Begin main content

2 - Using the ITC Application

Date: January 28, 2009
Result type: Resources

2.1 Overview

2.1.1 Introduction

This chapter introduces the user to the user interface and basic functionality of the ITC application.

2.1.2 In this chapter

2.2 Starting the ITC Application

2.2.1 Introduction

The ITC application is launched either from the disk or from your desktop.

  • Locate and double-click the ITC application icon.

2.3 User Interface

2.3.1 Introduction

The ITC application has been developed in Visual Basic to assist with various calculations associated with indirect comparisons.

 

When entering data, you can tab from one cell to the nex

 

2.3.2 Screen 1: Main Window

The ITC application consists of two windows. On the first window, the effect measure of interest is identified. Information for each consecutive pair of treatments is entered in terms of the point estimate and 95% confidence interval (CI) of the effect measure for each direct comparison involved in the indirect comparison. The resulting indirect comparison estimates for the effect measure and the 95% CI as well as the P value for the test of association corresponding to this effect measure are provided.

Figure 1: Main Window

2.3.3 Effect Measure

  • The desired Effect measure is chosen by selecting the corresponding option button

2.3.4 Number of Treatments

  • The Number of Treatments (k) is entered by typing directly in the appropriate field or by using the arrows to scroll to the desired value. The highest number you may enter is 10.
  • For each consecutive pair of treatments, provide the direct estimates of the measure of association and the 95% lower confidence limit (95% LCL) and 95% upper confidence limit (95% UCL).

    The order of entry of the treatment pairs must follow the exact sequence indicated with the bridging comparison groups linking the treatment pairs.
  • To reverse the order of a treatment comparison, select the appropriate checkbox.

    For example, if 1 = Treatment A, 2 = Treatment B, and 3 = Treatment C, then (1,2) is (A,B), and for (2,3) we can enter (C,B) and use the reverse option to switch it to (B,C). This option can be useful when B is a placebo and the results are given as the active treatment versus placebo (i.e., A versus Placebo[B] and C versus Placebo[B]).

 

2.3.5 Toolbars

The toolbars on the main window allow you to calculate, clear, import, and save your data as well as exit the application.

 

Table 1: Main Window Toolbars

Button Description
Calculate

When all data has been entered, returns the desired calculations.

Clear

Clears all data from the main window.

Save

Allows the user to save the data (*.txt file).

Open

Allows the user to import previously saved data.

Exit

Closes the application without saving the data.

2.3.6 Screen 2: Requested Weights Window

If the test of association is needed, the weights used for the calculation of each weighted average estimate from the main window are required for its calculation. The requested weights window is accessed by clicking the arrow to the right of the corresponding information for each consecutive pair of treatments.

The requested weights window calculates the weighted effect measure for the direct comparison on the main window. These weights are needed to calculate the test statistic for the test of association. There are various formats in which the information to calculate these weights can be provided, and these formats are identified through the weight selections (direct versus derived; fixed versus random), and the specific information for each study involved in the direct comparison is then identified and requested.

  • For a direct treatment (i, i + 1), enter the number of studies on which the estimate is based by typing directly in the corresponding field or by using the arrows to scroll to the desired value. The highest number you may enter is 20.
  •  
  • The desired Weight is chosen by selecting the corresponding option button.

    • The option is available to enter the weights directly. In particular, if the effect estimate is based on one study, then a single weight of 1 can be entered.

    • The weights can also be computed from first principles, based on the frequencies (for RR, odds ratio, and risk difference) or the standard errors (for mean difference), using either the fixed or random effects model, or the general inverse method.

Figure 2: Requested Weights Window

2.3.7 Toolbar

The toolbar on the requested weights window allows you to save your data and close the window, or clear your data.

Table 2: Requested Weights Window Toolbar
Button Description
Close Saves data and closes the window.
Clear Clears all data from the window.