Skip to main content

Table 1 Summary of the analysis plan

From: Helmet noninvasive ventilation for COVID-19 patients (Helmet-COVID): statistical analysis plan for a randomized controlled trial

Variables Intention-to-treat cohort Per-protocol cohort
Baseline characteristics No statistical comparisons None
Intervention and co-interventions Chi-square, Fisher’s exact test, Wilcoxon–Mann–Whitney test, t-test as applicable None
Primary outcome 1. Primary analysis: chi-square or Fisher’s exact test. Report relative risk.
2. Secondary analyses: unadjusted Cox proportional analysis, KM curves, adjusted logistic regression
1. Primary analysis: chi-square or Fisher’s exact test. Report relative risk.
2. Secondary analyses: unadjusted Cox proportional analysis, KM curves, adjusted logistic regression
Secondary outcomes-categorical Chi-square or Fisher’s exact test. Report relative risk Chi-square or Fisher’s exact test. Report relative risk
Secondary outcomes-continuous Generalized linear mixed models. Report beta estimate Generalized linear mixed models. Report beta estimate
Safety outcomes and other variables Chi-square or Fisher’s exact test. Report relative risk. For serial measurements, generalized linear mixed effect models None
Subgroup analyses Chi-square or Fisher’s exact test. Report relative risk. Tests of interaction None
180-day follow-up Chi-square or Fisher’s exact test for 180-day mortality. Report relative risk
Generalized linear mixed models for EQ-5D-5L ad VAS. Report beta estimate
Chi-square or Fisher’s exact test for 180-day mortality. Report relative risk
Generalized linear mixed models for EQ-5D-5L ad VAS. Report beta estimate