Skip to main content

Table 2 Correlations between PROMIS scales and OHS and SSWB

From: Measurement properties of PROMIS short forms for pain and function in total hip arthroplasty patients

  Correlation with OHSa,b Correlation with SSWBa,c
PROMIS PAIN
 Baseline −0.83 (− 0.88 to -0.77)
−0.84 (−0.88 to −0.78)e
 12 months − 0.84 (−0.89 to −0.78) 0.69 (0.58 to 0.77)
−0.86 (−0.90 to −0.80)e
 Change −0.81 (−0.87 to -0.74)d
PROMIS PI
 Baseline −0.84 (−0.89 to −0.78)
−0.78 (−0.81 to −0.71)e
 12 months −0.84 (−0.88 to −0.77) 0.66 (0.55 to 0.76)g
−0.81 (−0.86 to −0.74)e
 Change −0.76 (−0.82 to −0.67)d
PROMIS PF
 Baseline 0.82 (0.75 to 0.87)
0.80 (0.73 to 0.86)f
 12 months 0.79 (0.70 to 0.85) −0.53 (−0.65 to −0.38)h
0.80 (0.72 to 0.86)f
 Change 0.72 (0.63 to 0.80)d
OHS
 12 months −0.65 (−0.74 to −0.53)
 Change
  1. PROMIS Patient Reported Measurement Outcome Instrumentation System, OHS Oxford Hip Score, SSWB Symptom-specific well-being, PAIN Pain intensity, PI Pain interference, PF Physical function
  2. aCorrelation coefficients are shown for the total sample. Analysis was done for male and female subsamples as well. If not indicated (g), correlation coefficients for male and female subsamples are above the required threshold (confidence intervals do not include 0.5)
  3. bPearson’s correlation coefficient (r)
  4. cSpearman’s rank correlation coefficient (rs)
  5. dCorrelation OHS change score
  6. eCoefficients for OHS pain subscale
  7. fCoefficients for OHS function subscale
  8. gLarge correlation was not confirmed for the male subsample (absolute values of the confidence interval limits were required to be ≥0.5)
  9. hLarge correlation was not confirmed for the total sample and male subsample (absolute values of the confidence interval limits were required to be ≥0.5)