Dealing with Domestic Abuse with Databases and Risk Assessment

ACORN ACORN International Community Organizing
Facebooktwitterredditlinkedin

Little Rock       I don’t like to think about domestic abuse.  It makes me sad.  It makes me mad.  I can watch the blood-and-gore on police procedurals on TV or Netflix or whatever, and pretend it’s make believe, but harming women and children is a head turner and channel switcher for me.

Nonetheless, I read about the book No Visible Bruises:  What We Don’t Know About Domestic Violence Could Kill Us by Rachel Louise Snyder.  A number of publications rated it as one of the ten best books of the year and top five nonfiction.  A little over a year ago, I worked with an intern at ACORN from Albania who spent a month with us in New Orleans, and when I visited her in Tirana, Albania, among the many things I did was visit a women’s abuse shelter where she volunteered, in the limited area where men were allowed in this nondescript, anonymous space.  I knew it was time for me to man-up and learn more about this, so I read the book.

The book is invaluable, tragic, and a call for action.  Much of it is platformed around a tragedy where a husband killed his wife and children, and them himself, after trying to set fire to the house in Billings, Montana.  I felt like I knew the family and the neighborhood, though it could have been anywhere in any city in the United States or the world for that matter.

Much of Snyder’s reporting is about the gulf in information and understanding about what constitutes abuse in the intersection of families with police, courts, social services, and health centers.  Almost all of them earn failing grades.  Many do so not for lack of effort, but limited understanding.  Good hearts failed by poor brains.

Social workers have developed a Danger Assessment tool which is excellent in predicting the likelihood of abuse, and fortunately has been widely adapted in many areas of the country.  In looking at risks, the tool highlights:

…. substance abuse, gun ownership, extreme jealousy. Others were more specific: threats to kill, strangulation, and forced sex. Isolation from friends and family, a child from a different biological parent in the home, an abuser’s threat of suicide or violence during pregnancy, and stalking all added lethality. Access to a gun, drug or alcohol abuse, and controlling daily activities are among the risk factors, as are threats to children, destruction of property, and a victim’s attempt to leave anytime within the prior year. The sole economic factor … identified was chronic unemployment.

There’s been progress, but it’s not enough, and not synchronized with services, even when risks are identified, nor have laws and law enforcement sufficiently caught up with recognition of the warning signals.

Spending a lot of time with a database whiz, as he and I construct the Voter Purge Project, I was particularly struck by an observation of Synder’s that, “we still seem unable to create a database that speaks across state lines and across civil and criminal courts when it comes to violent people and their histories.”  In looking at one tragedy involving Arizona and Utah, as I recall, it turned out that the abuser – and killer – had a history of similar activity in a previous marriage in Texas.  When dealing with complaints against him, the police and the courts were clueless, so in a too common occurrence, they let him go.  Montana now keeps a history of DUIs even after they have expired.  Part of what advocates argue is that in addition to a national shared database, there needs to be a retained history of restraining orders.

Domestic abuse is so difficult, because its too personal, and the witnesses who are spouses and children are often silent for their own security and in hopes of stability.  Nonetheless, there’s no reason, any of us, including our government, should make abuse even easier and less transparent.

I’ve got a guy, and, believe me now, I’m going to be talking to him and others about why we can’t make this happen and do our small part.

Facebooktwitterredditlinkedin