Figure 4

This figure shows the results of a hypothetical trial that compares the effects size after an effective intervention in a treatment group (T) to a control group (C) after an effective intervention. The treatment group had experienced a large effect. The control group had experienced, on average, a small improvement compatible with a placebo effect. The effects are normally distributed and not every patient improved; indeed, some deteriorated. The vertical line shows the hypothetical MID for patients participating in this trial. A certain proportion of patients in the control group show improvement greater than the MID [labeled P(Imp|C), the area under the curve to the right of the MID line in the control group]. In the treatment group, a larger proportion [P(Imp|T),] of patients show improvement greater than the MID (the area under the curve to the right of the MID line in the group receiving therapy). The difference between these two proportions [P(Imp|T) - P(Imp|C), the difference in the two areas under the curve] is the proportion of patients who improved above the MID after accounting for placebo or control group effects.