Background/Purpose: The minimum clinically important improvement (MCII) and patient acceptable symptom state (PASS) are clinically relevant measures that report the patient response and condition. We aimed at estimating the MCII and PASS cut-off values for BASDAI and BASFI in patients with Ankylosing Spondylitis (AS).

Methods: A multinational study including patients with AS receiving NSAIDs for 4-weeks has been used to define PASS and MCII, by using external anchor questions for patient global assessment for BASDAI and functional impairment for BASFI. For PASS, patients were asked to consider the ways AS has affected them during the last 48 hours and if this would be an acceptable state for the rest of their life. Patients who answered “acceptable” met criteria for PASS analysis. For MCII, patients were asked to compare how they felt in the past 48 hours to the start of the study. Secondly, if they felt improved, to consider how important the improvement was.  Those reporting moderate or slightly important improvement met criteria for MCII analysis. Subgroup analysis was performed for gender, age, baseline BASDAI and BASFI scores, disease duration, HLA-B27 status and presence/history of SpA extra-articular manifestations. Continuous variables were stratified according to the median value.  For MCII in BASDAI, a separate analysis was done on patients with a baseline BASDAI≥4 to represent patients recommended to receive treatment in clinical practice. The 75th percentile approach was used to establish cut-off values.

Results: 283 patients with AS (76% males, 64% HLA-B27 positive, mean (SD) age: 43(14) years and mean disease duration: 13(10) years) were included. Mean baseline BASDAI and BASFI values were 5.0 and 4.6 respectively. Cut-off values for PASS values were 4.1 for BASDAI and 3.8 for BASFI (Table 1). The MCII cut-off was an absolute change of 0.7 BASDAI and 0.4 in BASFI. Subgroup analyses revealed differences between groups if stratified for age, disease duration and baseline value, with differences larger for PASS than MCII. For the sub-analysis of the patient group with a baseline BASDAI score ≥4, the MCII was 1.1 in BASDAI and 0.6 in BASFI for absolute change.

Conclusion: PASS for both BASDAI and BASFI are highly variable based on important patient characteristics such as age, disease duration and baseline value. Frequently the PASS value is even higher than the recommended cut-off for start of treatment. Therefore, no uniform PASS can be proposed and this makes this instrument less useful. This applies similarly to the MCII but to a lesser extent. As MCII will mostly be applied in patients who start with new treatment (and a BASDAI ≥4), we recommend a cut-off value for MCII of 1.1 for BASDAI and 0.6 for BASFI.

^a: median Baseline BASDAI = 4.9 ^b: median baseline BASFI = 4.6