Logo of bmcmudisBioMed Centralsearchsubmit a manuscriptregisterthis articleBMC Musculoskeletal Disorders
PMC full text:

Table 4

Final multivariate post-treatment pain-prediction models and performance metricsa

Responders (N = 249/93)b Future pain intensity (N = 262/93)b
Independent variablesOR(95 % CI) P-valueβ(95 % CI) P-value
Dose (per 6 spinal manipulation visits)1.14(0.95, 1.37)0.150−0.07(−1.35, 1.21)0.910
Time (in weeks)1.08(1.02, 1.14)0.004
 Pain intensity0.64(0.51, 0.80)<0.00110.7(8.84, 12.56)<0.001
 Days with pain (last 4 weeks)0.57(0.46, 0.70)<0.001
Objective Physical Exam
 LBP: right – left lateral bending0.76(0.63, 0.92)0.005
 LBP: right lateral bending2.95(1.21, 4.69)0.001
Performance metricsc AUC(95 % CI)RMSE(95 % CI)R2 (95 % CI)
Training set0.75016.3.366
Test set0.665(0.58, 0.74)17.5(15.0, 20.1).261(7.5, 43.2)
OR Odds ratio, PC part correlation, β regression coefficient, ROM range of motion, AUC Area under the curve (receiver operating characteristic curve), RMSE root mean squared error (SD of prediction error), R 2 coefficient of determination, LBP low back pain

aVariables were selected into the regression models using forward selection among variables with p  < .05 in the univariate analysis; dose was forced into the models. Independent variables were standardized except for dose (scale unit = 6 visits) and time (scale unit = 1 week). Lower scores were favorable for pain and days with pain

bThe first number is the sample size for the model in the training set and the second number is the N for the test set

cChance performance is indicated by 0.5 for AUC. RMSE is the standard deviation of the error in prediction of future pain intensity evaluated on the 0 – 100 pain scale. R 2 is the proportion of the variance in pain intensity explained by the independent variables in the model. Confidence intervals are given for the test set only