative impact can be observed even in patients with early disease (1–3). Given that loss of working capacity constitutes an important outcome of RA, the effectiveness of treatment strategies should be evaluated not only using traditional clinical, functional, and radiographic measures, but also by determining how well the patients receiving them are able to maintain employability.

The purpose of the present analysis was to compare employability between early and long-standing RA and examine the relationships between improvement in employability and disease stage after adjustment for demographic characteristics, disease activity, physical functioning, and response to therapy. We analyzed the employability of patients with RA from 2 large clinical trials designed to investigate the efficacy and safety of infliximab, a tumor necrosis factor α inhibitor, with concomitant methotrexate (MTX) therapy.

PATIENTS AND METHODS

The study population consisted of patients from the Anti–Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy (ATTRACT) and the Active-Controlled Study of Patients Receiving Infliximab for the Treatment of Rheumatoid Arthritis of Early Onset (ASPIRE). In ATTRACT, eligible patients had active RA, defined as ≥6 tender and swollen joints and ≥2 of the following: morning stiffness for ≥45 minutes, an erythrocyte sedimentation rate ≥28 mm/hour, or a C-reactive protein (CRP) level ≥2 mg/dl, despite receiving stable dosages of MTX (≥12.5 mg/week). Patients were randomly assigned to receive placebo or infliximab 3 or 10 mg/kg every 4 or 8 weeks after a 3-dose induction regimen (n = 428). All patients received concomitant MTX (≤25 mg/week) throughout the study. Most of the patients in this trial had long-standing disease, with 81% of patients having a disease duration >3 years. The trial was conducted at study sites in Austria, Canada, Denmark, France, Germany, The Netherlands, Sweden, the UK, and the US. Employability data from this trial have been previously analyzed (13). Patients from this trial will hereafter be referred to as the MTX-incomplete responders cohort.

Patients were eligible to participate in ASPIRE if they had never taken MTX and had persistent synovitis for ≥3 months but not >3 years, had ≥10 swollen joints, and had ≥12 tender joints. In addition, patients must have had radiographic erosions of the hands or feet, a positive test for serum rheumatoid factor, or a CRP value ≥2.0 mg/dl. Patients were randomly assigned to receive placebo or infliximab 3 or 6 mg/kg every 8 weeks through week 46 after a 3-dose induction regimen (n = 1,004). Patients also received concomitant MTX (≤25 mg/week) throughout the study. The trial was conducted at study sites in Austria, Belgium, Canada, France, Ireland, Israel, Italy, Germany, The Netherlands, Norway, Spain, Sweden, the UK, and the US. Employability data from the ASPIRE trial have been analyzed previously (3). Patients from this trial will hereafter be referred to as the MTX-naive cohort.

In both studies, patients reported their employment status by answering yes or no to the questions, “Are you currently actively employed?” and “If not, do you feel well enough to work if a job were available?” Each patient was categorized as unemployable if the patient was unemployed and felt unable to work even if a job were available or as employable if the patient was employed or felt well enough to work if a job were available. Physical functioning was assessed using the Health Assessment Questionnaire (HAQ) (4). Clinical response was determined based on the American College of Rheumatology 20% improvement criteria (ACR20) (5). The Disease Activity Score in 28 joints (DAS28) was used to evaluate disease activity at baseline (6–8). Radiographic damage was assessed using the van der Heijde modification of the Sharp score (SHS) (9). All patients in both studies were age ≥18 years according to the entry criteria. Employability analyses were based on data from the subset of patients who were age <65 years and had evaluable employment data at baseline. Early RA was defined as a disease duration ≤3 years; all patients in ASPIRE and 19% of patients in ATTRACT met the criterion for early RA. For patients with missing data during the study, the last observation was carried forward in the analysis.

