A senior U.S. behavioral scientist questioned the role of mental health care in preventing gun crimes, although the tragic Connecticut school shootings have drawn public attention to the role of mental illness in gun violence.

Terry Schell at the nonprofit, nonpartisan think tank RAND Corporation, wrote in her rand.org blog Thursday that many policymakers and commentators in the media have suggested that mental health care has a substantial role to play in reducing gun violence.

Indeed, Schell wrote, one of the prominent criticisms of the recently announced presidential plan to address gun violence is that it focuses efforts too much on guns and not enough on mental illness.

"Unfortunately, those suggesting that mental health treatment is the key to preventing gun crimes often mischaracterize the current state of the science in two ways," Schell wrote.

Commentators have widely assumed that psychiatric disorders are good predictors of gun violence and causally related to that violence, the expert wrote.

While there is some limited evidence in the literature that psychiatric disorders are associated with violence, empirical association is weak and not generalizable across disorders, according to the expert.

The very limited research investigating this association has found weak correlations, suggesting that psychiatric diagnoses may not be useful for real-world predictions of who will be violent, the expert wrote.

By comparison, gender, geographic region and race are all better predictors of gun violence than mental health. Knowing someone is male or from a particular state gives us more information about their likelihood to perpetrate gun violence than knowing that they have a diagnosed mental illness, the expert wrote.

To the extent that mental illness is associated with violence, the effect appears to be driven by a small subset of conditions for which criminal behavior is a partial basis for diagnosis, according to the expert.

Furthermore, some psychiatric disorders are associated with a lower risk of perpetrating gun violence, such as disorders that make it difficult for individuals to leave their home or that cause other substantial functional impairments, the expert wrote.

A second way that commentators have mischaracterized the science is by assuming that mental health professionals have some treatment that would effectively reduce the likelihood of violence among those with a specific mental illness associated with violence, such as paranoid schizophrenia, or among those who show a propensity toward violence, according to the expert.P As of today, practitioners and researchers lack data suggesting that standard mental health treatment reduces violence, the expert stressed.

Even when a perpetrator has an identifiable psychiatric disorder, it is highly speculative that mental health treatment could have prevented the crimes, the expert said.

Providing better mental health treatment for perpetrators of violence or those at high risk of committing such crimes is undoubtedly a good and humane goal, but there is no adequate evidence that it will prevent violence, the expert noted.