Skip to main content

Table 5 Performance compared over models (n = 220)

From: Predicting change in symptoms and function in patients with persistent shoulder pain: a prognostic model development study

Model

Adj. R2 (95% CI)

AIC value

Root MSEb

Final model derived from adj. R2

32% (22; 42)

1881

17.0

Final model derived from AICa

32.5% (22;43)

1880

16.9

Parsimonious model using only baseline QuickDASH as predictor

19% (10; 29)

1909

18.47

Model without independent variables

–

–

20.58

  1. Abbreviations: Adj.R2 adjusted coefficient of determination, CI confidence interval, AIC Akaike’s information criterion, Root MSE Root mean squared error, PI Prediction interval
  2. aDuration of symptoms was added in this model along with the predictors from the adj. R2 model
  3. bModel standard deviation