Table 2 Testing hypothesis of correlation (Spearman, r) between the mRS scores generated using different assessment methods and the different patient-reported outcome measures

mRS vs. EQ-5D-5L total score

–0.4 to − 0.8

–0.546

(n = 58)

–0.443

(n = 124)

–0.611

(n = 62)

Hypothesis supported

Yes

Yes

Yes

mRS vs. RAND-36 PCS

Higher correlation with PCS than with MCS

–0.439

(n = 60)

–0.465

(n = 128)

–0.599

(n = 64)

mRS vs. RAND-36 MCS

–0.574

(n = 60)

–0.288

(n = 128)

–0.506

(n = 64)

Hypothesis supported

No

Yes

Yes

mRS vs. SS-QoL total score

–0.4 to − 0.8

–0.671

(n = 56)

–0.417

(n = 121)

–0.699

(n = 61)

Hypothesis supported

Yes

Yes

Yes

Hypotheses for the convergent validity of different assessment methods:

The correlation of mRS-SI and mRS-SA with SS-QoL would be higher than the correlation between mRS physician and SS-QoL

Yes

Yes

Yes

Total number of hypotheses supported

3 out of 4

4 out of 4

4 out of 4