In contrast to the relatively age-invariant differences by gender, disparities by race/ethnicity did differ by age. For AUD and TUD, White participants had higher rates than Black participants in young adulthood, but then had similar or lower rates at older ages. A similar crossover was also observed for White versus both Black and Latino participants for OUDs. This is consistent with prior research that has documented a crossover effect in substance use (Chen and Kandel, 2002; Geronimus, 1993; Kandel et al., 2011; Ensminger et al., 2016). However, the ages at which this crossover occurs differ by substance; a clear crossover occurs in midlife for OUD and TUD, whereas AUD is similar for Black and White participants from about ages 30 to 60, with higher rates for Black compared to White participants in older adulthood. Note that these crossovers occur later than those observed for substance use (e.g., Chen and Jacobson, 2012; Chen and Kandel, 2002), for typically occur in the late twenties or early thirties. This may reflect a delay in when use escalates into diagnostic-level problems or differences in
