Designing cost-efficient randomized trials by using flexible recruitment strategies

Table 1 Simulated results for single stage and two stage screening procedures ^&

		Calculated costs		Simulated costs		Simulated actual power		Simulated actual type I error		Marginal treatment effects (λ₁)
$\frac{C pre}{C pre + C scr}$ †	ρ ‡	Single stage	Two stage	Single stage	Two stage	Single stage	Two stage	Single stage	Two stage	Single stage	Two stage
0.10	0.70	$117.5 K	$90.8 K	$118.3 K	$92.0 K	0.91	0.91	0.05	0.05	0.27	0.33
0.20	0.70	$117.5 K	$99.9 K	$118.4 K	$101.5 K	0.90	0.90	0.06	0.05	0.27	0.31
0.30	0.70	$117.5 K	$105.6 K	$118.3 K	$108.3 K	0.89	0.91	0.05	0.05	0.27	0.30
0.40	0.70	$117.5 K	$109.5 K	$118.3 K	$112.0 K	0.89	0.91	0.06	0.06	0.27	0.30
0.50	0.70	$117.5 K	$112.4 K	$118.4 K	$113.4 K	0.90	0.91	0.04	0.05	0.27	0.29
0.33	0.30	$117.5 K	$116.8 K	$118.5 K	$118.9 K	0.90	0.91	0.05	0.05	0.27	0.28
0.33	0.45	$117.5 K	$114.5 K	$118.5 K	$115.8 K	0.89	0.90	0.04	0.05	0.27	0.28
0.33	0.60	$117.5 K	$110.5 K	$118.5 K	$113.3 K	0.89	0.91	0.05	0.05	0.27	0.30
0.33	0.75	$117.5 K	$105.2 K	$118.3 K	$107.6 K	0.90	0.91	0.05	0.06	0.27	0.30
0.33	0.90	$117.5 K	$98.0 K	$118.5 K	$99.8 K	0.90	0.91	0.06	0.04	0.27	0.31

&“Single stage” screening procedure is defined in Section 2.1 where the screening process is viewed as one inseparable part of the clinical trial, this is in contrast to “two stage” screening procedure where the screening process is broken into a prescreening stage and screening stage (usually intensive or expensive).
†Ratio of prescreening costs to total screening costs. Higher ratio means prescreening costs make up a greater percentage of total screening costs.
‡Correlation coefficient between prescreening and screening variables.

ISSN: 1471-2288