In addition, we tested the proportional hazards assumption that risk remains constant over time, which tests whether the predictor variables have the same effects at all points in time (i.e., an interaction with time). This test permitted us to explore whether the contribution of a given risk factor to initiation of alcohol use is greater at certain developmental stages than others, e.g., whether its effects are specific to early initiation. This was assessed using the Grambsch and Therneau test of the Schoenfeld residuals (Grambsch & Therneau, 1994).
