Dropout Rate Impact Calculator

Work out dropout rate impact instantly with clear inputs, formula shown and shareable results.

About Dropout Rate Impact Calculator

Attrition erodes power. Losing 54 of 300 participants is an 18% dropout rate, leaving 246 for analysis and cutting power from the planned 90% to about 83%. Beyond roughly 20% loss, CONSORT guidance warns that bias from differential dropout becomes a bigger concern than the loss of power itself.

Formulas

Dropout rate

rate = dropouts / randomised

Retained power

recompute power with the analysed sample against the originally targeted effect

Worked example

With the default values below, Dropout Rate Impact Calculator returns Dropout Rate of 18.00%. Change any input above and every figure here is recalculated in your browser.

Inputs used

Participants randomisedminimum 1: 300
Participants lost to follow-upminimum 0: 54
Power the study was designed foraccepts 50–99 %: 90 %

Results

Dropout Rate: 18.00%
Power Actually Retained: 83.53%
Participants In The Analysis: 246
Power Lost To Attrition: 6.47%

Step by step

1Values used
2Dropout rate
3Retained power
4Dropout Rate= 18.00
5Power Actually Retained= 83.53
6Participants In The Analysis= 246
7Power Lost To Attrition= 6.47

Assumptions and limits

The result is derived only from the 3 inputs listed above. Anything not among them is not part of the calculation.
In particular, this calculator does not take inflation, tax, fees or charges or compounding frequency into account.
Figures are rounded for display to 2 decimal places, so totals built from several rounded parts can differ slightly from the unrounded arithmetic.

How these calculators are built and tested is described on the Methodology page. If a figure looks wrong, please report it.

Frequently Asked Questions

What dropout rate is acceptable?

Under 5% is unlikely to bias results and over 20% is a serious threat. Between the two, sensitivity analyses are needed.

Why is differential dropout worse than the total rate?

Because if one arm loses more participants, or loses different kinds of participant, the comparison itself becomes biased - which no sample size can fix.