To the Editor We read the excellent cohort study by Albaugh et al¹ with great enthusiasm and appreciation. The authors reported that cannabis use during adolescence is associated with increased thinning in the left and right prefrontal cortices. This is an important study, and we wish to contribute a limitation.

A 2021 study used co-twin control design (n = 436 twins aged 24 years) to separate alcohol or cannabis exposure associations with familial (genetic liability and environment) risk on cortical thickness.² This study found that more alcohol, but not cannabis, exposure was associated with thinning of prefrontal cortices and frontal medial cortices.² In addition, premorbid cortical thickness reduction in inhibitory control brain regions may increase the risk of excessive alcohol use during emerging adulthood.² Importantly, alcohol and cannabis are the most commonly used substances among adolescents, and the pattern of co-use of alcohol and cannabis may affect the associations of alcohol or cannabis with the cortical thickness. A study including 84 805 adolescents examined the patterns of co-use of alcohol and cannabis (simultaneous use or not), revealing that 11.2% were simultaneous use with binge drinking, 21.6% were simultaneous use without binge drinking, 10.7% were concurrent use without overlapping, and 56.4% were alcohol use without cannabis use.³ In the study by Albaugh et al,¹ the confounding association of alcohol was statistically adjusted for, and the analyses did not address the patterns of possible co-use of alcohol and cannabis. Moreover, the aim of adjusting for a covariate in a statistical model is to remove a noise but not to remove a main signal.⁴ For separating the cannabis and alcohol associations, the authors could categorize the participants into different co-use patterns or conduct mediation analysis.