Nonsteroidal anti-inflammatory drug–sparing effects. (A) Change in ASAS-NSAID score from baseline to week 8 in patients in the etanercept and placebo groups (primary analysis of primary endpoint). Analysis of covariance, LOCF, with imputation of missing diary data, in the ITT population. (B) Proportion of patients in the etanercept and placebo groups achieving other NSAID-sparing endpoints at week 8 of the double-blind period. LOCF, with imputation, ITT population (n = number of patients achieving endpoint; N = number of patients with analyzable data). (C) Mean ASAS-NSAID scores (± SD) in the etanercept/etanercept and placebo/etanercept groups in the double-blind and open-label phases. ASAS-NSAID scores were calculated for observed cases, with no imputation of missing diary data. ASAS-NSAID, Assessment of SpondyloArthritis International Society nonsteroidal anti-inflammatory drug use; ETN, etanercept; ITT, intent-to-treat; LOCF, last observation carried forward; NSAID, nonsteroidal anti-inflammatory drug; PBO, placebo; SD, standard deviation; SE, standard error.