Figure 3From: Accounting for individual differences and timing of events: estimating the effect of treatment on criminal convictions in heroin users Additive hazards regression models. Univariate and multiple Aalen’s additive hazards regression models for days with criminal convictions as a recurrent event dependent variable and treatment as a time-dependent covariate, and the non-parametric components of a multiple semi-parametric additive hazards regression model.Back to article page