Skip to main content
. 2013 Nov 1;8(11):e78926. doi: 10.1371/journal.pone.0078926

Table 9. Multiple Linear Regression Models to Estimate the Relationship of percent (%) predicted spirometric values (FVC and FEV1) with scoring of radiographic abnormalities among patients with cured pulmonary tuberculosis.

Variables Model for percent predicted FVC*
Model for percent predicted FEV1*
Parameter estimates (95% CI) P value Parameter estimates (95% CI) P value
Scoring of radiographic abnormalities* -2.48 (-3.45 to -1.50) 0.000 -2.92 (-3.87 to -1.97) 0.000
Smoking status (former vs never) -0.27 (-11.28 to 10.74) 0.961 1.82 (-8.82 to 12.46) 0.735
Pack-years of smoking -0.31(-2.76 to 2.14) 0.804 -0.98 (-3.34 to 1.40) 0.417
MRC Dyspnea Scale
MRC 0 Reference Reference
MRC 1 -1.16 (-9.35 to 7.03) 0.517 -2.39 (-10.30 to 5.53) 0.551
MRC 2 -2.42 (-14.36 to 9.52) 0.689 -3.92 (-15.45 to -7.62) 0.503
MRC 3 -23.84 (-46.09 to 0.88) 0.042 -16.10 (-38.29 to 6.08) 0.153
MRC 4 -24.18 (-68.32 to 19.96) 0.280 -13.29 (-55.94 to 29.36) 0.538
Oxigen pulse saturation 0.04 (-1.14 to 1.06) 0.943 0.26 (0.80 to 1.33) 0.619
Adjusted R2 0.24 0.30
*

Multivariate linear regression model.

CI: Confidence interval