Tuesday, December 21, 2010

An Innovative System to Combat Homelessness

Over at the New York Times Opinionator blog, David Bornstein has a post describing an innovative system to provide "permanent supportive housing" to chronically homeless individuals who are most at risk. Of note is that certain chronic homeless people are much more likely to die in their 40s and 50s due to problems related to their homelessness, such as having their medicine lost or stolen, lack of refrigeration for insulin for their diabetes, or the inability of doctors to follow up for cancer patients.

In response to such issues, an organization known as Common Ground has developed a "vulnerability index" using an algorithm that ranks homeless people by risk of death. The ranking is then used to prioritize the allocation of "permanent supportive housing."

One interesting thing noted by the article is how previous approaches to allocating permanent housing to homeless people only if they were "housing ready," meaning that they were drug and alcohol free. If you think about this, this is a completely ridiculous approach. It is nearly as though under previous approaches someone really needed to have to be perfect and have their act completely together (in a way that many non-homeless people do not; how many non-homeless people have issues with alcohol or drugs, for example?) before they can get access to a stable environment. But really, what is the probability that someone would be homeless in the first place if they didn't have difficult life issues? And why would anyone possibly think that someone will be better able to overcome problems with alcohol and drugs in an unstable environment than a stable environment? If life is unstable and harsh, as it is for homeless people, wouldn't this make the refuge of alcohol and drugs as a temporary respite more difficult to resist? It seems to me that the idea that people must be drug free and alcohol free before being offered the stability of permanent housing isn't actually based on pragmatic considerations of what is likely to work, but rather more of a morality play. In fact, this is approaching things from the wrong direction; for someone to have a higher probability of overcoming a drug or alcohol problem, they need to live in a stable environment.

Regardless, it is good to see innovative data-intensive approaches to the longstanding problem of homelessness.

No comments:

Post a Comment