A multivariate logistic regression analysis was used to assess the relationship between baseline employability and disease stage (i.e., MTX-naive patients with early RA versus MTX-incomplete responders with early RA or MTX-incomplete responders with long-standing RA) after adjustment for age, sex, physical functioning, and disease activity. Multivariate logistic regression was also used to assess the relationship between disease stage and change of employment status after 1 year of treatment by adjusting for age, sex, and baseline HAQ score separately among ACR20 responders and nonresponders.

RESULTS

Patient characteristics.

Although baseline tender and swollen joint counts were similar among all groups, MTX-incomplete responders had longer disease duration, greater HAQ scores, greater SHS, and greater prior use of disease-modifying antirheumatic drugs (DMARDs) compared with MTX-naive patients (Table 1). MTX-incomplete responders with early RA had a mean disease duration, mean SHS, and number of prior DMARDs that were between MTX-naive patients and MTX-incomplete responders with long-standing RA.

Table 1. Characteristics of study populations and change in employment status at week 54 by baseline employment status*
AssessmentMTX-naive patients with early RAMTX-incomplete responders
Early RALong-standing RA
  • *

    MTX = methotrexate; RA = rheumatoid arthritis; HAQ = Health Assessment Questionnaire; SHS = Sharp score as modified by van der Heijde; DMARDs = disease-modifying antirheumatic drugs.

  • P < 0.01, compared with MTX-naive patients with early RA.

  • P < 0.05, compared with MTX-naive patients with early RA.

No. patients randomized1,00482346
Patients age <65 years, no. (%)856 (85)73 (89)293 (85)
Age, mean ± SD years47.1 ± 10.747.0 ± 9.750.1 ± 9.9
Women, no. (%)615 (71.9)56 (76.7)228 (77.8)
Disease duration, mean ± SD years0.9 ± 0.71.7 ± 0.711.8 ± 7.4
Tender joint count, mean ± SD33.3 ± 14.631.2 ± 15.932.1 ± 16.5
Swollen joint count, mean ± SD21.6 ± 10.621.0 ± 11.322.7 ± 11.7
HAQ score, mean ± SD (range 0–3)1.5 ± 0.61.7 ± 0.61.7 ± 0.6
SHS, mean ± SD10.6 ± 14.926.6 ± 29.482.6 ± 70.9
No. prior DMARDs, mean ± SD0.4 ± 0.612.1 ± 1.183.6 ± 1.60
Patients evaluated for employment status, no.85673293
Employable at baseline, no.56945138
 Employable at week 54, no. (%)509 (89)40 (89)126 (91)
Employed at baseline, no.54242122
 Employed at week 54, no. (%)490 (90)39 (93)113 (93)
Unemployable at baseline, no.28127152
 Employable at week 54, no. (%)108 (38)4 (15)22 (14)
Unemployed at baseline, no.30831169
 Employed at week 54, no. (%)44 (14)2 (6)10 (6)

Factors associated with employability and actual employment at baseline.

Patients with early RA were significantly more likely to be employable and actually employed than those with long-standing RA (Table 1). Among patients with early RA, 66% of MTX-naive patients and 62% of MTX-incomplete responders were considered to be employable at baseline compared with 47% of MTX-incomplete responders with long-standing RA (P < 0.01).

The results of a logistic regression model showed that disease stage (MTX-naive patients with early RA versus MTX-incomplete responders with long-standing RA) was significantly associated with baseline employability (P < 0.001; odds ratio [OR] 1.75) after adjusting for age, sex, baseline HAQ score, and DAS28. Age (P < 0.001; OR 0.97 per year of age) and baseline physical functioning (P < 0.001; OR 0.28 for HAQ score) were also significantly associated with baseline employability in the logistic regression model. Similar results were observed when actual employment status was used in the model.

Effect of treatment on employability and actual employment.

After treatment at week 54, the majority of patients who were employed or employable at baseline maintained their employment status across 3 groups (P > 0.5) (Table 1). However, among patients who were unemployed or unemployable at baseline, a greater percentage of MTX-naive patients with early RA changed their employment status from unemployable at baseline to employable at week 54 or from unemployed at baseline to employed at week 54 compared with MTX-incomplete responders with long-standing RA (P < 0.01) (Table 1).

Therefore, the factors associated with improvement in employability after the treatment of patients who were unemployable or unemployed at baseline were further evaluated using a separate logistic regression model. After adjustment for age, sex, baseline physical functioning, and clinical response to treatment (ACR20), MTX-naive patients with early RA were found to be more likely employable than MTX-incomplete responders with long-standing RA at week 54 (P < 0.001; OR 3.23), among patients who were unemployable at baseline. Baseline HAQ score (P < 0.001; OR 0.31) and response to treatment (P < 0.001; OR 4.97) were also associated with employability. Similar correlations were observed when examining actual employment data. Also noteworthy is the finding that MTX-naive patients with early RA (disease duration 0.9 year) had significantly greater improvement in employability than MTX-incomplete responders with early RA (disease duration 1.7 years) (P = 0.040; OR 3.45).

Among patients who were unemployable at baseline and had an ACR20 response after 54 weeks of treatment, 87 (53%) of 165 MTX-naive patients with early RA became employable by week 54 compared with 15 (22%) of 67 MTX-incomplete responders with long-standing RA (P < 0.01; OR 4.0) (Figure 1). Although the number of MTX-incomplete responders with early RA who were unemployable at baseline was small, 4 (36%) of 11 patients became employable by week 54 (P > 0.05; OR 1.5, compared with patients with long-standing RA from the same cohort).

Figure 1.

Odds ratios (95% confidence intervals) for A, becoming employable or B, actually being employed at week 54 among patients who were unemployable or unemployed at baseline and achieved an American College of Rheumatology 20% improvement criteria (ACR20) response at week 54. Odds ratios were adjusted for age, sex, and baseline Health Assessment Questionnaire score using logistic regression. Data from the infliximab plus methotrexate (MTX) and placebo plus MTX treatment groups were included. RA = rheumatoid arthritis.

The results for actual employment mirrored the results for employability (Figure 1). Among patients who were unemployed at baseline and had an ACR20 response after 54 weeks of treatment, 33 (18%) of 180 MTX-naive patients with early RA became employed by week 54 compared with 5 (7%) of 74 MTX-incomplete responders with long-standing RA (P < 0.05; OR 2.7).

A similar trend was observed among ACR20 nonresponders when comparing employability and actual employment between MTX-naive patients with early RA and MTX-incomplete responders with long-standing RA. Twenty-one (18.6%) MTX-naive patients with early RA and 7 (8.2%) MTX-incomplete responders with long-standing RA were ACR20 nonresponders who became employable at week 54. Similarly, 11 (8.9%) and 5 (5.3%) patients, respectively, were ACR20 nonresponders who became actually employed, but the differences between the 2 groups were not statistically significant after adjustment for age, sex, and baseline HAQ score (P < 0.05). The number of MTX-incomplete responders with early RA who became employable or employed was too small to include in the analysis.

DISCUSSION

We found substantial work disability in patients with both early and advanced disease, based on baseline data from 2 large clinical trials. Although the overall results were consistent with findings from previous studies (1, 2), a key finding of the present analysis was the substantial difference between patients with early and long-standing disease in the improvement of employability after responding to treatment. MTX-naive treatment responders with early RA were 4 times more likely to become employable than those who were initially MTX-incomplete responders with long-standing RA but attained response after 1 year of therapy with infliximab plus MTX or placebo plus MTX. We also observed a trend demonstrating that MTX-naive patients with early RA (mean disease duration 0.9 years) who responded to treatment after 1 year were more likely to become employable than those who were initially MTX-incomplete responders with early RA but who later attained response after 1 year of study treatment (mean disease duration 1.7 years), emphasizing the importance of early effective treatment intervention. This finding is consistent with the results of the Finnish Rheumatoid Arthritis Combination Therapy trial (10), in which ACR20 response at 6 months predicted rates of work disability and loss of productivity.

