Risk Assessment - Prediction of Future Domestic Violence

My post the other day dealt with efforts to improve the ability to evaluate future risk of violence, based on an article indicating future trends may include biological/genetic markers as data points.  I read this article today which encourages the use of data already collected by medical facilities to try and determine who might be at risk for experiencing domestic violence.  The article indicates an analysis of data accumulated from over 500,000 patients over at least a four-year period yielded certain data points useful in the prediction of later victimization.  From the article:

Particular risk factors were strongly linked with a future diagnosis of abuse. The risk was highest after being seen in hospital or the emergency department for these events:

For women:

• injuries
• alcoholism
• poisoning

For men:

• mental health conditions such as depression and psychosis

The article I linked to did not offer any specifics regarding the statistics used, the predictive strength of the data, etc.  Regardless, for me the larger point is that with ever-increasing technology, there are larger, more sophisticated databases being developed and maintained (aka "data mining").  This data can serve all sorts of useful purposes in terms of giving clinicians direction in screening, treatment, etc.  In the research above, for example, a doctor at a hospital, normally under tremendous time constraints, may be alerted that a particular patient who has arrived at the emergency room meets the criteria for high-risk abuse victim.  The doctor may then screen this particular more thoroughly, or create an environment more conducive to the patient divulging the abuse, agreeing to a consult, etc.  The point is not that certain issues shouldn’t be screened with certain people; it is that with limited time and resources, we want to direct services and resources to individuals who need them most.  The ever-increasingly capacity to gather data allows for the identification of patterns, as well as to whom those patterns apply.