Page 138 - Big Data Analytics for Intelligent Healthcare Management
P. 138

5.8 RESULTS, INTERPRETATION AND DISCUSSION            131






                Table 5.15 Trend Line Models
                Model formula                                 Period   Techniques   (ln(Duration)+intercept)
                Number of modeled observations                231
                Number of filtered observations               21
                Model degrees of freedom                      20
                Residual degrees of freedom (DF)              211
                SSE (sum squared error)                       40.5473
                MSE (mean squared error)                      0.192167
                R-Squared                                     0.521327
                Standard error                                0.438369
                P-Value (significance)                        <.0001




                Table 5.16 Individual Trend Lines
                     Panes           Line                          Coefficients
                Row     Column   P-Value  DF  Term       Value       StdErr   t-Value    P-Value
                EMGav   BaseLine  .0730867  25  ln(Duration)  0.130076  0.0695203  1.87105  .0730867
                                              intercept  1.56326     0.174409  8.96318   <.0001
                EMGav   1month   .0081169  21  ln(Duration)  0.414192  0.141668  2.92369  .0081169
                                              intercept  0.687232    0.327455  2.09871   .0481139
                EMGav   3months  .0473046  21  ln(Duration)  0.36091  0.171282  2.10711  .0473046
                                              intercept  0.593559    0.352508  1.68382   .107025
                EMGav   6months  .340869  19  ln(Duration)  0.366373  0.375016  0.976952  .340869
                                              intercept   0.0115886  0.635114   0.0182464  .985633
                EMGav   12months  .605505  14  ln(Duration)  0.101835  0.192727  0.528389  .605505
                                              intercept  0.694253    0.331293  2.09559   .0547784
                GSRav   BaseLine  .94721  25  ln(Duration)  0.0093822  0.140287  0.066879  .94721
                                              intercept  1.54445     0.327587  4.71461   <.0001
                GSRav   1month   .0030268  26  ln(Duration)  0.30595  0.0935613  3.27005  .0030268
                                              intercept  0.759532    0.180279  4.21308   .0002677
                GSRav   3months  .0036585  23  ln(Duration)  0.456081  0.140981  3.23507  .0036585
                                              intercept  0.45989     0.232347  1.97932   .0598718
                GSRav   6months  .232588  23  ln(Duration)  0.265693  0.216708  1.22604  .232588
                                              intercept  0.291709    0.332022  0.878584  .388713
                GSRav   12months  .156609  14  ln(Duration)  0.577587  0.385843  1.49695  .156609
                                              intercept  0.195963    0.547597  0.35786   .725786


                  Also, the analysis showed that the majority of the data came under the quadrant of low duration and
               low intensity with the exceptions of three subjects out of which two were having average intensity but
               low duration and one was having high intensity but low duration at the end of 1year. Therefore, GSRav
               therapy showed an improvement in duration.
   133   134   135   136   137   138   139   140   141   142   143