The differential effect of treatment on employability in the 2 trial populations may relate to the differences in the magnitude of structural damage at baseline. Patients with long-standing RA had ∼8 times more extensive structural damage at baseline compared with the MTX-naive cohort. Assuming that severe structural damage leads to high degrees of irreversible disability (11, 12), regaining employment would have been difficult for these patients, even after receiving therapy that reduces disease activity. These findings are further supported by a previous analysis from ATTRACT (13), revealing a threshold of ∼50 SHS units, above which the likelihood of full-time employment was low. These results provided employment-related evidence supporting the window of opportunity concept of RA treatment (14, 15). Patients with more advanced disease may be limited in their ability to regain employment even if they clinically respond to therapy; specifically, they may have lost the potential for the reversibility of unemployability.

Given the differences in economics and social services between different countries, it is possible that differences in employability could be influenced by region. Therefore, we conducted a sensitivity analysis in which we included region (US versus Europe, Canada, and Israel combined) as an independent variable in the model. The results of the model after adjustment for region were similar to those before adjustment.

In the present study, we presented both employability and actual employment data to provide a complete assessment of the employment-related outcomes of patients in these studies. Employability is a subjective patient self-assessment of whether the patients are actually employed or feel able to work when a job is available. Theoretically, patients who were employable but not currently employed may have subsequently gained actual employment or may have reduced indirect costs from caregivers. These concepts have not yet been fully validated. However, of the patients in this study who were employable but not actually employed at baseline (n = 46), 20% became actually employed at week 54 compared with 10% of patients who were unemployable at baseline and became actually employed at week 54. These preliminary results support our hypothesis.

Although actual employment may be a more objective measure than employability, it also has important limitations that must be considered. When unemployable patients become employable, it can take an extended period of time before they can actually find a job and reenter the work force. Others such as homemakers may choose not to reenter the work force immediately, even if they could. Moreover, evaluating the change in employment status only in those patients who are actively employed will ignore the loss of employability among patients who are able to maintain a job but are not actively employed. Using employability rather than actual employment allowed us to examine the potential impact of the disease and its effective treatment on this outcome over the short term in the absence of nondisease factors that can influence a patient's ability to gain employment. The employability concept has also been successfully used to evaluate employment outcomes in other disease states (16, 17). In our study, we found that actual employment paralleled employability, adding to the validity of our conclusions.

It is important to note that effective treatment of patients with a chronic disease such as RA may only postpone the onset of disability. In a long-term study, one could evaluate the amount of time disability is delayed after effective treatment. However, such an analysis is not feasible in a 1-year study.

In summary, our results herein suggest that early intensive intervention in RA can favorably impact employability. Further studies will be required to determine if such benefits can be sustained over a longer period of time and significantly reduce the overall costs of this disease.

AUTHOR CONTRIBUTIONS

Dr. Han had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Han, Smolen, Kavanaugh, Baker, Bala.

Acquisition of data. Smolen, Kavanaugh, St.Clair.

Analysis and interpretation of data. Han, Smolen, Kavanaugh, St.Clair, Baker, Bala.

Manuscript preparation. Han, Smolen, Kavanaugh, St.Clair, Baker, Bala.

Statistical analysis. Han, Bala.

ROLE OF THE STUDY SPONSOR

This study reports the results of analyses conducted using data from the ASPIRE and ATTRACT clinical studies. These studies were funded by Centocor Research and Development, Inc. Drs. Han, Baker, and Bala were employees of Centocor when this analysis was conducted, and all 3 were involved in the design of the analysis. Drs. Han and Bala conducted the statistical analysis. All authors, including Drs. Han, Baker, and Bala, contributed to the manuscript development, reviewed the final manuscript before submission, and approved the submission of the manuscript.

Acknowledgements

The authors would like to thank the investigators, patients, and study personnel who made the ATTRACT and ASPIRE studies possible. The authors also acknowledge Scott Newcomer, MS, of Centocor for his assistance in preparing the manuscript.

Ancillary