Study | Intervention | Comparator | Outcome assessment | Statistical analysis | Results |
---|---|---|---|---|---|
Authors Study design Country | Ntotal (Nt), Nemployed (Ne) (♂: %); Age, mean (SD); Disease duration (Dd), mean (SD) | Ntotal (Nt), Nemployed (Ne) (♂: %); Age, mean (SD); Disease duration (Dd), mean (SD) | Domain (outcome measure) Recall period and disease attribution Assessment interval and follow-up | Method Confounders | Treatment versus comparator Corrected for confounders if any |
Barkham18 Double-blind, placebo-controlled, single- centre trial UK | Etanercept + DMARD continuing DMARD failures Nt=Ne=20 (75.0% male) Age: 40.8 (9.7) Dd: 11.0 (range 2.08–45) | Placebo + DMARD continuing DMARD failures Nt=Ne=20 (85.0% male) Age: 39.4 (10.1) Dd: 20 (range 0.58–30) | Presenteeism (AS-WIS*); 1. Work instability: change in score for work instability (group level). 2. Risk for job loss: change in category of the score for work instability (individual level). Absenteeism; proportion of patients losing at least some work hours owing to arthritis (% absent in study period). AS-WIS Work diary, self-report; At this moment, AS attributed; Assessment interval: 12 weeks Follow-up: 12 weeks | Method: ITT-LOCF; ANCOVA Mann–Whitney U test; comparison AS-WIS category between groups Presenteeism 1. Mean AS-WIS score 2. Change in AS-WIS category Absenteeism Proportion of patients losing working hours (χ2 analysis). Confounders: ‘baseline values’ | Presenteeism 1. Work instability score – ETN: Δ −2.75 points (95% CI −4.63 to −0.87) – PLB: Δ −0.68 points (95% CI −2.56 to 1.21) – Δ F=2.47, p=0.1092. Risk for job loss (ETN vs PLB)· ETN: 11 patients had decreased risk, 1 patient had increased risk – PLB: 7 patients had decreased risk, 3 patients had increased risk – Δ Mann–Whitney U=153, p=0.160 Absenteeism
|
Van der Heijde19 Double-blind, placebo-controlled, multicentre trial (ASSERT) USA, Canada, Europe | Infliximab NSAID failures, anti-TNF naive Nt=201 (78% male) Ne=123 (61%) Ne full-time=94 (49%) Age: 39.6±10.6 Dd: median 7.7 (IQR 3.3, 14.9 | Placebo NSAID failures, anti-TNF naive Nt=78 (87% male) Ne =41 (53%) Ne full-time=28 (38%) Age: 40±9.4 Dd: median 13.2 (IQR 3.7, 17.9) | Presenteeism: ‘How much has your disease affected your daily productivity in the past 6 weeks?’ (VAS 0–10) Absenteeism; 1. Work days missed 6 weeks before baseline and week 24, compared with baseline 2. Mean days of work lost during study. Self-report; past 6 weeks; not AS attributed Assessment interval: weeks 2, 6 and every 6 weeks through week 24 Follow-up: 24 weeks | Method: Presenteeism: productivity score change at week 24 compared with baseline (McNemar's test) Absenteeism 1. Comparison work days missed by full-time employed patients 6 weeks before week 24 and baseline (paired t test). 2. Comparison of mean days of work lost during trial between INF/PLB (t test). Confounders: none | Presenteeism
Absenteeism 1. Work days missed, compared with baseline – INF (n=94): Δ −1.0 days (59%) per 6 weeks, (p<0.01) – PLB (N=28): Δ −0.5 days (29%) per 6 weeks (p=0.60)2. Mean days of work lost – INF (n=94): 2.8 days (SD 13.3) per 24 weeks – PLB (N=28): 6.8 days (SD 17.6) per 24 weeks – Δ −4.0 days per 24 weeks (p=0.07), INF compared with PLB |
*AS-WIS (validated) with a score of 0–20, where <11 indicates low risk of job loss, 11–18 indicates medium risk, 19–20 indicates high risk.
AS, ankylosing spondylitis; ANCOVA, analysis of covariance; ASSERT, Ankylosing Spondylitis Study for the Evaluation of Recombinant Infliximab Therapy; AS-WIS, Ankylosing Spondylitis Work Instability Scale; DMARD, disease modifying antirheumatic drug; ETN, etanercept; INF, infliximab; ITT-LOCF, intention to treat-analysis with last observation carried forward for missing values; NSAID, non-steroidal anti-inflammatory drug; PLB, placebo; TNF, tumour necrosis factor; VAS, Visual Analogue Scale.