Tuesday, December 11, 2012

Looking at Statistics in Isolation Can Lead to the Wrong Conclusions

One aspect of public health is violence.  For example, at the Johns Hopkins Bloomberg School of Public Health we have a Center for Gun Policy and Research.  And, recently, there has been a lot of violence in the local news.  That, of course, is on top of numerous stories about gun violence nationally and internationally over the past several years.  So, it may seem like we should be looking for more ways to incentivize against violence and to find cost-effective solutions to preventing gun violence in particular and all types of violence in general.

In contrast, if we looked at the fact that the rate homicides appears to be going down, we might wonder, "Well, is there a problem with our perception of violence?  Maybe it is just lots of anecdotes and we don't need to look for more solutions or allocate more resources."

A piece in the Wall Street Journal recently entitled "In Medical Triumph, Homicides Fall Despite Soaring Gun Violence," (which is not available for free) they looked at this issue.  What the article reported on was an improvement in the survival after being shot in recent years.

Now, we could ask whether heroic treatment after gunshot wounds is cost-effective.  It is definitely expensive.  The trauma centers that have been set up to treat such patients (and that contribute to the much higher survival rate) have high fixed costs and high variable costs.  The WSJ article indicates that they deal with a great amount of uncompensated care in our current system and are reported to lead to losses for the hospitals at which they are located.  Thus, the demand for resources from society and resources that are taken away from the hospital are notable.  Importantly, I am not saying that we should not be saving lives.  Only that we need to be cognizant of the costs of saving the life and cognizant of the additional costs during any rehabilitation phase afterwards.

Then, once we are aware of the costs, we might ask whether the medical triumph is the most economically favorable way to decrease deaths from guns or whether we might be better off economically by trying to limit gun violence as a way to limit deaths rather than needing to treat those who do suffer gunshot wounds.

The contrasting inferences one might make based on different pieces of the data and even the contrasting steps one might take in structuring a cost-effectiveness analysis (e.g., comparing only prevention methods or only treatment methods or treatment with prevention) illustrate how important it is to try to assemble all relevant data before making a decision on further resource allocation and the need for and usefulness of economic evaluations.  

3 comments:

  1. This comment has been removed by a blog administrator.

    ReplyDelete
  2. Great post. I just located your blog and wished to let you know that I have certainly loved reading your blogs. At any rate I’m going to be subscribing to your feed and I really hope you are writing again soon.

    ReplyDelete
  3. Great post. I just located your blog and wished to let you know that I have certainly loved reading your blogs. At any rate I’m going to be subscribing to your feed and I really hope you are writing again soon.

    ReplyDelete