Fig. 1From: A competing risk joint model for dealing with different types of missing data in an intervention trial in prodromal Alzheimer’s diseaseNumber of subjects with measurements eligible for efficacy analysis of the primary endpointBack to article page