When folks say that mental illness is linked to over-representation in the criminal justice system, a reasonable common observer in the public could not be faulted for thinking that such mental illnesses are the severe ones predominately treated in state institutions decades ago: schizophrenia, bipolar disorder, etc. And it may be true that some of the people with these serious illnesses do cycle in and out of our jails. But what is also worth noting is that studies like the current one in the American Journal of Psychiatry include substance abuse as a classified mental illness. Moreover, almost all studies which have examined the link between mental illness and criminality have shown substance abuse -- not mental illnesses such as schizophrenia or bipolar disorder -- are the largest risk factors for incarceration within these samples (similar to the general population). In fact, the current study reveals that depression and substance abuse during adolescence was associated with a nearly 15 fold increased likelihood for arrest for a severe crime as an adult. The comorbidity between substance abuse and depression is well known and researchers often debate whether drug use leads to depression or if depressed people are more likely to use illicit drugs to "self medicate."

The current study also examined the link between adolescent diagnosis of conduct disorder and adult criminality. Yet this was only significant when paired with substance abuse. Many have argued that substance abuse is not a disease , and presumably, not a mental illness on par with schizophrenia. While those are interesting arguments, a more fundamental question remains: When it comes to risk for incarceration, are the mentally ill really that different than the general population? If not, should the criminal justice response be different for them? If it should, the reasons need to be more than just the circular argument that they have mental illnesses and therefore are different.