Figure 3
Trial sequential analysis of constraint-induced movement therapy (CIMT) versus control for disability (FIM). Diversity-adjusted required information size of 142 patients calculated on basis of 7.50 MID of FIM (see text); variance of 126.31; alpha = 5%; beta = 20%, and diversity (D or rather D2) of 50%. The blue cumulative Z-curve crosses the inward sloping red trial sequential monitoring boundary for benefit and the required information size. The horizontal green lines illustrate the conventional level of statistical significance (two-sided alpha = 0.05), which was intersected after the third trial. Although with 199 patients randomized we have sufficient evidence to accept an effect of CIMT on FIM, after the seventh trial, the cumulative Z-score comes back toward the null effect, just above the trial sequential monitoring boundaries score. The analysis and the figure were performed with TSA software. Legend. Square symbol: z-score for single study; Diamond symbol: trial sequential monitoring boundary for benefit score for single study. Abbreviations: DARIS, diversity-adjusted required information size; MID, minimal important difference; var, variance; D, diversity; a, alpha; b, beta.