The most widely used domains for the assessment of therapeutic efficacy in ankylosing spondylitis (AS) include patient global, pain, function, fatigue and inflammatory activity.1 A recently proposed composite score includes back pain, patient global, peripheral pain/swelling, duration of morning stiffness, fatigue and acute phase reactants.2 The selection of patient-reported outcomes in trials and clinical care should be consistent with outcomes that are important to patients,3 because health professionals and patients may have diverging opinions.4

The aim of this post-hoc exploratory analysis was to examine priorities for improvement in dimensions of health across inflammatory rheumatic diseases, within the setting of the Norwegian disease-modifying antirheumatic drugs (NOR–DMARD) register.5 The study was approved by the regional ethical committee and all patients signed written informed consent. This analysis focused on 2138 consecutive adult patients with inflammatory arthropathies (table 1) who responded to a seven-item questionnaire on priorities for improvement: (‘Below you see a list of 7 frequent health problems in patients with rheumatic diseases. Please select the 3 problems in which you consider it most important to obtain improvement. Rank these problems from 1 to 3 according to importance’). The listed health problems were pain, morning stiffness, physical functioning, fatigue, sleep, social function and emotional wellbeing.

The reported frequency of priorities for improvement for each health problem was compared across diseases. Pain was most frequently perceived as the top priority for improvement, and was one of three most important problems for improvement in between 80% and 90% of all patients (table 2). Patients with AS reported priority for improvement in sleep problems significantly more frequently than the other diagnostic groups, both as top priority and among the three prioritised dimensions (p=0.007/p<0.001). A similar finding was reported for fatigue among patients with juvenile inflammatory arthritis. Otherwise, priorities for improvement appeared to be similar across diagnoses (table 2).

Sleep disturbances have been reported to be more frequent in both rheumatoid arthritis (RA) and AS than in the normal population.6 7 The most important observation in the current analysis was that the priority for improvement in sleep was significantly different between groups (table 2), with significantly higher priority in AS patients compared with both RA patients (χ² p=0.001/p=0.002 for top one/top three priority, respectively) and psoriatic arthritis (PsA) patients (p<0.001/p=0.001). Findings were consistent within each gender (data not shown). Logistic regression analyses revealed that OR (95% CI) to report priority for improvement in sleep was 2.5 (1.7 to 3.6) for AS versus RA and 2.7 (1.9 to 4.0) for AS versus PsA after adjustment for age, gender, disease duration, patient global, erythrocyte sedimentation rate, MHAQ score and SF-6D.

Pain was as expected the most important dimension for improvement. The high priority of fatigue among all patient groups is consistent with increasing awareness of fatigue as an important patient-reported outcome that should be reported in clinical trials.8

Sleep disturbance has been shown to be a responsive measure in RA9 and AS.10 This study highlights the fact that sleep is considered a prioritised dimension for improvement, more frequently by patients with AS than by patients with other inflammatory arthropathies. These data encourage the assessment of sleep disturbances in AS.