Posted by: Knightbird | September 14, 2013

Use of Control Charts in Suicide Prevention

Suicide is a rare occurrence, but it’s an act that releases strong emotions. The State of Alaska experiences a high rate, particularly in a number of rural areas. And despite millions of dollars spent on suicide prevention activities, the suicide rates, overall, have not changed much. A comment made in a report by the Alaska Suicide Prevention Council was provocative. A report stated: “The decrease in the number of lives lost in 2011 is promising.” A caution was included about the historical range of suicides. I decided to construct a control chart using the data given. The results are interesting.


After calculating the average, the range and entering the data into a formula for calculating Upper and Lower Control Limits, I ended up with the chart shown here. What’s interesting is that there is no discernible trendlines indicating either a worsening or improvement in the suicide rate. All data points are stable and most appear to be within the 1 sigma range. According to this analysis, there is no support for the optimism contained in the report statement. Improvement is not happening.

How can improvements occur? I have thought about it for the past few years and came up with a theory. Using my lean training, I observed that current suicide prevention programs rely on identifying individuals with suicidal thoughts or suicide attempts, and attempting to counsel the individual. It’s similar to having inspection, and when you discover a defective product, fix or repair it. Real improvement comes when you attempt to find the reason, or root cause, for the defect in the first place. (Please understand that I am not referring to suicide thoughts or attempts as defects) I had a real insight about what the root cause for suicidal behaviors might be when I was introduced to the Adverse Childhood Experience Study (ACES) in 2008. For study participants who had experienced 7 or more adverse experiences as children, as a group 35% attempted suicide at one time during their life. For study participants with 0 adverse experiences, the rate was around 2%. Brain research has demonstrated a increased fear response among traumatized children that increases stress, tension and anxiety among the population. It’s my belief that unresolved childhood trauma causes the increase in suicidal behavior, which gives us an opportunity to address the behaviors much earlier, and most likely with better results. By identifying children of trauma, we can help them identify the symptoms of stress, and teach them tools to reduce it. I won’t go into a full explanation of the strategy I developed here. My purpose in writing this is to explore how a tool like a control chart may be able to assist us in understanding how to address some of the behavioral challenges that face us.


  1. That’s a great example of needing to look for a root cause — keeping people healthy (and this includes mentally and emotionally) rather than just reacting to an event, such as a suicide threat or attempt.

    Your SPC example is a wonderful example of how, once you learn statistical thinking, that you’re less quick to just blindly accept a statement about whether something is improving or not. As they say, two data points do not make a trend, but people make this mistake all the time.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: