Table 3 Logistic regression of Models for prediction of ECT response.

From: Resting state functional connectivity predictors of treatment response to electroconvulsive therapy in depression

Logistic regressiona

Remission (HDRS-24 ≤ 7)

Model 1b

Model 2c

Model 3d

OR per 1 SD

p-val

OR per 1 SD

p-val

OR per 1 SD

p-val

Step 1

DLPFC(p9-46v)-VIS(MT+)

    

0.09

0.008

Intra-VIS(ventral)

0.14

0.01

0.15

0.01

0.17

0.014

Step 2

Intra-DMN(10r)

0.10

0.005

    

DLPFC(46)-DMN(s32)

  

11.56

0.005

  

% Accuracy

Step 1

Step2

Step 1

Step2

Step 1

Step2

Globale

77.8%

100%

77.8%

94.4%

83.3%

94.4%

Non-remitter

77.8%

100%

77.8%

100%

77.8%

100%

Remitter

77.8%

100%

77.8%

88.9%

88.9%

88.9%

  1. aStepwise method. Step1: 1-factor; Step2: 2-factor. Firth, D. (1993) Bias reduction of maximum likelihood estimates. Biometrika 80, 27–3.
  2. bModel 1: Intra-VIS(ventral) and intra-aDMN(10r).
  3. cModel 2: Intra-VIS(ventral) and DLPFC(46)-aDMN(s32).
  4. dModel 3: DLPFC(p9-46v)-VIS(MT+) and intra-VIS(ventral).
  5. eAverage accuracy non-remitter and remitter; cut-point 0.5.
Back to article page