PMC

A (hypothetical) sigmoid attrition curve

A proposed “run-in and withdrawal” design

Attrition curves from on a logarithmic scale (y-axis is the natural logarithm of the proportion of users completing a module)

An example for logarithmic “attrition curves” in a hypothetical eHealth trial. In the intervention group (INTV), a proportion of participants will be lost to follow-up (INTV dropout), as will be in the control group (CTRL dropout). In addition, even within those not lost to follow up, there might be a proportion of nonusers

Nonusage attrition curves for two studies [,] published in this issue of the Journal of Medical Internet Research. Plotted are the number of completed modules from two Web-based interventions against the proportion of participants completing them. From the two Christensen/Moodgym curves, the upper one refers to a trial setting, while the other (lower one) refers to an “open” situation with casual Internet visitors.