1) To confirm the association between delay for assessment (DFA) and probability of first-time attendance in outpatient substance use disorder (SUD) treatment. 2) To evaluate whether this association varies by the type of primary substance for which the assessment was requested. 3) To assess the possibility of estimating differential DFAs to conform to equal probabilities of attendance across various types of primary substance. A prospective observational cohort of consecutive patients (N=1015) who requested a first-time assessment appointment at a publicly funded outpatient SUD treatment center in France between January 2014 and December 2015 was conducted. Logistic regression analyses were performed to evaluate associations between DFA (after log-transformation) and attendance and to provide estimates of attendance probability over time by the type of primary substance. After adjusting for gender, age and referral status, the attendance rate was observed to decrease significantly with longer DFA (OR=0.54; 95%CI: 0.44-0.66). The strength of this association differed across types of primary substance (p for heterogeneity <0.0001), with the strongest association being found for opioids (adjusted OR=0.21; 95%CI: 0.10-0.45). DFA was also associated with attendance for alcohol (OR=0.51; 95%CI: 0.37-0.71) and cannabis (OR=0.60; 95%CI: 0.37-0.96), but not for tobacco (OR=0.95; 95%CI: 0.60-1.50). Differential DFAs reflecting equal probabilities of attendance across types of substance could be estimated. Our study suggests that the approach of stratifying DFAs by the type of primary substance could be helpful to improve the probability of first-time attendance in outpatient SUD treatment services.