Table 7.

The predicted minimum sample size for a hypothetical trail of three-year duration assuming a balanced design with DSEG measurements taken annually to test a hypothetical treatment effect of 30, 25, 20 and 15% on the rate of DSEG θ change. For comparison, MD NPH and WMH volume values are also shown, taken from Benjamin et al. (2015) and Zeestraten et al. (2016).

	Hypothetical treatment effects
	30%	25%	20%	15%
DSEG θ sample size	73	104	163	290
MD NPH (Zeestraten et al., 2016)	128	185	325	578
WMH volume (Benjamin et al., 2015)	124	178	279	496